UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Design principles for homecare documentation based on classifying and modeling workarounds Al-Masslawi, Dawood 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2019_may_al-masslawi_dawood.pdf [ 2.36MB ]
JSON: 24-1.0378570.json
JSON-LD: 24-1.0378570-ld.json
RDF/XML (Pretty): 24-1.0378570-rdf.xml
RDF/JSON: 24-1.0378570-rdf.json
Turtle: 24-1.0378570-turtle.txt
N-Triples: 24-1.0378570-rdf-ntriples.txt
Original Record: 24-1.0378570-source.json
Full Text

Full Text

DESIGN PRINCIPLES FOR HOMECARE DOCUMENTATION BASED ON CLASSIFYING AND MODELING WORKAROUNDS by  Dawood Al-Masslawi  B.Sc., Al-Mustansyria University, 2008 M.Sc., Tarbiat Modares University, 2011  A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Electrical and Computer Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2019  © Dawood Al-Masslawi, 2019 ii   The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Design principles for homecare documentation based on classifying and modeling workarounds  submitted by Dawood Al-Masslawi in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Electrical and Computer Engineering  Examining Committee: Sidney Fels Co-supervisor Leanne M. Currie Co-supervisor  Konstantin Beznosov Supervisory Committee Member Guy Dumont University Examiner Alison Phinney University Examiner  Additional Supervisory Committee Members:  Supervisory Committee Member  Supervisory Committee Member iii  Abstract  Computer systems are being used in healthcare at an increasing rate, especially in homecare nursing.  However, mismatch between the technology and the clinical work has been a concern for clinicians and system designers.  This mismatch is a barrier to nurses’ work and results in the need to work around the technology.  The purpose of this dissertation is to identify design principles for interactive computer systems that reduce the mismatch.  This study had four phases: 1) identification and classification of workarounds; 2) modeling and mapping of workarounds to design features; 3) creation of the mapped design features; and 4) refining and evaluation of the design features.  An ethnographic study of homecare nurses who provide care for patients with wounds in Vancouver (n=33, 120 hours), indicated that they create and use workarounds.  It is possible that this is a manifestation of unsuccessful adoption of an implemented wound documentation system.  A user-centred design process was created to identify design principles for such interactive systems.  A model from the literature was adapted, the work situation model, to identify and describe the most common workaround situations and their attributes such as tasks, and resources.  The results were validated using a questionnaire (n=58).  Furthermore, the identified workarounds were mapped to design principles from the literature, with the use of the workaround situation model attributes.  This mapping used measures developed for applications of the technology acceptance model in healthcare, to identify a mapping fit for the workaround situations to a dimension of usefulness or ease of use.  These dimensions include items such as increased productivity, and lowered mental/physical effort.  The mapped design principles were evaluated and refined in iterations of exploratory prototyping (n=15), and experimental prototyping (n=12).  A set of 9 design principles were used iv  to create features for a prototype.  This prototype used features such as speech recognition, wearable technology, and smart mobile devices.  Results of qualitative data analysis (n=27) and questionnaires (n=11) indicated that the prototype was perceived to be useful, easy to use, and a good task-technology fit.  This showed that the design informed by homecare nurses workarounds addresses key aspects of technology acceptance.   v  Lay Summary  This study identified barriers to homecare nurses’ work in Vancouver, especially ones related to use of computer technology, identified and classified how nurses overcome these barriers, what can be learned from nurses’ creative problem solving strategies, and how what was learned can be used to create new solutions that support the nurses’ work.  Participants were 121 nurses.  First 33 nurses were accompanied on their patient visits.  The observations suggested that nurses create and use workarounds to overcome barriers.  Next workarounds were classified and a conceptual description of them was created.  Most common of those characteristics were used to find related design guidelines for creation of a new wound documentation system.  Then a series of participatory design and feasibility testing sessions were carried out, to refine, evaluate, and create new design features.  The results indicated that nurses find the newly designed wound documentation system useful and easy to use. vi  Preface  All the user studies discussed in this dissertation were approved by the University of British Columbia Behavioural Ethics Research Board and the Vancouver Coastal Health Authority.  Ethics approval certificate number for the initial fieldwork is H14-01982, and for the exploratory prototyping is H15-02892, and for the experimental prototyping is H16-00308.  All of the participants in the user studies signed a consent form. Chapters 1 and 2: the work including literature review, formation of research questions and ideas, and setting of the methodology were completed by the author of this dissertation.  Various sections of these chapters were also used in the introduction sections of the publication made during this work. Chapter 3: the work including ethnographic observations, data collection, analysis, topical coding, identification and classification workarounds were done by the author.  The results of the topical coding, the identification and classification of the workarounds were reviewed by Dr. Leanne Currie during this phase of this dissertation.  The below publications were written based on this chapter. 1. Dawood Al-Masslawi, Sidney Fels, Rodger Lea, Leanne M. Currie, “Nurse-Centred Design: Homecare Nursing Workarounds to Fit Resources and Treat Wounds,” 21st International Conference on Engineering Design, Vancouver, British Columbia, 2017. 2. Dawood Al-Masslawi, Leanne M. Currie, Sidney Fels, Rodger Lea, “Use of Documentation Systems and Community Nurses’ Problem Solving for Wound Care Management,” e-Health Annual Conference and Tradeshow, Vancouver, British Columbia, 2016 (winner of top 10 e-posters). vii  Chapter 4: work including data collection and analysis, topical coding, and mapping of design principles were done by the author.  The results of the topical coding, and mapping of the design principles were reviewed by Dr. Leanne Currie.  The below publication was written based on this chapter. 3. Dawood Al-Masslawi, Sidney Fels, Rodger Lea, Leanne M. Currie, “User‐Centered Mapping of Nurses’ Workarounds to Design Principles for Interactive Systems in Home Wound Care,” 5th IEEE International Conference on Healthcare Informatics, Park City, Utah, 2017. Chapter 5: the initial mapped design features were primarily created by the author.  The prototyping sessions were carried out with the help of Lori Block.  The clinical scenario for the prototyping sessions were created with help of Shannon Handfield, Lori Block, and Charlene Ronquillo.  The collected data during the prototyping sessions was analyzed by the author, Lori Block, and Charlene Ronquillo.  The final results were based on team consensus.  The below publication was written based on this chapter. 4. Dawood Al-Masslawi, Lori Block, Charlene E. Ronquillo, “Adoption of Speech Recognition Technology in Community Healthcare Nursing,” 13th International Congress on Nursing and Allied Health Informatics, Geneva, Switzerland, 2016 (winner of top 8 projects). Chapter 5, and 6: the medium and high fidelity prototypes were created and refined by the author.  The prototyping sessions were carried out with the help of Lori Block.  The clinical scenario for the prototyping sessions were created with the help of Shannon Handfield, Lori Block, and Charlene Ronquillo.  The collected data during the prototyping sessions was analyzed by the author, Lori Block, and Charlene Ronquillo.  The final results were based on team consensus.  The below publication was written based on this chapter. viii  5. Dawood Al-Masslawi, Charlene E. Ronquillo, Lori Block, Shannon Handfield, Sidney Fels, Rodger Lea, Leanne M. Currie, “SuperNurse: Nurses’ Workarounds Informing the Design of Interactive Technologies for Home Wound Care,” 11th EAI International Conference on Pervasive Computing Technologies for Healthcare, Barcelona, Spain, 2017. Other related publication and presentation: the infrastructure capable of supporting the backend technology for the developed prototype was developed by the author.  Early investigations to conduct user studies to evaluate this infrastructure were carried out by the author.  The below publications were written based on this phase of the dissertation. 6. Dawood Al-Masslawi, Sidney Fels, Rodger Lea, Leanne M. Currie, “Recording Events, Interactions, and Annotations to Communicate Reasoning in Medical Situations,” Workshop on Ubiquitous Technologies for Augmenting the Human Mind, at the 14th ACM International Joint Conference on Pervasive and Ubiquitous Computing, Seattle, USA, 2014. 7. Dawood Al-Masslawi, Sidney Fels, Rodger Lea, Leanne M. Currie, “Ubiquitous Capture and Augmented Access (UbiCA), A Study for Decision Support in Chronic Care,” 5th Annual GRAND Conference, Ottawa, Canada, 2014.  Table o AbstLay PrefTabList List GlosAckDedCha1. Contentract ...........Summary ..ace .............le of Contenof Tables ...of Figures .sary ...........nowledgemication ........pter 1: Intr1  Wounds2  Nurses, W3  Technol4  Motivati5  Problem6  Contribu7  Approac  The1.7.1  Res1.7.2  Me1. s .........................................................ts ........................................................................ents ...............................oduction .... and Woundound Manogy Acceptaon .............. Statement .tions..........h ................sis Statemeearch Questthodology ..Design .................................................................................................................................................................................................... Care ..........agement annce Model ................................................................................nt ................ions .....................................................................................................................................................................................................................................................................d Health Infand Nurses’.........................................................................................................................................................................................................................................................................................................................................................................ormation Te Workaroun.........................................................................................................................................................................................................................................................................................................................................................................chnology ...ds ......................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................ix .... iii ......v .... vi .... ix ....xv .. xvi  xviii .. xix .. xxi ......1 ..... 3 ..... 3 ..... 5 ..... 5 ..... 7 ..... 8 ... 10 ... 11 ... 11 ... 13 ... 13  Cha2. pter 2: Rela1  Workaro2  Technol3  Decreasi4  Increasin5  Lowerin6  Summarpter 3: Iden1  Overview  Wo3.1.12  Methods  Dat3.2.1  Dat3.2.23  Results:   The3.3.1  The3.3.2  The3.3.3  Bar3.3.44  Results:   Cla3.4.1  Val3.4.2DevelopmeEvaluation ted Work ..unds as Instogy Acceptang Mobilityg Naturalneg Informatioy ................tification a of the Sturkarounds a..................a Collectiona Analysis .Home and C Nurses’ W Nurses’ Da Nurses’ Wriers to NurWorkaroundssification oidation of Wnt .......................................................ruments to Ence and the Work for Nss of Interacn Load for ....................nd Classificdy ...............nd Problem .................... ......................................ommunity orkstation ...ily Routineork Activitiesing Activitis Classificaf Workarouorkaround .........................................................licit Requir Role of Woomadic Wotive SystemMemory Int...................ation of W...................Solving in N.........................................................Nursing Wo......................................s ................es ...............tion in Homnds .............Classificatio.........................................................ements for rkarounds ..rkers ..........s ................ensive Work...................orkarounds...................ursing ................................................................rk Characte............................................................................e and Comm...................n .........................................................................Systems ............................................................... .................................... .................................................................................................................ristics .....................................................................................unity Nurs..................................................................................................................................................................................................................................................................................................................................................................................................................................ing ...................................................x ... 15 ... 16 ....18 ... 18 ... 21 ... 23 ... 25 ... 27 ... 28 ....31 ... 31 ... 32 ... 33 ... 36 ... 37 ... 38 ... 39 ... 42 ... 45 ... 46 ... 50 ... 52 ... 56  3.3.Cha4.4.4.4.  Wo3.4.35  Opportu6  Summar  Con3.6.1pter 4: Wor1  Overview2  Methods  Dat4.2.1  Dat4.2.23  Results:   Wo4.3.1  Wo4.3.2  Wo4.3.3  Wo4.3.4  Tem4.3.5  Acc4.3.6  Act4.3.74  Results:   Cor4.4.1  Cha4.4.2  Ind4.4.3  Vie4.4.4  Con4.4.5rkaround Qunities in Usey ................siderations karound S of Study ...................a Collectiona Analysis .Workaroundrkaround Acrkaround Agrkaround Acrkaround Reporal and Sountability,ual ProductMapping Wrespondencrting Compividualized Cw Structuretrol of Currestionnaire of Workaro....................for Next Chituation Mo........................................ ...................................... Situation Mtions ..........endas ........tors ............sources ......patial Orga Division ofs or Outcomorkarounds e of Data anleteness andare Plannin and Status Cent Status ... Results ......unds Towa...................apter ..........del and Ma............................................................................odeling of ............................................................................nization of W Labour, Poes ...............to Design Pd Reality .... Compreheng ................hanges ............................................rds User-Ce......................................pping of W............................................................................Homecare N............................................................................orkaroundwer, and Di...................rinciples ........................siveness ................................................................................ntred Design......................................orkaround............................................................................urses’ Wor............................................................................ ...................scretion .............................................................................................................................................................. .......................................................s .............................................................................................karounds ..........................................................................................................................................................................................................................................................xi ... 60 ... 62 ... 63 ... 64 ....66 ... 66 ... 66 ... 68 ... 68 ... 70 ... 72 ... 74 ... 76 ... 76 ... 78 ... 80 ... 82 ... 84 ... 88 ... 92 ... 94 ... 95 ... 96  4.Cha5.  Sim4.4.65  Summar  Con4.5.1pter 5: Cre1  Overview2  Methods  Dat5.2.1  Dat5.2.23  Results:   Dai5.3.1  Inte5.3.2  Wo5.3.3  Pat5.3.4  Inte5.3.5  Wo5.3.64  Results:   Stic5.4.15  Results:   Dec5.5.1  Incr5.5.2  Low5.5.3  Fit 5.5.46  Summarplified Datay ................siderations ation of the of Study ...................a Collectiona analysis ..Iterative Crly Patients Sractive Patiund Photo Cient Data Floractive Supund Care PlCreation of ky MultimeLow Fidelitreased Timease in Chaering Efforto Workflowy ................ Entry ............................for Next Ch Mapped D........................................ ......................................eation of thechedule andent Contact apture ........w Sheet .....ply List .......an ...............the Emergendia ..............y Prototype e to Chart ...rt Accuracyt in Charting............................................................................apter ..........esign Featu............................................................................ Mapped De Summary .Information............................................................................t Design Fe...................Evaluation .................... ................... ................................................................................................................res .........................................................................................signs ............................ ...............................................................................................atures .......................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................xii ... 98 ... 99 . 100 ..102 . 102 . 104 . 108 . 110 . 112 . 113 . 115 . 116 . 118 . 120 . 122 . 124 . 125 . 127 . 129 . 130 . 131 . 132 . 133  Cha6.  Con5.6.1pter 6: Refi1  Overview2  Methods  Dat6.2.1  Dat6.2.23  Results:   Des6.3.1  Dis6.3.2  Ret6.3.34  Results:   Lik6.4.1  Unl6.4.2  Ver6.4.35  Results:   Dec6.5.1  Low6.5.2  Con6.5.3  Com6.5.4  Tec6.5.56  Summar  Con6.6.1pter 7: Considerations ning and E of Study ...................a Collectiona Analysis .Refining theign Featurecarded Desiained DesigAcceptanceely to Adopikely to Adoy Likely to Mid and Hirease of Timering Effortrol of Statupleteness ohnology Acy ................siderations clusion .......for Next Chvaluation o........................................ ...................................... Mid-Fidelis of Interestgn Featuresn Features .. of the Hight Design Feapt Design FAdopt Desiggh Fidelity Pe to Chart .t in Chartings ................f Chart .......ceptance Su....................for Next Ch...................apter ..........f the Design............................................................................ty Prototype......................................................... Fidelity Protures ..........eatures.......n Features .rototype Ev................... .......................................................rvey ..............................apter ................................................ Features .............................................................................. ...........................................................................totype .................................................................aluation .......................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................xiii . 135 ..137 . 137 . 138 . 139 . 140 . 140 . 141 . 142 . 143 . 144 . 146 . 147 . 148 . 149 . 151 . 152 . 154 . 155 . 156 . 157 . 158 ..160 1  Contribu  The7.1.1  The7.1.2  The7.1.3  Key7.1.42  Reflectio3  Limitatio4  Future W5  Concludrences ........endix A:  Dendix B:  Dendix C: Thendix D: Thendix E: Deendix F: Deendix G: Teendix H: Utions.......... Workaroun Workaroun Mapped D Dimensionns .............ns .............ork ...........ing Remark...................emographiemographie Workaroe Experimmographicmographicchnology Asefulness an....................d Categoried Situation esign Princips of Usefuln............................................................s .....................................cs of the Etcs of the Vaund Questiental Protos for the Exs for the Excceptance d Ease of U...................s .................Model ........les .............ess and Eas...............................................................................................hnographiclidation Stuonnaire .....typing Queploratory PperimentalModel and se Design P............................................................................e of Use .................................................................................................... Study ........dy ................................stionnaire ..rototyping PrototypinIts Variatiorinciples ............................................................................................................................................................................................................................................................................. ..................g ................ns .......................................................................................................................................................................................................................................................................................................................................................................................xiv . 161 . 161 . 162 . 163 . 164 . 166 . 167 . 168 . 169 ..171 ..178 ..179 ..180 ..181 ..183 ..184 ..185 ..193 xv  List of Tables Table  1.1. Summary of the mapped design principles. ................................................................... 2 Table  3.1. Frequencies of workarounds in a daily routine of a homecare nurse. ......................... 44 Table  3.2. Logged activities of home and community healthcare nurses. .................................... 46 Table  3.3. Causes of workarounds. ............................................................................................... 49 Table  3.4. Initial categorization of the identified workarounds. ................................................... 51 Table  3.5. Classifications of workaround instances. ..................................................................... 53 Table  3.6. Descriptive statistics of the validation questionnaire for workaround classification .. 56 Table  3.7. Results of the workaround measurement survey. ........................................................ 61 Table  4.1. Adaptation of work situation model to the workaround situation model. ................... 71 Table  4.2. Actions identified involving the instances of workarounds (n=489). .......................... 74 Table  4.3. Agendas of the identified workaround instances (n=472). .......................................... 75 Table  4.4. Resources used to support workarounds (n=681). ....................................................... 77 Table  4.5. The identified outcomes of the workaround instances (n=469). ................................. 83 Table  4.6. The mapped design principles from workarounds and the created features. ............... 91 Table  5.1. Questions asked during the exploratory prototyping sessions. .................................. 107 Table  5.2. Highlights of feedback received during exploratory prototyping. ............................. 124 Table  6.1. Results of the experimental prototyping survey ........................................................ 156  xvi  List of Figures Figure  1.1. Dissertation flow diagram to study workarounds. ...................................................... 13 Figure  3.1. Dissertation flow diagram (identification and classification). ................................... 31 Figure  3.2. A common layout of home and community healthcare nurses’ workstation. ............ 39 Figure  3.3. Example of a nurse's daily routine to visit patients with wound at home [88]. .......... 43 Figure  3.4. Frequencies (and percentage) of barriers causing workarounds (total 287). .............. 49 Figure  3.5. Frequencies (and percentage) of workaround classifications (total 464). .................. 54 Figure  3.6. Averages of workaround classification validation results. ......................................... 58 Figure  4.1. Dissertation flow diagram (modeling and mapping). ................................................. 66 Figure  4.2. The approach to map design principles that support workarounds. ........................... 67 Figure  4.3. The workaround mapping to design principles. ......................................................... 87 Figure  4.4. The designs for the patient data flow sheet (top) and the progress note (bottom). .... 89 Figure  4.5. The supply list design. ................................................................................................ 93 Figure  4.6. The design for the patient contact information feature. ............................................. 96 Figure  4.7. The design for the daily patient schedule (top) and the summary (bottom). .............. 97 Figure  4.8. The design for the wound care plan. .......................................................................... 98 Figure  5.1. Dissertation flow diagram (creation of mapped designs). ........................................ 102 Figure  5.2. The setup for the exploratory prototyping session. .................................................. 106 Figure  5.3. Modifications made by the user on the prototype features. ...................................... 109 Figure  5.4. Exploratory prototyping, assessments entries (middle), assessment page (right). ... 112 Figure  5.5. Exploratory prototyping user feedback. ................................................................... 128 Figure  6.1. Dissertation flow diagram (refining and evaluation of designs). ............................. 137 Figure  6.2. Experimental prototyping, assessments entries (middle), assessment page (right). . 141 xvii  Figure  6.3. Wound cleaning (left), patient assessment (middle), assessment entries (right). ..... 145 Figure  6.4. Experimental prototyping user feedback .................................................................. 150   xviii  Glossary Term Definition Conceptual model A set of concepts that are used to distinguish, describe, and predict occurrences of a phenomena (e.g. workarounds) Design feature An artifact that supports a functionality and/or has an appearance both informed by design principles. Design principle  An actionable description that guides the creation and modification of functionality and appearance of design features. Dimensions of a conceptual model Measurable concepts that comprise the conceptual model. Perceived ease of use The degree to which a person believes that using a particular system would be free of effort Perceived usefulness The degree to which a person believes that using a particular system would enhance his or her job performance. Workaround workarounds are defined as “intentionally using technology in ways the technology was not designed for, or relying on alternatives which conflict with the formal ideology of the used technology” Workaround classification Describing patterns in the identified workarounds that have common and unique actions, resources, times, locations, outcomes, and other characteristics. Workaround identification Differentiating instances of workarounds. Workaround mapping  Using attributes that characterize a workaround classification to identify a relevant design principle and revise the design principle to include and support those characteristics. Workaround modelling Articulating instances of workarounds using attributes such as actions, resources, times, locations, outcomes, and other characteristics.   xix  Acknowledgements It is utmost important to acknowledge that this work was carried out on the traditional, ancestral, and unceded territory of the Musqueam, Squamish and Tsleil-Waututh First Nations.  This work would have not been possible without the funding and support through various grants, fellowships, and awards, from the University of British Columbia, MITACS, St. John’s College, and CIHR.  In addition, I especially appreciate the support and understanding from my employers during the completion of this work, including TTT Studios, Collaborative for Advanced Landscape Planning (CALP), and the Clinical and Systems Transformation (CST) project at Provincial Health Services Authority. I am immensely grateful to my supervisors Dr. Sidney Fels and Dr. Leanne M. Currie.  I truly appreciate Dr. Fels’s insight, guidance, and the opportunities he gave me to explore, find, examine, and learn.  The invaluable support I received from Dr. Currie was crucial to the progress of my research. Her constant words of encouragement were a source of confidence, and her meticulous attention to detail were a source of rigour during my research. I would like to deeply thank Dr. Rodger Lea for his advisory role and providing feedback for various manuscripts I wrote for publication.  I also would like to thank Lori Block and Dr. Charlene Ronquillo for their collaboration which truly enriched this work, as well their friendship that became very dear to me.  I would like to express my great gratitude to all the homecare nurses who participated in my research, as well Shannon Handfield who made that possible. I have never ending appreciation for all who made this journey easier in one way or another.  Dr. Nima Kaviani for his great conversations and friendship, Dr. Roberto Calderon for his great advice, members of the St. John’s College community that were my support network especially xx  during early years of my research, and many other loved ones and friends who I crossed path with. Last but not least, I am very thankful to my parents and to my sisters, who always believed in me during the thick and thin of this.  Their love, sacrifice, wisdom, and perseverance is incredibly inspiring to me, and without their support this work would have not been possible.  xxi  Dedication To Mama, Baba, Zahra, Yasamin, and Nasim, you are the light and truth in my life.   1  Chapter 1: Introduction The majority of computing technology serves to support work as a resource or a tool, however at times technology may also act as barrier to work [1].  In such situations users develop alternative paths when they perceive that the technology is preventing them from accomplishing their work [2].  These alternative paths are referred to as workarounds [1], [2].  Workarounds are common in work environments where tasks are complex and workers need to constantly negotiate and find the best fit between different dimensions of work, such as activities involving, time, location, policies, and technology used [1], [2].  Workarounds have been identified in many different domains and, as such, several different definitions of workarounds exist [2]–[4].  In the context of technology adoption, workarounds are defined as “intentionally using technology in ways the technology was not designed for, or relying on alternatives which conflict with the formal ideology of the used technology” [1].  This definition overlaps with other concepts such as appropriation [5], [6], mobility work [7], [8], and paper systems [9], [10], which, in general, refer to the use of alternative means in order to accomplish tasks when the intended means are not accessible or deemed appropriate.  Healthcare has been an active research space for the study of workarounds and evidence shows workarounds are common in health information technology (HIT) implementations [6], [9], [11], [12]. The purpose of the work presented in this dissertation was to first identify and classify workarounds, then map these to a model of how workarounds are used in situational contexts via the workaround situation model.  The work involved identification of steps and creation of tools in a user-centred approach for the design of interactive systems.  In this approach workarounds were mapped to design principles, and design principles were used to create a new interactive system.  The interactive system was evaluated in a series of user studies in which the system was 2  refined from a low fidelity prototype to a high fidelity prototype.  The primary stakeholders in this study were home and community healthcare nurses who, as point-of-care care providers, interact with patients more than other clinicians [13].  Nurses manage and treat patients, however compared to patient-centred, physician-centred, or administration-centred literature, there is limited evidence regarding the design and the use of technology in nursing, or nurse-centred design [13].  The table below is the summary of the mapped design principles.  These design principles will be discussed in chapter 4. Table  1.1. Summary of the mapped design principles. Mapped design principles for usefulness  Correspondence of data and reality  The design should provide access to previous and current data entries, and support edit of current and past entries.  The design should provide prompts during wound assessment and documentation, care planning and execution, reviewing of records, and assessment of supplies to re-evaluate the current and past entries. Charting completeness and comprehensiveness  The design should support complete and pertinent capture of patient information, wound assessment items, executed care plan, necessary supplies, the care schedule, and the sources where this information is obtained.  The design should support capture of data during wound assessment and documentation, care planning and execution, reviewing of records, and assessment of supplies. Individualized care planning  The design should make care planning more individualized and/or accurate by allowing the nurse to capture data about the extra supplies left at the patient's home or in the nurse's car.   Mapped design principles for ease of use  View structure and status changes  The records system should support a structured view of patient information and treatment supplies where status information, especially changes to those items, are visible. Control of current status  The records system should support a dashboard view that summarizes the executed care plan and the used supplies.  The records system should support a dashboard view that summarizes the supplies needed for the patients' visits. Simplified data entry  The records system should simplify data entry with offering pre-analyzed and prepared patient information and wound care supply items that are needed for the current visit and can be entered interactively.   3  1.1 Wounds and Wound Care As the population ages and as countries move towards ‘aging in place’, there is increased demand for homecare services [14] and in Canada the home and community healthcare sector is responsible for providing homecare services [14].  In Canada and elsewhere, the prevalence of wounds within the population is approximately 1.7% [15] with an increased prevalence as people age.  Thus, wounds are an increasing problem and the estimated annual cost of wound care in Canada is $3.9 billion per year [16].  While there are many types of wounds (e.g., surgical wounds, burn wounds, diabetic foot ulcers, venous insufficiency wounds) many of these are not healed quickly and will become classified as ‘chronic wounds’.  Chronic wounds are wounds that do not heal through a timely and orderly reparative process to produce functional and anatomic integrity within 3 months [17].  Several reports have indicated that between 7.3%-35.5% of patients receiving health care services at home in Canada have chronic wounds, with an average healing time of 6 months [18]–[20].  A subgroup of patients with wounds are those who have diabetes.  In 2010, it was estimated that 15% of all patients with diabetes in Canada (approximately 345,000 people) would develop chronic wounds on their feet in their lifetime, with an estimated 4,000 annual limb amputees, and a 69% mortality rate within five years [21].  Wounds are a burden on the Canadian patient population, therefore understanding and improving wound care is a key priority in healthcare.   1.2 Nurses, Wound Management and Health Information Technology Nurses are the largest group of professional healthcare providers [22] and often are the bedside clinicians who manage and provide care for patients with chronic wounds [23].  Especially in Canada homecare nurses often are the clinicians who manage and provide care for 4  patients with chronic wounds, while support workers, physical therapists, physicians and families manage other patient needs [14].  Homecare nurses are the primary providers of wound care in community health centers and in patients’ homes [14].   Health information technology (HIT) has the potential to improve practice for homecare nurses who care for patients with wounds [15], however, adoption of HIT varies [24].  Several interactive systems have been developed for nurses to track wound care and show promise for adoption in home and community settings, especially for collaboration, coordination, and learning among the care team [11], [15], [25]–[27].  Researchers have investigated technology adoption and workarounds in nursing work [2], [28], [29], however workarounds in the homecare setting have not been studied to our knowledge [3].   Some evidence indicates that the use of technology in patients’ homes can be successful [30], [31].  While other studies suggest that there is insufficient evidence to support the use of technology in homecare of patients with wounds [15], [32].  Some researchers suggest that it is possible that the extensive use of paper systems is a manifestation of the unsuccessful adoption of the clinical information systems in homecare and other care settings [33], [34].  Research within certain industries such as banking, e-governance, and health suggests that there is a growing trend wherein work historically performed manually, is transitioned into computerized systems which workers are mandated to use [35]–[37].  The fieldwork carried out for this dissertation showed that mandated use of technology did not lead to complete technology adoption, which was the intended outcomes from the perspective of the healthcare organization.  Instead, workarounds were used to bypass barriers to the nurses’ work even if that meant the mandated system was not used.    5  1.3 Technology Acceptance Model and Nurses’ Workarounds  The technology acceptance model (TAM) was first introduced in 1989 by Davis [38].  TAM is based on theory of reasonable action [39], and theory of planned behavior [40].  These are theories in the field of psychology to explain and predict human behaviour based on existing attitudes.  The primary two constructs included in TAM were usefulness and ease of use.   TAM has been used to measure and identify workarounds in nurses’ work, with particular focus on two dimensions that define TAM, usefulness of perceived benefits and ease of use perceived for features [33].  These two dimensions have been used most frequently and consistently in various studies, especially in healthcare setting [24], [41].  Especially these two dimensions are the only ones that were used in the literature to identify and measure workarounds [33].  For these reasons the study presented in this dissertation used these two dimensions of TAM [38], perceived ease of use and perceived usefulness [33] to measure the workarounds that homecare nurses are creating and using in their work.  Perceived usefulness is defined as "the degree to which a person believes that using a particular system would enhance his or her job performance”; and perceived ease of use is  defined as "the degree to which a person believes that using a particular system would be free of effort” [33], [38].  In short, perceived usefulness and ease of use directly impact technology adoption and the perceived need for workarounds [33], [38].     1.4 Motivation An initial fieldwork investigation was carried out in January 2015 and involved 120 hours of ethnographic observation of homecare nurses who were performing wound care and documentation.  The fieldwork study is fully reported in chapter 3 in this document.  The 6  fieldwork study focused on community health units in the province of British Columbia (BC).  In this setting a wound documentation system (WDS) has been in use since 2009 in conjunction with a community electronic health records system (community EHR).  This made such an environment a suitable setting to study workarounds [35].  In the fieldwork investigation, it was observed that wound care involved assessment of 7 types and 6 stages of wounds, while wound treatment involved selecting from 120 medical products and documenting 25 characteristics of wounds.  These point to the complex nature of wound care.  The work of the homecare nurse involved driving or walking to multiple patients’ homes each day, which required planning and coordinating appointments that may change unpredictably throughout the day.   The findings showed that workarounds were very frequent when nurses used the WDS to record nursing activities (assessment and treatment) for each patient and for each visit.  In addition, the largest portion of interaction with the system was for data entry rather than data access.  Copied entries (i.e., cut and paste from previous documentation) were also common.  In addition, the use of paper systems was extensive.  Many nurses used a paper travel notebook during their visits to patients’ homes and wrote down items that required follow-up later that day.  Nurses used the same paper notepad later in the day as the data source to transfer the patients’ information into the WDS.  The notepad was part of an entire paper system that homecare nurses had created as a workaround.  The ‘workaround’ system consisted of the patient’s paper records, which was a collection of printouts from the WDS, the community EHR, faxes, hand written notes, sticky notes, and sometimes wound care supplies.  Using this paper ‘workaround system’, nurses were able to have an overall look at the patient’s status in one place while they were on the move.  As one of the nurses noted about the paper file, “it’s all there in front of you!” 7  Based on these observations, the complexity of wound care documentation may have led to  information overload, especially given the nomadic nature and the loosely defined workflows of the nurses in home and community healthcare [25].  While the healthcare organization expected complete adoption of the current WDS during nurses’ visits to patients’ homes, the fieldwork study demonstrated that nurses had not fully adopted the system, and instead were developing workarounds.   The findings of the fieldwork indicated that the nurses perceived the WDS to be easy to use, but not as useful as they would like.  Indeed, they perceived the community EHR system to be useful but difficult to use.  Despite that the use of these systems was mandatory, the nurses used some of the features because they perceived them as useful or easy to use.   As much as workarounds are prevalent in HIT implementations [2], [28], [29], where and how to use them for guidance towards successful adoption remains understudied [3], [42].  In order to shift from a designer-centred to a user-centred system, it may be possible for these workarounds to be identified, measured, and used in design, to move closer to a user-centred system [28].  In this dissertation, workarounds informed design decisions for WDS at patients’ homes.    1.5 Problem Statement During home visits, the main function of workarounds was to allow nurses to complete the treatment of patients, despite the barriers imposed by the WDS.  The nurses’ job required documentation and use of patient data.  Often nurses could only document and use the patient data that the workarounds allowed them access to, even though their WDS provided features to 8  for documentation and use of those data.  This resulted in delays, inconsistencies, and loss of data as noted by the nurses who participated in the fieldwork investigation.    In use of the WDS, the nurses indicated that some of the features, while easy to use, were not perceived to be useful.  For example, some parts of the system used subjective labels such as high or low, which resulted in the nurses being confused as to the severity assessment (i.e., sometimes ‘high’ meant ‘better,’ and at other times ‘high’ meant ‘worse’).  As a result, nurses indicated that they frequently use the WDS as a “tick sheet” to check if an item was documented rather than using it as a tool to plan patient management.  Instead, important patient data would be stored in the community EHR or on paper records.  The nurses reported that retrieving patient history from the free-text entries, such as from ‘case notes’ in the community EHR and the paper records, was challenging and at times impossible. Workarounds were common in the work of homecare nurses.  These workarounds may have been due to a mismatch between the mandated system and the nurses’ work.  Therefore, in this dissertation a series of studies were performed to identify a match between the users’ needs and the technology by using workarounds to guide design decisions.    1.6 Contributions Completion of this dissertation gained 4 primary contributions. I. Identification and classification of workarounds: Identification and classification of workarounds, their causes, and some of their consequences in wound documentation of patients at home was performed (chapter 3).  The results of this contribution were corroborated in a survey study (chapter 3).  Nurses’ workarounds in the context of home and community healthcare have not been studied before, and their identification as well their causes and 9  consequences not only contributes to research focused on home and community healthcare, but also can be used in similar settings where dimensions and attributes of workarounds correspond to home and community healthcare.  II. Modeling of workarounds: A conceptual model, the Workaround Situation Model (WaSM), was developed, based the work situation model [1], from the literature (chapter 4).  The WaSM assumes a similar perspective to articulation work [43], and mobility work research [44].  The WaSM includes 11 attributes of workarounds including tasks, agendas, actors, and outcomes.  Each instance of a workaround was given these attributes.  The workaround situation model allowed to identify the most common workaround situations and their attributes, which were necessary for the mapping of workarounds to design principles (chapter 4).  The results were in agreement with the results of the follow up done in the ethnographic study, especially for parallel system use (chapter 4).  III. Mapping of workarounds to design principles (ease of use and usefulness):  Design guidelines to improve aspects of usefulness and ease of use in home and community healthcare were mapped (chapter 4).  The design guidelines were used to create wound documentation prototypes in a series of iterated user studies (chapter 5 and 6).  The iterations started with exploratory prototyping towards a low fidelity prototype (chapter 5), and continued with experimental prototyping towards a high fidelity prototype (chapter 6).  The designs were evaluated with valid and reliable measures of usefulness and ease of use.  Applications in contexts similar to home and community healthcare also benefit from these design guidelines. 10  Especially applications designed for nomadic workers who carry out mission critical work such as healthcare, education, energy, transportation, environmental preservation, and public safety.  IV. Validation of mapping workarounds to design principles (usefulness and ease of use):  The dimensions of usefulness and ease of use were identified and validated.  This included using measures needed to create a user-centred approach in the design of interactive systems using workarounds, with wound documentation applications in home and community as a case.  Valid and reliable measures from the literature were selected that were shown to predict aspects of usefulness and ease of use (chapter 3).  A study (chapter 3) was conducted to corroborate the results of the use of these measures in identifying workarounds in the home and community healthcare setting.  The identified dimensions and measures can be used in other contexts similar to home and community care, especially if attributes of their workarounds correspond to home and community healthcare (chapter 7).  1.7 Approach The goal for this dissertation was to identify and measure workarounds using validated and reliable tools, and to use the identified and measured workarounds as feedback for the design of an interactive system used for wound care.  The feedback process involved steps to identify and map attributes of the most common workarounds, such as tasks, resources, and outcomes, to system design requirements.  This research was conducted as a case study for wound care documentation of patients in the Vancouver area where community health units provide in-home care for patients with wounds.  Investigation of the research questions listed below achieved the research goals set for this work to prove the thesis statement detailed as follows.   T1.7.1The Identhe most systems. evaluatedof use thameasures  R1.7.2The Whadimensio(WDS) inAnswto the prisecondar How durin Whatnurse Whatto clahesis Statethesis statemtification ancommon wo When appl in this disst are key to and the maesearch Qprimary quet are the dimns, needed t home and ering this qmary questiy questions do home ang their visits are the keys, and gener are the comssify them? ment ent for thisd measuremrkaround siied in the deertation resu successful apping towaruestions stion that thensions of uo use nursescommunity uestion depeon will be dare discussed communit?  measures toate feedbackmon and un dissertationent of worktuations andsign of a wolted in the idoption of tds a user-ceis work wassefulness a' workarounhealthcare?nds on answetailed in chd in other chy nurses wo assess work in the desiique charac is: arounds can user-centreund documedentificationechnology,ntred design trying to finnd ease of uds in the deering the seapter 6 and apters. rk around syarounds cregn of WDS?teristics of g be used to cd design printation syst of dimensiand validati.  d answers tse, and the msign of a wocondary quthe conclusistem limitaated and us  roups of woreate a mapnciples for iem, as a casons of usefuon of the uso is as folloeasures setund documeestions beloon after the tions to use ed by homerkarounds tping betweenteractive e, the approlness and eae of the ws:  within thosntation systw.  The answanswers to tpatient data and commuhat can be u11 n ach se e em er he  nity sed 12  Identification and classification phase of this work provided answers to these questions.  Chapter 3 of this dissertation discusses these questions and their answers, which is also the first contribution of this work listed in section 1.6.   In chapter 4, the following questions will be discussed, which refer to the modeling and mapping phase of this work.  What is a conceptual description for workarounds created and used by home and community nurses, such that they can be mapped to design principles for WDS at the point of care? The answer to this question is the work situation model, which is the second contribution of this work.  The next questions are as follows:  How can the work situation model be adapted for use in the mapping of workarounds to design principles for WDS in the home and community healthcare setting?  How to identify relevant design principles and use the adapted work situation model in the mapping of workarounds to design principles for WDS in the home and community healthcare setting? The answers to these questions are the third contribution of this work, the mapping of workarounds to design principles.  The fourth contribution is discussed in chapters 5 and 6.  In these chapters the following question, and consecutively the primary research question will be answered.  How to create, refine, and evaluate design features based on the mapped design principles?       M1.7.3The process.  the develphase condissertatithe flow The created aEthnograwas modprinciplePrototypi[49]–[51wounds aeasy to uclassifica(chaptersethodologmethodologIn the user-opment phasisted of teon.  This diaof this disseDesign first phase ond used worphic studieseling and ms were used ng has been].  The partict their homese, and praction, model 3 and 4).  TIdentclassy y to investigcentred desise consistedsting the program is incrtation that cFigure  1f design wakarounds by were succeapping of thto create de a successfuipants of ths, and they tical.  The ining, and maphe Technoloification and ification ate the abovgn process t of creating totypes in pluded at the hapter is lo.1. Dissertations an ethnogr home and cssful in simie workarounsign featuresl method fore prototypinprovided feeitial design ping of comgy AcceptaModeling amappinge research qhe design phlow to high atient scenabeginning ocated.  flow diagramaphic studyommunity lar contextsds to design using explo distilling ug sessions wdback to enideas were imon workance Model (Cremdend  uestions waase consistefidelity protrios.   Belowf each chapt to study work, which idenhealthcare n [3], [34], [4 principles.ratory protoser feedbackere nurses wsure the desnformed byrounds in thTAM) was ation of apped signs s a user-cend of prototyotypes, and  is the flower and it indarounds. tified and clurses (chapt2], [45].  Th  Then thesetyping [46] in healthcaho visit paign features identificatioe ethnograpselected forRefining aevaluationdesignstred design ping sessionthe evaluati diagram of icates wherassified the er 3).  e second ph design –[48].  re settings [tients with  were usefun, hic study  the nd  of  13 s, on this e in  ase 6], l, 14  measurement of the workarounds since it has been validated in related work and in similar settings [33], while there is less evidence for use of other approaches for the measurement of workarounds.  A validated model called the Work Situation Model (WSM) [1], was extended using findings from the identification and classification of the workarounds and related studies on articulation work [43], and mobility work [44] (chapter 4).  The new model was called the Workaround Situation Model (WaSM) The prototypes used during the exploratory prototyping were built using a rapid prototyping software and were printed on cardboard [52] (chapter 5).  This made the communication and the testing of ideas easier as modification of the prototype required only pen and paper.  There was one patient scenario adapted from real life cases.  The patient scenario had the common attributes of workarounds identified in the ethnographic fieldwork.  Exploratory prototyping is best for exploring different design options quickly and it starts with a set of functions permitting the users to perform one of their work tasks with the help of the prototype.  After building a low-fidelity prototype based on the feedback received during the exploratory prototyping, experimental prototyping phase was conducted.  Experimental prototyping can be used in the design process to refine and evaluate the design features [46]–[48] (chapter 6).  Similar to exploratory prototyping experimental prototyping had three steps: functional selection, construction, and evaluation.  With a similar patient scenario, the nurses performed the same tasks as those performed previously; however, the execution of the patient scenario was formal rather than informal.  The duration of the scenario was set for a specific number of tasks to be completed using a wound documentation prototype.  The experimental prototyping sessions were audio and video recorded for later transcription and analysis.  The participants verbalized their thoughts as they provided care for a fictitious patient.  This is known as the Think Aloud method 15  [53].  Content analysis was performed for coding of the transcriptions, which was similar to the approach used in the initial fieldwork.  Potential workarounds that nurses generated to resolve obstacles faced while using the prototype were identified by comparing the obstacles faced by the nurses while using the prototype with those observed during the initial ethnographic fieldwork. Development The prototypes created for use in the exploratory prototyping phase were cardboard models designed with a rapid prototyping software [52].  Customization of the cardboard model prototypes were done on-the-fly during the prototyping sessions using pen and paper.  The development of the low fidelity prototypes replicated mobile applications with most of the interactions included, but data aggregation and analysis features were incomplete.  According to the findings in the initial fieldwork, the main aspects of care documented by nurses during home visits included, wound care plans, wound pictures, and a patient summary (i.e. patient flow sheet).  It was therefore necessary that the technology developed in subsequent prototypes addressed these aspects of care.  As new prototypes were developed or customized through subsequent iterations, the design principles were refined based on the results of feedback received from home and community nurses who provide care for patients with wounds.  These refinements then determined the design recommendations, and subsequently the features and the functionalities of the high-fidelity prototype developed (chapter 7).    16 Evaluation The methods and tools used in this dissertation were evaluated and validated at every stage of the research.  The results of the identification and classification of workarounds were validated in a follow up study (chapter3).  In the follow up study, the participants were given a survey to validate the identified workaround patterns that were used by nurses and were common in home and community healthcare.  The results of the workaround mapping, which were the design principals, were evaluated in the prototyping sessions (chapter 5, and chapter 6).  User feedback was collected during the prototyping sessions using design ideas and features based on the mapped design principles.  The feedback collected was in the form of discussions during the prototyping, as well as refinements made using pen and paper on cardboard models (in exploratory prototyping), and in the form of survey responses to evaluate the usefulness and ease of use for the discussed ideas. The participants were of mixed gender, and of similar knowledge and experience with the clinical and technological aspects of home and community healthcare.  This was to account for vastly different levels of knowledge and experience, which may affect participants’ decision-making.   In the study setting nurses were mandated to use the electronic record systems purchased or developed by the health authorities.  This is shown in other studies to introduce power dynamics that may impact the nurses’ intention to adopt the mandated systems [35].  In regards to the power dynamics, this study followed recommendations made by others, which included keeping the measurement of intention and behaviour separate.  That was making sure that the measured items address intentions regardless of mandate of policies enforced, and also to operationalize 17  measures which are consistent with theory, and carrying out research on a homogenous population using the same technology [35]. 18  Chapter 2: Related Work The literature review in this work included multiple disciplines, due to the multidisciplinary nature of the study.  In particular, fields of nursing informatics, and ubiquitous healthcare were included, as well the field of human-computer interaction.  Within these fields the review was narrowed down to subfields that include work on topics such as wound care at home, workarounds in nursing, technology design approaches for homecare nursing, ubiquitous technology for homecare nursing, and sociotechnical research methodologies in nursing.  2.1 Workarounds as Instruments to Elicit Requirements for Systems One of the earliest and most commonly used theories to study workarounds is the social activity theory [43], which is extended to research in information and communication technology [1].  This perspective has informed much of the literature that discusses workarounds or their related concepts, which will be discussed in the next sections of this chapter.  This research area started with looking at social organization of medical work [54], specifically explicit description of work in terms of its components and the people involved [55].  Identifying components of work such as workers, actions, interactions, accountability, and division of labour allowed to describe and distinguish what is work, who defines trajectory of work, who and how carries out the work, and the accountability systems that are involved [55].  These studies introduced the concept of, "articulation work", which viewed workarounds as articulation work in which actors must negotiate the best fit in addressing various dimensions of a "work situation" in order to accomplish their work [56].  In this kind of work articulation of work becomes important to the actors in order to control the flow of resources, establish and maintain an equitable and flexible division of labour, establish and control quality standards for 19  work and resources, set and revise short-term and long-term goals, create and maintain interactions that support and guide working relationships, to supervise and monitor the work and rectify errors, to routinize and revise the work in forms of policies and procedures, and ultimately to shape the present and the future evolution of structural and organizational condition of the work, be it historical, political, economic, or legal [54], [56]. The concept of a work situation was described and studied based on characterization of work in the social activity theory [1].  Work situation was described with components such as work task, agenda, actors, resources, temporal organization of work, place and special organization of work, division of labour, power, discretion, accountability, and actual products or outcomes [1].  Articulation work was specifically studied using this concept since it described different dimensions of work, and it was found that adaptations of work can be identified and described using dimensions of work situation [1]. This motivated research about workarounds towards identifying and predicting situations wherein workarounds can happen.  For example, workarounds occur in work situations where actors must frequently find resources such as time, people, and equipment to accomplish their work, while these resources are stored at different locations.  Further research indicated that user adaptations to solve conflicts between a user’s work requirements (practice frame) and system requirements (system frame) can be in fact workarounds [34].   There are variety of workarounds identified in the literature, some are known as system workarounds [57], or computer-based workarounds [10], automated backup system [1], and system approaches [58].  Other types of workarounds have been noted in the literature as well, such as procedural adjustment workarounds [1], procedural deviations [28], paper-based workarounds [10], organizational process workarounds [59].  However, the vast majority of 20  workarounds can be grouped into two categories: technology (system) workarounds, and process workarounds [29], [59].  In another study, similar categorizations were proposed as system approaches and person approaches [58].  More closely to the context of clinical documentation systems some of the identified themes included free-text entry workarounds, duplicated order entry workarounds, human reminders (instead of system reminders), paper-based workarounds, and convenience entry [60], [61].  Identifying such workarounds used by nurses engaged in in-home wound care, contributes to the findings of the research community in a setting not previously studied.  In order to identify these workarounds what must be assessed is how nurses both work around system limitations to use the patient data that they need, and provide care.   In one aspect of this problem, the causes of workarounds are a topic of study frequently in the literature and they are well known.  A common perspective suggests that blocks are the primary causes of workarounds [28]; Blocks include design flaws, component failures, the inability of the system to address the problem or task at hand, and design limits or constraints that block the path to the goals that users are trying to achieve.  Others posit that the source of blocks can be expanded to policies, laws, regulations, protocols, process, design, flow, technology, people, and environment [2], [3], [45], [57], [59], [60].   In the current literature, the outcomes and consequences of workarounds are not well researched.  This is mainly due to difficulty faced in clearly attributing positive or negative outcomes to specific workarounds, as other factors also may affect the outcomes.  However, a number of studies do suggest that outcomes can include inadequate use of the system, and development of local adaptations [1], [3], [59]. Adaptations may become the new best practice over time.  This necessitates both identifying and supporting adaptations [34].  The most recent research demonstrates that studying 21  workarounds to collect feedback for system improvement can go beyond improving the software usability but it can trigger innovation in system design [42].   Current research on workarounds, and related concepts such as user adaptations, indicates that it may be possible to use these concepts to elicit requirements for interactive systems.  Much of the research currently focusing on workarounds aims at identification of workarounds; their causes and effects [2], [3], [29], [45], [57], [59], [59]–[62].  However a recent trend is starting to explore use of workarounds in system design [3], [28], [34].  In system design research use of related concepts to workarounds is more frequent, and they will be discussed in the next sections.    2.2 Technology Acceptance and the Role of Workarounds While some research groups identify workarounds as risk factors [2], [57], [59], in a different perspective towards workarounds other research groups note that user adaptations should be considered during the design process, and that can lead to a user-centred system, which is the opposite to a user-hostile system created by a designer-centred approach [28].  A positive view towards workarounds is a recent trend in the literature [28], [61], [62], which is also cited in reviews [63], where workarounds are seen as enabling the execution of patient care, and at times necessary, despite their potential to compromise care.   Several studies have shown that usefulness and ease of use, as two of the main constructs that make up the Technology Acceptance Model (TAM), can predict technology acceptance and use [38], [64]–[67].  Current systems may be perceived as less easy to use or less useful [24].  However, the hybrid of workarounds and the electronic records systems fill the gap between the systems’ frame of work and the nurses’ frame of work [34].  Other studies suggests that workarounds may represent creative problem solving strategies that can be used to provide 22  feedback in the design, development, and evaluation cycle of electronic systems, to move closer to a user-centred system [28], [42].   While there is a large body of evidence that validates use of TAM to evaluate technology systems [38], [64]–[70], especially in nursing [24], [35], [71]–[74], there is less evidence in regards to measuring of workarounds.  However, evidence in literature suggests that dimensions of usefulness and ease of use in TAM can be used to measure nurses’ workarounds, especially workarounds related to technology barriers [33].  TAM has been expanded over the years to add new constructs as predictors for user behaviour.  In TAM2 the new predictors included voluntariness, experience, subjective norm, image, job relevance, output quality, result demonstrability [64].  Later on in Unified Theory of Acceptance and Use of Technology (UTAUT) new predictors were added such as performance expectancy, effort expectancy, social influence, and facilitating conditions [65].  TAM3 added following predictors, computer self-efficacy, perceptions of external control, computer anxiety, computer playfulness, perceived enjoyment, objective usability [66].  In UTAUT2 which is the most recent version more predictors were added which included, hedonic motivation, price value, and habit [67]. TAM and its expanded versions are detailed in appendix G. In various versions of TAM, including TAM2-3, and UTAUT1-2, usefulness and ease of use always have been used as measures to predict user behaviour [38], [64]–[67]. These two constructs especially have been used frequently in healthcare and nursing informatics research [24]. Furthermore, these were the only two constructs used to develop, evaluate, and validate a tool to identify and measure workarounds in nursing informatics [33].  For this reason in this dissertation usefulness and ease of use are used for measuring, identifying, and mapping of design principles. 23  This dissertation built upon work to measure workarounds, and kept consistency in the measurement of constructs and concepts as a strategy to ensure that design decisions informed by specific concepts were evaluated using those concepts.  In this dissertation, the constructs and concepts of TAM were used as tools to identify and measure workarounds, as well to evaluate the success of the design decisions informed by the workarounds.  Concepts such as mobility work as an extension to articulation work, naturalness, and lowering information load are related to usefulness and ease of use.  The aforementioned concepts have been used to identify occurrences similar to workarounds, such as appropriations or user adaptations, and use them to inform system design.  These works are discussed in the remaining sections of this chapter.  2.3 Decreasing Mobility Work for Nomadic Workers Nomadic workers travel for most of their work day, are not associated strongly with any single home office, and are responsible for carrying, managing, and reconfiguring their own work resources [5].  Associated with nomadic workers, mobility work is defined as, “work needed to achieve the right configuration of people, resources, knowledge and places in order to carry out tasks” [7].  As one of the dimensions of the work situation, location determines what resources are available for use.  Researchers have studied how to support the mobility of the medical workers by identifying how they work around problems when they have to be mobile in their work (i.e. nomadic work) [5].  In a sense, how to reduce the need for mobility in order to save the time and effort required to travel to different locations to access resources.  For example one group notes that future interactive systems should be capable of dealing with deviations from plans, which can include workarounds [75].   24  Another line of inquiry, with a focus on adapting the communicated contextual data for nomadic workers, examines the usefulness of mobile context-aware systems in an EHR implementation [76].  The findings are similar to previous work [75], in addition, they recommend that as a solution, HIT implementations should use mobile and handheld devices to support the work of nurses.  They conclude that it is very difficult for users to master context-awareness as a style of interaction.  Hence, they recommend that context-aware technologies should support the appropriation of tools, and the learning process created and used by users, to create a fit between technology and work processes [77].   To more directly address the issue of workarounds in healthcare, one group studied the work practices of staff who are required to be mobile within a hospital ward [8].  They recommend new design principles for context-aware systems after identifying patterns of artifact use, such as a paper system [8].  Their design recommendations include, decoupling information from representation to adapt views to the context, bringing the “object of work” back to the real world, such as personal notes, to augment existing ways of coordinating work.  Though their reference to the established trajectories of work through continuous readjustment and given that articulation work is nearly identical to workarounds [1], strategies to use these work trajectories in a user-centred design process are not offered.  Meanwhile this dissertation envisioned that the designed applications would be used as work objects, or "kits", in order to provide care for the patients.  To this end, this research extended the work situation model for analysis of mobility work to include workaround elements that were identified in the initial fieldwork [7]. The works discussed in this section employ user-centred approaches to study how useful some of the provided benefits are to clinicians.  The conclusion is that the provided benefits, such as decreasing mobility work, are useful, but not necessarily easy to use.  There need to be 25  trade-offs that allow flexible appropriation, as the designed systems for mobility work might not be very easy to use [7].  Some have made early efforts to use workarounds or related concepts in the design process, but the introduction of a complete design process and measures of workarounds to inform design, requires further study.  2.4 Increasing Naturalness of Interactive Systems To study adoption and appropriation in support of the medical practitioners’ workflows, one research group has looked at modeling the natural workflow of medical staff in a neurological care hospital in Brazil [6].  These researchers designed a Natural User Interface (NUI) solution that supports natural workflows, and to establish measures for the degree of “naturalness” of the solution.  The rationale for an NUI is to lower the amount effort needed to interact with the system and therefore improve ease of use [6].  Their recommendation for a NUI-based solution in healthcare is that it should not interrupt existing workflows, and that it should allow for appropriation so that users can use the solution in different contexts.  In their observations, they identify successful appropriation strategies including the use of cell phones to both take pictures of injuries and to communicate those images to colleagues.  However, the practitioner also has to print a patient history timeline to better contextualize the pictures.  Although practitioners used the solution for appropriation in line with its design, it did not support communicating the context of the patient's data which lead to the use of the print outs as workarounds to system limitations.   Other research focuses on augmented paper-based systems in a hospital using ubiquitous technology [78].  In this example, a tangible user interface developed to input data into an EHR utilizes sensor-enabled augmented paper and digital pens.  The design of the system is to support 26  a situational overview using auditory and visual feedback.  In comparison with the work previously discussed in this section [6], the research which focuses on augmented paper-based systems [78] is the most successful in using the natural workflows in a user-centred design process.  However, the authors note that the users have to perform additional tasks not included in their natural workflow to use the system.  For example, the users need to ensure data consistency for both physical and digital interfaces.  In other words, the users have to perform workarounds to get the job done; which is acknowledged by the authors of the study who suggest that further studies can help determine the best methods to provide feedback on this issue [78]. In the work discussed in this section, users employed appropriation strategies to work around some of the system limitations [6], [78].  For this reason, some authors suggest that appropriation should be studied further to get feedback from and address design problems [78].  Researchers also note that there is a need for better ways of measuring appropriation to use in the user-centred design process [6].  Research indicates that clinicians find themselves performing additional tasks, similar to workarounds, to ensure both data consistency across systems and effective communication.  Overall, the effort for the additional tasks seemed low, which indicates that the systems are still easy to use; however, some of the tools were abandoned as they were perceived less useful.  Research suggests that the additional tasks performed should be studied and used as feedback in design [78]. The findings from the fieldwork conducted in this dissertation showed that in many cases homecare nurses use the appropriation of technology as a workaround; namely using the systems to convey information that they are not designed or purposed to convey.  This dissertation identified such workarounds to leverage them in informing the design of wound documentation applications. 27   2.5 Lowering Information Load for Memory Intensive Work Medical professionals often use pen and paper to decrease the information load in their work [10], [79], and in some cases they use these artifacts in conjunction with technology as workarounds [10].  For example, one group studied the way in which therapists make use of pen and paper to collect and communicate data about videos captured during therapy with an individual with autism [9].  This usage of pen and paper to bypass system limitations is noted as a common type of workaround in the literature, also known as “paper systems” [10], [59].  The researchers recommend a system design to automate adding annotations to the video.  Their recommendations for design principles of an interactive system include: capturing as much automatic data as possible and within reason; minimal manual data capture; multiple levels of detail to reduce information overload; and effective privacy mechanisms while preserving data integrity.  These researchers have designed, developed, and evaluated a system that both adds annotations to the video based on a digital pen input on a paper form, as well it supports speech recognition.  This improved timeliness and confidence in decision-making.   In the context of home and community nursing, one research group studied the use of technology in wound care to decrease information load [25].  This work investigated the design of technology that provides sufficient information for decision-making, remote assistance by experts to patients’ homes, thereby increasing satisfaction and confidence in treatment, and reducing the number of hospitalizations and nurse home visits [25].  Their design principles include three dimensions: monitoring, cooperation, and human computer interaction.  In the participatory design phase of their process, they found that continuous monitoring of patients 28  using electronic monitoring equipment was perceived as less useful, since in most cases wounds are stable between home visits. A common principle among the studies discussed in this section is the automated capture of the maximum amount of data, and providing easy access later on to that data when it is needed [9], [25].  This supports medical practitioners in their decision-making by lowering the mental and physical effort needed to communicate and access data, hence improve ease of use.   One group suggests investigating timeliness and the degree of confidence in decision-making to measure the usefulness of lowered information loads [9].  In wound care nursing at home, since the wound care plan is the main subject of decision-making [80], [81], wound documentation to manage the care plan is another factor important to measuring usefulness.  That is, assessment items documented at the point of care within a given time during a patient visit. 2.6 Summary In this chapter an investigation of related work indicated that early research on workarounds as instruments to elicit requirements for systems started with viewing the workarounds as articulation work [1] taken from the social activity theory [43].  From this perspective, workarounds required negotiating dimensions of a work situation to get the job done [1].  In line with that perspective, other research showed adaptations to solve conflicts between users’ work requirements and system requirements are in fact workarounds [34].  Research showed that obstacles to users’ work, or blocks, cause the workarounds [2], [3], [45], [57], [59], [60].  In addition, some studies indicated that workarounds can cause inadequate use of the system [1], [3], [59], while others suggested that workarounds can become new best practices in use of 29  systems, and should be identified and supported in new designs [34].  Using workarounds as feedback in design can go beyond improving usability, instead, they can trigger innovation [42].    The next step towards use of workarounds in design of systems was to identify and measure them.  Ease of use and usefulness as the main constructs of TAM were shown to predict technology acceptance [38], [64]–[67], and can be used in measuring workarounds [33].  This dissertation built upon these works. Further investigation about usefulness and ease of use led to the next category of related work in the context of healthcare and design which introduced mobility work as an extension to articulation work [7].  There was an argument to decrease mobility work that is associated with nomadic workers, such as home healthcare workers [5].  The result was supporting mobility work to make it easier for users to find a negotiated solution to get a job done [75].  Providing easy access to contextual data [75], [76], augmenting objects of work [8], or use of mobile devices [77], were examples of support for mobility work that users perceived as useful.   Another category of the investigated related work focused on increasing ease of use.  To increase ease of use, support for appropriation of technology was suggested and evaluated for “naturalness” [6].  That led to increasing naturalness of the design systems [6].  Natural and tangible user interfaces such as public displays, sensor-enabled augmented paper systems, and digital pens, supported and encouraged appropriation of technology [6], [78].  The difference between appropriation using NUIs or tangible user interfaces, and other design strategies was the support for users’ creativity and complex work dynamic.  Support for users’ creativity enabled them to appropriate technology and use it for purposes not considered during design, at times purposes that are not directly related to clinical work [6].  The results suggested that users have to carry out extra tasks to ensure consistency of data but that did not prevent the users from using 30  the technology [6], [78].  The reason for this was that users perceived that the design can support their complex work dynamics [6].   The amount of mental work that went into ensuring consistency of data created an information load for users, especially during memory intensive work [82].  Therefore, an investigation of another category of related work focused on lowering information load for memory intensive work.  The collected data during the initial fieldwork suggested that most of the workarounds involved use of pen and paper, instead of the electronic records systems, especially for memory intensive tasks such as patient documentation and follow up.  Identifying these types of workarounds in other research [10], [59], suggested that they can be used to design new interactive systems such as augmented videos [9], wound monitoring and remote assistance using mobile devices [25].   This dissertation extended and followed this direction in enabling users to be creative and appropriate technology in ways they see fit.  For this purpose the identification, classification, modeling and the mapping of most common workarounds to design principles was carried out and completed.  The initial fieldwork demonstrated that many of the workarounds used by the nurses who provide care for patients at home address complex aspects of their work.  Using these workarounds as feedback to inform new designs led to more usefulness and ease of use measured during this work.  Chapte In thpresented 3.1 OvIn orshadowinbarriers tthe worksimilar ap6 full-daypatients wtechnologtasks reqadministedocumenclinical pwound car 3: Identis chapter, a.  Then a deerview of thder to identg, interviewo use of techarounds alsoproach was (approximith woundsy educationuired for carring treatmtation systemractices in wre.   Identificaand classificaification aFigure  3.1. Din overview tailed presee Study ify workarous, in additionology and are identifi used in thisately 36 hou.  The traini.  The clinice of differenent.  The tec.  Participaound care, tion tion Momnd Classissertation flowof the identintation of thnds many rn to surveys system limied and meas dissertationrs) training ng sessions al aspect oft types of whnology trating in theseas well of prdeling and apping fication of diagram (idenfication ande study willesearchers h [33], [34], tations that ured, in add.  The initiasessions formainly cons the trainingounds incluining focuse training seactices estaCreationmappedesign Workarotification and  classificatio follow.  ave used fie[61], [83].  Ican cause wition to soml fieldwork  nurses whoisted of clin focused onding woundd on traininssions proviblished arouRev of d s unds classification)n of workaldwork metn these typeorkarounds e of their coconsisted of provide carical educati training nur assessmentg nurses to uded firsthannd use of teefining and aluation of designs . rounds will hods such as of studiesare identifiensequences participatioe at home foon, and ses to perfo, and se the wound exposure tchnology in31  be s , d; .  A n in r rm d o   Nextin the mecompleteactivity tprovides purpose oworkaroumeasure distribute  W3.1.1Accetechnologconstructnurses usmanifestaidentify thealthcarto complin the des   , a month lotro Vancouvd.  During thracking softwtopical codif this analynds used bythe workarod to the nurorkarounptance and y’s usefulns leads to cre other meation of unsuhe types ande nurses in cete a concepign of wounng observater area durie observatiare [84].  Dng, theme gsis was to id nurses whounds develoses, of whicds and Probuse of techness and easeeation of wons to get theccessful ad frequency ompleting wtual analysid care docuional study wng which obons, field noata were aneneration, anentify the ca visit patienped by Halbh 12 questiolem Solvinology in par of use [35]rkarounds  ir job done, option [33], of workarouound care s using the Tmentation sas conductservation otes were takalyzed usind statisticalusative barrts with wouesleben, Rannaires werg in Nursinticular is aff.  Perception[33].  To wosuch as alte[34].  The pnds created documentatiAM model ystems. ed at differef nurses in cen and eveng a data ana analysis caiers, types, nds in their thert, and Be completedg ected by nu of inadequrk around thrnative papeurpose of thand used byon.  Furtherin order to unt communommunity hts were logglysis softwapabilities (Nand consequhomes.  A qennett in 20 and returnerses’ percepacies regardese systemr systems, wis part of th home and c, this part ofse identifieity health unealth was ed using anre which Vivo) [85]. ences of theuestionnaire13 [33] was d. tion of the ing these  limitations,hich is a e study was ommunity  the study ad workaroun32 its   The   to also  the to imed ds 33  3.2 Methods The majority of the studies that have investigated workarounds in different contexts have used mixed methods [2], [63], [86].  Mixed methods have been shown to develop rich insights into issues that cannot fully be investigated and understood using only quantitative, or qualitative methods [86].  Mixed methods also were used for this study.  Qualitative observations were complemented with quantitative measurements and the analysis was reviewed by the research team multiple times to ensure agreement among the team members.  This part of the study was conducted in two phases, the first phase was focused at the identification and classification of workarounds, and the second phase was focused in validation of the results. As part of the ethnographic investigation, nurses were shadowed to collect data about their workarounds.  Events were logged using a mobile application [84].  During fieldwork, event logs and field notes were recorded to identify: blocks causing workarounds, the types of workarounds used, and some of the workaround consequences.  The topical codes used to analyze the data of those three aspects were adapted from the literature [29], [33], [42].  For blocks, the topical codes were adapted from three research groups which study blocks that cause workarounds [1], [2], [57].  The codes included: blocks in design, flow, people, policies, procedures, protocols, regulations, technology, resources, guidelines, and environment.  The field notes were coded topically based on codes adapted from the literature [1], [2], [28], [33], [62].  A code was assigned to each instance of blocks, workarounds, and their consequences.  Based on the identified instances of the codes, the frequency of blocks, workarounds, and their consequences were calculated. 34  Questionnaire data obtained from the participant nurses were analyzed using descriptive statistics.  The nurses perceived technology as the most important factor that affects workarounds in home and community healthcare, which corresponded to findings from acute care [33]. In the next phase, themes of workarounds were identified by analyzing the most frequent workarounds.  The initial topical codes for analyzing workarounds were adapted from another body of research that grouped the workarounds into system and process workarounds [34].  It should be noted that technology related blocks caused both system and process workarounds, as a nurse might work around a technology block by creating a process workaround rather using the technology.  The comparison between process workarounds and system workarounds was an indicator to technology adoption, where much higher instances of process workarounds suggested that the technology was not fully adopted and process means were used to accomplish work.  The system and process workarounds were further divided into classifications based on common patterns of action and attributes of the workarounds.  These most common patterns resulted in seven workaround subgroups, which included the following items:  Pre-emptive information use: when a nurse provides cues (e.g. verbal exchange, in writing, use of specific clinical supplies, or data entry in the system), informing the need to complete certain activities and preempt or reduce occurrence of workarounds in the future.  Preventive information use: when a nurse provides cues (e.g. verbal exchange, in writing, use of specific care supplies, or data entry in the system) informing the occurrence of barriers to activities, their workarounds, and the need to prevent risks associated with those workarounds. 35   Parallel system use: when a nurse uses more than one system for the same intended activities for those systems, and in the intended ways for those systems, to complete the same activities.  Appropriative system: when a nurse uses a system to complete activities in which the available system is not intended for completion of those activities.  Appropriative resource use: when a nurse uses resources to complete activities that the resources are not intended for completion for those activities.  Adaptive system use: when a nurse uses resources to complete activities that the resources are not intended for completion for those activities.  Adaptive resource use: when a nurse uses resources for the intended activities but in unintended ways to complete those activities. The literature suggests that it is important to measure workarounds [2].  Here classification of workarounds, the prevalence of workarounds, and the degree to which workarounds relate to successful adoption of WDSs in home and community healthcare, were drawn from research done on technology adoption in healthcare [24], [33].  A questionnaire developed and validated by another group [33], was used to measure the perception of the block, and the need to alter processes to work around a block.  These were the dimensions of the ease of use construct.  In addition, the preference to follow procedures, and the motive to assist the patient were measured.  These were dimensions of the usefulness construct.  The measured constructs related to factors such as technology, policies, equipment, work processes, and caregivers.  The questionnaire focused on technology and non-technology blocks, providing valuable data about the nurses’ perception of the technology compared to other determinants that cause workarounds [33].  For measurements of the workarounds in the next phases of this dissertation, the technology related  items of tuse from dimensioIn additioto these dcare kit in  D3.2.1A 6-whealthcarhealthcarnotes wequestionnshadoweused to cused was[33].  Mowere assifrom 2 toA folto identifthem wheidentifiedhe instrumethe literaturns of TAM n, the most imensions.  home and ata Collecteek long obe units in me units werere taken andaire to mead, of which omplement  a validated st nurses pagned to wou 8 hours.  Tlow-up study and classither if they patterns arnt were extee [24].  Theare key detecommon pa Those workcommunity ion  servationaletro Vancou shadowed f events weresure the wor12 questionnand validatetool to identrticipated innd care duthey are to oby was done fy workarou agree with te most commnded to incl results of thrminants fortterns of woarounds infhealthcare.   study was cver area.  Aor approxim logged usinkarounds [3aires were c the data colify the prev the study foies in the samserve wounusing a quesnds.  Nurseshe identifiedon, and to ude additione measurem WDS adoprkarounds gormed desig onducted at total of 33 nately 120 hg an activit3], was alsoompleted alected from alence of wor a full shife day.  In gd care by thtionnaire to were asked workarouncomment onal dimensioent made cltion in homrouped in cln principles different hourses in 7 hours.  Duriny tracking s distributed nd returned.the observarkarounds at of 8 hours,eneral, obse author of  validate the to fill in a qd patterns, w the identifins of usefuear which oe and commassification for prototypme and comome and cog the observoftware [84]to the nurse  The questitions.  The qnd their un however, mervations vathis disserta findings ofuestionnairhether if thed patterns. lness and eaf the measurunity healths were mapping a wounmunity mmunity ations, field.  A s who were onnaire wasuestionnairderlying cauultiple nursried in lengttion.   the data anae that askedey agree the36 se of ed care.  ed d   e ses es h, lysis     D3.2.2Duricausing wconsequefrom the in researc[85].  Forwhich stubarriers rresourcesadapted fA coBased onconsequeanalyzingwere adaprocess wworkarouworkarouThe technologtechnologata Analysng the obserorkaroundsnces.  The tliterature [2h done to st the identifidy barriers elated to de, guidelinesrom the litede was assig the identifinces were c the most frpted from anorkaroundsnds involvends involvecomparison y adoption,y is not fullis vations, eve, the types oopical codes9], [33], [42udy workarcation of baor blocks thsign, flow, p, and envirorature [1], [2ned to eached instancesalculated.  Iequent workother body  and it was m using the te use of othebetween pro where mucy adopted ant logs and f workaroun used to ana].  The selecounds.  The rriers the topat cause woreople, policnment.  The], [28], [33] instance of of the coden the next starounds.  Tof research ost commochnology itr tools such cess workarh higher instnd process mfield notes wds used, anlyze the datted topical ccoding was ical codes wkarounds [1ies, procedu field notes , [62].    barriers, wos, the frequeep, themes ohe initial topthat groupednly used toself to workas pen and pounds and sances of proeans were ere recorded some of tha of those thodes were odone using ere adapted], [2], [57]. res, protocowere coded rkarounds, ncy of blockf workarouical codes f the workar study work around barraper, or verystem workcess workaused to accod to identifye workarouree aspects nes that wea data analy from three The codes ls, regulatiotopically baand their cos, workarounds were ideor analyzingounds into sarounds [34iers, and probal exchangarounds warounds suggmplish wor, barriers nd were adaptere most comsis software research grincluded, ns, technolosed on codensequences.nds, and thntified by  workarounystem and ].  System cess e. s an indicatoested that thk.  The syste37 d mon  oups gy, s   eir ds r to e m 38  and process workarounds were further divided into classifications based on common patterns of the observed activities and attributes of the workarounds. The collected quantitative data, including demographics, and workaround surveys were analyzed using descriptive statistics.  The frequencies of events were counted in the event logs.  For the quantitative analysis a statistical analysis software was used [87].  3.3 Results: Home and Community Nursing Work Characteristics  The demographic characteristics of the ethnographic study participants are shown in Appendix A.  The participants were separated into 2 groups.  The first group consisted of participants who completed the workaround survey, and the second group consisted of the remainder of participants.  All 33 participants completed the demographic survey, while only 12 participants completed and returned the workaround survey. In the fieldwork study it was observed that wound care involved assessment of 7 types and 6 stages of wounds, while wound treatment involved selecting from120 medical products, and documenting 25 characteristics of wounds (e.g. wound etiology, wound measurements, exudate, wound bed, peri-wound skin).  The work of the homecare nurse involved driving or walking to the homes of multiple patients each day, which required planning and coordinating appointments that may change unpredictably throughout the day.   The nurses used a wound documentation system to chart patient data, and communicate with other clinicians in their unit.  The wound documentation system had seven primary components, including the patient profile, the wound profile, wound assessment, wound treatment, a summary, data viewer, and patient administration.  Usage varied across different health units, and on an individual level as well.  The nurses were provided with laptops, digital cameras, and a  USB donimplemenuse the wand nursehome visentry wit  T3.3.1Figuused the and treatmvery nomspent tim Alththe laptopgle to provitation, nursound docums instead usits.  In dailyh the patienthe Nurses’re 3.2 illustrtools and artent.  Althoadic, nursese with patieFigure ough nurses with themde internet ces were expentation syed the lapto home visits profile, if n Workstatiates the typiifacts highliugh the natu found themnts.  3.2. A commo were provid.  Obstacles onnection oected to takstem in patieps as workst, the nurses eeded, thenon cal set up oghted in there of nursinselves spenn layout of homed with a lasuch as unren cell phonee this equipmnts’ homesations at thecharted at th they finish f a nurse’s d figure on a g work in hding almost e and commuptop to use liable intern networks.  ent with th.  However, ir desk whie end of thetheir submisesk observedaily basis tome and comas much timnity healthcarat patients' het connectioIn the intendem to homethis usage wle taking their workday,sion at the sd during thio manage pmunity heae behind th e nurses’ workomes, nonen, the addeed  visits, and tas not adop cameras on starting datummary. s study.  Nuatient recordlthcare waseir desk as tstation.  of them tood burden of 39 o ted  the a rses s  hey k 40  carrying the laptop, and mismatches between the electronic records system and nurses’ own workflow were noted by participants as some of the reasons they would not take the laptops to their patients’ home.  Instead, nurses used a notepad to write down items that need checking for their daily visits.  This was typically done first thing in the morning before starting patient visits.  The items were written down based on patients’ electronic and paper records.  More items were added sometimes after making updates and collecting the relevant data during the home visit, which are also written down in the notepad.  The same notepad was also used as a data source to document patients’ wounds.  To manage weekly or monthly activities some nurses used a second notepad.  Although the electronic records system had alerts and reminders, the nurses reported that there were too many alerts and reminders which discouraged nurses from using them, and caused them to employ workarounds to prevent triggering the alerts and the reminders.  This is a common barrier in the use of electronic records systems in healthcare [82].  For example, in one case the electronic records system triggered, “low healing rate” alerts and reminders for a patient with a chronic non-healable wound who has been under care for 7 years.  The nurse documenting the patient’s wound entered slight changes in the status of the patient’s wound just enough to prevent triggering the low healing rate alert.  The nurse noted that the wound clinician was aware of the condition of the patient.  In addition, the patient’s case was very complex.  Any newly assigned nurses would have to talk to the other nurses who had previously cared for the patient in order to be able to provide care for the patient.  The slight changes entered in the electronic records system did not impose a risk to the patient’s health; otherwise, the care team would get the alerts, “all the time”. The patient files, as components of the paper records, in most cases consisted of a collection of printouts from patient’s electronic records in various systems, in addition to faxes, hand 41  written notes, sticky notes, and even wound care supplies.  In practice, the patient file provided the mobility that nurses need for their work.  Using the patient file, nurses were able to have an overall look at the patient’s status in one place while being on the move.  As one of the nurses noted, “it’s all there in front of you!”.  Annotations made in the patient’s file were most commonly used to highlight details about the case while hand written and sticky notes were used to summarize patient data or to communicate asynchronously with nurses who are visiting the patient in the future, specifically to remind or bring to their attention an item that needed to be dealt with.  In some cases the reminder included a wound treatment item from the wound care plan, a wound care product that was small enough to fit within a binder. The results of this study indicated that the use of the electronic systems was especially motivated by nurses’ perception of how much the system was useful and easy to use.  For example, it was common for nurses to use an electronic records system for wound documentation purposes, even though that system neither supported nor required wound documentation.  The case note component was used in this workaround.  That was a free-text case report about the patient’s health condition.  For this reason, many nurses reviewed the most recent case notes as part of their wound management activities.  Most nurses noted that another electronic records system, which was a wound management system, was easier to use as it employed a relatively modern interface with emphasis on dropdown menus, radio buttons, and automatic propagation of data.  However, it was a common perception that the wound management system was less useful.  Using the system as a tick sheet was a workaround that had emerged as a result of this perception.  That was when nurses used the checkboxes, radio buttons, and menus but did not make much use of other fields that needed more interaction, such as the wound care plan, which required searching for items and textual entry.  Nurses commonly  used the avoid repentries thThe common much of based oninvolved record prthe ones perceived  T3.3.2To enurse is snurse hasutilizing how nursThe 8:30 AMdifferent 3.1 exhibautomatic preating entryat are as lesobservationpractice, regthe paper sy organizationusing paperintouts.  Thethat nurses p as easy to uhe Nurses’laborate on elected and  characteristhe availablees use worknurse’s work and finishelocations wit this reconopagation o of unchangs useful for s suggested ardless of tstems; in othal and indiv artifacts suc most commerceived asse, such as  Daily Routthe context odiscussed.  tics of speed resources tarounds daiday is recod approximaithin the arestruction. f data in theed data.  Homanagementhat the hybrhe fact that ter words toidual factorh as notepaon electron useful, suchthe checkboine f nurses’ wThe selected and agility o overcomely to providenstructed bately at 5:00 a covered by system, whwever, in sot of the patieid use of pahe electroni go “paperls.  The mosds, handwriic records c as case notxes and radork outside  example isin completi barriers.  In care for pased on true PM.  The nu the commuich propagame cases thnt’s woundper and elecc records syess”.  This tt common thtten and sticomponents ue free-text eio buttons foof their offic a nurse obsng the activi other wordtients. events.  Therse visited nity health ted data fromis made it e. tronic systestems were ype of mixeemes of woky notes, ansed in workntry, or oner data entrye the daily erved duringties during s, this was a nurse startea total of 4 punit.  Figure past entrieasier to avoims was a meant to repd use differrkarounds d electronicarounds wes that nurses.    patient visit  the study. shifts, as wen example od her work atients at  3.2 and tab42 s, to d lace ed  re  of a  This ll f at le  At 8the assignpatient vibeginnindistance (typicallythe time bto get a h Tablthe daily :30AM, the ed patients sits for one g of the day from the off a wound cletween endead start onFigure  3e 3.1 showstravel of thebeginning oby the clinicday orderednurses plannice, patient’inician) for  of one visit the patient d.3. Example othe frequen nurse.  Thef the nursesal coordina by time of aed and ordes preferencea joint visit, and start ofocumentatif a nurse's dailcies and one duration of’ workday, ttor.  Figure rrival at thered their as, personal pr traffic, and  the next vison. y routine to vi example of patient visihe nurse pic3.3 demonst patient’s hosigned visitseference, scparking conit, the nursesit patients wit workarounts varied depked up the rrates the rouuse or at th based on fahedule of a siderations. would go b h wound at ho ds for each lending on tecord bindete of the nue office.  Atctors such acolleague   Dependingack to the ome [88]. ocation durihe patients’43 rs of rse’s  the s,  on ffice ng  case 44  and the personal workflow of nurses.  The frequencies were identified for each location and the time was the approximate time of entry to the location.    Table  3.1. Frequencies of workarounds in a daily routine of a homecare nurse.  The first activity a nurse would do at the beginning of the workday was to collect information about the patients that were scheduled to be visited that day.  A clinical coordinator in collaboration with the nurses usually did the scheduling.  In the example here, the nurse had reviewed the records of some of the patients the day before and already was familiar with some of the patients, so she spent less time reviewing their electronic records before the start of the visits.  This could be anticipated based on the number of workarounds observed during the first office visit at 8:30 AM, comparing to next 2 office visits, at 10:45 AM and 1:00 PM, when she was there again to document the wounds of the patients she visited.  After collecting the patients’ files and preparing the wound care supplies (8:30 AM office visit), she drove to see two patients (patient 1 and then patient 2), after the second patient since she was early for the next appointment she drove back to the office to chart the visited patients (10:45 AM office visit).  The use of the electronic records systems during the office visit explained the high number of Time Workaround example Workaround frequencies Location 8:30 AM Summary sheet on the patient file to prep supplies 3 Office 9:30 AM Leaves extra supplies 2 Patient 1 9:55 AM Asks about the dressing, to see why it was used 4 Patient 2 10:45 AM Sticky note to update the care plan in the patient file 10 Office 11:30 AM Will use a product next time since she didn’t bring it 5 Patient 3 12:15 AM Writes wound measurements on travel notepad 4 Patient 4 1:00 PM Copy and pastes from one electronic records system to another 21 Office  Total 49   workaroubarriers rAfter doc3).  Afterdocumenplan.   T3.3.3Basetablet [84using thethe summBaseactivitiesConsiderin the prethe bedsitime, on event logelectronitime).  Nwere not nds at this oelated to tecumenting so visiting thetation includhe Nurses’d on the act], the logge electronic rary of the ld on the obs while the ming that the vious sectiode.  In manyuse of techns indicated tc records syurses did noable to lookffice visit.  hnology whme of the p last patient ing making Work Actiivity logs cod activities oecords systeogged activiervations, thajority of tinurses spentn, it is possi cases, nursology, rathehat nurses dstem on datat use the sys for the dataThese obserile documenatient data t(patient 4) t notes to comvities llected usinf nurses prim to documties. e majority me at the nu a substantiable that moes expressedr than providedicated the entry (31%tem at the p that they nevations suggting patienthe nurse drohe rest of thmunicate tg an activitymarily involent and manof time at thrses’ office l amount ofst of the pati that they aing care for largest port of total timoint of care eded to comested that is using the eve to the nee working do others abo tracking soved providiage patiente patients’ bwas spent o their time aent documere spending  patients.  Iion of time e) rather tha(i.e. the patiplete their t was commlectronic rext patient’s ay was dediut the patieftware runnng care for ps’ records.  Tedside was n technologyt the office,ntation was just as muchn addition, tspent on intn data accesent’s home)activities.  Inon to face cords systemhome (patiecated to patints’ future cing on a moatients, andable 3.2 shspent on car use.   as demonstnot complet, or even mhe recordederacting wits (8% of tot; therefore tstead, they 45 s.  nt ent are bile  ows e rated ed at ore  h the al hey  created acare.     Type of A  Care a  Techno  Other   B3.3.4As dand codesoftware [85].  Inifrom the common categoriend used worNctivity  ctivities 4logy use 38Total 9arriers to Nescribed eard topically, which provitially the noliterature [2workarounds were addekarounds, suTable  3.2. Log (%) Ex 60 (51) ChClApPrsuCo57 (40) EndrEnsyVi6 (9) WCaResu03 (100)  ursing Aclier, the ethnthen analyzedes topical tes taken du].  Thematics used by nud as they wech as paperged activities amples anging the deaning the suplying the hoeparing the spplies and wllecting the wtering woundainage, exudtering vital sstem. ewing past eriting notes alling patientsviewing patirgeons. tivities ographic dad thematicacoding, themring the obs content anarses duringre discovere systems, inof home and cressing (wrarface of the wney treatmenterile wound ound dressinound treatm assessmenate, and odouigns and othentries from thbout items th to schedule ents’ binders ta collectedlly.  This wae generatioervations welysis was pe their work.d.    order to doommunity heapping) of the ound with dt patch on thtreatment kitg tools such ent supplies t data such ar in the wour health indice patients’ elat need followand confirm to find faxed  in the forms done usinn, and statisre analyzedrformed, to  As the topicument and lthcare nurseswound with nisinfectants. e infected wo, which includas scissors. from the sups wound meand documentators in the eectronic reco up at later va home visit.documents f of notes weg N-Vivo, atical analysi using topic extract pattcal coding pmanage wou. ew dressing.und. es wound drply room. surements, ation systemlectronic recrds. isits.  rom physicianre transcrib data analyss capabilitieal codes adoerns and rogressed, n46 nd  essing . ords s or ed is s pted ew 47  The initial topical codes identified barriers causing the workarounds [2].  The topical codes included factors such as policies, regulations, guidelines, procedures, workflow, people, resources, environment, and technology.  These factors were also referred to as “blocks” [2].  Policies were typically at organization level, regulations were at province or health authority level, procedures were at the health unit level, and workflows were at the individual level.  The care process was carried out with consideration of all these levels.  Barriers at each level caused the nurse to perceive a need to create and use a workaround.  The guidelines were communicated with nurses through educational sessions or other material in print or in multimedia form.  Many participants found these guidelines useful however guidelines might not always be up to date with the changing requirements of the care process.  At times people involved in the care process, including colleagues, the patient, or other care givers were barriers to a nurse completing a wound care activity.  During the observations this was commonly identified in the form of missed actions by others and it was addressed by following up on the activity at a later time.  Resource limitations were also barriers to nursing activities.  In most cases this was manifested as lack or misfit of resources.   In addition, homecare nurses’ work environment can be a source of barriers to their work [89], [90], and nurses often had to provide care for patients in occupationally unsafe or hazardous environments such as if a patient has a guard dog, or the neighbourhood might have high rates of car theft.  In the province of British Columbia health authorities had educational and occupational support initiatives to minimize risk for homecare nurses.  However, on the individual level nurses still had to navigate risks.  Workarounds to these risks were used to mitigate any negative consequences for the nurse, the patient, or other individuals involved in the care process.   48  Technology barriers are commonly cited as cause of workarounds [2].  Especially as the implementation and use of HIT becomes widespread across the healthcare industry.  The observations revealed that nurses used between 2 to 3 HIT systems at any given time while they were managing patients’ cases.  These systems were targeted towards different applications, such as patient documentation, case management, and pharmacy records.  There were also communication channels that nurses used in these systems to communicate with colleagues.  It was not always possible to use these systems according to their intended application.  For example, nurses felt that a specific feature in the system does not offer them the tools they need to get their job done.  In such cases the technology became a barrier to nursing work and nurses created and used workarounds to overcome the barriers they perceived in technology.  Table 3.3 and figure 3.4 detail examples of the identified causes of workarounds used by nurses who provide care for patients with wounds.   Based on the observations made during the ethnographic study the 3 top sources of barriers to the work of nurses at home and community healthcare were technology, people, and workflow.  Figure 3.4 illustrates the identified number of instances and percentage of these barriers.  As figure 3.4 shows, the majority of the observed barriers that led to creation and use of workarounds were technology related.  This observation indicated that nurses faced obstacles related to technology with a higher frequency than other factors; however other obstacles were equally as challenging to address (e.g. obstacles related to regulations and policies).  The higher frequency of technology related obstacles suggested that reducing these obstacles will have a substantial positive impact on nurses’ daily practice as they were consistently trying to overcome these obstacles.  49  Table  3.3. Causes of workarounds.  Example Type of Barrier   Regulations A nurse says it would be nice if the systems they use were more integrated (province-level regulation limitation).  Guidelines A nurse says sometimes the system provided care guidelines mean 'jack' to her, since products change all the time (irrelevant care guidelines).  Environment A nurse says they walk to most of their patients since the neighborhood is small, except for places with safety concerns in which they drive (unsafe environment).  Policies A nurse says, as an LPN she does not see the patient as consistently as she would like (organization-level policy limitation).  Resources A nurse is using a new trial version of the wound vacuum system, says the one with the foam was like doing arts and crafts, difficult to cut properly; they had to remember and visualize deep wounds, as well not to leave bits and pieces of the foam inside the patient (resource limitation).  Procedures A nurse mentions a patient as an example for discontinuity of care, where different nurses and clinicians had provided care for the patient while a potential success option for treatment was left unexamined (unit-level practice limitation).  Workflow A nurse says the camera is not here so she does not bother otherwise she would take photos of the wound (nurse’s workflow execution obstacles).  People A nurse notes the size for one of the supplies was not mentioned in the wound care plan, so she has to guess the size from the wound photo in the system (missed action by colleagues).  Technology A nurse says there are too many notifications that the WOC receive, which easily pile up to over a 100 (inefficient technology design). LPN=Licensed Practical Nurse, WOC=Wound Ostomy Clinician   Figure  3.4. Frequencies (and percentage) of barriers causing workarounds (total 287). 1(0) 2(1) 4(1)18(6) 20(7) 21(7)52(18)63(22)106(37)02040608010012050  3.4 Results: Workarounds Classification in Home and Community Nursing The initial topical codes for identifying workarounds were based on the recommendations made in a related study [34].  The two recommended categories were system and process workarounds.  System workarounds involved using the technology to work around barriers.  Process workarounds involved using manual alternatives to work around barriers.  Table 3.4 presents the results of the topical coding which identified the frequency of the system and process workarounds in the notes taken during this study, as well examples of the identified instances of these workarounds.  Analysis of the results show that 384 instances of process workarounds (79%) and 115 instances of system workarounds (80%) were caused by barriers related to technology.  This was an indicator of less successful adoption of technology; it also represented a manifestation of rejection of the technology itself by nurses.  On the other hand, nurses’ use of technology in ways that were not intended is considered as appropriation, which is studied by the Human-Computer Interaction community in a positive light [6], [91].  In some cases, appropriation of technology is a workaround.  An example of this was when one system is used to work around obstacles caused by another system.  Both system and process workarounds were often creative problem solving strategies that provide valuable feedback for design of HIT systems.  However, considerations for patient safety and quality of care both remain issues that need to be addressed.  Especially the system workarounds that were pre-emptive, preventative, appropriative, adaptive, and parallel to the technology point to possible solutions that can be designed using these workarounds as user feedback.    51  Table  3.4. Initial categorization of the identified workarounds.  N(%) Example Workaround Type    Process 331(71) A nurse looks at a patient's file and writes in scrap note about the GP visits, referrals, medications, infection symptoms, and assessments that have changed (scrap note use). A nurse mentions for many patients they bag supplies for multiple visits in the clinic, to be used in the future (batch work). A nurse has a notepad with notes on concerns, and follow-ups, that he writes before the visit and adds other items that might come up in the patient's home (notepad use). A nurse mentions everyone has extra supplies in their car, or at the patient’s home (use of additional resources).  System 133(29) A nurse copies and pastes some case notes from another system, since WOC does not look at that system as often (copied entry). A nurse says the wound documentation system is like a tick sheet, so she writes all that needs explanation in another system (partial use). Nurses chart at the end of the work day when they are back from home visits (late entry). A nurse notes WOC is aware of the patient’s condition and changes some of the measurements slightly just so it shows improvement and it would not trigger a low healing alert to the WOC (data entry to bypass alerts). A nurse asks from her colleague at the next desk to look up a patient’s record, since her system is not working (activity delegation).  Total 464(100)  WOC=Wound Ostomy Clinician, GP=General Practitioner.   The process workarounds highlighted creative ways in which nurses solved problems with the resources that were available to them, as lack of availability of resources at the bedside was common.  Furthermore, process workarounds had a great role in lowering the mental load of nurses by providing easier ways to negotiate a solution for their work.  To identify classifications of workarounds that involved use of process means and system means the identified system and process workarounds were further analyzed to extract themes.  The extracted themes were organized into classifications.    C3.4.1In orworkarouanalysis rworkarousystem uresource appropriausing sysconsistenresourceswith adapethnograpFor examsheet so sassessinghomecarerisks homhave to cmight dowith the wchange olassificatioder to provinds to guidevealed 7 cnds and incse, appropriause.  The tertion and adtems and ret with appro to complettation.  Thehic data baple, a homehe wound n their woun nurses havecare nursehange the ca this when vound clinicf the care pln of Workade workaroue design declassificationlude: pre-emtive systemms appropraptation [6],sources to copriation.  The activities i two other tsed on nursecare nurse mot have to tad.  Howevere to use wors create andre plan of aisiting the pian later is an before corounds  nd classificisions, addits of workaroptive inform use, appropiative and ad [92].  The dmplete actie data also n ways that erms, pre-ems describingight use noke addition, even with karounds th use prevent patient at thatient and coa preventivenfirming wiations with mional themaunds based ation use, priative resouaptive wereata from thevities that thshowed thatwere not intptive and p ‘why’ theytes written aal clinical suthe use of that can have ative workae point of cnfirm with  strategy to th the wounore detail,tic analysis on commonreventive inrce use, ada derived fro fieldwork ey were not nurses werended for threventive, w create and ut the beginnpplies to the pre-emptirisks for patrounds.  As are after assthe wound cmitigate anyd clinician.  and the potewas conduc actions performation uptive systemm the work showed that intended foe using systeat use, this were derivedse specific ing of a shife patient’s hve strategiesients.  To man exampleessing the wlinician late risks assoc Parallel sysntial to useted.  This fuformed in thse, parallel  use, adaptin system  nurses werer, this was ms and as consiste from the workaroundt as a cheat ome after  sometimes anage these, a nurse migound, the nur.  Confirmiiated with thtem use is a52  rther e ive  nt s.   ht rse ng e lso 53  derived from the literature that focuses on parallel use of paper systems, such as paper patient records [10].  Table 3.5 shows the definition and examples of the workaround classifications.    Table  3.5. Classifications of workaround instances. Pre-emptive information use (n=98, 21%)  Description: when a nurse provides cues (e.g. verbal, in writing, use of specific clinical supplies, data entry), informing the need to complete certain activities and preempt or reduce occurrence of future workarounds.  It can be either of system and process workarounds. Example: A nurse says that based on the electronic records of a patient she creates a paper cheat sheet as reminder for when she visits the patient.  Preventive information use (n=31, 7%)  Description: when a nurse provides cues (e.g. verbal, in writing, use of specific care supplies, data entry) informing the occurrence of barriers to activities, their workarounds, and the need to prevent risks associated with those workarounds.  It can be either of system and process workarounds. Example: changing the care plan and confirming with the wound clinician at later times.  Appropriative system use (n=35, 8%)  Description: a system workaround when a nurse uses a system to complete activities in which the available system is not intended for completion of those activities.   Example: use of the wound photo component to document use of care supplies. Appropriative resource use (n=38, 8%)  Description: a process workaround when a nurse uses resources to complete activities that the resources are not intended for completion for those activities.   Example: a nurse uses her personal smart phone to take photos of wounds. Adaptive system use (n=65, 14%)  Description: a system workaround when a nurse uses a system for the intended activities but in unintended ways to complete those activities.   Example: a nurse says they might chart some information in multiple systems so it cannot be missed. Adaptive resource use (n=38, 8%)  Description: a process workaround when a nurse uses resources for the intended activities but in unintended ways to complete those activities.   Examples: a nurse leaves extra supplies with the patient, or in her car. Parallel system use (n=159, 34%)  Description: when a nurse uses more than one system for the same intended activities for those systems, and in the intended ways for those systems, to complete the same activities.  It is often a process workaround.   Example: a nurse looks at the sticky note on the cover of the patient’s paper record with the summary care plan on it and packs care supplies (parallel paper system).   Figure 3.5 illustrates the number of instances and the percentage of the workaround classifications.  The results presented in figure 3.5 indicated that the classifications parallel 54  system use, appropriative resource use, and adaptive system use are much more prevalent than other classifications.  Figure  3.5. Frequencies (and percentage) of workaround classifications (total 464).  Although the health authority initiated the transition from paper to electronic records almost 10 years ago, substantial amounts of work continued to be done using paper systems.  Based on the data collected this was consistent across participant groups of varying age and technology use experience.  A direct implication of mixed use of paper systems and electronic systems was the increase in workload for nurses.  During the study nurses with many years of experience noted that they spent less time on patient documentation and more time with patients when only paper records were required.  As an example, they noted that the number of patients that they are able to visit now has decreased.  This is consistent with the data collected during the study.  It was observed that all participating nurses returned to their office for 2 to 3 hours before the end of their workday to transcribe and submit their paper records to the electronic records system.  31(7) 35(8) 38(8) 38(8)65(14)98(21)159(34)02040608010012014016018055  Meanwhile, resources for nurses remained the same even as the population under their care has rapidly grown given recent initiatives to discharge patients from acute care to home and community care in shorter timeframes. In resource appropriation, the human element played a key part, since many nurses gained access to information and other resources through their colleagues.  In other words, the persons with whom nurses worked or provided care to were also a key resource for their work.  In many cases, a nurse asked a patient or their family about recent changes to the patient’s care plan.  This was a workaround to limitations that nurses faced in accessing electronic records at the bedside. The prevalence of adaptive system use suggested that with the used electronic records systems reliability of patient data was of concern.  Observations indicated that nurses did not fully follow the intended or the preferred method of use of a component in the system if they felt they could not rely on the patient data provided in an electronic records system.  For example, nurses often found themselves having to double or triple chart in order to make sure reliable data was recorded for future use of colleagues or themselves.  In another example a nurse documented a cluster of wounds in the proximity of each other as one wound when the nurse felt that was a better representation of the wounds assessment, especially since the system did not provide components for documenting such wounds.  A closely related category of the identified system workarounds was appropriative system use.  This category was observed less commonly during this study, however it is a studied category of user behaviour in human-computer interaction [6], [91], [93].  Appropriation can be seen as closely related to adaptation, with the main difference being that appropriation is to use technology for an entirely different purpose than what it is designed and intended for.  As an example, a nurse brought with her several sizes of wound wrapping supplies since the patient’s electronic record did not mention the size.  The reason for  this was telectronisupply inpatient prThe section d  V3.4.2The questionnmultiple table withThis wasclassificapoint Likthe first qtheir worin home identifiedand commphase are he lack of oc records sy the free-texogress. identified cliscusses the alidation ofindings in taire.  In thechoice quest a definitio to ensure thtion and areert scale, 1 buestion askk.  The secoand commun classificatiunity healt detailed in Table  3.6.ptions to selstem.  As ant componenassificationsvalidation rf Workaroable 3.5 and questionnaiions and a fn, a rationale participan able to relaeing stronged if the parnd questionity healthcaon.  During hcare.  In toAppendix B Descriptive stect a size fo appropriatits of the ele of workaroesults. und Classif figure 3.5 wre, each woree text quee, and examts are awarete that to thely disagree ticipant had  asked if there.  The thirthe validatiotal, 58 were.  Table 3.6 atistics of the vr that specifon method, tctronic recounds were vication ere validaterkaround clastion.  For eples of the id of the meanir own nursand 5 being used the idey believed td question an study 70  completed shows the ralidation quesic wound whe nurse dords system, alidated in ad in a follossification wach classificentified claing and theing work.  Tstrongly agrntified classhe identifiedsked if theyquestionnairand returnedesults of thetionnaire for wrapping supcumented thwhich was i second phaw up study uas assignedation in the ssification w context of ehe questionee.  For eacification of  classificati had commees were dist.  The demo validation sorkaround claply in the e size of thentended to rse.  The nexsing a  3 questionquestionnaias includedach naire was inh classificatworkaroundon was comnts about thributed to hgraphics fotudy. ssification 56  eport t s, 2 re a .   a 5-ion, s in mon e ome r this 57  Questionnaire items Mean Std. Dev. Pre-emptive information use    I have used pre-emptive information use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 4.4 0.9  To-do cues are common in home and community healthcare. 4.53 0.63  Preventative information use    I have used preventative information use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 3.48 1.27  Follow-up cues are common in home and community healthcare. 3.66 1.19  Parallel system use    I have used parallel system use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 4.21 1.07  Paper system use is common in home and community healthcare. 4.37 0.92  Appropriative system use    I have applied appropriative system use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 3.65 1.2  Appropriative system use is common in home and community healthcare. 3.82 1.09  Appropriative resource use    I have applied appropriative resource use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 3.48 1.33  Appropriative resource use is common in home and community healthcare. 3.45 1.49  Adaptive system use    I have applied adaptive system use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 3.77 1.28  Adaptive system use is common in home and community healthcare. 3.82 1.27  Adaptive resource use    I have applied adaptive resource use when problems with technology, equipment, rules/policies, people, and work processes prevent me from completing my activity. 4.23 0.91  Adaptive resource use is common in home and community healthcare. 4.26 0.92  Table 3.6 and figure 3.6 demonstrate that the findings of the validation phase were consistent with the findings of the initial fieldwork.  All identified classifications of workarounds received a score higher than 3.  The participants gave the highest scores (mean above 4) to the 58  pre-emptive information use, parallel system use, and adaptive resource use.  This is an indicator of the importance of these classifications.    Figure  3.6. Averages of workaround classification validation results.  The nurses gave the highest score to the pre-emptive information use category, or as one participant noted in the free-text portion of the survey “FYIs” (for your information).  The importance of this classification was when pre-emptive information use workarounds were used to work around limitations of the electronic records systems in support of nomadic work and communication that was required between nurses.  Literature points out that nomadic work requires accommodation by technology [5].  The prevalence of this classification indicated that the used electronic records systems did not accommodate and support nurses in their nomadic work.  In addition the observations made during this study indicated that a great number of disposable paper scraps were used to create prompts to remind nurses about activities that they 00.511.522.533.544.55Avg. MeanAvg. Std.59  were expected to carry out in the future.  In fact, it was observed that at the start of the day, before home visits, nurses prepared scrap notes that listed the activities that needed to be done for the patient visits.  Some nurses noted the scrap notes were their “cheat sheet” to remind them of activities that they needed to perform for the multiple patients that they visited during the day.  This workaround classification allowed nurses to compensate for the lack of mobility support by the electronic records system.   The primary use of the parallel system use classification of workarounds was for patient documentation and care planning; hence, consistency and accuracy were important factors in using this workaround.  During the observations, this parallel track of patient documentation became more apparent when patients with chronic conditions accumulated records from various arms of the healthcare system.  As system integration remains a complex issue in many healthcare settings, nurses face greater challenges to get their job done, and it is possible that their reliance on workarounds will increase.   Some participants noted that access to records for patients who came from other health authorities was limited, so they had to, “hunt down” faxes and printouts of patient records to get a better picture of the patient’s condition.  In other instances sheets of paper on the cover of the patients’ records binder was referred to as the, “convenience chart” and was used to chart a summary of the patients’ condition, as well to communicate and access this information.  The prevalence of parallel systems during the observations and the score given by the nurses during the validation phase highlighted the amount of additional work that nurses had to do in order to keep both the paper records and the electronic records accurate and up-to-date.  Another important factor motivating workarounds identified during this study in addition to mobility, consistency, and accuracy, was reliability.    A cowhich adsystems. proactiveFor ipatients’ supplies”measuresParticipapatient w  W3.4.3Durito measuparticipanimpact onThe technwhich a pperformausing a pusefulnesmotive toease of ummon themaptive strate Adaptive re measures tonstance, pachomes or in, as one par allowed thents also noteas not recenorkarounng the obserre nurses wots (survey r workarounology relaterson belience”; and particular sys that were  assist patiese were the e in the collgies proactisource use w work arounking extra w nurses’ carticipant statem to addresd that they lt or had not d Questionnvation phasrkarounds iespondents ds, as the hied factors wves that usinerceived easstem would measured wnts when altperception oected data vvely increasas one sucd any obstaound care ss were used d.  Some pas unreliabiliike to be prbeen updateaire Resule of the studn the acute cin appendixghest numbere identifieg a particule of use, debe free of effere the prefeering use off the technoalidated by ed reliabilityh classificatcles that nuupplies, leato ensure thrticipants aty in data onepared for sid. ts y a questionare setting  A).  Technoer of measurd as perceivar system wfined as, “thort” [38].  Trence to use technologylogy blocksthis phase o in the use ion of workarses might fving extra wat the they also noted tha patients’ etuations in wnaire that w[33] (appendlogy relateded barriers ed usefulneould enhance degree to wo items fr technology.  Two items, and the perf the study wof the resourrounds thatace in the fuound supplre not “caugt taking thelectronic or hich the caas developeix C), was g factors hadwere technoss, defined ae his or herwhich a perom the dim when poss from the dception of has the way ces and the  involved ture.   ies at the ht out withose proactivepaper recordre plan of thd and validaiven to  the biggestlogy relateds, “the degr job son believesensions of ible and the imensions oow much w60 in ut  s.  e ted  .  ee to  that f ork 61  must be altered due to those blocks.  Table 3.7 summarizes the results of the workaround questionnaire. The demographics for this phase are in Appendix A. The questionnaire measured nurses’ perception of having to engage in workarounds when facing barriers related to one of the 5 factors indicated in table 3.7.  The results of the survey in table 3.7 showed that dimensions of usefulness had more impact on workarounds compared to dimensions of ease of use.  The overall mean score of the two items used to measure usefulness was much higher than that of the two items used to measure ease of use; which indicated nurses’ perceptions were more in agreement in regards to the items measuring usefulness.    Table  3.7. Results of the workaround measurement survey.  Ease of Use Usefulness   Perception  of a block Altering process to work around a block Preference for following  procedures Motive to assist patient   Mean SD Mean SD Mean SD Mean SD Overall MeanA. Technology 3.23 1.42 2.91 1.11 4.69 0.72 4.46 0.63 3.82 B. Equipment 2.75 1.01 2.75 1.01 4.76 0.57 4.53 0.49 3.70 C. Rules/policies 3.08 1.03 2.83 0.89 4.84 0.53 4.38 0.92 3.78 D. Other people 3.00 0.96 2.92 0.99 4.38 0.48 4.46 1.00 3.69 E. Work processes 3.50 1.11 3.16 0.89 4.30 0.99 4.46 0.84 3.86 Overall Mean 3.11  2.91  4.59  4.46  3.77 N=12  The overall average scores indicated that technology and work processes had more impact on workarounds.  This is consistent with findings from other research [33], as well observations made during this study, which identify that obstacles in technology and work process are more common. 62  In comparison to the study conducted in the acute care setting [33], responses to the questionnaire for measurement of workarounds used by nurses in home and community healthcare have a higher score than score given by nurses to the same questionnaire in the acute care setting [94].  The mean score for collected data was 3.77, (Std.=0.88), higher than the mean score of 2.54 (scaled from 1-7 to 1-5, Std.=1.11) reported in other work [94].  This comparison showed that according to the perception of nurses, in home and community healthcare they had to engage in workarounds more frequently than their counterparts in acute care settings.    3.5 Opportunities in Use of Workarounds Towards User-Centred Design  The work presented in this chapter confirmed that dimensions of usefulness and ease of use can be used to measure home and community nurses’ workarounds in wound care, especially workarounds related to technology blocks.  Many of the participant nurses perceived the use of electronic records systems as not easy to use or less useful.  However, the hybrid of workarounds and the electronic records systems filled the gap between the systems’ frame of work and the nurses’ frame of work [34].  The data collected and analyzed in this chapter suggested that workarounds affected the home and community nurses’ intention to use the systems positively, as they would not be able to get their job done without the creation and use of workarounds.  The literature suggests that workarounds as creative problem solving strategies can be used to provide feedback in the design, develop, and evaluate cycle of electronic systems, to move closer to a user-centred system [28].  The study presented in this dissertation suggested system designs informed by workarounds can lead to implementations that are in better alignment with the goals and the practice frames of the nurses.  Workarounds once identified can be modeled as articulation work.  Articulation work is defined as reorganization and maintenance of work when 63  conflicting forces exist [1].  Once modeled, the workarounds patterns and attributes will become clear.  The common patterns of workarounds and their attributes can inform design for tools that the home and community healthcare nurses use.  Workarounds are valuable user feedback that are created and used by the home and community healthcare nurses, as well in other settings [1], [28], to solve problems.  Use of workarounds as feedback in the design process led to solutions inspired by the problem solving strategies of the users in a step closer to user-centred design.  The tools developed were easier to use and more useful for the home and community nurses’ work.  The next section discusses the next steps to investigate further the use of workarounds as feedback in user-centred design.  3.6 Summary Adoption of new health information technology is challenging.  However, it is predictable by measures of usefulness and ease of use; therefore, in this study these two constructs were investigated.  Consistent with evidence in the literature from similar settings, this work found that the nurses had not fully adopted the implemented system in the home and community healthcare setting.  Manifestation of this was workarounds employed by the nurses to bypass obstacles that kept them from getting their work done.  Without the use of workarounds, the nurses were not able to carry out the care activities in the planned care for management of patients with wounds.  The hybrid of workarounds enabled them to use the electronic systems despite barriers.  On the other hand, if the nurses would have relied solely on workarounds their work would get closer to manual work.  The workarounds presented an opportunity to use the workarounds as feedback in the design process.  Identification and mapping of workarounds to design principles is possible.  Use of workarounds in user-centred methodologies addresses  design gaadoption highlightWhaWha  C3.6.1In thadaptatio[43], andprocess fps in key diare achievabed items in tt is already  Work Work Adopt In thedocumt this dissert Collechome  Althou Dimenhome  Some whichonsideratie next chaptn of the Wo the most coor a wound mensions ofle using suhis chapter.known abouarounds are arounds canion of techn case of homentation syation is addted and anaand commugh workarosions of useand commuworkaround can inform ons for Nexer the identirk Situationmmon workdocumentati usefulness ch user-cent t the topic:common in  compromisology can be and commstems have sing to the bolyzed data snity healthcunds can cafulness andnity healthcs fill in desinew design t Chapterfied categor Model (i.e.,around situon tool.  Thand ease of red methodoacute care se care and ce challenginunity healthhown promdy of knowuggested thaare settingsuse risks the ease of useare settings.gn gaps to adecisions toies of worka the Workarations were e most commuse.  Higherlogies.  Theettings. ause patient g in the precare for patise. ledge: t workarounthan in the ay enabled c can be used ddress usefuwards user-rounds weround Situatidentified toon workar success rat following asafety riskssence of woients with wds were mocute care seare for patie to measurelness and ecentred syste analyzed uion Model ( use in a useound situaties in technore the . rkarounds. ounds patiere commontting. nts.  workaroundase of use, ems. sing an WaSM)) [1]r-centred deons were 64 logy nt  in s in , sign 65  mapped to existing design principles from the literature.  Participants were recruited from home and community healthcare nurses who visit patients with wounds at home for exploratory and experimental prototyping sessions [95].  The prototyping sessions made use of mock up models with pen and paper to discuss the initial ideas, and then initial ideas that received positive feedback were used to create the interfaces and the interactions.  The initial fieldwork, the results of the workaround measurements, and mappings informed the design features.  For example, during the initial fieldwork it was found that there were number of primary data elements that were documented and used by the nurses, such as the wound care plan items, the wound photos, and the progress note details.  These data elements were in use with workarounds as well and they informed the designs discussed in this dissertation.  Some of the patterns of the workarounds occurred more commonly and they related to the dimensions of TAM, such as using a notepad to write down items that need checking during the visit.  This related to ease of remembering tasks, within the ease of use construct.  After the completion of the design and development of the prototypes, the prototypes were evaluated in experiments.  Similar to the initial fieldwork the next phases used a mix of qualitative and quantitative data that was collected and analyzed.  Effective and successful adoption of the developed technology needed validation by measures relevant to the clinical outcomes.  The study observations indicated that the wound care plan items were important data elements used by the nurses and they affected clinical outcomes.  This is consistent with findings of other studies done in wound care and nursing settings [80], [81].  Therefore, changes made by the nurses to the care plan were of importance to the clinical outcome, as well to technology adoption, as noted by others [35].  The workaround measurements clarify the degree of technology adoption.  Chapte 4.1 OvIn thfeedbackprincipleanalyzedelementstasks, agediscretionOnceabove weease of ucorresponindividuastructure  4.2 MeThe attributesr 4: Workerview of Sis chapter ex for design [s from the li using an ad of the workndas, actors, accountab the most cre identifiedse from the dence of dalized care pand status cthods objectives o of workaroIdentifanclassifaround SFigure  4.1tudy isting resea6], [7], [9], terature.  Thapted modelarounds for , resources,ility, and thommon elem they were literature. Thta and realitlanning. Forhanges, conf the work punds, and keication d ication Mituation M. Dissertation rch on using[25], [33], we workarou from the liteach catego temporal ane outcome.  ents of wormapped to ce design pry, charting c ease of usetrol of curreresented in y dimensionodeling anmappingodel andflow diagram ( the workaras extendednds that wererature [1], ry.  These ed special or karounds inorrespondininciples for ompletenes the mappednt status, anthis chapter s of usefulnCreatmapdesd  Mappingmodeling andounds creat by mappine identified [7] to determlements inclganization,  each of theg design priusefulness ws and comp design prind Simplifiedwas to identess and easion of ped igns  of Work mapping). ed by technog of workaroin the previine the mouded the modivision of l seven categnciples for uere to imprrehensiveneciples were  data entry.ify the mose of use thatRefining anevaluation odesigns arounds logy users aunds to desous phase wst common st common abour, authoories listed sefulness anove ss, and to improve   t common  can be usedd f 66  s ign ere rity, d view  in  mapping sequentiaUsinanalysis, approach Exis[28], [34The workmodel froelementsattributesworkarouAfter theof workaroul mixed metg both qualiwith the goa used for theFigting research] was extendarounds prem the litera of the work and the chands each in most commnds to desighods approatative and ql to improv mapping oure  4.2. The a on using thed here by msented in thture (the woarounds for racteristics stance of theon elementsn principlesch was useduantitative me the validitf workarounpproach to mae workarouapping woe previous crk situation each categoof the worka workaroun of workaro for wound  to collect aethods alloy of results. ds.   p design princnds created rkarounds tohapter weremodel) [1], ry.  The adarounds [1]. ds was articunds in eachdocumentatnd analyze wed triangu Figure 4.2 iples that suppby end-user design prin analyzed us[7] to determptation was  To apply thulated using of the seveion applicatdata [96].   lating sourcillustrates thort workarouns as feedbacciples froming an adapine the moused to idenis model to the elemenn categoriesions.  A es of data ane steps to th ds. k for design the literaturted conceptust common tify the  the instancets of the mo of workaro67 d e  [7], e.  al s of del.  unds  were idenuse from   D4.2.1Therworkarou  D4.2.2The analyze twas initianegotiateIn thworkarouwhich incassess pasystem oinstancesThe the key ddocumenthe nursinthe userstified, they the literaturata Collecte was no prnds that weata Analyswork situatihe workaroully introduc in order to is work, thends.  Theseluded attribtient or docur sticky note of workaroTechnologyimensions otation applicg setting [2.  The resultwere mappee.   ion imary data cre reported iis on model [1nds and ideed to indicaget their job work situat attributes wutes such asment wouns for the resunds were c Acceptancef usefulnessation.  TAM4], [33], [35s of the workd to correspollection forn the previo], was adaptntify their mte the dimen done [1], thion model were added b verbal exchd for the ageources dimelassified.  Model (TA and ease of is frequen], and it is saround queonding desi this phase us chapter wed into the wost commonsions of wois is called aas adapted based on worange or wrinda dimensnsion.  UsinM) was use use to adoptly used to shown to prestionnaire togn principleof the disserere the focuorkaround  elements. rk that the wrticulation wy includingkarounds obting on papeion, and theg these attrid to measurtion of the dtudy technodict the inte measure ws for usefulntation.  Instes of analyssituation mo The work sorkers haveork [43].    attributes oserved in thr for the tas wound docbutes, the ide workarounesigned wology adoptiontion to useorkarounds ess and easad, the is.  del (WaSMituation mod to somehowf the e fieldworkk dimensionumentation entified ds and idenund n, especiall technology discussed in68 e of ) to el  , , tify y in by  the 69  previous chapter also identified usefulness and ease of use as factors that affected the homecare nurses’ intention to use the wound documentations systems.  For these reasons the design principles recommended in the literature for usefulness and ease of use were selected for the workaround mapping process. Once the most common attributes of the workarounds and the category they belong to was known, the design principle, the TAM dimension (e.g. completeness), and the TAM construct (e.g. usefulness) that they can be mapped to were identified.  For example, use of notepad as a cheat sheet is a pre-emptive information use workaround created by the nurses to easily view patient status changes without much memorization and that was mapped to, support for a structured view of patients' information where status information, especially changes in status, are visible.   The design principle used from the literature for this instance is one of the 10 Nielsen’s usability heuristics [97].  That is, easily view information without memorization and visibility of system status where status information is visible especially changes [97].  This pattern of workaround is to lower the mental effort, which is a dimension of the ease of use construct in TAM.  The mapped design principle informed a flowsheet feature that the nurses can use to summarize patient information and indicate changes easily.    Once the mapped design principles were identified, a set of design features were created as mock ups with use of a wire framing software (Axure) [52].  These design features were used in iterations of prototyping to get user feedback [46].  To ensure interrater reliability at least 2 researchers checked the summary results of the initial coding and theme analysis during the application of the WaSM on the collected data in the initial fieldwork.  In addition to cross 70  checking among different individuals the qualitative data was triangulated with quantitative data, such as questionnaires.  4.3 Results: Workaround Situation Modeling of Homecare Nurses’ Workarounds In order to elicit design requirements from workarounds we needed to establish a path that can be applied consistently.  After identifying workaround instances, the first step was to use a conceptual model to identify the attributes of the workarounds.  Such a model was applied to each instance of a workaround.  Once all the attributes of all the workaround instances were known, commonalities among the attributes clarified the patterns in workarounds.   These attributes indicated, for example, what actions were taking place across instances of workarounds, who carried out those actions, what were the resources used, what was the purpose, what was the time and location, what was the outcome, and other attributes.  These attributes belong to 11 dimensions that make up WaSM.  These dimensions originally were defined in the work situation model, and in WaSM they were given attributes different than the original model.  In addition, the definitions of the 11 dimensions were changed to appropriate definitions that were informed by the data collected during the fieldwork discussed in the previous chapter. These dimensions were used to describe each instance of a workaround.  Originally these dimensions described work situations in the literature [1], [7], however these dimensions were adapted to account for attributes that characterize workarounds identified in the previous chapter.  While the definitions of these dimensions remained the same in WaSM, their attributes were informed by the workarounds we identified in the previous chapter.  The dimensions of WaSM included, workaround tasks, agendas, actors, resources, temporal organization, place and spatial 71  organization, division of labour, discretion, accountability, and actual products or outcomes.  Inclusion of these dimensions allowed for characterization of the technological, social, organizational, and personal aspects of workarounds.  In this section each dimension is discussed.  The adaptation from the work situation model to WaSM is shown in table 4.1.   Table  4.1. Adaptation of work situation model to the workaround situation model. Dimension Definition WSM Attribute Examples WaSM Attribute Examples Workaround task/action Action being performed. Computing use Writing on paper, Verbal exchange, Data entry Agenda Program, goal, intended outcome of the action. Report generation, scheduling Execute care plan, Set care plan, Document wound Actors Persons or groups performing the action. Programmer, sales administrator  Homecare nurse, Wound clinician, Clinical coordinator Resources Supporting resources with which the action is performed. Information, time, support staff Record printouts, Records systems, Patient’s family Temporal organization of workarounds Description of workaround actions over time. Cyclical, persistent Concurrent to barriers, Before barriers, After barriers Place and spatial organization of workarounds Description of the workaround actions over physical locations. Workplace Clinic, Home-bedside, Same place as the barriers Division of labor Expectation from an actor to carry out an action.   Actor expected to do the action, actor not expected to do the action Part of the actors job, Not part of the actors job Power Access to key actors, and knowledge of procedures (authority). -- Part of actor’s authority, Not part of actors authority Discretion Ability to create and exploit flexibility. -- Actors can change task execution, Actors cannot change task execution Accountability The degree of responsibility for the outcomes. -- Actors required to justify tasks, Actors are not required to justify tasks Outcomes The actual products Reports, schedules Patient record, Care schedule, Executed care plan      W4.3.1The underlyinaction sinnot presethat was other settThe similar acof actiondata entryIdentifyinworkarouactions cliteraturenoted as “A nmeds, infThe visit patieto do or tfuture visin consecorkarounwork situatig action beice nurses arnt [63].  Thecarried out tings as wellfinal list of ations.  Quas included: v, inspectingg and consond instancearried out du [63].  The mfollows.   urse looks aection sympscrap note wnts at homehey need to its.  In effecutive home d Actions on model inng performee knowledg majority ofo overcome.   ctions was litative data erbal excha wound, prelidating thes such as actring creatioost commot the patientoms, and asas used like they can usfollow up ot, this also pvisits whichcluded the dd.  For the We workers a the identifi a barrier.  Tidentified afanalysis sofnge, using mparing supp list of theseors, agendan and use ofn referred at’s file and wsessments t memory one the scrap nn.  This worrevented a c delays treatimension ofaSM, the nd the term ed workarouhese actionster accumultware (N-Viedical supplies, taking  actions allo, resources,  workarounction is writrites in a schat have cha paper.  Thiote as remikaround preascading efment.  This work task wdimension wtask impliesnd instance were commatively addivo) was usely, writing ophotos, placwed to idenand outcomds have beening on paperrap note thenged.”  s means thanders to higvented postpfect in whic was an imphich is defas renamed that clinicas had one pron tasks thng new actiod for this pun paper, reving record, tify other ates.  Some of documente.  An observ doctor visit when the hhlight itemsoning care h a care actiortant issue ined as the  to workarol judgement imary actioat can happens and merrpose.  The iewing recoand searchintributes of t the commod in the ed examplets, referralsomecare nu that they neactivities tovity was miin homecare72 und is n n in ging list rd, g.  he n  is , rses ed  ssed  73  since typically a different nurse visited the patient each time.  If the visiting nurse was unable to complete a care activity s/he would have to first recognize that the activity was missed, and then communicate with the next nurse who will visit the patient that the missed care activity needs to be completed.  A common form of communication that was observed during the study was verbal exchange.  At times this even surpassed the amount of communication nurses had using the communication channels provided in the electronic records systems, or the paper systems such as paper records and hand written notes.  An example of this is the following workaround. “A nurse notes sometimes the care plan in the patient’s file is not current, people forget to print, and hence she talks to the nurse who visited the patient last and “gets a load [of patient’s information] off her”.”   While much effort went into keeping the patients’ information up to date, it was not always clear that the records reflect the most recent status of the patient.  Verbal communication allowed for the most recent information to be shared.  Other actions that were identified during the analysis of the workaround instances are shown in table 4.2.   Table 4.2 shows that writing on paper, reviewing records, and data entry were the most common tasks carried out during workarounds.  This indicated that many workarounds involved collection, use, and communication of patients’ information, which the electronic records systems were intended for.  However, the paper-based systems were being used in parallel to the electronic records systems.  The parallel use of these systems allowed for patient assessment, documentation, planning, and execution of the treatment plan, as well communication of these items, with more consistency than what was perceived to be possible with any one of the electronic systems in use by nurses.  The next section discusses the extent of which workarounds were created and used to achieve these goals.  Type of AWrite on p29%) Review re(n=120, 2Data entry21%) Gather su9%) Verbal  exchange(n=45, 9%File record%) Inspect w2%) Search weTake photUse medi(n=4, 1%)WOC  W4.3.2Anogoals, or suggests wound asschedule4.3 showTabction aper (n=142cords 5%)  (n=102, pplies (n=46,  ) s (n=11, 2 ound (n=8, b (n=6, 1%) os (n=5, 1%)cal supply   =wound ostomorkarounther dimensiintended ou[98], and thsessment, w, executing ts the identifle  4.2. ActionExample , A nurse wrbefore the home. A nurse rewrites cheaA nurse chrecords sy A nurse prreminder foA nurse lealook at the patient’s reThe nurse the nurse tA nurse inswounds shwounds. A nurse Goshe doesn A nurse sahave not taA nurse brthat shouldvisit. y clinician d Agendas on of WaSMtcomes of this study suppound documhe care planied workaros identified invites in a notepatient visit aads the patiet sheet for harts some of stems so it wepares supplr the nurse vves a phonepatient’s recocord and thethat visited thhat is visitingpects a patiee thinks of thogles nursin't use the proys she will taken photos dings a specia have been d was the we workarouorted, that tentation, se, and commund agendasolving the insbook about cnd adds mornts electronicer scheduled the importanould not be my items and pisiting the pa message forrd earlier, si WOC referrae patient las the patient nnt’s wound aem as two clug lingo termsper lingo for tke photos of uring previoul removal equone in the laorkaround and instance.he primary tting a careunicating w in this studtances of workoncerns, ande items that m records at thvisits on a dat patient dataissed. laces them intient next.  the nurse wnce the visitinl until the dat puts the patext. nd says sincsters of wou, and jokes shose terms.the wound ins visits. ipment to rest visit.  This gendas.  Th  In the caseidentified ag plan (treatmith colleaguy. arounds (n=4 follow-ups.  ight come ue beginning y.  in multiple p the patient’sho is visiting g nurse will y of the visit. ient's file prine the leg is conds instead ohe would be  the current vmove stapleswas missed is referred to of wound cendas for went plan), ses about the89). S/he writes itp in the patieof the day anlaces and mu file as a the patient nenot look at thtout in the shvered with f multiple smfrowned uponisit since oth from a wounin the last pat the programare, literaturound care aetting a carese items.  Ta74 ems nt's d ltiple xt, to e elf of aller  if ers d ient s, e re:  ble 75  In table 4.3 it can be seen that documenting the wound, which involved both electronic records systems and paper records systems, along with executing the care plan make up 65% of the workaround agendas.  This indicates nurses found themselves having to create and use workarounds more often to collect patient information and provide treatment.  These were core goals of the nurses, while other goals such as setting the care plan sometimes was done in collaboration with the wound ostomy clinician (WOC).   Based on the observations nurses created and used these workarounds as ingenious solutions to solve problems that put barriers to practice what they are best at, which were their core competencies in managing and treating chronic or difficult to heal wounds.  Other groups of actors were also involved in workarounds.  These actors are discussed in the next section.  Table  4.3. Agendas of the identified workaround instances (n=472). Type of Agenda Example Document wound (n=159, 34%) A nurse notes that she forgot to take notes when visiting the patient, but she might remember the details in later days, then she will make a late entry. Execute care plan (n=146, 31%) A nurse asks from the home support worker how to carry out the treatment procedures for the patient since they have had joint visits with nurses who visited the patient in the past. Set care schedule (n=59, 12%) A nurse updates the visit schedule in the sheet on the cover of the patient's file, as well in the scheduling board sheets. Assess patient (n=42, 9%) A nurse asks from the patient who visited them last for wound treatment, even though the nurse has seen records of the last visit in the case notes of the electronic records system. Prompt colleagues (n=35, 7%) A nurse says nurses know the double bagged supply item means the supply may or may not be needed for treatment.  If not used the item can be returned to the supply room, otherwise it is considered contaminated. Assess supplies (n=18, 4%) A nurse assesses the extra supplies the patient needs and writes them down to bring in the next visit. Set care plan (n=13, 3%) In a joint visit with the WOC, the WOC cannot add the product she used to the care plan in the electronic records, so the accompanying nurse has to do that instead. WOC=wound ostomy clinician    W4.3.3In Wskills, peduring thhome andthe primaprocess fcoordinatduring thdecision workaroutraining rthat invoTypically  W4.3.4 Rand comm[99].  Thlimited recommon occupatiocare [100orkarounaSM actorsrsonal agendis study the  communityry care provor the treatmors (n=1, 1%ose visits Wmaking and nds, especiaelated to wolve clinical c, units that orkarounesources arunity healtis is a reasonsources for in workaroun satisfactio].  Howeverd Actors  were defineas, meaninghome and c healthcareiders.  In adent of patie).  WOCs OCs faced btreatment olly since Wund diagnooordinatorscover largerd Resourcee one of thehcare setting that causedthe best fit tnds, and it cn for nurse, using resod as personss, sentimenommunity n units (n=47dition, theynts.  Other asometimes aarriers simif patients it wOCs were thsis and treat is due to th geographicas  most impors where res many nurso get their jan have negs, as well it curces to sup or groups pts, attentionurses were t0, 98%).  Th were alwayctors involvccompaniedlar to nursesas commonemselves nment.  The le fact that nol areas havetant attributources are ses to find thob done [63ative effectan cause deport workaroerforming t.  Based on the primary ais was expes involved ied were WO nurses on p.  As they w that they aurses who hower numbet all units h a clinical ces of workarhrinking anemselves ha].  This types on patient creased patiunds is neche workarouhe observatctors of wocted since nn the decisioCs (n=6, 1atients’ homere activelylso would cave receivedr of workarave a clinicoordinator. ounds.  Espd demands aving to nego of resourcesatisfactionent safety aessary in rend task, haions made rkarounds inurses were an making %), and cline visits and involved inreate and us especial ound instancal coordinat  ecially in hore increasintiate betwee negotiation, job and nd quality osource limite76 ving:  lso ical   the e es or.  me g n  is f d 77  healthcare environments such as home and community healthcare [99].  Table 4.4 details the identified resources in this study that were most common to support workarounds.  Most of the resources used during workarounds were patient data and computational resources.  This is shown in table 4.4 and it is also noted in the literature [63].  Table  4.4. Resources used to support workarounds (n=681). Resources Example Patient's paper file (n=113, 17%) A nurse leaves the most relevant items in front of the patient's file so it will not be missed by nurses who might not look at the electronic resounds. Wound management records system (n=106, 16%) A nurse copies and pastes patient information from the wound management records system to the case notes component in the community health records system. Community health records system  (n=88, 13%) A nurse does not have time and goes through only a few (2-3) case notes that are most recent.  Even if she did she might not know what to look for in the older notes. Sticky note  (n=76, 11%) It is the first time a nurse is seeing a patient and she looks at the sticky note on the cover of the file with the summary care plan on it, in order to get a quick picture of the patient’s status. Wound treatment supply  (n=68, 10%) A nurse uses wound dressing from the extra supplies at the patient’s home, since she needed a smaller dressing the she did not have. Notepad  (n=49, 7%) A nurse writes in his notebook records of a blister for a week, to document that later in the electronic records.    These results showed the amount of effort nurses had to put into keeping patient data reliable, consistent, and accurate.  For example, when a nurse copied and pasted from the wound management records system to the community health records system she was trying to make sure that the nurses visiting the patient next have access to reliable, consistent, and accurate patient data.   During the observations, some nurses did not review the patient’s records in all of the systems in use, whether electronic or paper.  Workarounds such as copying and pasting patient information between electronic records system, and the use of the paper records systems such as  the patienparticipan Twork thagreater amwas likeltheir worsystems, work for Othepatients' nurses arthat takinpatients’ treatmentpatients'   T4.3.5 Tafter the show to whappen, oanalyzedt paper filet noted.   he prevalent nurses did,ount of timy that the elk.  Lack of swhich benefnurses.   r types of revisits, and lee not “caugg these proapaper or ele caused by uhome. emporal ao overcometime they what extent wr to limit th during this s, were usedce of paper  which posse on patienectronic recoupport left nited complesource use daving extra ht out withoctive steps actronic reconreliable innd Spatial O barriers to ere facing a orkaroundeir effects astudy the ma in such waysystems shoibly was thet documentards systemsurses no chmenting paturing worksupplies at tut supplies”llowed themrds; otherwiformation inrganizatiowork, nursebarrier.  This were used fter they havjority of wo to make suwn in Table greatest stration than the in use did noice but to uient data buarounds inclhe patients’, as another  to compense there wou records ann of Workas used works was an imto prevent be happenedrkarounds wre “nothing 4.4 indicatein on them,y spent timot accommse hybrid (it it also causuded bringi homes or inparticipant sate for the ld have beed lack of appround arounds befportant attriarriers, deal.  Based on ere either c is missed”,s the amoun as they spee on patientodate and su.e. paper+coed repeatedng extra sup nurses’ carstated.  Partiunreliabilityn delays in ropriate resore, concurrbute to iden with them wthe data colloncurrent to as one t of additiont equal or  treatment.  pport nursemputer)  and redundplies befores to ensure tcipants note in data on the patient ources in thently with, otify since it hen they ected and  barriers 78 nal It s in ant  the hat d e r can 79  (n=273, 59%), or prior to barriers (n=139, 30%).  For example, when nurses faced barriers in collecting patient information at the bedside they used paper systems such as notebooks or scrap notes to write down the data that they needed to collect at the bedside.  This was concurrent to barriers such as lack of support for patient bedside documentation by the electronic records systems.  An example of using workarounds before occurrence of barriers was when nurses left extra supplies in their cars or at patients’ homes.  Using this workaround they prevented a situation where lack of wound care supplies became a barrier to their work.  In this workaround the task was carried out before the barrier had occurred.  It was also possible that the workaround task be carried out after the barrier had occurred (n=54, 12%).  As an example, when a nurse did not have access to the patient’s doctor to get a new order for change of a catheter type, then the nurse changed the catheter type and as a workaround to get a fax order from the doctor later. Another attribute of workarounds was their spatial organization.  That was, the spatial order and the location of the workaround’s task relevant to the place of the barrier that caused the workaround.  Based on the observations made during this study the spatial organization was grouped into two categories, workarounds used at the same place as the barrier (n=318, 69%), and workarounds used at a different place than the barrier (n=146, 31%).  Similar to the temporal organization of workarounds the spatial organization of workarounds allowed to identify how workarounds were used to prevent or bypass a barrier to the nurses’ work.  The spatial organization of a workaround showed where workarounds were used to bypass or prevent a barrier to the nurses’ work, as well whether or not the workarounds were helping nurses in the place where barriers had occurred.  As an example, the data collected during the observations indicated that the majority of workarounds were used in the community health unit (n=332, 72%), while other workarounds were used in the patients’ homes (n=125, 27%), and the nurses’  cars (n=7health unprimary tthe workmobility   A4.3.6Reseaccountahealthcaraccountawhat expexternal fadministrregards towere foujustify thactions wsystems. doctor’s To cattributeswas divis, 2%).  Thesit were usedask that the arounds werin their worccountabilarch shows bility in heae settings, ability to arriense [101]. actor to theation, but ac workarounnd to either e performedhen they co In contrast,orders for a larify the ex should be cion of laboue results su to overcomnurses carrie used in thek.   ity, Divisionthat accountlthcare can lccountabilitve at conclu In this study care procescountabilityds, this wasbe required  task (n=405pied and pa nurses werepatient’s tretent in whiconsidered ar, which waggested thate barriers reed out in the patients’ h of Labourability has teads to issuy may not bsions about , the collecs that was en was a dyna very importo justify the, 87%).  Fosted, or docu required toatment. h these works well.  An as the assign the majoritylated to patiir units.  Atomes and nu, Power, ano be part of es [101].  Fue very clear,what care thted data sugforced by emic dimenstant.  Actors performed r example, nmented pati justify whearounds cottribute thated ownershi of workaroent docume the same timrses’ cars, wd Discretiothe care prorthermore,  as care provey should pgest that accxternal partiion of the n who createtask (n=59, urses were ents’ informn they chanmpromised  was closelyp of the worunds used intation, since a considehich allown cess and lacin home andiders have rovide, to wountability es, such as turses’ workd and used w13%) or nonot requiredation in muged an item nurses’ resp related to akaround tasn the commue this was thrable portioed nurses k of  communityto negotiatehom, and atwas not justhe health .  Especiallyorkaroundst required to to justify thltiple recordprescribed bonsibilities ccountabilitk.  In other 80 nity e n of     an  in   eir s y other y 81  words, whether or not the performed action was for the actor’s job.  The data collected and analyzed during this study indicated that the majority of the performed tasks were for the actors’ job (n=344, 74%), such as when a nurse made a late entry in the patient’s electronic record after s/he remembered new details from the patient’s home visit.  Other actions in the observed workarounds were performed for another person’s job (n=120, 26%), such as when nurses left extra supplies at the patient’s home so nurses who are visiting next would have the wound care supplies that they might need. Another attribute related to the nurses’ responsibility was the assigned authority to them in performing the workaround task.  In this study, findings showed that the vast majority of the observed workarounds were in situations where the actor had the authority to perform the task (n=434, 93.53%).  For instance, using web resources to get information about a patient’s condition when the electronic records system did not provide enough information.  Only a few tasks were performed where the actors did not have the authority (n=30, 6.47%), such as using a personal phone to take wound photos and emailing them to the patient’s doctor.  This did not mean necessarily that the nurse compromised patient’s information, in this case the photos were anonymized and any relevant information was discussed with the doctor in a phone call. Given the above attributes related to responsibility of nurses in providing care to patients, a remaining attribute identified the amount of flexibility the nurses have when they used workarounds.  This is defined as the latitude of action and control over how the workaround actors perform the workaround task, or more simply said ‘discretion’.  In this regard, either nurses could change the way the task was performed, or they could not change the way the task was performed.  The analysis of the data showed that in the majority of cases actors of workarounds in home and community healthcare could have changed the way the workaround  task was and did nsystem torecords.  usually inexample which thedoctor. The workaroupatient sawere not authoritythe limits  A4.3.7Whiidentifiedagendas. table 4.5,care planIdenworkaroucarried out (ot need to ju communicThe remainvolved lackof this was i nurse will attributes dinds used byfety.  In facrequired to j to use the w of their practual Prodle long term a set of dir Table 4.5 s where the m.   tifying thesends were acn=369, 80%stify using ate wound ming tasks tha of resourcenforming thbe accountascussed in th nurses wert, these resuustify their orkaround, ctice.  This ucts or Out outcomes oect outcomehows these ajority of w outcomes chieved.  Pat).  For examthe case notanagement t cannot be s and powere patient’s dble if the chis section he within scolts show thapractice, theand that theallowed ovecomes f nurses’ acts of the woroutcomes.  Torkaround larified the ient recordsple, a nursee componenitems, as opcarried out d, or could hoctor aboutange was noighlight the pe of practict nurse prefe workaroun workaroundrcoming barions were bkaround taskhe results inagendas invextent in wh in both form had the autt in the composed to usiifferently thave caused l changes to t communicfact that thee.  This warred to use d was for ths were subjriers to the eyond the sc that also re table 4.3 rolved patienich the intens of electrohority and tmunity healng the patiean they weriability for tthe prescribated with th majority ofs a key factoworkaroundeir job, theyect to their dnurses’ workope of this flect the woeflect the agt documentaded goals fnic and paphe resourcesthcare recornt’s paper e (n=95, 20he nurse.  Aed order, in e patient’s  the r that impacs when they had the iscretion w. study, this srkaround endas showtion and theor the er comprise82 , ds %) n ted  ithin tudy n in  d a 83  substantial portion of the workarounds outcomes.  The workarounds here were intended to allow nurses to collect accurate, reliable, and consistent data.  When contextualized, these outcomes with the other attributes of WaSM discussed in previous sections provided the data needed to distinguish patterns of workarounds.  Table  4.5. The identified outcomes of the workaround instances (n=469).  Type of Outcome Example  Patient record  (n=120, 26%) For a patient’s chart, a nurse asks another nurse how to describe the patient’s wound around the kidney as she is unfamiliar with it.  Executed care plan (n=105, 22%) A nurse asks the home support worker about treatment procedures since in the past they have had many joint visits to the patient’s home with other nurses.  Necessary supplies (n=60, 13%) A nurse asks a patient if they used a specific product and have any extra left.  The patient says they have extra supplies.  Patient information (n=58, 12%) A nurse says sometimes the care plan in patient’s file is not current, as people might forget to print, so she talks to the nurse who visited the patient before and 'gets a load [of information] off her'.  Care schedule  (n=58, 12%) A patient’s paper file has several calendars for different months, used as a history and a scheduling calendar.  Wound assessment items (n=47, 10%) A nurse writes wound dimensions and patient symptoms in a notepad at the car in order to enter them in the electronic records system later.  Care plan  (n=21, 5%) A nurse writes on a scrap note new updates to the care plan of a patient and puts that on the cover of the patient’s paper file.  These data provided answers to questions such as: For each instance of workarounds what resources are used?; What tasks are carried out, where, when, in which order, and by whom are they carried out?; and What are the goals and what are outcomes?.  Those answers informed what solutions were the best fit that can make use of workarounds as feedback.  The WaSM attributes were used to create a mapping from the most common patterns of workaround situations to design recommendation for user-centered solutions.  This is discussed in the next section.  84  4.4 Results: Mapping Workarounds to Design Principles The most common workaround patterns were the best candidates to be used for mapping to design recommendation for new solutions.  The attributes identified using WaSM contextualized workarounds and their patterns.  This refined workaround patterns to “workaround situations”.   Workaround situations had the attributes defined in WaSM.  Therefore, for the purposes of the mapping, workaround situations were defined as workaround patterns that had tasks, agendas, actors, resources, temporal and spatial organization, accountability, division of labour, power, discretion, and outcomes assigned to them.  Using these attributes a mapping process was created from the most common workaround situations to design recommendations within the literature.  The steps to the mapping process are discussed in this section. As described earlier, during the initial fieldwork a survey was given to participants to identify the causes and prevalence of workarounds, as well how they relate to usefulness and ease of use [33].  The survey was previously validated in a the acute care setting [33].  The results confirmed that workarounds were prevalent in home and community healthcare, barriers caused by factors such as technology, policies, work process, equipment, and other people led nurses to create and use workarounds, and most importantly aspects of usefulness and ease of use from TAM impacted the perceived need to create and use workarounds.   The observations also confirmed that homecare nurses commonly created and used workarounds when they perceived a feature in the electronic records systems as less useful or less easy to use.  This suggests that workarounds were created and used to overcome barriers that impacted usefulness and ease of use.  There exists a large amount of work in the literature on design recommendations that address issues related to usefulness and ease of use.  The key role that workarounds had was to provide feedback about which aspects of usefulness and ease of use 85  should be used for the mapping to design recommendations.  During the mapping process first the attributes of the most common workaround situations were used to determine which construct of TAM was most relevant to those situations (i.e. usefulness or ease of use).  The workaround agendas and outcomes were the main determinants in this step.  For example, documenting patient assessment and care was a common agenda for the parallel systems workaround.  Given that the majority of workarounds used in the community health units were parallel system uses, and that the nurses feedback to the workaround measurement questionnaire gave higher scores to usefulness factors, this was considered to overcome barriers related to usefulness of the electronic records systems used at the units.   Once the related construct was determined, whether if usefulness or ease of use, then the relevant aspect of that construct was identified (e.g. completeness for usefulness or simplified data entry for ease of use).  Other attributes of WaSM such as tasks, actors, and resources were used in this step.  For instance, the parallel system use workaround situations mostly were used to collect patient information that was accurate and effective.  Accuracy and effectiveness were aspects of usefulness.  The literature provided a pool of design recommendations for these aspects.  The WaSM attributes were used to further contextualize the matched design recommendations.  This process was repeated for all of the identified most common workaround situations, detailed in table 3.5.  This process resulted in a set of 9 design recommendations for usefulness and ease of use.  These design recommendations are discussed in the next section.  After identifying the related aspects, or dimensions, of usefulness that were relevant to the most common workaround situations, a pool of usefulness design principles were extracted from the literature (appendix H).  The identified relevant aspects of usefulness were accuracy, effectiveness, control over work, and productivity.  The two main workaround categories that 86  were used for the mapping to design principles for usefulness were parallel system use and adaptive resource use.  These workaround categories were identified as most common during the data analysis, as well by feedback received from the nurses.  Figure 4.3 illustrates an example of the mapping process.    Figure  4.3. The workaround mapping to design principles. 87   In thworkarouidentify aand suppusing thiThe During sewound cadocumendone on cArticles tselected aThispapers wrelevant preview pron how rare in theincluded   C4.4.1The Homecaris example tnd and the m relevant deort those mos process arefull list of tharch for there documentation, usefuommon acahat were mond downloa process wasere similar iublication iocess are thelevant they literature, ain the pool oorrespondparallel syste nurses usehe workarouost commosign principst common  detailed in e identifiedse design prtation, technlness, ease demic publist relevant tded.    repeated fon study settis included we publicatio are to the send how welf the candidence of Datem use was d parallel synd situationn attributesle from the attributes.  Tthe next sec design prininciples keyology accepof use, and hcation searco wound car the searchngs, methodhile as fewn shown in aarch keywol they descriate design pa and Realithe workarostems most model is ap of the resulliterature anhe design ption. ciples from twords suchtance, nursomecare doh engines sure or chronic keywords ds, and findin as possible ppendix H.rds, how cobe their desrinciples foty und classifily in the formplied to theting workarod revise the rinciples mhe literature as wound ce charting, wcumentationch as Goog conditionsetailed abovgs.  This ware missed.   The final rmmon theirign principlr mapping. cation used  of paper s parallel sysund situatiodesign princapped from  is includedare, nursingound chart were usedle Scholar a documentate until the das to ensure The final reesults were  suggested des.  The fina for this partystems to entem use n are used tiple to incluthe workaro in appendix documentating, patient .  The searchnd PubMedion were ownloaded  that the mosults of thisselected basesign princil results wer of the mappsure the da88 o de unds  H.  ion,  was .  st  ed ples e ing.  ta  they colleSupport faccuracynurses neOthetreatmentpaper sprworkarouthe patiendocumenfeatures bThe to currendesignedillustratect for their or correspon of the data eded the mor work note [71].  Manyeadsheets thnds when tht’s record.  tation by nuy these despatient data t and past da as a note bos these two fFigure  4.4own use anddence betwcollected forst recent ans that this ca of the paraat have dataey were facTherefore, trses.  Theseign principleflow sheet wta entries, ax that couldeatures. . The designs  to communeen data and patient docd accurate pn be achievllel system  from the cuing barriershere were tw design prins were the pas a featurend they wer be assignedfor the patienticate with o reality is rumentation atient data ied by prompuse workarorrent and pa in accessingo mapped pciples are lisatient data  similar to ae able to edi to entries i data flow sheether cliniciaecommende[102].  Thisn order to exts to re-evaunds used rst patient re and editingrinciples toted in table flowsheet an spreadsheet those entrin the patientt (top) and thens is as accud in the liter was importecute the caluate patientesources succords.  Nur current or  increase ac4.6. The infd the progret, in which ues.  The pro data flow s progress noterate as possature to incrant when there plan.    informationh as notepases used thepast entries curacy of paormed desigss notes.   sers had acgress note wheet.  Figure (bottom). 89 ible.  ease   and ds or se in tient n cess as  4.4  90   The exclamation mark bubble in the cells of the patient data flow sheet indicated that there is a progress note attached to that cell.  In these designs the users were able to press in any of the cells to edit, as well they were able to use the arrows to navigate the entries of past dates not shown in the view.  Progress notes were added by press and hold.  The red and highlighted rectangle around the exudate cell provided a prompt to assess that specific wound assessment item.  Using these designs the users could access and edit current and past entries, as well receive prompts during the wound management activities. 91  Table  4.6. The mapped design principles from workarounds and the created features. Mapped design principles for usefulness  Correspondence of data and reality The design should provide access to previous and current data entries, and support edit of current and past entries.  Mapped from [103]–[105]. The design should provide prompts during wound assessment and documentation, care planning and execution, reviewing of records, and assessment of supplies to re-evaluate the current and past entries.  Mapped from [71], [102]. Features: patient data flow sheet, progress note.  Charting completeness and comprehensiveness The design should support complete and pertinent capture of patient information, wound assessment items, executed care plan, necessary supplies, the care schedule, and the sources where this information is obtained.  Mapped from [72], [102], [104], [106]–[109]. The design should support capture of data during wound assessment and documentation, care planning and execution, reviewing of records, and assessment of supplies.  Mapped from [104], [108], [110]–[112]. Features: patient data flow sheet, wound photo capture using a wearable camera and speech recognition, supply list.  Individualized care planning The design should make care planning more individualized and/or accurate by allowing the nurse to capture data about the extra supplies left at the patient's home or in the nurse's car.  Mapped from [72], [104]. Feature: supply list.   Mapped design principles for ease of use  View structure and status changes The records system should support a structured view of patient information and treatment supplies where status information, especially changes to those items, are visible.  Mapped from [97], [113], [114].Features: patient data flow sheet, supply list, patient contact information.  Control of current status The records system should support a dashboard view that summarizes the executed care plan and the used supplies.  Mapped from [97], [113], [115]. The records system should support a dashboard view that summarizes the supplies needed for the patients' visits. Mapped from [97], [113], [115]. Features: daily patient schedule, summary.  Simplified data entry The records system should simplify data entry with offering pre-analyzed and prepared patient information and wound care supply items that are needed for the current visit and can be entered interactively.  Mapped from [113]. Feature: wound care plan.       C4.4.2Pre-ewere clascollect daas ‘cheatrecords stake photcommonrange of assessmeattributesnursing pall these documenand execfor compThe photo capphoto capsmall camto receivewhile theharting Comptive infosifications tta effective sheet’, use aystems so itos of drugs ly referencedpatient docunt items, ex of the WaSrocess, it shitems shouldtation shoulution, reviewleteness anddesign featuture using ature using aera that can verbal comy perform ompletenessrmation usehat were crely according flowsheet ’s not missea patient is u in the litermentation oecuted care M.  These dould be pert be supportd be supporting of reco comprehenres informed wearable c wearable c be attachedmands whether care act and Comp, parallel syated and use to the phasto check offd, and use thsing.  Increature as waybserved durplan, necessata should binent, and thed for comped during wrds, and assesiveness are by these damera and samera and s to the nursn needed toivities.  Therehensivenstem use, and by homece of their wo items that we wound phasing compls to improving the studyary suppliese collected e sources shleteness, andound assessssment of s listed in tabesign princippeech recogpeech recoges’ body andtake photos  supply list ess d adaptive aare nurses trk process. ere assesseoto componeteness ande effectiven included p, and the caraccording toould be cle for comprement and doupplies.  Thle 4.6. les were panition, and snition was a it could coof the wounfeature was nd approprio ensure tha They mighd, enter dataent of the re compreheness [106].  Tatient informe schedule,  different phar [102].  Dhensivenescumentatioe mapped dtient data flupply list.   feature thatnnect to thed when theidesigned asative systemt they can t use a notep in multiplecords systemsiveness arehe completeation, wouwhich wereases of the ocumentatios their n, care plannesign principow sheet, wThe wound  made use o nurses’ phor hands are  a list with t92  use ad  s to   nd  n of ing les ound f a nes busy wo  columns.and anothhomes.  F In figdoes not the patienbetween delete icoThe during that the patare at theallowed fThe patient da  One columer column figure 4.5 diure 4.5 the have at theirt has at theithe two coluns at the topdesigns of the fieldwork ients’ home patients’ hoor the compuse of the wta flow shen was for wor wound casplays the scolumn on t home, and r home.  Thmns, as wel right of thee patient dawere identifs.  Nurses almes and whlete and perearable camet, enabled tound care sure supplies upply list feFigure  4.5he left side the column is design alll to add or d supply list ta flow sheeied as necesso had notedich ones aretinent captuera and speehe nurses topplies that that the patiature. . The supply lis for woundon the right owed for theelete woundbox. t and the susary for wou that often  needed to bre of the woch recognit capture datthe patient aents need anist design.  care suppliside is for th nurses to d care supplypply list incnd documeit is not cleae brought tund assessmion, as well a during othlready has ad do not haes that the pe wound carag and dro items usinluded all thentation and r what wouno them.  Theent and carethe supply ler wound cat their homeve at their atient needsre supplies p items in g the add an items that providing cd care suppse features  items. ist and the re activities93 s,   and that d are lies .   For examwhile the  I4.4.3Adapamount omade conmade durcontrol ohome or in to barrrecommeThe attribtreatmentplan as antable 4.6The 4.5.  Nurdecrease supplies Simimapped famount ople, using thy update thendividualiztive resourcf control ovsiderable aming the studver work forin the nurseiers such as nds that recutes of the  supplies as outcome.  . design featuses commonin the accurcan help thelar to how dor ease of uf mental andese feature supply list.ed Care Plae use workaer their worount of effy showed th each indivi’s car ensuremissing or uords systemadaptive res the primaryThe mappedre informedly noted thaacy of the exm to individesign princise as well.  T physical efs they could nning rounds werk processes ort to ensureat nurse usedual patientd that they cnavailable s should alloource use w resource ha design prin by this desit not knowinecuted careualize the cples were mhe results ofort was ma continue to e often usedinvolving pl they had cod adaptive r.  For exampould carry otreatment suw for the caorkaround sd led to preciple for indgn principleg what wou plan, and hare plans forapped for uf the data ade by nurseassess the w to allow hoanning and ntrol over tesource use le, having eut the care pplies.  Resre planningituation indiventing barrividualized was the supnd care supaving an up patients mosefulness, denalysis indics to overcomound care smecare nursdelivering oheir work.  workaroundxtra supplieplan withouearch in lite to be indivicated that thiers and exe care planniply list, shoplies are nedated list of re accuratelsign principated that a se barrier toupply needses greater f care.  NursThe observas to ensure s at the patit having to grature dualized [72e use of cuting the cng is listed iwn on figureded leads towound carey. les were ubstantial  data review94  es tions ent’s ive ].  are n e    and  entry in twell the ewere useworkarouliterature  V4.4.4Manuse it wato use preless mentmemorizthe mentamapped dworkarouThe list, patieclear whaof the patpatient co he electronixecution ofd in order tonds are des and refereniew Structy nurses nots very diffic-emptive inal effort foration, and vil effort requesign princnd situationdesign featunt contact int the status ient was alsntact informc records sy the care pla overcome tcribed in tabced in frameure and Staed during thult to commformation u communicasible status ired for comiple based o is listed in res informedformation.  of the itemso included iation featustem.  This in.  A large ahose barrierle 4.6.  Easeworks suchtus Changee study thatunicate chanse workaroution.  In theinformationmunicationn the identiftable 4.6.  by this desThe emphas is and whatn the patientre. mpacted themount pre-s.  The desig of use desi as the Nielss  in the electrges to the pnds.  Using  literature, v, especially  that uses elied attributeign principlis in this m changes hav contact infir communiemptive infon principlesgn principleen and Normonic systematients’ statpre-emptiveiew informafor changes ectronic recd of the pree were patieapped designe occurred.ormation decation with rmation use mapped fros are often san usabilit that they wus.  For that informatiotion structuis recommeords system-emptive infnt data flow principle w  The personsign.  Figurecolleagues a workarounm these tudied in thy heuristics ere mandate reason, theyn use requirre without nded to lows [115].  Thormation us sheet, suppas to makeal informat 4.6 display95 s ds e [97]. d to  had ed er e e ly  ion s the   As sare structthe wounthe highllist also ihomes or  C4.4.5Notesummaryplan at a execute thown in figuured such thd assessmenighted itemsndicate the s needed to bontrol of Cpads used b items that pglance.  Thihe care planFigure  4.6.res 4.5 and at any chant items by l such as thetatus of the e brought inurrent Stay nurses forrovided infs decreased .  A summar The design fo4.4 the desiges are visibooking at as highlightedwound treat the next vitus  pre-emptiveormation abboth the mey of the exer the patient cogns of the sule.  Nursessessments m item in figument supplisit.  informatioout the currental and thecuted care pntact informatpply list ancan identifyade at differe 4.4.  Thees, whether n use workant status of  physical efflan and the ion feature. d the patient any changerent dates, o two columnif they are arounds oftenthe patient aort they neeused supplie data flow ss in the statur by lookings in the supt the patient contained nd their carded in orders lowered t96  heet s of  at ply s’ e  to he  mental efwhat treasupplies patient dosummarizfor loweris listed iThe summary The schedulelists and home pagespeciallyfort in memtment supplthan they necumentatioing informaing mental an table 4.6. design featu.  Figure 4.7Figuredaily patiend to be visitetheir wounde based on  provided aorizing thosies were neeed.  These wn.  In the littion that cand physicalres informed shows thes  4.7. The desigt schedule wd in a day.   assessmentthe order the dashboard e items for lded, the nurorkaroundserature use on be reviewe effort in co by these de two featurn for the dailyas designedThe home p.  These sum patients weto easily seeate entry or ses did not  allowed nuf dashboardd at a glancntrolling theesign principes.  patient schedu as a tab to iage also hadmaries werere schedule in one glanedit.  Whenmake the phrses to have views is ree [113].  Th current statles were dale (top) and thndicate the  summary v also designd to be visitce the impo the summarysical effort better contrcommendede mapped dus of patienily patient se summary (border in whiiews of theed to be orded.  The sumrtant patienty also inclu to bring mool over the  for esign principt documentachedule, andottom). ch patients w patients supered in the mary featu information97 ded re les tion   ere ply re   such as thnext visit  S4.4.6Verytheir timeof the obTo decreemptive ihas not cchangingcurrent adocumen[113].  Tthe attribThe is display e wound ca, and the woimplified D frequently , at times asserved nursease the amounformation hanged.  Us as the electssessment.  Atation systemhe mapped dutes of the pdesign featued in figurere supplies uund assessmata Entry nurses expre much as ths spent at lent of physicuse workaroing this worronic record design recs should siesign princre-emptive re informed 4.8. Fised in the cent items. ssed duringe time they ast 3 hours al effort neunds to notkaround thes system couommendatimplify data iple that useinformation  by this desigure  4.8. The are plan, the the study thspent providof their 8 hoeded for pate that some y were able ld auto-popon extractedentry with ps this designuse workarogn principledesign for the  wound carat data entring treatmeurs workdayient documeof the patiento change onulate the m from the litre-analyzed recommendund situatio was the wowound care ple supplies ny took a largnt for patien on patient ntation nurst assessmenly items thaost recent aserature indi and prepareation for thn is listed inund care plaan. eeded for the amount ots.  In fact, idocumentates used pre-t informatiot needed sessment tocated patiend answers e literature a table 4.6.  n.  This fea98 e f n all ion.  n  the t nd  ture  99  The design of the wound care plan feature included the ability to fetch the items added to the previous care plans to prepare a template for nurses to use and edit.  This design provided pre-analyzed and prepared items as a template that nurses were able to enter for their current patient visit.   4.5 Summary In the work presented in this chapter a new user-centered approach was used to map those workarounds to design principles.  In this approach WaSM was used to identify the attributes of the workaround categories.  The attributes characterized a workaround situation.  The attributes identified using WaSM and measures developed and validated for healthcare applications of the TAM were used to identify which workaround categories are more likely related to dimensions of usefulness and ease of use.  The usefulness measures were for effectiveness, accuracy, control over work, and the ease of use measure were for mental effort, and physical effort.  Then design principles within those dimensions were extracted from the literature for the mapping of workaround situations to new and contextualized design principles.  The workaround mapping using WaSM allowed to identify those design principles that can support workarounds.  New designs that support workarounds will fill a gap where nurses are facing barriers to get their job done.  The mapped design principles informed a nurse-centered design process in the next phase to develop and evaluate a wound documentation prototype.  The following are the highlighted items in this chapter. What is already known about the topic:  Work situation model describes articulation work.  Workarounds are recommended to be used as user feedback.  Wha  C4.5.1It is necessitaidentificabeyond imuser-centconcern tConsidersystem, w[28].  Mocauses anrecent trethe undernew persbehaviouindicated Usefuadoptit this dissert Work  Workimporonsideratipossible thates their idention and useproving usred approachat is practiing workarohich is the st research d the risk thnd is startinlying creativpective chanrs that do no that the nexlness and eaon. ation is addsituation moarounds cantant to technons for Next workarountification a of workaroability [28].hes does notcal, measuraunds as useropposite to athat investigey may havg to exploreity and ingges the wayt fit within t step in desse of use deing to the bodel can be e point to keyology adoptt Chapterds may becond support.  unds as feed  Other rese only involvble, involve feedback d user-hostilates workare on patient  a different peniousness i health infothe agreed uigning usersign principdy of knowxtended to  dimensionsion. me new besThe most reback for thearch adds the the look as all stakehouring the dee system creounds aim asafety or operspective n creation anrmatics resepon protoco-centred sysles are impoledge: describe wo of usefulnet practices ocent researc design of Hat improvinnd feel of Hlders, and isign phase lated by a det identificatierational fatowards word use of woarchers and ls.  The resutems was to rtant factorsrkarounds. ss and ease ver time anh suggests tIT allows fg usability aIT, but in fat drives innoeads to a ussigner-centon of workailures [63].  karounds wrkarounds [developers lts in this didentify and to technoloof use that ad this hat or innovationd followinct it is a mavation [116er-centred red approachrounds, theiHowever, a hich examin28], [34].  Tthink of userissertation  use such 100 gy re n g jor ].   r es his  101  behaviours in design.  Workarounds were found to enable the execution of patient care, and at times necessary, despite their potential to compromise care.  While other researchers have studied workarounds to inform changes at organizational [34], and technological levels [28], to the best of our knowledge this study was the first to create a formalized framework to identify, characterize, categorize, and map instances of workarounds to design principles for HIT solutions.  Furthermore, the mapped design principles were used to create a set of initial design features for a wound documentation application.  Using iterative prototyping sessions, the design features were evaluated and refined from a low fidelity prototype to a medium fidelity prototype, and then to a high fidelity prototype.  Use of workarounds as human factors that can inform design of new technology solutions has been less explored and other studies in similar or related settings to this study will expand further the knowledge we have gained.   Chapte 5.1 OvAftedesign feclarify retype of pnature anwork taskproposedThe userPartiprototypeincorporaexpectedconstructserve as mreal life cscenario of the patr 5: Createrview of Sr the workaratures for fuquirements rototyping id typically is completel proof of cos’ feedback cipants wer in the targeted in the d to be unstruion, and 3) eodel casesase with thethe nurse waient involviIdentifanclassifion of theFigure  5.1. Dtudy ound mappirther evaluaand desirabls best for ext started wity with the hncept solutioallowed refie informed tt wound doesign of the ctured.  Eacvaluation.   for demons help of clins required tng different ication d ication  Mapped issertation flong phase, thtion and refe features inploring diffeh a set of fuelp of the prn for the taning the deshat there wacumentationwound careh prototypinDuring functration wereical expertso perform caelements ofModeling anmappingDesign Few diagram (cre set of desiinement.  Ex the target srent design nctions permototype.  Usk that they igns to bettes no commi application applicationg phase hadtional select identified.   who were fre activities the workaroCreatmapdesd atures eation of mappgn principleploratory pystem [46]. options quicitting the users assessedwere given r suit the nutment to rep.  However,.  The explor three stepsion the relevA patient scamiliar with and make dund situatioion of ped igns ed designs). s informed rototyping w As the namkly.  It wassers to perfo the suitabiand for simirses’ work. roduce and  good ideas atory protot; 1) functionant work taenario was a the topic.  Iecisions abn while usinRefining anevaluation designs the creation as used to e suggests t informal byrm one of thlity of the lar work tasimplement twere yping phaseal selectionsks that coudapted fromn the patienout the treatg the wound of 102  of his  eir ks.  he  was , 2) ld  a t ment d 103  application prototype (e.g. actors, resources, temporal and spatial organization, etc.).  At this phase the execution of the patient scenario was informal and the functions were not implemented in detail, a non-functional prototype was used.  This enabled to easily test ideas based on the design principles and get feedback from the users. The non-functional prototypes were constructed using cardboard models with interfaces designed by a wire framing application [52].  Marker pens were used to make changes to the cardboard prototypes on-the-fly when needed (i.e. construction).  To collect feedback from the prototyping, questions about workarounds were asked as well.  The questions were adapted from TAM and were drawn from the observed workarounds during the fieldwork.  Questions were items related to usefulness and ease of use of features, in addition to items related to practicality of the wound care prototype for daily practice use by the nurses.  The questions specifically aimed to get feedback on dimensions of TAM that the workarounds were mapped to.  To maintain the informal nature of this phase the questions were asked and discussed with the participants in an open format while the researchers took notes and recorded the participants’ feedback (i.e. evaluation).  Notes taken during the prototyping sessions and the changes made to the prototype were taken as feedback from the users.  Specifically feedback about what the users perceive as useful, easy to use, and practical for their daily nursing practice.  The features that receive positive feedback were used in the next phase, and features that needed to be changed were modified during the prototyping until the nurses provided positive feedback regarding usefulness, ease of use, and practicality of the feature.  At the end of this phase and based on the received feedback components of the wound care application were identified as useful, easy to use, and in line with the practicalities of homecare nursing practice.  104  5.2 Methods User engagement and on-the-fly modifications of the designed features characterized the exploratory prototyping phase of this study.  As the name suggests, this type of prototyping is best for exploring different design options rapidly and is informal by nature [117].  Individual 30-minute, semi-structured prototyping sessions were carried out with nurses, to explore existing and desirable features, identify technology requirements, and evaluate the initial features of a mobile wound documentation application.  The application was named SuperNurse, inspired by a common sentiment among nurses that often they are asked to be “super nurses”, just as the prototype design was informed by their creative solutions (workarounds).   Three key steps made up this phase: 1) functional selection, 2) construction, and 3) evaluation.  During functional selection, the relevant work activities that can serve as model cases for demonstration were identified, with the help of clinical experts.  In this setting it was important that the selected work activities reflect the real-life situation of a nurse’s day.  According to the observations made during the initial fieldwork different nurses managed their day differently, however there were commonalities that were considered when the relevant work activities were selected.  The clinical experts who were consulted in selecting these activities also had a key role in creating a simulated daily work routine of a nurse’s work.  The reason for choosing to simulate an entire day of a nurse was to make sure the designs can be evaluated in a situation that is close to the routine of the nurses.  This approach allowed to make sure that the issues that were identified as important during the fieldwork are discussed during the prototyping sessions.  The research team agreed to limit the number of patient visits to one since simulating more patient visits for each nurse and during each prototyping session would require more resources and time than was allocated to each session.  Furthermore, the daily work routine was 105  created to cover the most important of the issues that nurses faced during their daily work activities.  The created work routine started in the morning with the nurses getting assigned patients and then reviewing the patients’ information.  After reviewing, the nurse hypothetically called the patient to confirm the visit.  Once the visit was confirmed the nurse gathered the clinical supplies needed for the patient visit, also hypothetically.  Typically in a nurse’s daily routine this took about half an hour to an hour, depending on the number of patient visits and the complexity of their condition which may have required more time for gathering the appropriate supplies and possible consulting the nurse who visited the patient last.  In the next step of the daily work routine the nurse drives to the patient’s home.  At this point the patient visit starts by the nurse asking questions related to the health of the patient.  These questions typically included general well-being, pain levels, medications taken, and information about previous visits.  In the next step the nurse proceeded to clean her hand and prepare the clinical supplies required for the patient’s treatment.  Then the nurse puts gloves on and starts removing the old wound dressing.  After removing the old dressing the nurse cleaned the wound, removed the gloves, and started the wound assessment process.  The wound assessment process included items such as wound measurements, odour, colour, exudate, skin, and other related factors.  This process also involved taking photos of the wound.  After wound assessment was done the nurse puts on new gloves and applied the new wound dressing on the wound.  At the end of the patient visit the nurses were told at this step that they would talk with the patient about any concerns that they might have.  In the created work routine after driving back to their office they started entering the information that they collected during the patient visits into the patient documentation systems until the end of their shift.  For tdesign feThey werThey incmenus, spand cut o5.1 illustrIn figshown.  Tincluded buttons.  constructnurse, anthem as the construcatures were e printed onluded graphreadsheets,ut of paper tates the setuure 5.2 all the paper pielements suConstructioions stage ad the nurse whey saw fit. tion of the pinformed by paper so thical element photos, ando be placedp for the prhe material eces seen inch as, text bn involved tn interface tas asked to The activitFigure  5.rototype a s the design ey easily cos that were f others.  A s where the nototyping seprovided to the trays aroxes, commhe active crehat was desi modify theies were giv2. The setup foet of initial dprinciples fruld be modiamiliar to thet of commurses needession before the nurses de the printedand buttonsation and mgned for a s prototype wen based onr the exploratesign featurom the prevfied during e nurses, suon graphicad on the pro it started.uring the pr and cut gra, switch buttodification pecific activith the mate the patient ory prototypines were creious phase othe prototypch as buttonl elements wtotype’s inteototyping sephical elemons, tables, of the protoity was presrial that wascenario desg session. ated.  Thesef the study.ing sessionss, drop dowere also prinrface.  Figussions are ents.  They and date pictype.  Durinented to thes provided tcribed abov 106    .  n ted re ker g the  o e.   107   The main goal here was to collect feedback about how the nurses think the design features can be made more useful, easier to use, and a better fit for their activities.  The use of speech recognition and wearable technology was also discussed to investigate how these technologies might be integrated in the prototype.  This was based on feedback received during the initial fieldwork as well as workarounds that used features such as voice messaging on mobile phones to leave audio messages for later documentation.  After completing the tasks the nurses we asked questions about usefulness, and ease of use.  The usefulness and ease of use questions that were asked are listed below in table 5.1.  Table  5.1. Questions asked during the exploratory prototyping sessions. Usefulness  What features of this documentation prototype would help you decrease time to chart?  What features of this documentation prototype would help you increase your charting accuracy? Ease of Use   How would you change this documentation prototype so it was easy to remember its functions, even if you have not used it for a long time?  What features would make this documentation prototype easy to use for experienced users?  What features would make this documentation prototype easy to use for inexperienced users?  The nurses were asked to continue making any changes to the prototype if they thought it could make the prototype more useful or easier to use.  Asking these questions allowed for evaluation of the usefulness and ease of use.  As well to guide the prototyping sessions with clear goals that nurses could think about when they completed the given activities in the patient scenario.  During the sessions, an expert nurse was also present.  This nurse had several years of experience in research and implementation of HIT in home and community healthcare.  The  expert nuthe approsessions,completeresearch collaboraand follo  D5.2.1A tohealth unpresentedpresentedfidelity pthe protoduring threcorded notes higwere recoprototypemodificaIn thwrote comrse was infoach develop or review od with the hteam and astion with clwed from thata Collecttal of 15 nurits.  During  to the nurs, along withrototype.  Ntype and mae prototypinusing an elehlighted userded using s, post-it notions.  The de examples ments on trmed abouted for use inf the ethics aelp of this asisted with dinical profese clinical peion ses participathe sessionse.  A basic s accompanyurses were pke it more eg sessions fctronic audir feedback aphotos.  Thetes, user intemographicshown in fighe paper pro the previou this phase.pplication snd two otherata collectiosionals was rspective.  ted in this p, a low-fidelcenario (woing activitierovided maasy to use arom several o recorder, nd reactionse modificaerface elemes for this phure 5.3 the totype gives phases of t  For exampubmitted to nursing expn and analynecessary tohase.  The sity prototypund assessms that were terials, suchnd useful, assources.  Thin addition ts.  All of thetions were ints, and draase are in Anurses who n.  These cohe study anle, the patie carry out therts.  Thesesis of this a ensure valessions were (i.e. printoent for a hoto be compl as pens, ma they saw fie full lengtho notes take modified dn the form owings.  Figuppendix E.participatedmments vard had contrint scenario uis phase of t nurses wernd later phaidity of the ae held in locut of a usermecare patieted with thrkers, and pt.  Data wer of the sessn by the resesign featurf notes writre 5.3 show in the protoied from sugbuted as wesed during he study wee included ises.  This clopproaches sal commun interface) went) was thee help of theost-its to me collected ions was auearchers.  Thes by the nuten in the pas some of thtyping sessigestions for108 ll to the re n the se et ity as n  low odify dio e rses per ese ons   modificanew featuthat they or textboaudio rec Afteasked thewound docaptured displayedof use antions to impres.  The nuwanted to mxes to suggeordings, digr the scenari nurses to excumentatio(e.g. in the p, and 4) whd usefulnessrove the easrses were aodify and tost modificatital photogrFigure  5.3. Mo ended, a spress their n using speeatient’s homat additional in the protoe of use andsked to write use the proions to exisaphs, and thodifications memi-structuropinions abch recognite, the nurs characteristype (table  usefulness  their commvided user iting featurese notes wereade by the used interviewout: 1) the tyion, 2) the loes’ car, etc.)tics or featu5.1).  The seof the designents on or cnterface ele as well intr shared amoer on the proto was condupe of hardwcation wher, 3) how theres would cossions were features tolose to the ments such oducing newng the resetype features. cted, whereare that woe document informationntribute to  fully audio suggestionsdesign featuas spreadshe ones.  Thearch team.  the researchuld best capation would would be perceived ea recorded, 109  for re ets  er ture  be se   includingcollectedto provid  D5.2.2Durisessions,the nursewas doneEach stepsimulatedcharting codes useinstancesexample decreasesThis wasranked fethe studyspreadshnegative A secompleti the answer data were se access to tata analysng the analy listened to ts response,  in a spread was identif patient scethe patient dd were crea of user feedwhen a featu the time to to allow foratures were, which is dieet was to idinterest, in aparate spreaon of the pats to the questored on a phe data for ais sis two expehe audio recand their addsheet whereied by the rnario, for exata.  The rested based onback were are was men chart a code ranking of  refined andscussed in tentify featuddition to wdsheet was cient scenaritions asked assword proll the researrt nurses, oording of thitional com each step haesearcher diample to staulting sprea the questiossigned to btioned and t for decreasthe features  developed fhe next chapres that werehat their feereated for to.  These quafter the pattected sharech team mene of whome sessions aments for ead the three recting the nrt gatheringdsheet was ns in table 5elong to onhe user feeding time to based on thor further prter.  The ma of interest dback is abhe responsesestions, whiient scenarid drive accembers.    was presentnd took notech step of taforementiourse to start the wound later coded b.1.  In this ce of the 5 quback indicachart was aseir usefulnesototyping sin purpose to the nurseout the iden to the quesch are detaio was compssible throu during the  of the featuhe prototypined fields to the next stecare suppliey the reseaoding schemestions in tated that thatsigned to ths and ease oessions in thfor the creats, that is, botified featurtions asked led in table leted.  All thgh the internprototypingres mentionng session.  complete. p of the s, or to startrch team.  Te each of tble 5.1.  Fo feature at instance. f use.  The e next phasion of this th positive aes.   after the 5.1, identifie110 e et,  ed,  This   he he r  top e of nd d 111  which features were perceived by the participants as easy to use by experienced users, easy to use by inexperienced users, decreasing the time to chart, increasing the charting accuracy, as well how they would change the prototype to make its use easier to remember if not used for a long time.  These questions were based on the items in TAM, and the results of the previous phase where participants had expressed concerns about these items.  In addition, the questionnaires responses from the previous phase had confirmed that the dimensions of TAM are perceived as key factors in the adoption of the patient documentation systems.  Specifically, in the perceived need to create workarounds when nurses face barriers related to technology. In the next step the rest of the research team reviewed the coding done by the other team members.  Additions or edits were made until team consensus was achieved.  This was to finalize the coding step of the data analysis.  The codes assigned to segments of the recording from the prototyping sessions allowed to create a list of features that were discussed during those sessions, as well the kind of feedback that were given to those features by the nurses.  The research team then sorted the identified prototype features based on the feedback they received from the prototyping sessions.  Sorting the entire prototype features led to a ranked list that was used to identify the top priorities perceived by the nurses.  These priorities informed new designs and design modifications that made the prototype more useful and easier to use. Overall, there were three major refinement cycles to the prototype during the exploratory prototyping sessions.  At each cycle, the previous feedback received from the nurses during the sessions was consolidated and after making refinements to the prototype based on the consolidated feedback the next sessions were carried out using the refined version of the prototype.  This approach allowed for the re-evaluation of the suggested changes by the nurses  before inrefinemeThe prototypeand evalunext chap 5.3 ReOne created bfeatures devaluatio vesting timents were maranked list o that was fuation of dester. sults: Iteratpart of the rased on the uring the prns were othFigure  5.4 and resourcde on the paf features crrther refinedign featuresive Creatioesults of themapped desototyping ser parts of th. Exploratory es to createtient summaeated in this and evalua using expern of the Ma study in thiign principleessions.  In ae results.  Tprototyping, a a medium ory, supply l phase inforted in the neimental propped Desigs phase wers from the pddition to these resultsssessments entr high-fidelist, and the hmed the desxt phase.  Ttotyping andns e the designrevious phahe design fe are discusseries (middle), ity prototypome tabs.  ign of a mehe next pha will be dis features thase, and the atures the red in this secassessment pae.  Most of t dium fidelitse was refincussed in tht were initiarefined desisults of the tion.  ge (right). 112 he y ing e lly gn user   Therprevious patients swearableand a supinvolved setup and  D5.3.1Thisother woto the supnot be upsystem umight nohow mucresults ofchart is rthe bedsiopposed when thesupply lilist freque were sevephase.  Thechedule and camera andply list.  Bythe patient a the materiaaily Patien feature wasrds this was ply list sumdated frequesed by nurset be updatedh nurses per the initial feliable, for tde.  This meto recording nurse goes st in the protently.  One pral features se features, w summary, p speech reco the end of pssessment pl used for phts Schedule the first viethe home pamary compntly.  This ws and for th as frequentceive the suieldwork mahe most partans it was m that informback to the ootype madearticipant nthat were crehich were atient contagnition, pathase 1, 3 mage and entase 1.  and Summw the nursege.  The moonent.  Howas a commat reason somly by the nupply list infny nurses d because thouch easier tation on papffice at the  participantsoted: ated based discussed duct informatiient data floajor refinemries, and theary s interacted st positive fever, there won probleme of the parses who areormation to id trust that se charts aro update theer or in memend of her d feel that thon the mappring the proon, wound pw sheet, woent cycles w supply list.with during eedback forere concern in the wounrticipants an visiting thebe reliable. the supply le with the n supply list ory to updaily patient ey are more ed design fetotyping sehoto capturund care plaere carried  Figure 5.4 the prototyp this features that the sud care docuticipated th patients.  T However, bist informatiurse who viin the paperate the electvisits.  Similikely to upatures fromssions, are de using a n, progress  out, which illustrates thing session was attribupply lists mmentation at the supplyhis can affeased on theon in the pasits the patie chart as ronic recordlarly, havindate the sup113  the aily note, e s.  In ted ay  list ct  per nt at  g the ply 114  “If the most updated care plan was there and if you could actually change it while you’re out, that would be fantastic! For example, seeing the list of supplies that were last used.  The current electronic records don’t get updated accurately.  We have to change the care plan to the latest and print a copy and put in the paper chart and still the paper chart is the fastest thing, you just grab and you’ve got your care plan, you’ve got the medication the patient is on, and all their other data.  So if we didn’t have to print off that and it was on our device that would save some errors, because sometimes the chart is also old.”  Another important component in the summary was the wound assessment summary.  Especially it was noted by the participants that seeing changes is important and easy in the design.  Seeing a summary made the participants feel they can rebuild the context of the patients’ condition.  This is very important since they see many patients in between until their next visit to the same patient.  This is equally important for nurses who were visiting a patient for the first time and were not familiar with the details of a patient’s condition.  One of the participants noted as below: “I think having a patient summary whether or not for this example, like for an infection, all of that would be great, to see is there a change in the wound from yesterday to today, and I think the way you have it would be very easy to maneuver.”  Having the patient summary also was referred as a useful feature that can save nurses time.  Especially in the morning when they are preparing to gather supplies, set their patient visit schedules, and get as much information as possible about the patients in as little time as possible.  In fact, a nurse noted in reference to her work routine in the morning before she leaves for her daily visits: “I don’t usually look at the electronic records! I don’t have time to look at them.  I don’t choose to have time, and I guess you could, but I find getting that far into it to logon to the electronic records and so on taking too much time.  So it’s just easier for me to look at the last care plan, and the summary on the front of the chart.”   The treatmentstatus by  I5.3.2The start theirlocationsimportanhelpful bprototypiResefor a mobNursspeed diaclient is nthem, you Herecaretakerlocation ohelpful.  sometimesaid the f“Hacertain sanurse maconfidentsummary fe. This featu summarizinnteractive Presults of th daily visits.  For this ret.  For examy participanng sessions:archer: “Is ile device?e: “I think il, also inforot reachabl know, fami a nurse is rs, as well otf a patient’It was also ns it is not clollowing:   ving an imagfety aspectskes they caniality issuesature providre offered a g the patienatient Cone initial field is to scheduason being aple, having ts.  Below is there any so” t would be hmation aboue by other nly, friend, oeferring to iher homecars home markoted that haear for exame of the hou that the pe take a pictu, but a photoed a quick odashboard vt data and thtact Informwork indicale and plan ble to accesmultiple alte the feedbacrt of phone elpful, becat backup peumber so wr neighbor, ncluding pate providers ed on a mapving the locple where tse is really rson visitingre and save of the clienverview of iew to alloweir treatmenation ted that a btheir visits bs patients ornative wayk received fnumber, demuse this is spople and supe should havwhoever!”ient specificwho are inv was also aation on its he entrance helpful, wha might be co it, in additit, would hethe patients’ nurses easyt plan. ig part of theased on patr their caretas of contactrom one of ographics,ecific to thport, for me best alter contacts toolved in the feature thatown may noto the hometever buildncerned wion to the malp a lot in id condition a control of  nurses’ woients’ preferkers was peing a patienthe nurses d address thae patient, to e and the panative numb reach the p care proces nurses recot be enough is.  One of ing they liveth, so that fip.  Also, thientifying thes well their the current rk before thences and rceived as vt was noted uring the t you find uhave maybetient, mayber to contacatient, their s.  Having tmmended a, since the participa in, maybe erst visit thats might be am.” 115 ey ery as seful  a e this t he s nts ven  the    In thmay be athis, whicidentify tnurses tonurse is vhelpful to  W5.3.3The at the bedin order tfrom the records.  unnecessdue to prdevices.  The them to ceffort andbandaginwound cais session, thbsent mindeh were obsehe patient w easily haveisiting the p the nurses ound Phowound photside.  In theo take photodigital cameAll the nursary while mivacy laws aThe biggestwearable caarry a digita decreased g and envirore, and theye nurse wasd and hard trved as comas not alway access to thatient for thwork. to Capture o capture wa implements from patiera to their wes who partiobile phonend organiza concern wamera was anl camera.  Neffectivenesnmental sit preferred a referring too locate evemon in homs a straighte patients’ se first time ts a commonation used bnts’ woundorkstation ccipated in ths have digitational policis that patien alternative urses stateds.  For examuations they wearable ca patients whn if the addre healthcarforward actitatus.  Espeche interactiv request, esy the nursess.  Once bacomputer ane study expl cameras thes they weret informatioto the existi that the curple, a nurse wear a headmera mouno may haveess is easy te, being ablvity.  The inially when e patient inpecially to c they were gk at their ofd then uploaressed that tat can be us not allowen should remng system inrent digital  described thlamp for beted on their  mental heao find.  In se to reach, loformation inchanges hapformation wompare theiven digitalfice they copded to the ehey find thied much ead to use theiain secure use by nurcamera incrat in difficutter lightinghead or on tlth issues.  Tituations likcate, and thcluded helppen or whenas perceive wound prog cameras to ied the pholectronic s method sier.  Howevr own perso.  ses that requeased physiclt wound  to provide heir arm. 116 hey e en ed  a d as ress use tos er, nal ired al 117  Feedback received from the participants also indicated that nurses preferred a complete and pertinent capture of the wound photos while they are carrying out wound assessment.  Participants perceived that the captured photos can be incomplete or incomprehensive in the implementation that they were using.  Using speech recognition and wearable cameras were suggested by the literature [118] to improve capture of wound photos during wound assessments. One of the participants noted the following in regards to times when she had gloves on while doing wound assessment: “Voice command would help, but another issue would be if another nurse takes my phone, if the computer is adapted to only your voice they would have an issue because voice command would not recognize your voice as a new nurse, and that could be frustrating, to say it over and over again, and then the clients will ask what are you doing? I think if you could do both as a backup that would be beneficial.”  This shows the nurse’s perception about the challenges that they might have in the use of speech recognition.  These perceptions were informed by past personal experiences with speech recognition.  Therefore, it was very important that the design takes into consideration the limitations of the speech recognition technology for use in real-life settings like patients’ homes.  Using speech recognition to recognize voice commands while the hands of nurses are busy addressed this issue, since most speech recognition technologies are capable of recognizing predefined commands in variety of environments.  This also meant having a wearable camera is necessary to make sure nurses did not have to take their gloves off, then assess the wound, take photos, and put new gloves back on.  During the sessions most of the participants indicated that they prefer a wearable camera on their arms rather than a camera attached to a headband.  However, some of the participants noted that having the camera attached to a headband that also has a lamp can be useful in situations where patients’ homes did not have enough light to take photos that show the details of the wound completely and comprehensively.   P5.3.4The participanperceivedprovide aunderstanfollowing“Thathe surgethese thin The wound aschanges aalso ensustatus.  Ewas highthe proto“Any<commumean wewe use itoff, becautwo dayswriting mthis then there was Thisthe patienatient Dataflexibility thts in the pro ease of use clear picturd what the  about the pt’s awesomon is supposgs.” ability to viesessments. s needed inres that the specially wily desirable typing sessi kind of addnity health e’re supposed’s not as accse we migh ago, so thatore details yes, if there something  comment ret data flow  Flow Sheeat this desigtotyping se and usefulne of the paticurrent statuatient data fe! This woued to see thw, and editThey were  real-time.  Inurse who isth this desigby the nurseons: itional notelectronic re to chart evessible to lot be seeing t’s the generabout the wo were notes,there, yes, tfers to compsheet.  Witht n feature prssions.  Theess of the pents’ historys of the patilow sheet: ld be helpfulis wound in  current andable to revienformation  visiting then feature ths.  For exam would be ucords>, but erything herok at the whhem for woual piece andund in <co especially shat would beleteness anout a designovided rece design of thrototype by .  It was seeent is.  A pa because thithree days,  past entriesw at the bedsuch as men patient nexat allows to ple, a nurseseful! Honesif it was moe (wound caole picture nd but they we always mmunity heaomehow hig nice.” d comprehe that alloweived very pois feature hnurses.  Thin by the participant’s fs is where wor they’re g allowed forside the histioned abovt has a clearadd notes to explained atly, we mosre conveniere electronto see where have a toothave to looklth electronhlighted if tnsiveness ofd nurses eassitive feedbad substantis feature warticipants aseedback ince would puoing to get a nurses to cotory of the pe by one ofer picture o the items.  s quoted betly chart nurnt elsewhereic records),  they’ve (i.ehache, or th at that, so ic records>here was a n the informay access to ack from al impact ons designed t essential toluded the t in things li CT scan, antextualizeatient and m the participf the patientThis capabillow in one osing notes i we wouldnbut in the sy. patients) ley had a seioften we end, but if we hote to showtion includereview and 118  the o  ke ll of  their ake ants ’s ity f n ’t, I stem eft zure  up ad  that d in edit 119  entries while carrying out wound assessment and care they had preferred to abandon the entire feature and use another system.  Ensuring correspondence between data and reality by providing such access, and including items that completely and comprehensively capture the patient information during wound assessment and care addressed nurses concerns in regards to getting a clear picture of the patients’ status.  However, there were still concerns to how much having a clear picture of the patients’ status is useful for situations when a nurse visits a patient for the first time.  One of the participants explained this concern as follows: “This could be helpful in decision making, it depends if I know the patient well, if I know the patient well then the information I find is more helpful, but if I don’t then it’s not that meaningful to me, I will just follow the care plan, take photos, and do the assessment”  In this example the nurse explained that the information included in the patient data flowsheet can be less meaningful when they are subjective.  This is a common problem in patient documentation, especially in wound care [119].  For this reason less subjective patient data items were included in the prototype.  For example, wound photos, which were commonly referred to when other items were perceived as incomplete or less comprehensive, or the supply list which can include supplies with specific measurements and uses.  These additional items complemented the patient data flow sheet.  In discussion about situations that require such evaluations a participant made the below comment: Researcher: “How often does that happen?” Nurse: “About 30 percent of the time, only if I was alarmed by what I saw I would want to look at previous photos, if I see a black toe I would want to see if they were black the last time, if it looked pretty ok, I wouldn’t”  This comment and the previous comment made about the process of wound assessment by nurses exhibits the complexity involved in their decision making and care activities.  Nurses often noted that their work is not routine and contingencies are a big part of their job.  They find  themselva patient.observedsessions design feOthecommentcompleteuse by th“I likprogress  The nurses prdedicatedreviewinassessmeinexperiethe patien  I5.3.5Thissystems asupplies es having to  The design in such situshowed thatatures and dr commentss were mad and compree nurses.  Ae to go onlynotes go anconfusion abogressed in  to wound ag the notes int items.  Thnced particit data flow nteractive S design was nd the papeof patients.   be prepared of the patieations, and t the design iraw from th were made e about the nhensive capbout this fea to one placd where the out what tythe use of thssessment nn a separateis was percpants who wsheet worksupply Listperceived vr charts thatNurses need to deal witnt data flowhe feedbacks useful to neir personalabout the eaotes.  Whileture of patieture one of e to see the treatment npes of notese prototypeotes.  The p component eived to demere not fam.  ery positive they use doed to be ablh various sc sheet was i received frurses work, judgement bse of use of having notnt informatthe participaprogress nootes go, the f are stored w.  However, articipants hrather than and less effiliar with thly by the par not supporte to update enarios that nformed byom nurses d while in coased on the the patient es attached ion, it was pnts explainetes, right noewer placeshere in thethey still adad a prefereattaching noort from thee design of hticipants wh their needsthe treatmenmight happe workarounduring the prntingencies  informationdata flow shto items offeerceived to d the followw I’m confu to look, the prototype dded their nonce for entetes to indivim, especialow the noto felt the el in managint supply listn while viss that were ototyping they use oth they find.eet.  Specifirs flexibilitreduce ease ing: sed where t best for meecreased as tes to the fiering and dual woundly for es componenectronic recg treatment  as often as 120 iting er cally y for of he .” ld  t in ords 121  possible and in as many places as possible.  In their current practice sometimes, this is managed through paper notes as a workaround, but also a large amount of wound care supplies are left at the patients’ homes or in the nurses cars.  Once a wound treatment supply has left the supply room at their office, it cannot be put back due to contamination hazards.  The interactive supply list allowed the complete and pertinent capture of the information about the supplies that are needed and the supplies that are already at the patient’s home.  Nurses were able to capture this information throughout their wound assessment and care process.  As a result they were also able to individualize the care plan instead of guessing based on other information in the patient record what wound care supply they needed.  As feedback to the interactive supply list design one of the participants elaborated as below: “That’s great, because that’s one of our biggest issues, it’s the crazy amount of stuff that we carry, because care plans change, or if someone didn’t write on the outside of the bag what’s inside of that bag which was prepared for a patient, when I go to see a patient if the care plan says they need a wound care supply but it’s not written on the bag I will get that supply before I go to see the patient, I don’t want to be there without any supplies, because of this extra supplies get left behind, so if it’s easy to update the supply list there at the bedside, that’s good, because usually I write in my notebook an inventory of what’s there then I come back and I put it on the bag but other nurses might not do that, so things can easily get lost in the shuffle, so definitely having what’s in the home and what I need to bring is awesome!”  In the clinic in which this particular nurse was working, the larger population covered by the clinic required them to establish a coordination system for the wound care supplies.  In this system the nurse who visited the patient last has to update the care plan, pack the necessary wound care supplies in a bag, tag the bag with the patient’s information and a list of the content, and put the bag in a designated area in the supply room.  As the participant above noted, this system did not always worked as intended.  The nurses had to create other workarounds using extra supplies, paper, and verbal communication to make sure that they had all the supplies that they needed at the patients’ homes.  However, with the use of the interactive supply list they  were ablewound ca  W5.3.6Thiswound caSince it ibecome vFor this rprototypeused the treatmentneeds miplan: “Theat the wowe’re comisn’t partthere is uelse.”  The informatisupply lione of thpreferabl to add newre plan chanound Car was identifire plan is ess common toery compliceason simpl easier to usinformation  might be oght be in the only thing und care plaing next w of the wounsually how design of thon as well hst.  Nurses we participante and easier items easilyges. e Plan ed as a coresentially the have multiated with seifying the pre.  Nurses nin the wounn the wound notes.  Oneis when is thn and I seeeek for examd care planwe do the woe wound carave a templere able to s.  Having a to use as we, drag and d element in  directions ple treatmenveral directocess of revoted that in d care plan  care plan c of the nurse next visit, how often wple, to have in <wound und care, ae plan in theate based onadd informaccess to thisll.  Anotherrop them asthe managemto how to ust supplies inions and coniew and setthe wound ccan be fragmomponent bues explained I would likee are chang that, that wcare electrond we have  prototype a the treatmetion such as informatio aspect men they need, ent of patie the treatm a patient’ssiderations ting the wouare electronented.  Fort how many the below i to remind ming the wouould be goonic recordsthe frequenllowed to innt supplies  mentioned n without frtioned by thand change ents with woent supplies wound carefor the patiend care planic records s example ho visits in a n regards toyself, and snd and I’ll td, the visit f>, the wouncy of changeclude all ththat were adin the commagmentatione participanthe items asunds.  The  for a patien plan, it cannt’s conditi made the ystem that tw to use a week the pa the care woometimes I ell the patierequency, thd care plan  somewheree relevant ded to the ent above b was seen asts was when122  the t.   on.  hey tient und look nt is  y   and 123  where they would use the wound care plan feature.  Given the complexities involved in many of the cases nurses noted that they might need to assess additional information, or consult colleagues, before making their final decision about the wound care plan.  For this reason they preferred to have flexibility to where and when they set the wound care plan.  A participant noted the following:  “I wouldn’t change the care plan until I’m charting, because a lot of times I might liaise with the wound clinician, I come back and I say this is what I saw today and this is what it looked like, what do you think? I like to change it to this thing, and they say yes or no or whatever, then I go back and change the care plan when I’m doing my charting.”  The participant here was referring to finalizing the wound care plan for some patients when she was back at her office.  At that point she would start transferring the patient information to the electronic records system.  To finalize the wound care plan she might consult the wound clinician, or other nurses who might know the patient well, and after synthesizing that information with the wound assessment that she completed during the patient visit she will set the new wound care plan.  This care plan would be used by the nurse who is visiting the patient next, and it would be reassessed and modified if that nurse assesses substantial changes in the patient’s condition.  Overall, simplicity of the wound care plan feature was very important for the participants, to the extent that a participant commented as below about the new design feature: “If it actively worked it would be just a new habit to form, but the whole thing is to keep it simple, the simpler it can be the more efficient and the more likely that we will use it.”  The most important parts of the feedback detailed in this section (5.2.1) are highlighted in the below table.   124  Table  5.2. Highlights of feedback received during exploratory prototyping. Feature Feedback  Daily Patients Schedule and Summary  If the most updated care plan was there and if you could actually change it while you’re out, that would be fantastic!  It’s just easier for me to look at the last care plan, and the summary on the front of the chart.  Interactive Patient Contact Information  Having an image of the house is really helpful, whatever building they live in, maybe even certain safety aspects that the person visiting might be concerned with, so that first visit that the nurse makes they can take a picture and save it.  Wound Photo Capture  Voice command would help, but another issue would be if another nurse takes my phone.  Patient Data Flow Sheet  This is where we would put in things like the surgeon is supposed to see this wound in three days, or they’re going to get a CT scan, all of these things.  If I know the patient well then the information I find is more helpful.    Interactive Supply List  Having what’s in the home and what I need to bring is awesome!  Wound Care Plan  I would like to remind myself, and sometimes I look at the wound care plan and I see how often we are changing the wound.  I wouldn’t change the care plan until I’m charting, because a lot of times I might liaise with the wound clinician.  5.4 Results: Creation of the Emergent Design Features During the prototyping sessions the feedback received from participants resulted in new designs emerging that were not included in the original design.  Nurses were very engaged during the patient scenarios and when they were going through the wound care activities they realized that they can draw from their experiences and suggest features that were completely different than what was included in the prototype.  As these ideas emerged the research team also synthesized and refined them.  Once they were refined the resulting design features were discussed in the next prototyping sessions.  One such design was the sticky multimedia.  This design feature is discussed in this section.      S5.4.1Stickrecordingcollect inclinical dfacilitatinwas inspicollect inled to delmultimedcharting,question participan“I wday anyw In thof the nutheir worrecord inchart.  Duterminoloprofessioticky Multy multimeds that were formation, iocumentatiog the abilityred by the pformation foineation of wia options a but used as about how lts noted: ouldn’t keepay.” is comment rses similar k.  The nursformation wring the ethgy reflects nal image.  imedia ia was envisnot capturedn real time, n.  They we to review aaper systemr later use. hat this feand have ‘tema reminder tong an nurse them for mit became cto how the tes expressedithout worrynographic otheir level oioned as dis for the purto be used lare temporarnd complete workaroun Further expture would porarinesso support ac would keeore than a dlear that the ravel notepa that they uing about thbservations f expertise tposable or tpose of legater, sometimy in nature a the patientd which ofteloration of tlook like.  N’ as an impocurate charp the capturay, I alwayssticky multids, referred sed workaroe professiosome nurseo others andemporary phl documentaes after refnd used as s’ records atn involved uhis concept amely, thatrtant featureting at a lateed multimed throw my smedia desigto in the comunds such anal documens had noted  was in alignotos, text ntion, but inslection, in cnon-legal re later time. se of dispowith subseq it should su (i.e. not parr time).  In ria content oheets away n should sument as shs sticky notetation requithat writing ment with totes, or audtead used toomprehensiference  This concepsable paper uent particippport variout of the legaesponse to ne of the at the end ofpport the woeets, suppors to obtain red in the lespecific clinheir 125 io  ve t to ants s l a  the rk ted or gal ical 126   The sticky multimedia enabled recording audio or taking photos to quickly collect patient information similar to how nurses used sticky notes as workarounds.  The temporary and disposable nature of this feature appealed to nurses who felt the existing system did not support information that changed frequently.  Specifically, nurses perceived the photo capture of the sticky multimedia to be useful.  One of the participants commented as below: “If they are given a new medication then I don’t want to write it down, I can just take a picture of it and copy it down later, so something like that would be good, so it’s obviously useful.”  As feedback to the audio sticky another participant noted: “Audio should be good enough, video could be in case I need something, usually I’m always writing things down so visuals are not really necessary, the audio for written is good.”  Participants also perceived this feature to support their work while they did not have time to capture the patient’s information in detail.  In other cases they may not want to use the audio sticky note at the bedside, they preferred to use a mix of other mediums as well.  For example, one nurse elaborated on a possible scenario below: “I don’t want to say too much in front of the patient, I would want to just type it in, so I would probably do it last, when I’m out, so I can actually think about it, and review the photo and make my other choices, but talking into it would be really good, it could be for anything too, and it would be 30 seconds at most, maybe not even, because I would put in words that are brief and are the main words, and later I would remember it.”  In this scenario the nurse does not want to record an audio sticky note in the presence of the patient, so she preferred to type some of the information and take a sticky photo.  After the visit is finished, she would record an audio sticky when she left the patient’s home.  The participant perceived that this gives her enough information to reflect on later and make decisions about the patient’s case.  This comment also highlights the importance of saving time in patient documentation.  The comment reflects that the audio sticky does not have to be long or 127  descriptive to be useful to the nurses’ work, it only needs to have the main words and ideas.  Having the mix of these mediums received very positive feedback from participants.  The ability to capture information in multiple mediums supported nurses’ work while their work environment could be unpredictable.  A nurse commented about this issue as follows: “If I don’t want to speak then I’ll just take a picture, and I know that’s good, I also don’t know how clear this will be when it records my voice, there could be a lot of background noise, cause some clients like to watch the T.V. really loud when you’re at their home, I guess having both is good, because if I can’t hear myself talk then I can have a visual of it.”  The unpredictable nature of homecare nursing necessitates design solutions that support nurses in various scenarios throughout their patient visits.  The sticky multimedia design was perceived to offer such support to nurses.  During refinements this feature was added to the patient assessment, supply list, and the home view of the prototype.  This provided easy access to the functionality for different activities involved in wound assessment and care. The refined designs for all of the features discussed in this chapter were developed as medium fidelity prototypes, which were further refined and evaluated during the next phase of this study, experimental prototyping.  The refining and evaluation of the design features using experimental prototyping is discussed in the next chapter of this dissertation.  5.5 Results: Low Fidelity Prototype Evaluation The audio-recorded sessions were analyzed and coded with the aim of identifying when and which prototype features received more feedback, as well as noting nurses’ reactions and responses to those features.  After transcription, two researchers coded instances of user feedback during the sessions and a third researcher reviewed the coding.  The most common topical codes were based on dimensions of usefulness, such as time needed to chart (timeliness) 128  and corresponding data and reality (accuracy) [102], dimensions of ease of use such as ability to find and remember functions (view structure) [115], and dimensions of task-technology fit such as corresponding task characteristics and system usage (flexibility to fit workflow) [120].  The features studied in this phase were then ranked according to the amount of feedback received from nurses (see figure 5.5).  The top ranked features made up a refined list to evaluate in the experimental prototyping, and described in the next chapter.  The demographics for the exploratory prototyping phase are shown in Appendix E.   Figure  5.5. Exploratory prototyping user feedback.  The feedback received and shown in figure 5.5 is discussed in this section.  0102030TimelinessAccuracyLowering effortFlexibility to fitworkflowInstances of user feedbackSummary pageWound assessment pageSupply listEmergent feature (e.g. stickymultimedia)Speech recognition  D5.5.1Figuduring thwas that concernsThe and the wMany nuend of thmultiple The and at theto gatherAt thassessed timelinesmultimedelectronitime whisupported   ecreased Tre 5.5 showe prototypinit can save s about how wwound assesound care prses indicateeir shift, sinsources, elecsummary pa end of thei supplies, we end of thethe patient. s.  The feedia to quicklc records.  Ele they woul by the elecime to Chas that timeling sessions.  ubstantial amell speech sment pagelan, also wad that the dece they spentronic and pge was percr patient visithout havinir patient vi The sticky mback was foy capture inspecially aud have had ttronic recorrt ess and accThe most coount of timrecognition , which incls perceived sign of the t substantialaper, to updeived to savits.  At the bg to review bsits they weultimedia cused on sceformation thdio and phoo keep papeds systems turacy were mmon sente in nurseswould perfouded the woby the partiwound asse amount of ate the patie time at theeginning ofoth the patire able to updesign also rnarios wherat they needto sticky fear records ofhat they useprimary driviment for thwork.  Howrm.   und photos,cipants to sassment pagetime at the eents’ record beginning  the shift theents’ electrodate the supeceived pose nurses coued for decistures were p patient infod. ers of the de use of speeever, there w the patient ve time in t would savend of their ss.   of their nursy would usenic and papply list righitive feedbald use the sion making erceived asrmation thaiscussions ch recognitere also data flow shheir work.   them time ahift consultes work shi the supply er records.  t after they hck about itsticky or updating saving nurst was not 129 ion eet, t the ing fts list  ave   the es   I5.5.2The patient supatients’ at a glancsynthesizaccurate recognitiinformatithe necesthis whileto reducepatient inThe the increaneeded totreated bythe patienthe usualpatients’ participanregards tousually wncrease in Csummary pammary desfiles.  This we.  With que the summassessment on increasedon right whsary inform they have  the chancesformation asticky multise of chart  capture pat a new drugts’ status to charting of charts by rets expresse the treatmeere used to hart Accuge receivedign feature workaround ick access toary informatof the patien the chart acen they haveation that thgloves on an of missed ofter they havmedia and thaccuracy.  Tients’ inform prescribed be capturedpatients.  Thducing the cd the systemnt suppliesmanage supracy  most of theas informedallowed nur this informion and theits’ status.  Tcuracy.  Pa undressed ey need for d probably r forgotten e provided e supply lishe sticky mation as rem by their phy, especiallye Supply lishances misss that they u.  The reasonplies often i feedback fo by use of ases to have ation nursesr wound asshe feedbackrticipants nothe wound ftheir work.  have other eitems in the the new treat received multimedia winders.  Fosician.  Thi when there t was perceing to updatsed for thei was that intems were nr increasing summary sa dashboard’ feedback ressment at t received alted that the rom the old The design quipment inchart when tment and huch less feeas noted as br example, ws allowed foare items thived as conte the supplyr work were the electronot up to date the chart acheet on the  view of theeflected thahe bedside tso suggesteability to cadressing wafeature offer hand.  Thisthe nurses have closed tdback for theing usefulhen the par a more accat are not neributing to t list at the b less accuraic and pape.  The morecuracy.  Thcover of the patients’ stt they can o have a mod that speecpture patiens ideal to reed them to  was perceivad to capturhe wound.eir impact o when nursetient was beurate picturcessarily pahe accuracyedside.  Thete, especiallr records tha up to date t130 e  atus re h t cord do ed e the n s ing e of rt of  of  y in t he  supply liprototypethe protothat they   L5.5.3Manfor them.navigate could usewould haphotos, mgloves barecogniticamera fodocumenThe they dediusing varprototypeTheir feemanage asections tst was the m design suptype’s supplused in theiowering Ey participan  One of thethe prototyp voice commve had to unake sure nock on, applyon had here r each patietation. supply list acate a great ious paper a offered allodback suggend keep trache design oore nurses pported the ny list designr daily workffort in Chats commente common sce and carry and to takedress the wne of the eq wound carwas to elimnt at the bedgain receiveamount of end electroniwed the nusted that thik of the staf the supply erceived theurses inform feature wou. rting d that the uenarios wasout wound c a photo of ound, take thuipment is ce, and dressinate the neeside.  Particd positive feffort to manc systems.  rses to keep s feature elitus of wounlist feature w patients’ chation needsld be kept use of speech when nurseare and assethe wound weir gloves oontaminated the wound ad for nursesipants perceedback in tage and keeThe structurtrack of theminates the d treatment as perceiveart to be ac with less efp to date m recognitions were able ssment actiith a wearaff, prepare  or contamigain.  An e to take theiived this tohis categoryp track of wed view tha status of theneed for thesupplies.  Ad to primarcurate and rfort and theyore often th would simpto give voicvities.  For eble camera. their digital nating the wssential partr gloves off lower their .  The particound treatmt the supply  wound trea use of redus mentionedily decreaseeliable.  The perceived an the systemlify data ene commandxample, the Otherwise camera, takound, put th that speech and prepareeffort in woipants notedent supplieslist design itment suppndant system in the previ the time to 131  that s try s to y they e eir   the und   n the lies.  s to ous  chart, thetreatmentaccuracyfeedbackrecognitinurses.  Tregards to  F5.5.4For tsupportinsubstantiteam.  Thof circumsticky muthat the eother desperceivedfeedbackparticularnurses haproperly.from a pan also impa supplies as and lowerin for accuracon they werhere were n the lowerinit to Workhis categoryg the prototal feedback e sticky mustances thatltimedia fillectronic recign features the sticky m related to thly commenve to be res  At the begtient’s recoct accuracy  well.  The rg effort is my compared e the only two clear themg effort catflow  the sticky mype’s fit to wabout this caltimedia wa nurses had led a gap beords offered, that does nultimedia is category ted about whourceful andinning of therd retrieved and the amoeceived feeduch less thato other higo features tes identifieegory. ultimedia workflow oftegory and s specificallto deal withtween the in.  While theot mean theyas much strowas about there ones th creative to ir shift for eon their worunt of effortback suggen time to chhly ranked fhat thematicd by the resas the only the nurses. no clear they perceived  in their visiformation nre was not a did not fit nger fit for e sticky muat involved mcarry out thxample, thekstation, in  nurse makested that theart.  In fact,eatures; howally were suearch team f design feat The other fmes were idto fit very wts to patienteeds of the n clear themnurses’ wortheir workflltimedia.  Sost unpredeir wound my could takeorder to hav in managin impact of t this featureever, alongggested to lor other desure that waseatures did nentified by tell with thes’ homes.  Turses and the about fit tokflow.  The ow and moscenarios thaictability.  Ianagement  a photo as e access to g the woundhe supply li received le with speechower effort ign features perceived aot receive he research  unpredictabhe design oe informati workflowparticipantst of their t were n such scenaactivities a screenshothat screensh132  st on ss  by  in s ility f the on of  rios t ot 133  when they are on the bedside.  This is especially useful since not all electronic records systems are accessible by the bedside.  Some are accessible only in the nurses’ office.  The reasons are sometimes policy related and sometimes technology related.  Another example is when a patient is given a new prescription by their physician.  The nurse can take a photo, or record an audio sticky, of that prescription to update the patient’s records at the end of the work shift in the office. In scenarios such as these, nurses did not have to worry about what information is supported by the records systems, but rather what information they need to carry out their wound care activities and continue with their workflow.   5.6 Summary In this chapter the creation of the mapped design features using exploratory prototyping was discussed.  With participation of 15 nurses a series of prototyping sessions were carried out.  In these sessions one participant and two researchers were present.  The researchers guided the nurse through a simulated patient scenario.  These sessions were informal and semi structured.  The primary goals were to contextualize, discuss, and refine the design features that were informed by the mapped design principles from the previous phase of this study.  The feedback received from the participants was used to refine the designs and identify which features are perceived as most useful and easy to use.  The design features discussed during the prototyping sessions were summary page, patient schedule, interactive patient contact information, wound photo capture, patient data flow sheet, interactive supply list, wound care plan, sticky multimedia, use of speech recognition, and use of a wearable camera. The prototyping sessions were audio recorded.  The audio recordings were analyzed by the research team until consensus was achieved about the results.  The data were coded to identify 134  which features received more feedback from the participants about their perceived usefulness, easy to use, and flexibility to fit workflow.  The results revealed that speech recognition and the wound assessment page (i.e. patient data flow sheet, wound photos, and wound care plan) were perceived to support nurses in the timeliness of their work more than other features.  The summary page, the supply list, and the sticky multimedia also were perceived as supporting timeliness but much less than speech recognition and the wound assessment page.  The summary page, speech recognition, and the wound assessment page were perceived to increase accuracy of charting by the nurses, as were the supply list and the sticky multimedia, but less than the other three design features mentioned. Speech recognition was perceived as lowering the effort in the nurses work the most, followed by the supply list.  The supply list however received much lower scores in the results.  Other design features did not generate a clear theme in the results for the lowering effort category.  The sticky multimedia was perceived as supporting flexibility to fit workflow more than any other design feature, while there were no clear themes in the results about this category for other design features. The following are the highlighted items in this chapter: What is already known about the topic.  Prototyping is used to create and refine designs with participation of users.  Usefulness and ease of use positively impact the users’ technology adoption. What this dissertation is adding to the body of knowledge.  Informed by workarounds exploratory prototyping with an informal and semi structured setting resulted in new designs emerging.    C5.6.1Creaamount odesign feof this fehigher scfeatures i(e.g. stickchapter.  have the integratiothe next pand data Durithen a higis supporfunctionaor very s Timelwound Lowerand adonsideratition of the mf feedback. atures that aedback for tore were selncluded spey multimedThe mediumability to sun of the prohase, similafrom a real-ng the develh fidelity pted.  The chlity and inteimilar functiiness and ac documentaing effort anoption of thons for Nexapped desi The receivere more usehe next phasected to be ech recogniia) and supp fidelity propport user intotype with r to the explife case canopment of trototype it isoice of the draction.  Thons and intecuracy are ition prototyd flexibilitye wound dot Chaptergn features ud feedback ful, easier toe of the studdeveloped intion, woundly list.  Thetotype doesteractions.  the electronloratory pro be preloadehe low fidel necessary tevelopmente selected dractions to tmportant facpe.  to fit workcumentationsing explorindicated th use, and bey the designto a medium assessment y were discu not need toFor this studic records sytotyping phad to the proity prototypo ensure tha platform is evelopment he low fideltors to usefflow are imp prototype. atory prototat workaroutter fits for  features w fidelity prpage, summssed in the  be fully funy it was nostems that nse, a patientotype. e into a medt the same fessential to platform shity prototypulness and aortant facto yping providnds can be unurses’ workere ranked aototype.  Thary page, emevaluation sctional; howt necessary turses use.  t scenario caium fidelityunctionalityensure consould be ablee.  For this sdoption of trs to ease oed a great sed to infor.  To make nd those wiese design ergent featection of thever it shouo provide However, dun be adapte prototype a and interacistency in  to support etudy the 135 he f use m use th ure is ld ring d nd tion xact 136  Android platform was selected for development.  This platform allowed the flexibility to make changes quickly.  Chapte 6.1 OvThisinvestingevaluatinreceived refine theof the nuSimiconstructfeatures wdone prevInterpatients' prototypecapabilitiwith datascenario phase of r 6: RefinFerview of S phase emph time in impg a high-fidin the previo design prinrses about ular to the exion, and evaas used at tiously; howfaces were chomes.  The to a high fies, includin entry and rwas adaptedprototyping Identiaclassiing and Eigure  6.1. Disstudy asized on dlementationelity systemus phase.  Tciples and csefulness, anploratory prluation.  Thhis phase asever, the exreated for a design featdelity protog navigationetrieval featu from was uthe setting wfication nd fication valuationertation flow determination of a higher .  A mid fidehe prototypollect data ad ease of usototyping the same patie well.  The ecution of t mobile comures were evtype.  The m, however itres.  Pre-ensed to provias formal. Modeling amapping of the Deiagram (refini of the adeqfidelity systlity prototye was evalubout the effe of the desis phase hadnt scenario nurses werehe patient scputer devicaluated andedium fidel was not abltered data frde data to u That meantCreamadend  sign Featung and evaluauacy of the em, and thepe was deveated by expects of the pigns.    three stepsused in crea asked to peenario weree which nur refined fromity prototype to functioom the realse during th nurses weretion of pped signs res tion of designsproposed son implemenloped basederimental usrototype on, functional tion of the mrform the sa formal rathses would b a mediume provided sn as a comp life case thae prototypin asked to coRefining aevaluationdesigns). lution beforting and  on the feede.  This was the perceptiselection, apped desime tasks as er than infoe able to tak fidelity ome interaclete applicatt the patieng.  During thmplete the nd  of  137  e back  to on gn were rmal.   e to tive ion t is 138  patient scenario and the length of time to complete the scenario was captured.  Then the nurses completed a questionnaire which included items related to usefulness and ease of use adapted from TAM.  The decisions made by the nurses about the wound documentation items also were recorded by noting which wound care items they did or did not document in the wound care application.   Qualitative data were collected using audio and video recordings of the users while they used the prototype and verbalized their thoughts (i.e. Think Aloud protocol) [53].  Qualitative data were transcribed and coded based on codes developed during the identification and classification of workarounds.  The results of the qualitative analysis was be triangulated with the quantitative data from this phase.  6.2 Methods Experimental prototyping was used for enhancement of the target application's specification and to test a proposed solution for an identified problem [117].  Informed by the exploratory prototyping reported in Chapter 5, the top rated features were used to design an operational mid-fidelity wound documentation prototype for an Android device.  These features included: 1) summary of previous charting, 2) wound measurements and photos, 3) wound assessment items (e.g. exudate, odour, and pain), 4) sticky multimedia, and 5) supply list.  The prototype was an Android application using Android’s speech recognition to recognize voice commands for the capture of wound measurements and wound photos (wearable wireless camera) when the homecare nurses’ hands were occupied (i.e. when providing direct wound care).  The experimental prototyping included 2 phases.  During phase 1 a mid-fidelity prototype was refined into a high fidelity prototype that was evaluated in phase 2.    D6.2.1Thishowever,scenario help of cnurse parstart of thoutline oappendixThe measurinparametestatus, anOncescenario interviewand ease nurses, owere aud   ata Collect phase of res the sessionand 30-40 mlinical experticipants (toe session, ef the tasks th F. scenario incg and photors, 4) using d 5) finalizi the formal and use the  and asked nof use, as wf which 11 wio and videoion earch also is were strucinutes for ats.  The sesstal n=12, phach nurse wey would bluded five mgraphing thethe sticky mng the supplsession begrelevant prourses to comell task-techere comple recorded.  ncluded funtured to finin interview.ions were case 1=6, phas given a she asked to coain tasks: 1 wound witultimedia toy list based an, the nursetotype featuplete a quenology fit [ted and retu ctional selecsh within 10  The patienarried out inase 2=6) difort introducmplete.  Th) preparationh SRT, 3) d take informon the treatm was asked re as they sastionnaire [120].  A totarned.  The intion, constr minutes fort wound car local commferent of thotion to the pe demograp and reviewocumenting ation relateent plan.  to completew fit.  Final24] related tl of 12 questerviews lauction, and  a patient we scenario wunity healthse in Chaptrototype, anhics for this of patient hthe local wod to the pati a task at ealy, we condo perceptiontionnaires wsted 30 to 40evaluation; ound care as set with  units with er 5.  Prior td given a b phase are inistory, 2) und assessment’s medicch step of thucted an s of  usefulere given to minutes an139 the o the rief  ent al e ness  d   D6.2.2Threaudio andand evalucreation oprototypicoding eaand the padded to the rest orenamed coding dountil everreviewed 6.3 ReThe First the recorded the quantdiscussedDuriincluded ata Analyse of the rese video recoation of thef the mappeng.  The codch of the rerototype feathe codes fof the researcto match thene by each yone agreed by the reseasults: Refinresults in threfined protoin the form itative user  in this sectng phase 1, interest towis archers laterdings were design featud design fees were grosearchers coture that war usefulnessh team to ac correspondmember of  to the coderchers and ing the Mide refining antypes basedof video andresponse to ion. several themards the desr transcribed.  The codesres were deatures, as weuped into usded an instas involved.   and ease ofhieve conseing categorythe researchs assigned. team consen-Fidelity Pd evaluation on the user audio whica questionnaes of user rign features, coded topi for the analveloped basll feedbackefulness annce of a useThroughout use as theynsus on a fi that was ag team was re This processus was achrototype  of the desi feedback, sh was analyire given toeactions wer, frustration cally, and aysis of the ded on the fe received dud ease of usfulness and the coding  saw fit.  Thnal list of coreed upon bviewed by s was to maieved.  gn features wecond the qzed by the r the participe apparent tdue to malfunalyzed themata collecteedback recering the expe categories ease of use process the ese codes wdes.  The coy the researother membke sure that ere in threualitative usesearch teamants.  Thesehroughout. nctions, anatically thed during refiived from erimental .  During therelated issueresearchers ere shared wdes then wech team.  Ters and refinthe coding we categorieser feedback, and last w results are  These reactd surprise du140  ning  , ith re he ed as .   as ions e to  success.  and the cphase 1 s  D6.3.1Nursfeatures. comparedstarted tofeature.  “Duso many a whole vwhat to c One it would Participant are providedessions.  Figure  6.2esign Feates were hes Specifically to what the lay out a scShe said:  ring your cosupplies at herbal diaryheck.”  of the nursereplace paperesponses hi for the pati. Experimentaures of Inteitant, but int their interey use in theenario in whnversationsome, I don' of informats added thatr for her, evghlight the ient.  Figurel prototyping, rest erested in ust was notabir current mich she wou with patient have to pacion that you  she would uen if the prompact the d 6.2 illustratassessments ense of some ole for featuedical recorld be interets you end upk, etc., and can pick thrse the protototype did nesign had ones the setup tries (middle)f the mid-fires that wereds system.  Fsted to use t fumbling ithen from though and dtype in its cot synchron the compleand the mat, assessment pdelity protot less conveor examplehe sticky mnto…oh my e audio sticecide from wurrent stateize the colletion of taskerial used fo age (right). ype design ntional , one nurseultimedia gosh you haky you will hat to keep, and she thocted 141 s r the ve have  and ught  informatirecognitipatients. resonated“Spepeople’s  Thoube approppotential searchingto recordperiods (more comuse the rethe use obetter on  D6.3.2The disappoinphoto wisimple volife simuon back to ton seemed t The sense o with the wech recognihomes and hgh it appeariate in eveto increase  in audio re longer recoe.g. 20 seconprehensivecordings tof audio recoce they heariscarded Dvideo, audiotment, and th the wearaice commanlation sessiohe electronio trigger thof interest, cuork and needtion is mostard to recored that use ry homecareflexibility ancordings mirdings, evends).  They n wound asse chart later. rdings as sti back what tesign Featu, and writteperseverancble camera. ds and means.  For exac medical reughts aboutriosity, ands of the nur useful for hrd without spof speech re circumstand fit the woght seem tim though we oted that lossments wh Another pocky multimehey recorderes n commentse predomina While the ssurement numple, one nucord system the needs o anticipationses.  One nuand-free mereading gecognition ance, the nursrkflow, ande consuminoffered an onger recordile they havesitive aspectdia encourad.  contained mntly relatedpeech recogmbers, it prrse attempt.  The stickyf the nurses  about thesrse explaineasurement, rms or contd sticky mues’ reactions decrease thg, nurses prption to liming time can gloves on  expressed bges them tooments of  to use of spnition was roved less pred to take a  multimediwhile provie features ind:  often in diffiaminating nltimedia fea indicate the time to chaeferred to reit recording be used whand cannot ty nurses wa structure thfrustration, ceech recogneasonably aactical for twound phota and speechding care fodicate that icult positionotebooks.”tures wouldat there is grrt.  While tain the opts to shorter ten they neeype.  They cs that they feir thoughtsonfusion, ition for takccurate for he phase 1 ro and record142  r the t s in  not eat ion ime d an elt  ing a eal- a  wound mfix issuesend of th“I ofHer possibly accompliunsuccestype of tein the usefeature seIt alscases whuse the predesignistudying located into take ph  R6.3.3Whiprototypeeasurement such as came session, shten have troresponse sugreinforced dsh in the consful experiechnology co of speech remed the mo should beere use of thhone camerang these feathese featur hard to accotos. etained Dele nurses exp features thfive times eera being oe explaineduble with mgests previouring the phtext of homnces with spuld be an imecognition tost challeng noted that ne phone cam.  That leadtures, such aes for specifess body pasign Featurerienced fruat exceeded ach, and in ff, or other :  y own voiceus unsuccease 1 sessioecare nursineech recognportant cono recognizeing.   urses expreera is not ps to a realizas replacing ic useful casrts or cavitiees stration andtheir expectall cases, waissues such a recognitionssful encounn.  Expectatg were temition.  Ongosideration f voice commssed that theossible, howtion that futthe wearables as noted s that need  interest, thations.  For s unsuccesss recognitio.” ters with spions of whapered by preing unsucceor redesign ands for tak wearable cever in mosure studies e camera wiby the nursea wearable cey were alsoexample, afful.  Either dn of “to” vseech recognt speech recvious negatssful encouin later staging wound amera can bt cases theywill have toth a phone cs, for examamera next positively ter successfuue to easier. “two”.  Atition were ognition canive or nters with thes.  In particphotos, as the useful for  would opt t focus on eitamera, or ple, wounds to a headlamsurprised ablly complet143 -to- the  is ular, is o her  p out ing 144  the documenting wound measurements using speech recognition, one nurse exclaimed, “oh cool! totally worked.” Later on, when creating a sticky multimedia to serve as a reminder, she noted:  “mkay... (quiet laugher) um, I kind of forgot how to do this but we'll just try (laughing), um, ‘remember to contact the vascular surgeon for Mr. A?’ oh! It worked, cool.”  Surprise as a reaction can be viewed as an indicator of how well the features fit the original aims and intentions of the design.  For instance, the latter quote suggests that the user is unsure of how to complete the assigned task.  However, the surprised reaction at successfully completing the task suggests that the design functioned in the way it was meant to, especially in regards to ease of use.  That is, the design allowed for the user to navigate and complete the task intuitively, even after forgetting the specific feature required.  The nurses expressed that the sticky multimedia and the use of speech recognition for capturing wound measurements, would especially decrease the time to chart and increase the accuracy, since the data are being collected at the bedside when they are assessing the wound and while their hands are busy.   By the end of phase 1 three major refinement cycles were carried out that involved the patient assessment page and the speech recognition.  The final version included mostly functional features, however speech recognition functioning remained inconsistent.  6.4 Results: Acceptance of the High Fidelity Prototype The feedback received during phase 1 was used to refine the mid-fidelity prototype into a high fidelity prototype.  The high fidelity prototype did not use the wearable camera since the feedback received from phase 1 indicated that a wearable camera would be useful only for specific cases.  The prototype has the same 5 primary components similar to phase 1.  However, these features were modified according to the same experimental prototyping principles to reflect  feedbackfunctionias complThe nurswas to adanother.  FigPredprototypiphase.  EDuring thvisit, whionly diffe6.3) was portion o received frong wound peting the funes also weredress conceSome of theure  6.3. Woominantly sng sessions ach nurse pae session thle using Suprence was tused to bettef the sessionm phase 1. hoto capturections and i given a disprns expressese features und cleaninimilar to theusing the hirticipated ine nurses weerNurse.  That a medicar simulate t was dedica This includ, direct accenterfaces ofosable plasd in phase 1are shown ing (left), pati previous phgh fidelity p one sessionre asked to phe same patl mannequihe real-life ted to the paed better funss to sticky  the supply ltic bag that  about sprea figure 6.3. ent assessme ase, in phasrototype.  A that was aprovide wouient scenarion with a chrsituation of tient scenarctioning spmultimediaist and the pwas the samd of germs   nt (middle)e 2 we cond total of 6 nproximatelnd care for  from phaseonic wounda home visitio for 10 mieech recogn from the maatient assese size of thefrom one pa, assessmenucted furtheurses particiy 40 minutea fictitious p 1 was used on the butto for the nursnutes, and thition, fully in page, as sment page. phone.  Thtient’s hom t entries (rigr experimenpated in thiss in length.  atient in a h in phase 2.cks area (fies.  The firse rest was f145 well   is e to ht). tal  ome   The gure t or  the intervidentify wgiven to n  L6.4.1The would usof interachow she “I wThentyping whand andThis allosterile hastomach)example,“Wh“maintenwouldn't  Thisvalue, whwell.  ThAs for thiew.  Durinhich featururses whichikely to Adhigh-fidelitye the prototytion and wothought the ould take on she put theith the unglo her wrist onws the nursends, and use to collect p one nurse nat I would lance”, and have to say  means nursich is easieris will save e values, nug the intervies the nurse included itopt Design prototype fpe in their drkflows notprototype we glove off! phone betwved hand.  S the other hs to multita the other hatient data.  oted:  ike to probatouch woun“goal of cares can use o than havintime and redrses noted thew questions found mosems of TAM Features urther elicitaily home v discovered ould be used” een her stomhe took a pand.  She prsk.  They caand in combThis can infbly do is to bd bed and sae”.” ne hand to pg to verbalizuce the chanat minor mis about useft easy to use [24] and taed interest fisit routinesduring phas and said: ach and hehoto by holdessed to capn use one glination withorm new dee able to toy “70 perceoint to theire the data ences for incostakes can bulness and e or useful.  sk-technolorom the nurs.  This leade 1.  For insr forearm ofing the camture the phooved hand t other bodysigns that suuch goal of nt blah and target data try field narrect data ee negligiblease of use wA questionngy fit [120]es to envisi to identifyintance, one n the gloved era betweentos by her uo carry out t parts (e.g. opport this wcare and say 30 percent entry field ame and thenntry by spee since they ere asked taire also wa.   on how theyg other mourse describhand and sta the unglovngloved thuasks that reqther forearmorkflow.  F blah”, so I nd verbalize the value ach recognitican review t146 o s  des ed rted ed mb.  uire , or  the s on.  he  data entryof “by”.  “Whwhat it w OverOne nurs“I thhang of i  U6.4.2The primary cwhen colproperly.at the pat“Youwill be suthis sort  As apatients’ right afteAnother preventinscenario “I doputting d later and cOne nurse een it (speechas.” all, nurses ae even wentink [it woult and you genlikely to Alevel of frusoncern of nlecting patie  Nurses expients’ bedsid don't wantch that the pof stuff.”  solution, nubedside theyr they leavepoint of cong spread of one nurse non't want to pown on. orrect minorxplained:   recognitiogreed that th as far as sayd save time]t good at it,dopt Desigtration obseurses was honts’ informaressed conce.  One nur to spent tooatient is lyirses sugges can collect the patientscern for nurbacteria or vted:  ut it (the ph mistakes.  n) writes doe prototypeing the foll more than h then it will n Featuresrved during w much timtion.  Especern that it isse noted:   much time ng there, anted that to re some basic’ homes, or ses was keepiruses fromone) down For examplewn the item  would saveowing in regalf, maybe be very fast. phase 2 dece they mighially in case important tin there (patd they’ve goduce the am data at the bin their car bing the ster one patienton anything, “to” insteaif it's close e them time ards to timeinitially not” reased compt have to sps when the hat the protoient’s homet to lay therount of unnedside and efore they dile areas of t’s home to a, because yod of “two”,nough it wiand make th saving:   that, but onared to phaend at the ptechnology type will fu) because a e while youecessary timcomplete thrive to the nheir work cnother.  Duu don't kno or “buy” inll trigger meeir job easiece you get thse 1.  The atient bedsidoes not wonction proplot of wound finish doinge spent at teir data entrext patient.lean and ring the patiw what you'147 stead  to r.  e de rk erly s  all he ies   ent re  ReseNurs In adnew tech“I wThislife wouninformatipatients. made durprototypetheir man  V6.4.3Surpfeature.  substantithose datof the ses“I wwound, dto the off Ano“Thearcher: whae: You couldition to thenology, or aouldn't be so was in respd care practon systems, This has being this rese during theidated workery Likelyrised reactioThe overwhal amount oa are not supsions a nursould do everescribe whaice.” ther nurse no sticky multt if it was ond do that, buse reactionss one nurse  keen on usonse to a quice.  Her pe and more teen true accoarch.  Otherr real-life wflow.  This i to Adopt Dns in phase elming feelif time and itported by the explainedything witht I need to bted the follimedia, I thi the side oft then you'r it is expectput it:  ing a new pestion aboutrspective wachnology mrding to lite nurses felt ound care prs discussed esign Featu2 were mucng of the nu allows flexie current el:   voice! I wouring in nextowing:  nk that's the your sterilee contaminaed that someiece of techn whether if s that nurseeans more wrature [121]differently aactice if proin the next sres h most comrses was thability to colectronic meld just go to time, and ev main thing tray? ting the tray users will ology.”  she would us already usork that is , as well thend agreed thvided to theection. mon in react the sticky lect patient dical record speech (stierything wi, it does a li with your presist adoptise the protote several henot providin ethnographat they woum, even if ition to the stmultimedia data that the systems.  Acky audio) all be there wttle bit of evhone.” on and use oype in her ralthcare g care for thic observatild use the t is not part icky multimcan save they need event the end ofnd describehen I comeerything!” 148 f eal-e ons of edia m a  if  one  the  back 149  The concept of “little bit of everything” refers to the flexibility that the nurses perceive in the sticky multimedia feature.  However, this also extended to the speech recognition feature.  When asked what the prototype’s most useful feature is, one nurses answered:  “Uh, I think the voice recognition.  To be able to use that is fantastic.  Just the ease of using that rather than having to write it down…It's like a voice recognition notepad…that's a big selling feature for me.”  6.5 Results: Mid and High Fidelity Prototype Evaluation The audio and video recordings from this phase were topically coded by 3 researchers.  The topical codes were adopted from the literature to identify usefulness and ease of use issues [24].  The coding was reviewed by the researchers to achieve team consensus.  Figure 6.4 shows the results of the data analysis.  Additional concepts compared to figure 5.5 are visibility and control of changes in patient information (control of status) [115], and inclusion of all relevant and applicable information (completeness) [102]. The demographics for experimental prototyping are shown in appendix F. The experimental prototyping user feedback shown in figure 6.4 indicates that compared to the exploratory prototyping phase accuracy and flexibility to fit workflow received much less feedback.  This can be explained by the perception of the nurses towards the sticky multimedia, as well the supply list and the wound assessment page.  As noted in the previous section the nurses felt very strongly about how the sticky multimedia fits into their workflow.  All three features also were identified by the nurses as having a very positive impact on the accuracy of their job.  The nurses indicated that they are able to use the wound assessment page and the supply list to keep the patient’s record up to date, while using the sticky multimedia to collect any complementary data that they need. 150   Figure  6.4. Experimental prototyping user feedback  The larger number of instances coded in phases 1 and 2 is expected since the nurses were using a functioning prototype.  Furthermore, wound assessment required more time during the patient scenarios, which impacts issues related to the wound assessment page and the speech recognition active during wound patient assessment.  Given this fact, compared to figure 5.5, the speech recognition received better user feedback.  The wound assessment page elicited more feedback related to timeliness.  The primary cause for this was that nurses felt that other forms of input instead of typing can save them even more time.  That can include dropdown menus, or radio buttons.  However, a drawback is that nurses will be limited to the options available in the application.  The best solution can be a hybrid solution that affords free-text as well quick entry.  The lowering effort category issues discussed related to the wound assessment page were mostly 0510152025303540TimelinessLowering effortControl of statusCompletenessInstances of user feedback Summary pageWound AssessmentpageSupply listSticky multimediaSpeech recognition focused ahigher cofeedbackissues relclear.  Thwhich onFiguand not oand 6.4 efeatures treceived   D6.5.1In thfeedbackfeedbackfeatures. features. 3 the initpatients’ reduce thscenario when nurt level of famplexity an is expectedated timelinis is causedes are require 6.4 highlinly the feedxhibit what hey found ufor each of tecrease of is category,  than other d.  There wer This is con Timelinessial fieldworkconditions. eir time to cthat receivedses hands armiliarity wid the numbe.  Another cess have bee by differentred.   ghts the impback given tconcepts weseful and eahe design feTime to Chthe wound aesign feature no other csistent with  remained a  revealed th Speech recohart by coll positive fee occupied th the Andror of interfacomparison wn addressed interpretatiortance of to a specificre importansy to use.  Tatures and tart  ssessment pes.  Followelear themes the feedbackvery importat nurses spgnition wasecting patienedback was with woundid interfacee elements ith figure 5, at times chons of what he listed con feature.  Tht to the nurshe followinhe conceptsage and sped by the supidentified w received duant concept ent substant seen as a tot informatiothe use of sp treatment o elements usin the woun.5 indicates anges in thesupply itemcepts in relat is, the cones work andg subsection listed in figech recogniply list thatithin this caring the creto nurses woial amount ool that can n at the bedeech recogr assessmened in that pd assessmenthat while th supply lists are alreadation to the cepts listed not just whs discuss thure 6.4 tion receive had receivetegory for thation of therk.  As discf their timehelp the partside.  The mnition at thet. age.  Given t page moree supply lis items are noy in stock andesign featu in figure 5.at design e feedback d much mord much lesse other desi mapped deussed in cha documentinicipants to ost commo patients’ ho151  user t t d res, 5 e  gn sign pter g n mes  The was usedwas at threcords oinformatiback to thcompletemuch timpreferredchart didthe patienThe during thbeginnindaily visiperceived  L6.5.2The Three othsupply lias, wounto make swound asses during the pe end of nurf the patienton that theyeir office.   their patiene between v to complete not necessats’ bedsidefeedback rece creation og of the nursts without h as useful aowering Ewound asseser design fest, and stickyd photos, paure that thesment pageatient visitsses work shis.  Commen need at the This allows t documentaisits to com their patienrily mean th. eived for thf the mappedes work shiaving to revt the end of ffort in Chasment pageatures that a multimeditient data flo information received po to assess thft when theyts from partpatients’ bethem to refltion when tplete the patt documentat all of the e supply lis design feaft to gather tiew multiplethe patient vrting  was the primlso receiveda.  The wounw sheet, an captured insitive feedbe wound an go back toicipants indidside, and thect on the inhey visit theient documeation at the epatient docut design feattures.  The she wound tr records sysisits to updaary feature feedback fd assessmed the wound these compack in two sd collect pat their officecated that then complemformation a patient.  Thntation.  Innd of their mentation nure was simupply list weatment suptems.  Thiste the suppl that receivor this categnt includes  care plan.  onents are acenarios.  Fient data.  S to update they would lient them wnd not feel rey also mig such situatishift.  Decreeeded to beilar to the feas perceivedplies they n design featuy list. ed feedbackory were, suimportant coFor this reaccurate andirst was wheecond scenae electronicke to collechen they goushed to ht not have ons nurses aasing time t completed edback rece as useful aeed for theirre was also in this categmmary pagmponents sson, nurses h reliable, wh152 n it rio  t the  lso o at ived t the   ory.  e, uch ad ich 153  was discussed in the creation of the mapped design features of this study.  The refinement efforts in the experimental prototyping phase were focused on improving ease of use, and the feedback received was for lowering the effort in charting to improve ease of use.  Comments were also made about familiarity with the Android platform, which concerns ease of use and the amount of mental and physical effort nurses had to spend to use the prototype.  These comments were addressed by including a short introduction to how the prototype functions, before the prototyping sessions started.  The feedback received from participants who were familiar with the Android platform suggested that they found this feature to lower their mental and physical efforts at times when they are doing wound assessments, and during the final charting at the end of their work shift.  During the wound assessment the items included in the wound assessment page also function as reminders to what nurses need to document, and these would be items that they update at the end of their shift.  It was commonly noted that it becomes difficult to remember what to assess for which patient once they have seen their first or second patient.  Just as well it was difficult to remember what were the assessments at the end of the work shift without having a wound assessment feature.  The summary page design feature was given positive feedback in this category for its ease of use at the beginning of the work shift.  Participants noted that the summary page provides an easy way to have a quick overview of their daily patient visits.  This was while the summary page was not perceived as useful during the patient visits.  The supply list was perceived as easy to use at the end of patient visits.  When updating the list of wound treatment items that the patient has at their home and the wound treatment supplies that they need for their next visit.  Use of drag and drop and checklist items made it easy for nurses to update the supply list with less effort.  The sticky multimedia also received positive feedback in this category.  The main  scenario at the endobservedappointminformatiwere not   C6.5.3In thpatient inthe most That wassupplies of the patThe nurses wmost of tsticky muother comeasy to aview of tand curresticky muin which the of the patie that at the eent.  Duringon that laterclear themeontrol of Sis category tformation, apositive fee, wound treathat were neient visits, afeedback recere confusedhe items as altimedia coponents success and mhe status of nt entries.  Pltimedia de sticky multnt visits.  Dnd of patien that time th would be us identified tatus he recordinnd how easdback for altment suppleded for thes discussed eived for th about the sccessible.  Wmponent, wch as the woodify.  Simithe patient iarticipants sign featureimedia was uring the int visits nurse sticky mused for the fin this categgs of the sesy it was to alowing nursies that wer next visit.  in the previe wound asstatus of somhile somehich was incund photos,larly, the sumnformation, who were m easy to navperceived toitial fieldwoes were typiltimedia alloinal chartingory for othesions were cccess and ches to access e at the patieThey were aous section.essment pae of the comnurses strugluded in the wound assemary pagewhere nurseore familiarigate.  They  lower the nrk, discussecally in a ruwed to cap at the end r design featoded for feange.  The the status ofnts’ homeslso able to m ge in this caponent of tgled with a wound assessments, an provided ans were able with the Anwere able tourses effortd in chaptersh to arrive ture importaof the work ures. edback abousupply list f wound trea and wound odify this stegory sugghis feature, tccessing andssment pagd the wound easy to acc to navigate droid platfo identify th in charting  3, it was to their nextnt patient shift.  Theret the status eature receivtment suppltreatment tatus at the ests that thohey perceiv modifying e, they perc care plan aess dashboaeasily to parm found the status of th154 was   of ed ies.  end ugh ed the eived s rd st e e  recorded suggests platform navigatio  C6.5.4Comessential assessmeshift.  Thlater finavisits arealso werewound asfeedbackfrom clincompletetheir varipatients’ complete   sticky multithat over timthey will han and modifompleteneplete and pepart of the fnt page wasis feature allize the patie done.  Parti reminders sessment ite received duical experts ness of the pous wound cbedside, or  informationmedia whene with the ive easier acication of pss of Chartrtinent capteedback for perceived alowed nursents’ electrocipants alsoto what infoms includedring the creon the reseaatient chartare activitieright after th that they n there were ncrease of fcess to the satients’ info ure of patien this categors useful durs to capture nic records w commentedrmation the  in this desation of the rch team.  C, and nursess.  Especiale patient viseed to updatchanges, in amiliarity oftatus of the prmation t informatioy.  Based oning the patiepatients’ infhen they a that the comnurse mightign feature. mapped desapturing thi were able toly capturingit, ensured te the patienta sticky pho other particatients’ infn during wo the feedbacnt visits andormation wre back at thponents of have neede These itemsign principles informatio use the wo the wound hat the parts’ electronito for examipants to thormation forund care ack received  at the end ohile visitinge office afte the wound d to capture were selecs and basedn was impound assessmassessment icipants wouc records. ple.  This e Android  better tivities was the wound f nurses wo the patientsr their patieassessment p.  There werted based on on feedbacrtant to ent page duitems at the ld have the 155 an rk  and nt age e 12  k ring   T6.5.5To fugiven to nwhich sm UsefIt It It It It EaseIt It It It It It TaskIt It It It It It I iI p The and scorereactionsgiven to uechnology rther evaluurses were aller scoresulness saves time increases proenhances effmakes job eais useful to jo of use is easy to leais easy to comis clear and uis flexible to iis easy to acqis easy to use-Technologyis wise is beneficial is valuable makes me femakes me femakes me fentend to use redict I woulquestionnair 1 to, extrem and the resusefulness, eAcceptanceate the usefuanalyzed us show higheTable  6ductivity. ectiveness sier b rn plete tasks nderstandabnteract with. uire skills   Fit el happy el positive el good it d use it es were in aely unlikellts of the daase of use,  Survey lness, ease ing descriptr satisfactio.1. Results of le  7 point Liky.  The questa analysis and task-tecof use, and tive statisticsn.  The questhe experimenMean (sd)4.91 (1.874.73 (1.854.82 (1.835.09 (1.645.36 (1.43 5.91 (0.945.55 (1.215.82 (0.755.45 (1.045.82 (1.175.73 (1.01 5.09 (1.645.73 (1.565.18 (1.835.27 (1.1)5.27 (1.015.45 (1.215.55 (1.045.73 (0.79ert scale.  Stionnaire refrom the prohnology fit iask-technol.  The resulttionnaire is tal prototyping ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) core 7 was asponses are totyping sestems indicaogy fit the qs are shownincluded in  survey ssigned to, aligned withsions.  The tes higher inuestionnaire in table 6.1Appendix Dextremely li the nurses’positive scotention to a156 s in .  kely,  res dopt 157  the prototype [24], [120].  The scores given to usefulness items are likely affected by timeliness issues, as discussed earlier in this section.  The ease of use scores indicate that even though nurses experienced issues related to view structure and status visibility during the prototyping sessions, they ultimately perceived the SuperNurse easy to use.  Similarly, the task-technology fit scores indicate that nurses perceive SuperNurse as a fit into their workflow.  6.6 Summary Nurses are the primary care providers for patients with chronic wounds.  However, their problem solving behaviours are understudied as a form of end-user feedback in the design of HIT.  This study used identified themes of workarounds as a source of feedback in the design of SuperNurse.  The designs were evaluated and refined in iterative experimental prototyping sessions, in which the reactions of interest and curiosity were essential in further refining the sticky multimedia and speech recognition.  The nurses were interested and curious about designs aligned with their workarounds, such as sticky multimedia, which offered flexibility and fit to workflow, and speech recognition which supported accurate documentation.  The nurses were frustrated when a feature, such as malfunction of speech recognition or wearable cameras, created a barrier to their work which indicated circumstances that challenge usefulness, ease of use, and task-technology fit.  Positively surprised reactions in cases of success for speech recognition and sticky multimedia indicated improvements in usefulness and ease of use.  Identification of workarounds and using them to inform design as a new approach resulted in innovative technology that fit the circumstances of nurses’ work, and the evaluation of it in experimental prototyping showed that the design informed by nurses workarounds addresses key aspects of technology acceptance.  The WhaWha  C6.6.1The may be trshows thtechnologand accocautious workarounurses into fit wordocumenfollowing art is already  Experlife se Usefut this dissert Timelcomplsystem Survelikely onsideratievidence preue for similat pervasiveies becomeunt for creatto not treat wnds to be in regards to kkflow.  Thetation applice the highligknown abouimental protttings. lness and eaation is addiness, lowereteness are  by nurse. y results sugto be adopteons for Nexsented herear nomadic  technologie more prevaion and use orkaroundsstrumental tey aspects o end result oation that dhted items t the topic.otyping is lise of use poing to the boing effort inkey factors igested that d by nursest Chapter indicates thwork enviros can be deslent in healtof workarou as user erroo the delivef usefulnessf the designoes not docuin this chaptkely to revesitively impdy of know charting, con the adoptidesign featu in real-life at workarounments in higned to suphcare, interands by end-rs or mistakry of care.  W and ease of informed byment all aner: al issues thaact the usersledge. ntrol of infoon of a mobres informesettings. nds are comealthcare.  Fport workarctive systemusers.  Howes.  In fact,orkaround use, such a workaround every patit users migh’ technologrmation staile wound dd by workarmon in nururthermore,ounds.  As  designers ever, they s during this s revealed ths timelinessds was a ment data thatt face in reay adoption.tus, and ocumentatioounds are veses’ work, w this study pervasive should expehould be study we foe needs of  and flexibilobile wound the electron158 l-n ry hich ct und ity  ic 159  medical record might require, but it rather supports timely and flexible collection of patient data, and note-taking for nomadic workers.  160  Chapter 7: Conclusion Informed by a literature review, 120 hours of fieldwork were completed accompanying 33 homecare nurses in several community healthcare units in Vancouver.  This study focused on home healthcare clinicians working with patients with chronic wounds who received care at their home, and included a series of user studies at community healthcare units to design, develop, and evaluate a patient documentation system that supports homecare nurses’ work. Analysis of the qualitative and the quantitative data collected during fieldwork confirmed that user adaptations, appropriations, preventions, and redundancies were used as workarounds to barriers in the adoption of technology.  Results of the fieldwork identified that a primary characteristic of home healthcare clinicians, such as homecare nurses, was being highly nomadic.  That is, they visit patients’ homes to carry out care activities, such as providing treatment, care planning, and patient documentation.  Design and development of interactive computer technologies for nomadic work is a recent and active research area.  Literature pointed to, and evidence from my fieldwork suggested that workarounds created by home healthcare clinicians should be used to inform the design and development of interactive computer technologies to support the specific challenges that they face.  Following this, a user-centred approach for the design of interactive systems informed by user workarounds was created.  In this approach the creative problems solving strategies of the users were used as feedback in the design new features, or to refine existing features for the development of a mobile homecare patient documentation system.  This included identification and classification of workarounds, workarounds’ conceptual modeling and mapping to design principles, creation of design features based on the mapped design principles, as well evaluation, and refinement of the design features.  Duricommunplans forout with quantitatwere carropportunof a mobtechnologinternatio 7.1 CoDurithese conpervasive  T7.1.1Duriworkarouworkarouworkarouuse, parause, and ng the user ity health un a series of ea large numbive data weried out.  Thities for inteile patient dy.  The resunally.   ntributionsng this studytributions w and ubiquihe Workarng the initiands, in ordends were usnds.  Thesellel system uadaptive resstudies, alonit managersxploratory aer of partice collected fis provided ractive comocumentatiolts of this re  several conere to the fitous computound Catel fieldwork r to later ideed to group categories ise, approprource use.  Dg with a tea, the author nd experimipants from rom 118 usecritical knowputer systemn applicatiosearch weretributions welds of huming. gories data were contify pattern the workaroncluded preiative systemuring the am of researcof this disseental prototycommunity rs across alledge abous in homecan that uses s reported, pere made toan-computerllected and s in these inunds in cate-emptive inf use, appronalysis to idhers and in rtation had sping sessiohealth unitsl phases of tt the informre settings, peech recogublished, an the body o interactionanalyzed to stances (chgories.  Theormation uspriative resoentify thesecoordinationet scope, obns which lat (n=27).  Quhis work, anation needs leading to tnition and wd awarded lf knowledge, health infoidentify insapter 3).  Th results were, preventaturce use, ad categories i with jectives, aner was carrialitative andd 27 sessionand he developmearable ocally as we.  Especiallrmatics, andtances of e patterns oe 7 categoriive informataptive systenitial topica161 d ed  s ent ll y  f es of ion m l  codes weliteratureidentifiedwere valito particia questiosecond itduring thcategory chapter 3limitationto assess workarouanswered  T7.1.2The adapted tworkarouworkaroupatterns, of the wore assigned .  As the ana.  This was dated in a fopants, whichn about wheem was anoeir work, anof the identi.  The identis to use patthese workands were id the first 3 she Workarconceptual mo create anond situationnd situationwhich wererk situationto the instanlysis progrerepeated untllow up stu included thther if they ther questiond the third ified workarfied workarient data durrounds and entified andecondary reound Situaodel from ther conceps and their a model.  Du identified in model.  Theces of workssed new thil team consdy (chapter ree items foagree that th about whetem was for ounds. The ounds were ing their visgenerate fee used to catesearch questtion Modelthe literaturetual model tttributes (chring the ada the initial f result was arounds.  Themes emergensus was a3).  In the for each of the identified ther if they hcomments. results of thhow homecits.  Usefulndback for dgorize workions listed i , known as hat was usedapter 4).  Tptation procieldwork, wthe adapted ese codes wed and the wchieved.  Tllow up stude workarounworkaroundave used th These iteme follow up are nurses wess and easesign.  The carounds intn 1.7.2. the work sit to identifyhis adapted ess attributeere assignedmodel whicere adoptedorkaround he workarouy a questiod categorie category ise category os were assigstudy were dorked aroune of use meaommon chao 7 categoriuation mode the most comodel was cs of the wor to the relevh had dimen from the categories wnd categorinnaire was gs.  One item common, thf workarounned for eachiscussed ind system sures were racteristics es.  These l [1], was  mmon alled the karound ant dimenssions and 162 ere es iven  was e d   used of ion  attributeswas repeThe allowed fand existthe most identifieddesign prThe they can were usethe fourth  T7.1.3Anomost commapped. design ofexploratousefulnescorresponcompreh that could dated and adjapplication or a mappining design pcommon wo during the inciples for workaroundbe mapped td to adapt th and fifth rehe Mappedther contribumon worka These desig the wound ry and expes and ease odence of daensiveness, escribe instusted until tof this conceg to be crearinciples frorkaround siinitial fieldwpatient docu situation mo design prie work situasearch ques Design Prtion made dround categn principlesdocumentatrimental prof use detaileta and realitone design pances of woeam consensptual modeted betweenm the literatuations werork, and thmentation aodel was thenciples.  Thtion model tions listed inciples uring this stories (chapt were informion prototyptotyping sesd in chaptery, two desigrinciple for rkarounds aus was achil to the colle the identifiture for usefe identifiedeir attributespplications  conceptuale most commto the workain 1.7.2. udy were ther 4).  Thereed by worke that later wsions.  They 4.  There wn principlesindividualizs workarouneved. cted data dued most comulness and e within the w were used from the lite descriptionon charactround situae mapped d were 9 desiarounds anas refined  were withiere two des for compleed care pland situationsring the initmon workaase of use. orkaround to map and rature.  for workaroeristics of wtion model. esign princign principled were usedand evaluaten the dimenign principlteness and ning, one d.  This proceial fieldworround situat In the mappcategories, acontextualizunds such torkarounds  These answples from ths that were  to create thd in the sions of es for esign princip163 ss k ions ing lso e hat ered e e le  for view design prThe dimensioin iteratiophase theThese quthe literaprototypiparticipanquestionnresults ofThe and onescommon design pr  K7.1.4The selected fdimensio(accuracyThe dimestructure aninciple for smapped desns of usefulns of explo participantestions incluture.  The reng phase a vts to measuaire also in this step werelevant des that addressattributes ofinciples.  They Dimensmapped desrom severalns included ), charting cnsions for ed status chanimplified daign principleness and easratory and es were askedded items asults of this alid and relre their perccluded itemsre discussedign principled usefulne the workarese answereions of Useign principle dimensionsthe followinompletenesase of use inge, two desta entry. s were usede of use.  Thxperimental questions about the dimstep were diable questioeived ease o about the d in chapter es were idenss and ease oound situatiod the sixth fulness ands from the m of usefulneg for usefuls and comprcluded viewign principl to create dee design fe prototypingbout the easensions of uiscussed in cnnaire fromf use and uimensions o6. tified from tf use were ns were useresearch que Ease of Usost commoss and ease ness, corresehensivene structure aes for controsign featureatures were .  During the of use andsefulness ahapter 5.  In the literatusefulness of f usefulnesshe literatureselected for d to contexstion listed e n categorieof use withipondence bess, and indivnd status chl of currents that aimedfurther refine explorator usefulnessnd ease of u the experimre was giventhe prototyp and ease of by carryingthe mappingtualize and min 1.7.2. s of workaron TAM [41]tween data idualized caanges, contr status, and  to improveed and evaluy prototypin of the protose adopted fental  to the e.  The  use.  The  out a revie.  The mostap the releunds were .  These and reality re planningol of curren164 one  the ated g type.  rom w  vant .  t 165  status, and simplified data entry (chapter 4).  Design principles within these dimensions were used to create the prototype that was refined and evaluated in exploratory prototyping.  During the exploratory prototyping phase additional dimensions received more feedback and were suggested as more important by the nurses who participated in the study (chapter 5).  The dimensions that users suggested as important during the creation of the mapped design features for usefulness were timeliness, and accuracy, and for ease of use, lowering effort, and the fit workflow.  The prototype was refined based on these dimensions in the received feedback.  During the experimental prototyping phase, a mix of the aforementioned dimensions of usefulness and ease of use received feedback that suggested nurses perceive them as important (chapter 6).  These dimensions for usefulness were timeliness and completeness, and for ease of use they were lowering effort and control of status.  These answered the seventh and the primary research questions listed in 1.7.2. These dimensions were suggested by the collected and analyzed data to be relevant and key dimensions of usefulness and ease of use needed to create a user-centred approach in the design of interactive systems using workarounds, with wound documentation applications in home and community healthcare as a case.  Both qualitative and quantitative data were collected and analyzed to validate the use of these key dimensions in the design of the wound care documentation application. The qualitative data were collected in various forms such as audio, video, notes, artifacts, free form questions, and photos.  These data were analyzed by a team of researchers as described in chapters 3, 4, 5, 6.  Analysis of the qualitative data included topical coding and theme analysis, and for phases that involved analysis by the research team, the analysis was continued until team consensus was achieved.  Quantitative data were collected in the form of questionnaires.  These questionnaires were given to large numbers of participants to 166  triangulate data sources and increase validity of results.  The results of these questionnaires were analyzed using descriptive statistics and they were discussed in chapters 3, and 6.  7.2 Reflections Since parts of this work involved observations of homecare nurses work it is important to note the impact this observation had on the author of this dissertation as well on other collaborators who contributed to the data collection and analysis during this work [122].  Before the start of this work the author had done research in healthcare settings; however that was in hospital settings with physicians and surgeons.  Past exposure and research in healthcare provided the necessary knowledge and experience to identify and record elements of clinical work during observations.  The observations themselves added to that knowledge and experience, and as a result the author became more familiar with the work of homecare nurses, especially their wound care work.   At the same time two members of the research team (L. B. and C. R.) who contributed to data collection and analysis during prototyping were nurses who were completing their graduate degrees.  This resulted in balancing perspectives in data collection and analysis.  Of those two team members one (L. B.) had previous knowledge and experience with wound care and nursing.  The various levels of knowledge and experience with the context of the study benefited the results in that analysis was done with more than one perspective to reduce bias.  The results were more reliable since consensus was achieved during data analysis. 167   7.3 Limitations This dissertation was meant to study the research questions detailed in chapter 1 in the context of home and community healthcare in Vancouver.  There were a few limitations to the study presented in this dissertation.  As one can expect there are limitation to the conclusions made from this study.  While the sample size for the initial fieldwork and the validation phase were substantial, the sample size for the exploratory and the experimental prototyping were less than the other phases.  The main reason was that it was not possible to schedule sessions with many nurses who were needed by patients.  The community healthcare unit managers were very considerate and generous to allocate time for nurses to participate in the study in the numbers that they did.  Larger number of participants can be attained by longer term studies which were not feasible with the timeline and resources available to this study.  The demographics of the studies are another limiting factor.  Nursing by nature is mostly a female dominated profession, even though there were male participants in the study iterations but a larger number can address this limitation.  Other demographic factors such as age and level of experience with computer technology was varied, but it is possible that younger and better computer technology skilled participants were advantaged. The feedback received from nurses during the prototyping sessions was based on a simulated patient scenario, questions asked by the researchers, and what participants anticipated about the real-life use of the design features.  With a real-life deployment of the prototype nurse will be able to provide feedback about ease of use and usefulness of the prototype in uncontrolled conditions involving real patients.  It should be noted that a real-life deployment can face many more difficulties since it can impact the lives of patients. 168   7.4 Future Work Future research to this dissertation should further explore the role of user-centred approaches in the design and implementation of new technologies, especially in healthcare and other similar settings, particularly use of workarounds in user-centred methodologies.  While workarounds can be risks but they also can be opportunities for improved designs. The aim of this work and future work is to identify workarounds that can be used in system improvements and use them in design.  This can possibly lead to reducing workarounds that are risks to patients or users, and support workarounds that improve usefulness and ease of use.  Building on the research findings presented in this dissertation, future research questions that need to be investigated are as follows:   How to use workarounds to inform design decisions in fields other than healthcare?  What are the types and frequencies of nurses’ workarounds in other healthcare settings (i.e., examine if the workaround classification is consistent in other settings)?  How to design and develop interactive computer technologies in nonconventional environments, such as nomadic work environments?  How users create and use problem solving strategies, such as workarounds, in the adoption of technology?  What are the key factors that affect adoption of technology in nonconventional type of work?  What are the adaptive, appropriative, preventative, and redundant work patterns that users create to overcome barrier to technology adoption? 169   How user-centred approaches can be used in the design and development of novel technologies, such as wearables, internet of things systems, and ubiquitous computing systems?  How these user-centred approaches can be used to make progress towards smart communities (e.g.  cities, districts, buildings, homes) with a healthy amount of skepticism towards what technology can realistically achieve? Within cities future research to this dissertation also should focus specifically on marginalized and vulnerable communities in larger cities that also face challenges in access to services.  For example, the population receiving care from the community health unit in Downtown Eastside of Vancouver, and the populations in subsidized housing units across metro Vancouver area who receive care from other community health units, both of which were included in my past research.  User-centred approaches can benefit how these communities access services.  Future research should explore the role user-centred approaches can have in solving the socio-technical problems in the aforementioned communities.  7.5 Concluding Remarks This research in home and community healthcare suggested that involving the end-users in the design and implementation of computer systems positively impacts technology adoption, and possibly benefits patient outcomes.  Evidence from the literature suggested that problem solving strategies of users in form of workarounds are important user behaviour that should be included in the design framework.  The results in this dissertation, consistent with the literature, demonstrated that in the current implementations, especially in healthcare systems, end-users feedback in the form of workarounds is often not taken into account while designing and 170  developing computer systems.  As new technologies such as ubiquitous computing and internet of things are introduced, there remains a critical gap wherein the role of user-centred approaches are not being explored and clarified in the implementation of these technologies.  This dissertation identified, articulated, categorized, and mapped important user behaviours known as workarounds to design principles.  These design principles improved ease of use and usefulness of a mobile wound documentation application.  The methodology used for this work and the results can be applied and used in similar contexts where users carry out mission critical work while being mobile outside of the office environment.  171  References [1] L. Gasser, “The integration of computing and routine work,” ACM Trans. Inf. Syst. TOIS, vol. 4, no. 3, pp. 205–225, 1986. [2] J. R. Halbesleben, D. S. Wakefield, and B. J. Wakefield, “Work-arounds in health care settings: literature review and research agenda,” Health Care Manage. Rev., vol. 33, no. 1, pp. 2–12, 2008. [3] M. Kobayashi, S. R. Fussell, Y. Xiao, and F. J. Seagull, “Work coordination, workflow, and workarounds in a medical context,” in CHI’05 Extended Abstracts on Human Factors in Computing Systems, 2005, pp. 1561–1564. [4] J. M. Morath and J. E. Turnbull, To do no harm. San Francisco, CA: Jossey-Bass, 2005. [5] G. Mark and N. M. Su, “Making infrastructure visible for nomadic work,” Pervasive Mob. Comput., vol. 6, no. 3, pp. 312–323, 2010. [6] J. Anacleto and S. Fels, “Adoption and Appropriation: a design process from HCI research at a Brazilian neurological hospital,” in Human-Computer Interaction–INTERACT 2013, 2013, pp. 356–363. [7] J. E. Bardram and C. Bossen, “Mobility Work: The Spatial Dimension of Collaboration at a Hospital,” Comput. Support. Coop. Work CSCW, vol. 14, no. 2, pp. 131–160, Apr. 2005. [8] J. E. Bardram and C. Bossen, “A web of coordinative artifacts: collaborative work at a hospital ward,” in Proceedings of the 2005 international ACM SIGGROUP conference on Supporting group work, 2005, pp. 168–176. [9] J. A. Kientz, “Embedded capture and access: encouraging recording and reviewing of data in the caregiving domain,” Pers. Ubiquitous Comput., vol. 16, no. 2, pp. 209–221, Feb. 2012. [10] M. E. Flanagan, J. J. Saleem, L. G. Millitello, A. L. Russ, and B. N. Doebbeling, “Paper- and computer-based workarounds to electronic health record use at three benchmark institutions,” J. Am. Med. Inform. Assoc., vol. 20, no. e1, pp. e59–e66, Jun. 2013. [11] J. Clemensen and S. B. Larsen, “Cooperation versus coordination: using real-time telemedicine for treatment at home of diabetic foot ulcers,” J. Telemed. Telecare, vol. 13, no. suppl 3, pp. 32–35, 2007. [12] J. Bardram and A. Doryab, “Activity analysis: applying activity theory to analyze complex work in hospitals,” in Proceedings of the ACM 2011 conference on Computer supported cooperative work, 2011, pp. 455–464. [13] D. W. Dowding et al., “International priorities for research in nursing informatics for patient care,” 2013. [14] Canadian Home Care Association, “Home care in Canada: Advancing quality improvement and integrated care,” Canadian Home Care Association, 2015. [15] C. Chanussot-Deprez and J. Contreras-Ruiz, “Telemedicine in wound care: A review,” Adv. Skin Wound Care, vol. 26, no. 2, pp. 78–82, 2013. [16] Wound Care Alliance Canada, “Pre-budget consultations 2012,” 2012. [17] K. Järbrink et al., “Prevalence and incidence of chronic wounds and related complications: a protocol for a systematic review,” Syst. Rev., vol. 5, no. 1, p. 152, Dec. 2016. [18] K. Denny, L. Christina, and S. D. Perry, “Compromised Wounds in Canada,” Healthc. Q., vol. 17, no. 1, pp. 7–10, May 2014. [19] M. G. Woodbury and P. E. Houghton, “Prevalence of pressure ulcers in Canadian 172  healthcare settings,” Ostomy Wound Manag., vol. 50, pp. 22–39, 2004. [20] M. G. Woodbury and P. E. Houghton, “The extent of chronic wounds in Canada,” Wound Care Can., vol. 3, no. 1, p. 18, 2005. [21] “Statistics - Canadian Association of Wound Care.” [Online]. Available: http://cawc.net/index.php/public/facts-stats-and-tools/statistics/. [Accessed: 14-Nov-2014]. [22] E. Mossialos, M. Wenzl, R. Osborn, and C. Anderson, “2015 International Profiles of Health Care Systems,” 2015. [23] L. V. Nordheim, M. T. Haavind, and M. M. Iversen, “Effect of telemedicine follow-up care of leg and foot ulcers: a systematic review,” BMC Health Serv. Res., vol. 14, no. 1, p. 565, 2014. [24] R. J. Holden and B.-T. Karsh, “The Technology Acceptance Model: Its past and its future in health care,” J. Biomed. Inform., vol. 43, no. 1, pp. 159–172, Feb. 2010. [25] J. Clemensen, S. B. Larsen, and J. E. Bardram, “Developing pervasive e-health for moving experts from hospital to home,” in Proceedings of the IADIS e-Society Conference, 2004, pp. 441–448. [26] J. Clemensen, S. B. Larsen, M. Kyng, and M. Kirkevold, “Participatory Design in Health Sciences: Using Cooperative Experimental Methods in Developing Health Services and Computer Technology,” Qual. Health Res., vol. 17, no. 1, pp. 122–130, Jan. 2007. [27] B. Mapar, Y. Lam, A. Mehrnia, B. Bates-Jensen, M. Sarrafzadeh, and W. J. Kaiser, “Wearable sensor for continuously vigilant spatial and depth-resolved perfusion imaging,” in Proceedings of the conference on Wireless Health, 2012, p. 15. [28] P. Koopman and R. R. Hoffman, “Work-arounds, make-work, and kludges,” Intell. Syst. IEEE, vol. 18, no. 6, pp. 70–75, 2003. [29] F. Cabitza and C. Simone, “‘Drops Hollowing the Stone’: Workarounds as Resources for Better Task-Artifact Fit,” in ECSCW 2013: Proceedings of the 13th European Conference on Computer Supported Cooperative Work, 21-25 September 2013, Paphos, Cyprus, O. W. Bertelsen, L. Ciolfi, M. A. Grasso, and G. A. Papadopoulos, Eds. London: Springer London, 2013, pp. 103–122. [30] G. Litzinger, T. Rossman, B. Demuth, and J. Roberts, “In-Home Wound Care Management Utilizing Information Technology,” Home Healthc. Nurse, vol. 25, no. 2, pp. 119–130, 2007. [31] D. Visco, “Use of telehealth for chronic wound care: A case study,” J. WOCN, vol. 28, no. 2, pp. 89–95, Mar. 2001. [32] J. White, “Telewound Care Network.” [33] J. R. B. Halbesleben, C. Rathert, and S. F. Bennett, “Measuring Nursing Workarounds: Tests of the Reliability and Validity of a Tool,” JONA J. Nurs. Adm., vol. 43, no. 1, pp. 50–55, Jan. 2013. [34] L. L. Novak, R. J. Holden, S. H. Anders, J. Y. Hong, and B.-T. Karsh, “Using a sociotechnical framework to understand adaptations in health IT implementation,” Int. J. Med. Inf., vol. 82, no. 12, pp. e331–e344, Dec. 2013. [35] P. Rawstorne, R. Jayasuriya, and P. Caputi, “Issues in predicting and explaining usage behaviors with the technology acceptance model and the theory of planned behavior when usage is mandatory,” in Proceedings of the twenty first international conference on Information systems, 2000, pp. 35–44. [36] F. K. Chan, J. Y. Thong, V. Venkatesh, S. A. Brown, P. J. Hu, and K. Y. Tam, “Modeling 173  citizen satisfaction with mandatory adoption of an e-government technology,” J. Assoc. Inf. Syst., vol. 11, no. 10, pp. 519–549, 2010. [37] V. Venkatesh, S. A. Brown, L. M. Maruping, and H. Bala, “Predicting different conceptualizations of system use: the competing roles of behavioral intention, facilitating conditions, and behavioral expectation,” Mis Q., pp. 483–502, 2008. [38] F. D. Davis, “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology,” MIS Q., vol. 13, no. 3, p. 319, Sep. 1989. [39] I. Ajzen and M. Fishbein, “Attitude-Behavior Relations: A Theoretical Analysis and Review of Empirical Research,” Psychol. Bull., vol. 84, no. 5, p. 31, 1977. [40] I. Ajzen, “Behavioral Interventions Based on the Theory of Planned Behavior,” Action Control, p. 6, 1985. [41] N. Marangunić and A. Granić, “Technology acceptance model: a literature review from 1986 to 2013,” Univers. Access Inf. Soc., vol. 14, no. 1, pp. 81–95, Mar. 2015. [42] P. Vassilakopoulou, V. Tsagkas, and N. Marmaras, “Workaround identification as an instrument for work analysis and design: a case study on ePrescription,” Work J. Prev. Assess. Rehabil., vol. 41, pp. 1805–1810, 2012. [43] A. L. Strauss, Ed., Social organization of medical work. New Brunswick, N.J., U.S.A: Transaction Publishers, 1997. [44] J. E. Bardram, S. Jeuris, and S. Houben, “Activity-based computing: computational management of activities reflecting human intention,” Ai Mag., vol. 36, no. 2, pp. 63–72, 2015. [45] J. R. Halbesleben, G. T. Savage, D. S. Wakefield, and B. J. Wakefield, “Rework and workarounds in nurse medication administration process: implications for work processes and patient safety,” Health Care Manage. Rev., vol. 35, no. 2, pp. 124–133, 2010. [46] C. Floyd, “A systematic look at prototyping,” in Approaches to prototyping, Springer, 1984, pp. 1–18. [47] W. R. Bischofberger and G. Pomberger, “Prototyping-Oriented Software Development: Concepts and Tools,” 2011. [48] L. Berkowitz and C. McCarthy, Eds., Innovation with Information Technologies in Healthcare. London: Springer London, 2013. [49] R. Calderon, S. Fels, J. Leite de Oliveira, and J. Anacleto, “Understanding NUI-supported Nomadic Social Places in a Brazilian Health Care Facility,” presented at the Brazilian Symposium on Human Factors in Computing Systems, Porto Alegre, Brazil, 2012, pp. 76–84. [50] T. Andersen, F. Kensing, L. Kjellberg, and J. Moll, “From research prototypes to a marketable eHealth system,” Context Sensitive Health Inform. Many Places Many Users Many Contexts Many Uses, vol. 218, p. 9, 2015. [51] J. E. Bardram, “Clinical Proof-of-Concept–A Evaluation Method for Pervasive Healthcare Systems,” Ubiquitous Comput., p. 51, 2008. [52] “Interactive Wireframe Software & Mockup Tool - Axure.” [Online]. Available: http://www.axure.com/. [Accessed: 29-Mar-2015]. [53] K. A. Ericsson and H. A. Simon, Protocol analysis. MIT press, 1985. [54] A. L. Strauss, Ed., Social organization of medical work. Chicago: University of Chicago Press, 1985. [55] A. Strauss, “Work and the division of labor,” Sociol. Q., vol. 26, no. 1, pp. 1–19, 1985. 174  [56] J. M. Corbin and A. L. Strauss, “The articulation of work through interaction,” Sociol. Q., vol. 34, no. 1, pp. 71–83, 1993. [57] R. Koppel, T. Wetterneck, J. L. Telles, and B.-T. Karsh, “Workarounds to Barcode Medication Administration Systems: Their Occurrences, Causes, and Threats to Patient Safety,” J. Am. Med. Inform. Assoc., vol. 15, no. 4, pp. 408–423, Apr. 2008. [58] J. Scalise, “Working THROUGH the workaround: FIX why your nurses can’t adopt Health Information Technology and STOP why they won’t,” in ANA 7th Annual Nursing Quality ConferenceTM, 2013. [59] A. A. Vogelsmeier, J. R. B. Halbesleben, and J. R. Scott-Cawiezell, “Technology Implementation and Workarounds in the Nursing Home,” J. Am. Med. Inform. Assoc., vol. 15, no. 1, pp. 114–119, Oct. 2007. [60] X. Zhou, M. Ackerman, and K. Zheng, “CPOE workarounds, boundary objects, and assemblages,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2011, pp. 3353–3362. [61] S. A. Collins, M. Fred, L. Wilcox, and D. K. Vawdrey, “Workarounds used by nurses to overcome design constraints of electronic health records,” in NI 2012: Proceedings of the 11th International Congress on Nursing Informatics, 2012, vol. 2012. [62] E. H. Ferneley and P. Sobreperez, “Resist, comply or workaround? An examination of different facets of user engagement with information systems,” Eur. J. Inf. Syst., vol. 15, no. 4, pp. 345–356, Aug. 2006. [63] D. S. Debono et al., “Nurses’ workarounds in acute healthcare settings: a scoping review,” BMC Health Serv. Res., vol. 13, no. 1, p. 175, 2013. [64] V. Venkatesh and F. D. Davis, “A theoretical extension of the technology acceptance model: Four longitudinal field studies,” Manag. Sci., vol. 46, no. 2, pp. 186–204, 2000. [65] V. Venkatesh, M. G. Morris, G. B. Davis, and F. D. Davis, “User acceptance of information technology: Toward a unified view,” MIS Q., pp. 425–478, 2003. [66] V. Venkatesh and H. Bala, “Technology acceptance model 3 and a research agenda on interventions,” Decis. Sci., vol. 39, no. 2, pp. 273–315, 2008. [67] V. Venkatesh, J. Y. Thong, and X. Xu, “Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology,” MIS Q., vol. 36, no. 1, pp. 157–178, 2012. [68] B. Dj, V. S. P, S. Ds, and C. Wa, “Evaluating a Spoken Dialogue System for recording clinical observations during an endoscopic examination,” Inform. Health Soc. Care, vol. 28, no. 2, pp. 85–97, Jan. 2003. [69] P. J. Hu, P. Y. Chau, O. R. L. Sheng, and K. Y. Tam, “Examining the technology acceptance model using physician acceptance of telemedicine technology,” J. Manag. Inf. Syst., pp. 91–112, 1999. [70] P. Van Schaik, J. A. Bettany-Saltikov, and J. G. Warren, “Clinical acceptance of a low-cost portable system for postural assessment,” Behav. Inf. Technol., vol. 21, no. 1, pp. 47–57, Jan. 2002. [71] D. R. Berlowitz, C. Ratliff, J. Cuddigan, G. T. Rodeheaver, and others, “The PUSH tool: a survey to determine its perceived usefulness,” Adv. Skin Wound Care, vol. 18, no. 9, pp. 480–483, 2005. [72] D. E. McDowell, T. W. Dillon, and D. Lending, “Perceived quality benefits influenced by usefulness and documentation accuracy of information systems,” Comput. Inform. Nurs., 175  vol. 26, no. 6, pp. 350–357, 2008. [73] J.-H. Wu, S.-C. Wang, and L.-M. Lin, “Mobile computing acceptance factors in the healthcare industry: A structural equation model,” Int. J. Med. Inf., vol. 76, no. 1, pp. 66–77, Jan. 2007. [74] W. Maass and U. Varshney, “Design and evaluation of Ubiquitous Information Systems and use in healthcare,” Decis. Support Syst., vol. 54, no. 1, pp. 597–609, Dec. 2012. [75] T. B. Iversen, L. Melby, A. D. Landmark, and P. Toussaint, “Managing variations from surgical care plans: Challenges for coordination,” Int. J. Med. Inf., vol. 82, no. 1, pp. 47–57, Jan. 2013. [76] J. Kjeldskov and M. B. Skov, “Exploring context-awareness for ubiquitous computing in the healthcare domain,” Pers. Ubiquitous Comput., vol. 11, no. 7, pp. 549–562, Oct. 2007. [77] C. Bossen and L. W. K. Jensen, “Implications of Shared Interactive Displays for Work at a Surgery Ward: Coordination, Articulation Work and Context-Awareness,” 2008, pp. 464–469. [78] M. Bang and T. Timpka, “Ubiquitous computing to support co-located clinical teams: Using the semiotics of physical objects in system design,” Int. J. Med. Inf., vol. 76, pp. S58–S64, Jun. 2007. [79] G. Marcu et al., “Why do they still use paper?: Understanding data collection and use in autism education,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2013, pp. 3177–3186. [80] J. M. Levine and E. Totolos, “Pressure ulcers: a strategic plan to prevent and heal them.,” Geriatrics, vol. 50, no. 1, pp. 32–37; quiz 38–9, 1995. [81] A. Ehrenberg and C. Birgersson, “Nursing documentation of leg ulcers: adherence to clinical guidelines in a Swedish primary health care district,” Scand. J. Caring Sci., vol. 17, no. 3, pp. 278–284, 2003. [82] S. Sendelbach and M. Funk, “Alarm Fatigue: A Patient Safety Concern,” AACN Adv. Crit. Care, vol. 24, no. 4, pp. 378–386, 2013. [83] C. Lalley, “Workarounds and Obstacles: Unexpected Source of Innovation,” Nurs. Adm. Q., vol. 38, no. 1, pp. 69–77, 2014. [84] aLoggers, “aTimeLogger 2.” [Online]. Available: http://www.atimelogger.com/. [Accessed: 11-Mar-2015]. [85] QSR International, “NVivo 10.” [Online]. Available: http://www.qsrinternational.com/products_nvivo.aspx. [Accessed: 11-Mar-2015]. [86] V. Venkatesh, S. A. Brown, and H. Bala, “Bridging the qualitative-quantitative divide: Guidelines for conducting mixed methods research in information systems,” Mis Q., vol. 37, no. 1, pp. 21–54, 2013. [87] Free Software Foundation, “GNU PSPP.” [Online]. Available: https://www.gnu.org/software/pspp/. [Accessed: 01-Nov-2016]. [88] Government of British Columbia, “City of Vancouver’s Map,” Government of British Columbia, 2016. [89] M. M. Quinn et al., “Occupational health of home care aides: results of the safe home care survey,” Occup. Environ. Med., vol. 73, no. 4, pp. 237–245, 2016. [90] E. E. M. Maurits, A. J. E. de Veer, P. P. Groenewegen, and A. L. Francke, “Dealing with professional misconduct by colleagues in home care: a nationwide survey among nursing staff,” BMC Nurs., vol. 15, no. 1, Dec. 2016. 176  [91] A. Dix, “Designing for appropriation,” in Proceedings of the 21st British HCI Group Annual Conference on People and Computers: HCI... but not as we know it-Volume 2, 2007, pp. 27–30. [92] S. Y. Park, Y. Chen, and S. Rudkin, “Technological and Organizational Adaptation of EMR Implementation in an Emergency Department,” ACM Trans. Comput.-Hum. Interact., vol. 22, no. 1, pp. 1–24, Feb. 2015. [93] S. Bødker, “Third-wave HCI, 10 years later—participation and sharing,” interactions, vol. 22, no. 5, pp. 24–31, 2015. [94] J. R. B. Halbesleben, C. Rathert, and E. S. Williams, “Emotional exhaustion and medication administration work-arounds: The moderating role of nurse satisfaction with medication administration,” Health Care Manage. Rev., vol. 38, no. 2, pp. 95–104, 2013. [95] R. Budde, K. Kuhlenkamp, L. Mathiassen, and H. Züllighoven, Eds., Approaches to Prototyping. Berlin, Heidelberg: Springer Berlin Heidelberg, 1984. [96] J. W. Creswell, Research design: Qualitative, quantitative, and mixed methods approaches. Sage, 2013. [97] J. Nielsen, Ten usability heuristics. http://www. nngroup. com/articles/ten-usability-heuristics/(acc-essed December 19, 2013), 2005. [98] K. Bumpus and M. A. Maier, “The ABC’s of Wound Care,” Curr. Cardiol. Rep., vol. 15, no. 4, Apr. 2013. [99] E. O. Siegel, H. M. Young, L. Zysberg, and V. Santillan, “Securing and Managing Nursing Home Resources: Director of Nursing Tactics,” The Gerontologist, vol. 55, no. 5, pp. 748–759, Oct. 2015. [100] E. Papastavrou, P. Andreou, and G. Efstathiou, “Rationing of nursing care and nurse-patient outcomes: a systematic review of quantitative studies: RATIONING OF NURSING CARE,” Int. J. Health Plann. Manage., vol. 29, no. 1, pp. 3–25, Jan. 2014. [101] S. Jerak-Zuiderent, “Accountability from Somewhere and for Someone: Relating with Care,” Sci. Cult., vol. 24, no. 4, pp. 412–435, Oct. 2015. [102] A. Thoroddsen, G. Sigurjónsdóttir, M. Ehnfors, and A. Ehrenberg, “Accuracy, completeness and comprehensiveness of information on pressure ulcers recorded in the patient record: Accuracy, completeness and comprehensiveness of information on pressure ulcers,” Scand. J. Caring Sci., vol. 27, no. 1, pp. 84–91, Mar. 2013. [103] M. M. Shepherd, D. D. Wipke-Tevis, and G. L. Alexander, “Analysis of Qualitative Interviews About the Impact of Information Technology on Pressure Ulcer Prevention Programs: Implications for the Wound, Ostomy and Continence Nurse,” J. Wound. Ostomy Continence Nurs., vol. 42, no. 3, pp. 235–241, 2015. [104] C. T. Hess, “The art of skin and wound care documentation,” Home Healthc. Nurse, vol. 23, no. 8, pp. 502–513, 2005. [105] E. N. Munyisia, P. Yu, and D. Hailey, “Does the introduction of an electronic nursing documentation system in a nursing home reduce time on documentation for the nursing staff?,” Int. J. Med. Inf., vol. 80, no. 11, pp. 782–792, Nov. 2011. [106] M. Fossum, M. Ehnfors, E. Svensson, L. M. Hansen, and A. Ehrenberg, “Effects of a computerized decision support system on care planning for pressure ulcers and malnutrition in nursing homes: an intervention study,” Int. J. Med. Inf., vol. 82, no. 10, pp. 911–921, 2013. [107] N. Wang, D. Hailey, and P. Yu, “Quality of nursing documentation and approaches to its 177  evaluation: a mixed-method systematic review: Quality of nursing documentation and audit,” J. Adv. Nurs., vol. 67, no. 9, pp. 1858–1875, Sep. 2011. [108] E. R. Gjevjon and R. Hellesø, “The quality of home care nurses’ documentation in new electronic patient records,” J. Clin. Nurs., vol. 19, no. 1–2, pp. 100–108, Jan. 2010. [109] M. Fossum, G. L. Alexander, K. E. Göransson, M. Ehnfors, and A. Ehrenberg, “Registered nurses’ thinking strategies on malnutrition and pressure ulcers in nursing homes: a scenario-based think-aloud study: Nurses’ thinking strategies,” J. Clin. Nurs., vol. 20, no. 17–18, pp. 2425–2435, Sep. 2011. [110] E. Törnvall, L. K. Wahren, and S. Wilhelmsson, “Advancing nursing documentation—An intervention study using patients with leg ulcer as an example,” Int. J. Med. Inf., vol. 78, no. 9, pp. 605–617, Sep. 2009. [111] K. M. Romagnoli, S. M. Handler, F. M. Ligons, and H. Hochheiser, “Home-care nurses’ perceptions of unmet information needs and communication difficulties of older patients in the immediate post-hospital discharge period,” BMJ Qual. Saf., vol. 22, no. 4, pp. 324–332, Apr. 2013. [112] L. Gunningberg, M. Fogelberg-Dahm, and A. Ehrenberg, “Improved quality and comprehensiveness in nursing documentation of pressure ulcers after implementing an electronic health record in hospital care,” J. Clin. Nurs., vol. 18, no. 11, pp. 1557–1564, Jun. 2009. [113] P. Chang, C.-L. Hsu, Y. Mei Liou, Y.-Y. Kuo, and C.-F. Lan, “Design and Development of Interface Design Principles for Complex Documentation Using PDAs:,” CIN Comput. Inform. Nurs., p. 1, Oct. 2010. [114] M. F. Andresen, A.-S. H. Henriksen, M. D. Johansen, and O. K. Hejlesen, “A Decision Support and Documentation System for Treatment of Chronic Ulcers,” p. 5. [115] M. L. Rogers, P. S. Sockolow, K. H. Bowles, K. E. Hand, and J. George, “Use of a human factors approach to uncover informatics needs of nurses in documentation of care,” Int. J. Med. Inf., vol. 82, no. 11, pp. 1068–1074, Nov. 2013. [116] Z. J. Nagykaldi, M. Jordan, J. Quitoriano, C. A. Ciro, and J. W. Mold, “User-Centered Design and Usability Testing of an Innovative Health-Related Quality of Life Module:,” Appl. Clin. Inform., vol. 5, no. 4, pp. 958–970, 2014. [117] J. Favela, M. Tentori, and V. M. Gonzalez, “Ecological Validity and Pervasiveness in the Evaluation of Ubiquitous Computing Technologies for Health Care,” Int. J. Hum.-Comput. Interact., vol. 26, no. 5, pp. 414–444, Apr. 2010. [118] G. Aldaz et al., “Hands-Free Image Capture, Data Tagging and Transfer Using Google Glass: A Pilot Study for Improved Wound Care Management,” PLOS ONE, vol. 10, no. 4, pp. 3–21, Apr. 2015. [119] S. Coleman et al., “Development of a generic wound care assessment minimum data set,” J. Tissue Viability, vol. 26, no. 4, pp. 226–240, Nov. 2017. [120] T.-C. Lin, “Mobile Nursing Information System Utilization: The Task-Technology Fit Perspective,” CIN Comput. Inform. Nurs., vol. 32, no. 3, pp. 129–137, Mar. 2014. [121] N. Clynch and J. Kellett, “Medical documentation: Part of the solution, or part of the problem? A narrative review of the literature on the time spent on and value of medical documentation,” Int. J. Med. Inf., vol. 84, no. 4, pp. 221–228, Apr. 2015. [122] R. Berger, “Now I see it, now I don’t: researcher’s position and reflexivity in qualitative research,” Qual. Res., vol. 15, no. 2, pp. 219–234, Apr. 2015. 178  Appendix A:  Demographics of the Ethnographic Study  Survey RespondentsN(%) The Remainder N(%) Sex    Female 10(%83) 20(%95)  Male 2(%17) 1(%5) Age group    21-30 1(%8) 3(%14)  31-40 3(%25) 8(%38)  41-50 4(%33) 6(%29)  51-60 3(%25) 3(%14)  61-70 1(%8) 1(%5) Education    Certificate 3(%25) 2(%10)  Diploma 3(%25) 5(%24)  B.Sc. 4(%33) 13(%62)  M.Sc. 2(%17) 1(%5) Profession    RN 8(%67) 4(%19)  LPN 4(%33) 17(%81) Nursing experience    Average 13.40 years 13.7 years  Std. 10.13 10.02 Community health experience    Average 5.7 years 6.3 years  Std. 5.61 6.8 Computer use outside of work    Several times/day 7(%58) 17(%81)  Once/day 3(%25) 2(%9)  Several times/week 2(%17) 1(%5)  Never 0(%0) 1(%5) Demographics of the ethnographic study (RN= Registered Nurse, LPN= Lenience Practice Nurse).    179  Appendix B:  Demographics of the Validation Study  Survey Respondents N(%) Sex   Female 50(%86)  Male 8(%14) Age group   21-30 12(%3)  31-40 23(%14)  41-50 12(%21)  51-60 8(%40)  61-70 2(%21) Education   Certificate 2(%3)  Diploma 10(%17)  B.Sc. 40(%69)  M.Sc. 6(%10) Profession   RN 50(%86)  LPN 8(%14) Nursing experience   Average 12.69 years  Std. 10.06 Community health experience   Average 4.68 years  Std. 5.17 Computer use outside of work   Several times/day 38(%66)  Once/day 11(%19)  Several times/week 6(%10)  Several times/month 2(%3)  Never 1(%2)    180  Appendix C: The Workaround Questionnaire  This survey was developed, validated, and used by Halbesleben et al. [33].  Strongly DisagreeDisagree Neutral Agree Strongly Agree Don’t Know1. Problems with technology prevent me from completing tasks as well as I would like to. 1 2 3 4 5 6 2. Problems with equipment prevent me from completing tasks as well as I would like to. 1 2 3 4 5 6 3. Rules or policies prevent me from completing tasks as well as I would like to. 1 2 3 4 5 6 4. Other people prevent me from completing tasks as well as I would like to. 1 2 3 4 5 6 5. Poorly designed work processes prevent me from completing tasks as well as I would like to. 1 2 3 4 5 6 6. I have to alter my work process because of problems with technology. 1 2 3 4 5 6 7. I have to alter my work process because of problems with equipment. 1 2 3 4 5 6 8. I have to alter my work process because rules or policies keep me from doing my job efficiently. 1 2 3 4 5 6 9. I have to alter my work process because other people keep me from doing my job efficiently. 1 2 3 4 5 6 10. I have to alter my work process because my work processes are not well designed. 1 2 3 4 5 6 11. When possible, I follow procedures regarding use of technology. 1 2 3 4 5 6 12. When possible, I follow procedures regarding use of equipment. 1 2 3 4 5 6 13. When possible, I follow rules and policies at work. 1 2 3 4 5 6 14. When possible, I follow intended work processes even when they are poorly designed. 1 2 3 4 5 6 15. When given the choice between following procedures or taking a shortcut, I prefer to follow procedures. 1 2 3 4 5 6 16. When I have to alter my work process because of problems with technology, I do so to better assist a patient. 1 2 3 4 5 6 17. When I have to alter my work process because of problems with equipment, I do so to better assist a patient. 1 2 3 4 5 6 18. When I have to alter my work process because rules or policies, I do so to better assist a patient. 1 2 3 4 5 6 19. When I have to alter my work process because other people keep me from doing my job, I do so to better assist a patient.1 2 3 4 5 6 20. When I have to alter my work process because my work processes are not well designed, I do so to better assist a patient. 1 2 3 4 5 6 181  Appendix D: The Experimental Prototyping Questionnaire A. Perceived usefulness Please mark X on the scales provided. 1. Using the speech recognition prototype in my job would enable me to accomplish tasks more quickly.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  2. Using the speech recognition prototype in my job increases my productivity.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  3. Using speech recognition prototype would enhance my effectiveness on the job.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  4. Using speech recognition prototype would make it easier to do my job.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  5. I would find the  speech recognition prototype useful in my job.  Likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  B. Perceived ease of use Please mark X on the scales provided. 1. Learning to operate speech recognition prototype would be easy for me.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  2. I would find it easy to get the speech recognition prototype to do what I want it to do.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  3. My interaction with the speech recognition prototype would be clear and understandable.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  4. I would find the speech recognition prototype to be flexible to interact with.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  5. It would be easy for me to become skilful at using the speech recognition prototype.  182  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  6. I would find the speech recognition prototype easy to use.  likely |_______|_______|_______|_______|_______|_______|_______| unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  C. Place an X mark on each of the scales. Attitude (cognitive) Neutral 1. The speech recognition prototype is a ______ instrument in performing my tasks: Wise |__|__|__|__|__|__|__| Foolish 2. The speech recognition prototype is a ______ instrument in performing my tasks: Beneficial |__|__|__|__|__|__|__| Unbeneficial 3. The speech recognition prototype is a ______  instrument in performing my tasks: Valuable |__|__|__|__|__|__|__| Worthless  Attitude (affective)  Neutral 4. Using the speech recognition prototype makes me feel _____: Happy |__|__|__|__|__|__|__| Annoyed 5. Using the speech recognition prototype makes me feel _____: Positive  |__|__|__|__|__|__|__| Negative 6. Using the speech recognition prototype makes me feel _____: Good |__|__|__|__|__|__|__| Bad  D. To what extent do you agree with the following?  Intention to use3 Strongly disagree Moderately disagree Somewhat disagree Neutral (neither disagree or agree) Somewhat agree Moderately Agree Strongly agree 1. Assuming I had access to the speech recognition prototype, I intend to use it. 1 2 3 4 5 6 7 2. Given that I had access to the speech recognition prototype, I predict that I would use it. 1 2 3 4 5 6 7  E. Fit between task and technology 1. Please tell us about how this technology might fit your work needs. ______________________________________________________________________________________________________________________________________________________  183  Appendix E: Demographics for the Exploratory Prototyping  Survey Respondents N(%) Sex   Female 14(%93)  Male 1(%7) Age group   21-30 5(%33)  31-40 5(%33)  41-50 2(%13)  51-60 3(%20)  61-70 0(%0) Education   Certificate 0(%0)  Diploma 4(%27)  B.Sc. 11(%73)  M.Sc. 0(%0) Profession   RN 13(%87)  LPN 2(%13) Nursing experience   Average 13 years  Std. 9.19 Community health experience   Average 6 years  Std. 5.01 Computer use outside of work   Several times/day 12(%80)  Once/day 3(%20)  Several times/week 0(%0)  Several times/month 0(%0)  Never 1(%0)     184  Appendix F: Demographics for the Experimental Prototyping  Survey Respondents N(%) Sex   Female 12(%100)  Male 0(%0) Age group   21-30 2(%17)  31-40 2(%17)  41-50 5(%42)  51-60 3(%25)  61-70 0(%0) Education   Certificate 0(%0)  Diploma 3(%25)  B.Sc. 8(%67)  M.Sc. 1(%8) Profession   RN 10(%83)  LPN 2(%17) Nursing experience   Average 15.75 years  Std. 11.11 Community health experience   Average 8.61 years  Std. 8.22 Computer use outside of work   Several times/day 10(%83)  Once/day 2(%17)  Several times/week 0(%0)  Several times/month 0(%0)  Never 1(%0)       185  Appendix G: Technology Acceptance Model and Its Variations The technology acceptance model (TAM) was developed, validated, and used by Davis [38].  A. Perceived usefulness Please mark X on the scales provided. 1. Using [the system] in my job would enable me to accomplish tasks more quickly.  Likely |_______|_______|_______|_______|_______|_______|_______| unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  2. Using [this system] would improve my job performance.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  3. Using [this system] in my job would increase my productivity.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  4. Using [this system] would enhance my effectiveness on the job.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  5. Using [this system] would make it easier to do my job.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely  6. I would find [this system] useful in my job.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                           extremely     quite         slightly       neither      slightly       quite        extremely   B. Perceived ease of use Please mark x on the scales provided. 7. Learning to operate [this system] would be easy for me..  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  8. I would find it easy to get [this system] to do what i want it to do.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  9. My interaction with [this system] would be clear and understandable.  186  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  10. I would find [this system] to be flexible to interact with..  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  11. It would be easy for me to become skillful at using [this system].  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  12. I would find [this system] easy to use.  Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  TAM2 was developed, validated, and used by Venkatesh and Davis [64].  A. Intention to Use 1. Assuming I have access to the system, I intend to use it. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  2. Given that I have access to the system, I predict that I would use it. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  B. Perceived Usefulness 3. Using the system improves my performance in my job. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  4. Using the system in my job increases my productivity. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  5. Using the system enhances my effectiveness in my job. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  6. I find the system to be useful in my job. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  C. Perceived Ease of Use 7. My interaction with the system is clear and understandable. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  187  8. Interacting with the system does not require a lot of my mental effort. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  9. I find the system to be easy to use. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  10. I find it easy to get the system to do what I want it to do. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  D. Subjective Norm 11. People who influence my behavior think that I should use the system. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  12. People who are important to me think that I should use the system. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  E. Voluntariness 13. My use of the system is voluntary. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  14. My supervisor does not require me to use the system. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  15. Although it might be helpful using the system is certainly not compulsory in my job. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  F. Image 15. People in my organization who use the system have more prestige than those who do not. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  16. People in my organization who use the system have a high profile. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  17. Having the system is a status symbol in my organization. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  G. Job Relevance 18. In my job, usage of the system is important. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely 188                        extremely    quite         slightly       neither      slightly       quite        extremely  19. In my job, usage of the system is relevant. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  H. Output Quality 20. The quality of the output I get from the system is high. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  21. I have no problem with the quality of the system's output. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  I. Result Demonstrability  A. I have no difficulty telling others about the result so f using the system. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  B. I believe I could communicate to others the consequences of using the system. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  C. The results of using the system are apparent to me. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely  D. I would have difficulty explaining why using the system may or may not be beneficial. Likely |_______|_______|_______|_______|_______|_______|_______| Unlikely                       extremely    quite         slightly       neither      slightly       quite        extremely   UTAUT was developed, used, and validated by Venkatesh et al. [65]. The answers are in Likert’s scale similar to TAM and TAM2.  A. Performance expectancy 1. I would find the system useful in my job. 2. Using the system enables me to accomplish tasks more quickly. 3. Using the system increases my productivity. 4. If I use the system, I will increase my chances of getting a raise. B. Effort expectancy 5. My interaction with the system would be clear and understandable. 6. It would be easy for me to become skillful at using the system. 7. I would find the system easy to use. 8. Learning to operate the system is easy for me. C. Attitude toward using technology 9. Using the system is a bad/good idea. 189  10. The system makes work more interesting. 11. Working with the system is fun. 12. I like working with the system. D. Social influence 13. People who influence my behavior think that I should use the system. 14. People who are important to me think that I should use the system. 15. The senior management of this business has been helpful in the use of the system. 16. In general, the organization has supported the use of the system. E. Facilitating conditions 17. I have the resources necessary to use the system. 18. I have the knowledge necessary to use the system. 19. The system is not compatible with other systems I use. 20. A specific person (or group) is available for assistance with system difficulties. F. Self-efficacy 21. I could complete a job or task using the system 22. If there was no one around to tell me what to do as I go. 23. If I could call someone for help if I got stuck. 24. If I had a lot of time to complete the job for which the software was provided. 25. If I had just the built-in help facility for assistance. G. Anxiety 26. I feel apprehensive about using the system. 27. It scares me to think that I could lose a lot of information using the system by hitting the wrong key. 28. I hesitate to use the system for fear of making mistakes I cannot correct. 29. The system is somewhat intimidating to me. H. Behavioral intention to use the system 30. I intend to use the system in the next <n> months. 31. I predict I would use the system in the next <n> months. 32. I plan to use the system in the next <n> months.  TAM3 was developed, used, and validated by Venkatesh et al. [66]. The answers are in Likert’s scale similar to TAM and TAM2.  A. Perceived Usefulness 1. Using the system improves my performance in my job. 2. Using the system in my job increases my productivity. 3. Using the system enhances my effectiveness in my job. 4. I find the system to be useful in my job. B. Perceived Ease of Use 5. My interaction with the system is clear and understandable. 6. Interacting with the system does not require a lot of my mental effort. 7. I find the system to be easy to use. 8. I find it easy to get the system to do what I want it to do. C. Computer Self-Efficacy I could complete the job using a software package: 190  9. …if there was no one around to tell me what to do as I go. 10. …if I had just the built-in help facility for assistance. 11. …if someone showed me how to do it first. 12. …if I had used similar packages before this one to do the same job. D. Perceptions of External 13. I have control over using the system. E. Control 14. I have the resources necessary to use the system. 15. Given the resources, opportunities and knowledge it takes to use the system, it would be easy for me to use the system. 16. The system is not compatible with other systems I use. F. Computer Playfulness The following questions ask you how you would characterize yourself when you use computers: 17. …spontaneous 18. …creative 19. …playful 20. …unoriginal G. Computer Anxiety 21. Computers do not scare me at all. 22. Working with a computer makes me nervous. 23. Computers make me feel uncomfortable. 24. Computers make me feel uneasy. H. Perceived Enjoyment 25. I find using the system to be enjoyable. 26. The actual process of using the system is pleasant. 27. I have fun using the system. I. Objective Usability 28. No specific items were used. It was measured as a ratio of time spent by the subject to the time spent by an expert on the same set of tasks. 29. People who influence my behavior think that I should use the system. 30. People who are important to me think that I should use the system. 31. The senior management of this business has been helpful in the use of the system. 32. In general, the organization has supported the use of the system. J. Voluntariness 33. My use of the system is voluntary. 34. My supervisor does not require me to use the system. 35. Although it might be helpful, using the system is certainly not compulsory in my job. K. Image 36. People in my organization who use the system have more prestige than those who do not. 37. People in my organization who use the system have a high profile. 38. Having the system is a status symbol in my organization. L. Job Relevance 39. 1 In my job, usage of the system is important. 40. In my job, usage of the system is relevant. 191  41. The use of the system is pertinent to my various job-related tasks. M. Output Quality 42. The quality of the output I get from the system is high. 43. I have no problem with the quality of the system’s output. 44. I rate the results from the system to be excellent. N. Result Demonstrability 45. I have no difficulty telling others about the results of using the system. 46. I believe I could communicate to others the consequences of using the system. 47. The results of using the system are apparent to me. 48. I would have difficulty explaining why using the system may or may not be beneficial. O. Behavioral Intention 49. Assuming I had access to the system, I intend to use it. 50. Given that I had access to the system, I predict that I would use it. 51. I plan to use the system in the next <n> months. P. Use 52. On average, how much time do you spend on the system each day?  UTAUT2 was developed, used, and validated by Venkatesh et al. [67]. The answers are in Likert’s scale similar to TAM and TAM2.  A. Performance Expectancy 1. I find mobile Internet useful in my daily life. 2. Using mobile Internet helps me accomplish things more quickly. 3. Using mobile Internet increases my productivity. B. Effort Expectancy 4. Learning how to use mobile Internet is easy for me. 5. My interaction with mobile Internet is clear and understandable. 6. I find mobile Internet easy to use. 7. It is easy for me to become skillful at using mobile Internet. C. Social Influence 8. People who are important to me think that I should use mobile Internet. 9. People who influence my behavior think that I should use mobile Internet. 10. People whose opinions that I value prefer that I use mobile Internet. D. Facilitating Conditions 11. I have the resources necessary to use mobile Internet. 12. I have the knowledge necessary to use mobile Internet. 13. Mobile Internet is compatible with other technologies I use. 14. I can get help from others when I have difficulties using mobile Internet. E. Hedonic Motivation 15. Using mobile Internet is fun. 16. Using mobile Internet is enjoyable. 17. Using mobile Internet is very entertaining. F. Price Value 18. Mobile Internet is reasonably priced. 19. Mobile Internet is a good value for the money. 192  20. At the current price, mobile Internet provides a good value. G. Habit 21. The use of mobile Internet has become a habit for me. 22. I am addicted to using mobile Internet. 23. I must use mobile Internet. 24. Using mobile Internet has become natural to me. (dropped) H. Behavioral Intention 25. I intend to continue using mobile Internet in the future. 26. I will always try to use mobile Internet in my daily life. 27. I plan to continue to use mobile Internet frequently. I. Use 28. Please choose your usage frequency for each of the following: a) SMS b) MMS c) Ringtone and logo download d) Java games e) Browse websites f) Mobile e-mail Note: Frequency ranged from “never” to “many times per day.” 193  Appendix H: Usefulness and Ease of Use Design Principles Usefulness Home-care nurses’ perceptions of unmet information needs and communication difficulties of older patients in the immediate post-hospital discharge period [111].  The prototype should document disease/condition progression.  The prototype should document severity of condition.  The prototype should document Symptoms of worsening condition.  The prototype should document discharge information and medication. Analysis of qualitative interviews about the impact of information technology on pressure ulcer prevention programs [103].  The prototype should document patient specific information: current skin status, food preferences, fluid needs or restrictions, mobility choices, hygiene, and preferred timing of care.  The prototype should document skin care, pressure ulcer specific information: evidence-based protocols for pressure ulcer prevention, and specific guidelines for skin or wound care. Effects of a computerized decision support system on care planning for pressure ulcers and malnutrition in nursing homes: An intervention study [106].  The prototype should document with completeness.  The prototype should document with comprehensiveness. Accuracy, completeness and comprehensiveness of information on pressure ulcers recorded in the patient record [102].  The prototype should document with accuracy: correspondence between the data and reality.  The prototype should document with completeness: pertinent information.  The prototype should document with comprehensiveness: documentation according to the different phases of the nursing process.  The prototype should document systematic pressure ulcer risk assessment: sensory perception, moisture, activity, mobility, nutrition, friction/shear, incontinence (Braden scale). Advancing nursing documentation—An intervention study using patients with leg ulcer as an example [110].  The prototype should have clear documentation of the status of the wound, including size, appearance, secretion, odour, pain, localisation, wound edge and skin around the wound. The circumference of the ankle and calf should be measured for control of oedema. Quality of nursing documentation and approaches to its evaluation: a mixed-method systematic review [107].  The prototype should document with completeness taking in account quantity, legibility, appearance, plausibility, patient identification, abbreviations, correction of error, linguistic correctness (objective or factual language, and scientific terms), chronological report of events, the colour of the ink, blank spaces and gaps within the text, documentation in a correct section, the phrases of recording, succinct and clear language, avoidance of use of jargon or technical terms. 194  The art of skin and wound care documentation [104].  The prototype should document patient's chief complaint.  The prototype should document history of present illness.  The prototype should document past medical, family, and social history (including any dressings or modalities that were effective or ineffective in healing past wounds).  The prototype should document review of  any past wound sites.  The prototype should document physical assessment (including the appearance of the wound or presence of any skin conditions).  The prototype should document risk assessment tools used and information obtained.  The prototype should document manual assessment tools used and information obtained.  The prototype should document skin and wound assessment tools used and information obtained.  The prototype should document procedures performed.  The prototype should document supplies and tests ordered.  The prototype should document details of patient education provided.  Documentation should discard all “sticky notes” and scrap paper from the department.  Document all findings in a consistent place in the medical record.  The prototype should document concurrently with the patient’s visit to accurately record the care provided.  The prototype should ensure timely and complete documentation. The PUSH Tool: A Survey to Determine Its Perceived Usefulness [71].  The prototype should monitor pressure ulcer healing.  The prototype should detect clinically important changes in ulcer appearance over time.  The prototype should prompt nurses to reassess the patient and wound.  The prototype should prompt nurses to re-evaluate wound treatment. Does the introduction of an electronic nursing documentation system in a nursing home reduce time on documentation for the nursing staff? [105].  The prototype should support longitudinal format of data. Perceived Quality Benefits Influenced by Usefulness and Documentation Accuracy of information systems [72].  The prototype should document with accuracy: descriptive, timely and complete documentation by self and others, with the ability to correct errors easily.  The prototype should document with quality benefits: the perceived quality of care and the time spent with the patient, permit more time with the patient.  The prototype should support easily correcting errors when charting.  The prototype should document descriptively.  The prototype should document in a timely manner.  The prototype should improve the quality of others documentation.  The prototype should  make care planning more individualized and/or accurate. 195  Improved quality and comprehensiveness in nursing documentation of pressure ulcers after implementing an electronic health record in hospital care [112].  The prototype should document with comprehensiveness in regards to the below items.  The problem is described or interventions planned or implemented.  The problem is described and interventions planned or implemented.  The problem is described and interventions planned or implemented and nursing outcome is recorded.  The problem is described and interventions planned and implemented and nursing outcome is recorded.  All aspects of the nursing process are recorded. Good description of the problem and recording of the relevance for nursing. The quality of home care nurses’ documentation in new electronic patient records [108].  The prototype should document  with comprehensiveness in regards to below items:  The documentation is non-comprehensive when nursing diagnosis is not described and intervention is neither planned nor executed.  Nursing diagnosis is described or intervention is planned or executed.  Nursing diagnosis is described and intervention is planned or executed.  Nursing diagnosis is described and intervention is planned and/or executed. Results are described.  Nursing diagnosis is described and intervention is planned and executed. Results are described.  Nursing diagnosis is well described and all phases of the nursing process is present. The nursing documentation is completely comprehensive.            196  Ease of use Design and development of interface design principles for complex documentation using PDAs [113].  The prototype should easily view information structure without memorization.  The prototype should navigate easily with no more than two tabs.  The prototype should simplify data entry with term-based soft-keys.  The prototype should easily distinguish between answered and unanswered items.  The prototype should directly view all alternative choices for assessments and use tapping for data entry.  The prototype should simplify data entry with pre-analyzed and prepared answers so there is tapping not hand writing.  The prototype should better control the current status of the document with a dashboard for summary.  The prototype should simplify system navigation with hyper jumping. Use of a human factors approach to uncover informatics needs of nurses in documentation of care [115].  The prototype should support visibility of system status: make status information visible, especially changes so nurses would not have to document in multiple places.  The prototype should support matching between system and the real world: reduce over-reliance on memory when there are menu that don't apply to the documentation.  The prototype should support user control and freedom: increase observability of work to encourage a shared mental model of the work environment.  The prototype should support error prevention: support ability to plan/predict work memory prompts to give feedback for the functionalities.  The prototype should support recognition rather than recall: make status information for other services and patients visible so the entered data is propagated through the system.  The prototype should support flexibility and efficiency of use: make shortcuts visible to increase system flexibility.  The prototype should have aesthetic and minimalist design: allow for system modification and customization to reduce irrelevant data cluttering the screen. A decision support and documentation system for treatment of chronic ulcers [114].  The prototype should indicate items to document  The prototype should use pictures and clear comparison scales, like comparison to a coin, for objectivity 197  Ten usability heuristics [97].  Visibility of system status: the system should always keep users informed about what is going on, through appropriate feedback within reasonable time.  Match between system and the real world: the system should speak the users' language, with words, phrases and concepts familiar to the user, rather than system-oriented terms. Follow real-world conventions, making information appear in a natural and logical order.  User control and freedom: users often choose system functions by mistake and will need a clearly marked "emergency exit" to leave the unwanted state without having to go through an extended dialogue. Support undo and redo.  Consistency and standards: users should not have to wonder whether different words, situations, or actions mean the same thing. Follow platform conventions.  Error prevention: even better than good error messages is a careful design which prevents a problem from occurring in the first place. Either eliminate error-prone conditions or check for them and present users with a confirmation option before they commit to the action.  Recognition rather than recall: minimize the user's memory load by making objects, actions, and options visible. The user should not have to remember information from one part of the dialogue to another. Instructions for use of the system should be visible or easily retrievable whenever appropriate.  Flexibility and efficiency of use: accelerators -- unseen by the novice user -- may often speed up the interaction for the expert user such that the system can cater to both inexperienced and experienced users. Allow users to tailor frequent actions.  Aesthetic and minimalist design: dialogues should not contain information which is irrelevant or rarely needed. Every extra unit of information in a dialogue competes with the relevant units of information and diminishes their relative visibility.  Help users recognize, diagnose, and recover from errors: error messages should be expressed in plain language (no codes), precisely indicate the problem, and constructively suggest a solution.  Help and documentation: even though it is better if the system can be used without documentation, it may be necessary to provide help and documentation. Any such information should be easy to search, focused on the user's task, list concrete steps to be carried out, and not be too large.   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items