UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mapping nursing wound care data elements to SNOMED-CT Block, Lorraine Joy 2016

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

Item Metadata

Download

Media
24-ubc_2017_february_block_lorraine.pdf [ 1.65MB ]
Metadata
JSON: 24-1.0340679.json
JSON-LD: 24-1.0340679-ld.json
RDF/XML (Pretty): 24-1.0340679-rdf.xml
RDF/JSON: 24-1.0340679-rdf.json
Turtle: 24-1.0340679-turtle.txt
N-Triples: 24-1.0340679-rdf-ntriples.txt
Original Record: 24-1.0340679-source.json
Full Text
24-1.0340679-fulltext.txt
Citation
24-1.0340679.ris

Full Text

  MAPPING NURSING WOUND CARE DATA ELEMENTS TO SNOMED-CT by  Lorraine Joy Block  B.S.N., The University of British Columbia, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE IN NURSING  in The Faculty of Graduate and Postdoctoral Studies  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   December 2016  © Lorraine Joy Block, 2016 ii  Abstract Documentation is a professional responsibility in nursing because it facilitates communication, promotes good nursing care, and acts as a valuable method to demonstrate that legal and agency standards are followed.  Nurses are increasingly using health information technologies, such as electronic health records, to document care. To be able to measure and compare the impact of nursing on patient outcomes, standardized clinical terminologies compliant with international standards are necessary.  In British Columbia, Canada, nurses use a standardized wound care template to document their assessments and the care they provide to patients; however, the content of this assessment is currently not shared in a computable format between different electronic health records within the province.  The purpose of this thesis was to map wound care data elements from the BC Standardized Nursing Wound Documentation standard to SNOMED-CT.  To complete this “bottom-up” mapping activity, creation of a conceptual model of knowledge representation for nursing wound care was developed to inform three concurrent methods of mapping (manual, automated, and literature comparison) for 107 data elements.  These methods produced candidate lists, which were reviewed by two expert wound care clinicians who created an expert consensus list.  Results of this expert consensus list indicated that 40.2% of the terms had direct matches, 1.9% had one-to-many matches, and 57.9% had no matches. The outcome of this study was the creation of a conceptual model of nursing knowledge representation for wound care, a list of mapped wound care data elements to SNOMED-CT, identification of missing and duplicate concepts in SNOMED-CT, and application of concurrent mapping methods to inform the creation of an expert consensus list.  The advancement of standardized clinical terminologies to support semantic interoperability between disparate electronic health records is an important measure to ensure patient information iii  is shared throughout the continuum of care.  This thesis work provides a method to incorporate local nursing standards into SNOMED-CT, with the intent to ensure that nursing care is represented.    iv  Preface This thesis is the original work of Lorraine Block.  Throughout the paper, guidance and support was provided by Leanne Currie (supervisor), Sabrina Wong (committee member), and Rosa Hart (committee member).  The British Columbia (BC) Nursing Wound Care documentation standard is a product of the BC Nursing Skin and Wound Committee who gave Lorraine Block permission to use this standard for the purpose of mapping the data elements to SNOMED-CT.  The author was solely responsible for writing this thesis, under the supervision of the thesis committee.    Publication of conference proceedings occurred through presentation of the preliminary results of Chapter 4.  Both Leanne Currie and Shannon Handfield were contributors to this presentation. Contributions of their work are described above and account for approximately 10% of writing, each.   This work was published as Block, L. and Handfield, S. (2016). Mapping wound assessment data elements in SNOMED CT. Studies in Health Technology and Informatics. 2016, 225: 1078-79. v  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents ...........................................................................................................................v List of Tables ..................................................................................................................................x List of Figures ............................................................................................................................... xi Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiii Chapter 1: Introduction ................................................................................................................1 1.1 Introduction ..................................................................................................................... 1 1.2 Background ..................................................................................................................... 1 1.2.1 Digitalization of the Health Care System ................................................................... 1 1.2.2 Standardized Clinical Terminologies .......................................................................... 2 1.2.3 Clinical Terminology Mapping................................................................................... 3 1.2.4 Information Sharing and Wound Care ........................................................................ 5 1.2.5 Mapping Wound Care Parameters to Standardized Clinical Terminologies .............. 6 1.3 Purpose ............................................................................................................................ 7 1.4 Significance..................................................................................................................... 8 1.5 Research Questions ......................................................................................................... 9 1.6 Theoretical Framework ................................................................................................... 9 1.6.1 Rationale ................................................................................................................... 11 1.6.2 Theoretical Operationalization .................................................................................. 11 1.7 Summary ....................................................................................................................... 12 vi  Chapter 2: Literature Review .....................................................................................................13 2.1 Introduction ................................................................................................................... 13 2.1.1 Identification of Literature ........................................................................................ 13 2.2 Clinical Interoperability ................................................................................................ 14 2.3 Clinical Standardized Terminologies ............................................................................ 15 2.3.1 Systematized Nomenclature of Medicine – Clinical Terms ..................................... 17 2.3.2 Logical Observation Identifiers Names and Codes (LOINC) ................................... 18 2.3.3 International Classification of Nursing Practice (ICNP) .......................................... 19 2.4 Problems Facing the Operationalization of Standardized Terminologies .................... 20 2.4.1 Multiple Standardized Clinical Terminologies ......................................................... 21 2.4.2 Human-Computer Concept Interpretation ................................................................ 22 2.4.3 Limited Usage of Standardized Clinical Terminologies ........................................... 24 2.5 Clinical Impact of Using Standardized Terminologies ................................................. 27 2.6 Considerations Once Terminologies Have Been Implemented .................................... 30 2.7 Canadian Nursing Perspective about Standardized Terminologies .............................. 31 2.8 Current Wound Care Standardization Initiatives .......................................................... 33 2.9 Summary ....................................................................................................................... 35 Chapter 3: Methods .....................................................................................................................36 3.1 Introduction ................................................................................................................... 36 3.2 Research Questions ....................................................................................................... 36 3.3 Study Design ................................................................................................................. 36 3.4 Sampling Plan ............................................................................................................... 38 3.5 Procedures and Data Collection .................................................................................... 39 vii  3.6 Phase 1: Wound Care Conceptual Model of Knowledge Representation .................... 40 3.6.1 Clinical Meaning ....................................................................................................... 40 3.6.2 Unified Modeling Language ..................................................................................... 42 3.6.2.1 SNOMED-CT Hierarchy Types ....................................................................... 44 3.7 Phase 2: Mapping for Cross-Coverage ......................................................................... 46 3.7.1 Manual Mapping to SNOMED- CT ......................................................................... 46 3.7.2 Automated Electronic Mapping to SNOMED- CT .................................................. 46 3.7.3 Literature Comparison of Wound Assessment, Diagnosis, and Outcome Mapping 47 3.8 Phase 3: Expert Consensus Mapping ............................................................................ 48 3.9 Data Analysis ................................................................................................................ 48 3.10 Ethical Considerations .................................................................................................. 49 3.11 Plans to Navigate Possible Difficulties ......................................................................... 49 Chapter 4: Results........................................................................................................................50 4.1 Introduction ................................................................................................................... 50 4.2 Phase 1: Conceptual Model Findings ........................................................................... 50 4.2.1 Consensus of Model Development ........................................................................... 51 4.3 Phase 2: Mapping Findings ........................................................................................... 55 4.3.1 Manual Mapping ....................................................................................................... 55 4.3.2 Automated Mapping ................................................................................................. 56 4.3.3 Literature Mapping ................................................................................................... 57 4.4 Phase 3: Expert Consensus Mapping to find Rate of Equivalence ............................... 59 4.4.1 Rate of Equivalence between SNOMED-CT and Wound Assessment .................... 60 4.4.2 Rate of Equivalence between SNOMED-CT and Wound Diagnosis ....................... 61 viii  4.4.3 Rate of Equivalence between SNOMED-CT and Goal of Care ............................... 62 4.4.4 Inter-Mapping Comparison against Expert Consensus List ..................................... 63 4.5 Summary ....................................................................................................................... 64 Chapter 5: Discussion ..................................................................................................................65 5.1 Introduction ................................................................................................................... 65 5.2 Use of a Conceptual Model........................................................................................... 65 5.3 Implications of the Conceptual Model of Knowledge Representation for Nursing Wound Care .............................................................................................................................. 66 5.4 Mapping Methods ......................................................................................................... 67 5.4.1 Manual Mapping ....................................................................................................... 69 5.4.2 Automated Mapping ................................................................................................. 71 5.4.3 Literature Comparison Mapping ............................................................................... 72 5.4.4 Expert Consensus Mapping ...................................................................................... 73 5.5 Implications of Mapping Methods ................................................................................ 73 5.6 Missing Content Coverage ............................................................................................ 75 5.7 Implications of Missing Content ................................................................................... 78 5.8 Relationship to Theoretical Model ................................................................................ 79 5.9 Limitations of Study ..................................................................................................... 80 5.10 Implications and Recommendations for Nursing Practice, Policy and Research ......... 81 5.11 Summary ....................................................................................................................... 83 References .....................................................................................................................................85 Appendices ....................................................................................................................................92 Appendix A- Wound Care Parameters ..................................................................................... 92 ix  Appendix B – Wound Assessment: Terms and Matches by Different Mapping Methods ....... 97 Appendix C – Wound Diagnosis: Terms and Matches by Different Mapping Methods........ 111 Appendix D – Goal of Care: Terms and Matches by Different Mapping Methods ............... 119     x  List of Tables Table 3.1 UML Class Diagram Components and Definitions ...................................................... 43 Table 3.2 Matching criteria for mapping data elements to a standardized clinical terminology .. 48 Table 4.1 Manual Mapping of Nursing Wound Documentation Parameters to SNOMED-CT ... 56 Table 4.2 Automated Mapping of Nursing Wound Documentation Parameters to SNOMED-CT....................................................................................................................................................... 57 Table 4.3 Comparison of Mapped Nursing Wound Documentation Parameters to Harris et al. . 58 Table 4.4 Expert Consensus Mapping of Wound Documentation Parameters to SNOMED-CT 60 Table 4.5 Expert Consensus Mapping to SNOMED-CT for Wound Assessment Data Elements 61 Table 4.6 Expert Consensus Mapping to SNOMED-CT for Wound Diagnosis Data Elements .. 62 Table 4.7 Expert Consensus Mapping to SNOMED-CT for Goal of Care Data Elements .......... 62 Table 4.8 Differences between Expert Consensus Mapping and Manual, Automated and Literature Mapping ....................................................................................................................... 63          xi  List of Figures Figure 1.1 Theory of Wisdom in Action for Clinical Nursing ..................................................... 10 Figure 3.1 Overview of Study Methodology ................................................................................ 37 Figure 4.1 Conceptual Model of Knowledge Representation for Nursing Wound Care (overview)....................................................................................................................................................... 52 Figure 4.2 Conceptual Model of Knowledge Representation for a Nursing Wound Assessment 53 Figure 4.3 Conceptual Model of Knowledge Representation for a Nursing Wound Diagnosis ... 54 Figure 4.4 Conceptual Model of Knowledge Representation for a Nursing Wound Outcome .... 55  xii  Acknowledgements   Dr. Leanne Currie, you saw something in me that I didn’t know was there.  You so generously gave me your time, your ear, your mind, and your guidance.  Your expert knowledge in nursing informatics has inspired and guided me to truly enjoy this domain of nursing.  I am so very lucky and so very grateful of your support.  You exemplify what a good person, professor, and mentor is. Thank you.   A special thank you to Dr. Sabrina Wong, whose expectation for strong reasoning helped find the path to (nursing) visibility.  As well, to Rosa Hart, whose pragmatic strength kept my research focused and grounded.  Thank you.  Shannon Handfield, you have been the role model I have been in awe of from the start.  I will never forget your early words to me: “what we do will push the line – sometimes pushing it into a grey area that others will fight against – but push we must for nursing and for wound care”.  Those are powerful words I see you enact every day.  Thank you.  Finally, I am so eternally grateful to those who have loved me and kept me afloat all these years.  I love you all.  Chris, my husband and best-friend, you encouraged me, held me up, and kept me going; you are my heart.  Julia and Rosa, some of my best reflective thoughts happen while cradling you; you give me so much happiness, peace and laughter.  Mom and Dad, you have unconditionally loved and nurtured me from the start, my drive to never give up comes from you.  Kerry and Andy, you brought me into your lives from the beginning, I have always felt loved and protected.  To all family and friends, I really am such a lucky person- you all inspire me, guide me, and keep me sane.  Thank you.   xiii  Dedication    To my Jewel and Rose 1  Chapter 1: Introduction  1.1 Introduction Documentation is a professional responsibility in nursing because it facilitates communication, promotes good nursing care, and acts as a valuable method to demonstrate that legal and agency standards are followed (College of Registered Nurses of British Columbia, 2013).  Nurses are increasingly using health information technologies, such as electronic health records (EHR), to document care.  The Canadian Nurses Association (CNA) supports the use of health information technologies to prevent essential knowledge about the impact of nursing care from being lost and advocates that the collection, storage, retrieval and use of nursing data through information technologies are necessary (CNA, 2006).  To be able to measure and compare the impact of nursing on patient outcomes, standardized clinical terminologies compliant with international standards are necessary (CNA, 2006).  1.2 Background 1.2.1 Digitalization of the Health Care System For over a decade, the Canadian health care system has invested considerable time, money and energy implementing digital health solutions to capture clinical information at the point of service (Canada Health Infoway, 2014).  With this transformation, there is a heightened awareness that digital patient information needs to be: i) accessible to other clinicians, ii) used to improve clinical decision making, and iii) used in such a way as to not disrupt clinical workflows.  One of the most important mechanisms to achieve these goals is interoperability between health information systems.  Interoperability is the ability for related and disparate 2  clinical information systems to share data within and across different organizational boundaries to support the delivery of patient care (HIMSS, 2005; Sensmeier & Murphy, 2014).  In short, clinical interoperability is an essential measure to support high quality patient care (Infoway, 2014; Westra, Delaney, Konicek, & Keenan, 2008).    1.2.2 Standardized Clinical Terminologies Standardized clinical terminologies provide the design and structure to allow the meaning of clinical concepts to be used and shared in an EHR and meaningfully understood by computer system processing (Sensmeier, 2011).  This level of electronic information sharing is called semantic interoperability.  Standardized terminologies help support semantic interoperability, data retrieval and exchange, quality care monitoring, collection of data for research, and decision support logic (Hammond, Jaffe, Cimino, & Huff, 2014; Kim & Matney, 2014; Häyrinen, Lammintakanen, & Saranto, 2010; Cimino, 1998).  In addition, standardized terminologies support the ability to make nurses’ work and decision making visible through documentation of actions and clinical decisions (Nagle & White, 2016).  The CNA encourages nurses to advocate and lead initiatives to collect and code nursing care using standardized clinical terminologies to aid in the development of a pan-Canadian sharable and interoperable electronic health record (CNA, 2006).   Despite these recommendations, standardized clinical terminologies have not achieved widespread adoption in Canada or abroad (Westra et al., 2015; Thoroddsen, Ehrenberg, Sermeus, & Saranto, 2012).  Reasons for the lack of clinical adoption include the proliferation of standardized clinical terminologies with the paradox of not having one robust enough to represent the vast complexity of clinical practice, no mandate or policy levers (i.e., funding, 3  incentives) from the federal Canadian government requiring organizations and vendors to advance semantic interoperability requirements, and general concerns with the ongoing maintenance requirements to on-board and sustain standardized clinical terminologies within an EHR (Cimino, 1998; Hardiker, Bakken, Casey, & Hoy, 2002; Hardiker, Bakken, & Kim, 2011; Harris et al., 2015).  Though the integration and use of a standardized clinical terminology is complex and poses significant challenges, the potential to significantly improve the delivery of health care through interoperable and analytic solutions aided by the use of standardized terminologies, makes the effort meaningful.  As such, efforts to map nursing care to standardized terminologies has become a practice within clinical informatics for over 40 years, with researchers extending this work into decision support, patient care analysis, and information exchange frameworks (Harris et al., 2015; Westra et al., 2015; Kim, Hardiker, & Coenen, 2014; Kim & Park, 2012).   1.2.3 Clinical Terminology Mapping The practice of mapping, in its simplest form, is matching a clinical concept to an equivalent codified concept in a large database.  However, the mapping activities required for the systematic mapping of clinical content to standardized clinical terminologies requires scientific rigour and oversight.  For example, consideration of matching concepts as “pre-coordinated” (concepts match exactly) or “post-coordinated” (concepts in the database are pieced together to match the clinical concept), hierarchy placement and structure (“is-a” relationships and antecedents) and method of mapping (manual, automated, or semi-automated) are some of the ways researchers are approaching clinical mapping activities (Coiera, 2015; Rosenbloom et al., 2006).  Current activities to map nursing data elements to a standardized clinical terminology can 4  be broadly described in three categories: a) inter-terminology mapping, b) “top-down” mapping, and c) “bottom-up” mapping.  Inter-terminology mapping is the action of identifying equivalent concepts between existing standardized clinical terminologies.  As mentioned above, the proliferation of standardized terminologies presents a challenge to researchers and organizations when deciding which standardized clinical terminology to use, and how to incorporate them into an EHR.  As one method to help in this challenge, some of the larger organizations that control warehouses of codified clinical concepts are working together to laterally map (harmonize) to each other.  For example, the Systemized Nomenclature of Medicine – Clinical Terms (SNOMED-CT) has been harmonized and cross-referenced to other existing terminologies, such as the International Classification of Diseases (ICD) and somewhat to the International Classification of Nursing Practice (ICNP) (Kim & Matney, 2014; SNOMED International, 2014).   Second, “top-down” mapping occurs when previously mapped clinical data sets (or reference sets or value sets) are added to an existing EHR.  As an example, if an organization chooses to incorporate the Canadian Health Outcomes for Better Information and Care (C-HOIBC) data set in their EHR, they would have a set of 24 standardized clinical assessment parameters already matched to SNOMED-CT and ICNP (C-HOBIC, 2015).   Finally, “bottom-up” mapping occurs when clinical concepts existing in a health information system, or domain of knowledge, is initially mapped to a standardized clinical terminology (i.e., data elements in an EHR are mapped to SNOMED-CT) (Harris et al., 2015).  Once a “bottom-up” mapping activity is completed, and a set of mapped data elements are produced and shared, that list may be used in future “top-down” mapping activities in other 5  health information systems.  The research presented in this thesis is a “bottom-up” mapping activity. 1.2.4 Information Sharing and Wound Care  The clinical practice of wound care is an important consideration for nurses because wound management represents a significant challenge to the health care system.  The prevalence of patients with wounds is reported at >40% in acute care (Hurd & Posnett, 2009) and >35% in home care (Woodbury & Houghton, 2005).  Measures to facilitate standardized clinical documentation of wounds and their management have become a focus for many researchers and organizations to support communication, inform clinical decision making, and support safe patient care (Handfield, 2013; Kinnunen, Saranto, Ensio, Iivanainen, Dykes, 2012; Harkier, Bakken, & Kim, 2011; Kim & Park, 2011).  Recently, a large American nursing collaborative mapped skin, wound, and pressure ulcer data elements to SNOMED-CT and the Logical Observation Identifiers Names and Codes (LOINC) (Harris et al., 2015).  Their purpose was to standardize nursing wound content between six organizations, determine content coverage between the nursing wound care data elements and SNOMED-CT / LOINC, and inform those regulatory standards development organizations of any missing content.  They determined that 261/320 (82%) of the nursing wound assessment concepts were mapped to SNOMED-CT (Harris et al., 2015).   Locally, the College of Registered Nurses of British Columbia (CRNBC) has noted that wound care is a nursing practice responsibility and has recently updated the Scope of Practice for Registered Nurses (2016) to include a statement that nurses may carry out wound care without a physician or nurse practitioner’s order.  As one measure to support nurses in this responsibility, the British Columbia (BC) Provincial Nursing Skin and Wound Committee has created 6  standardized nursing wound documentation parameters to be used in all provincial health institutions in paper or electronic format (e.g., Pixalere™) (Handfield, 2013).  These parameters represent the nursing care related to a wound assessment, diagnosis, outcome, intervention, and evaluation.  This thesis focuses on the standard nursing parameter concepts related to wound assessment, diagnosis, and outcome developed by the BC Provincial Nursing Skin and Wound Committee.  As patients move across the continuum of care in BC, the sharing of this pertinent wound care information is often fragmented because it is not semantically interoperable between systems.  As such, the wound care findings are often shared between different health care settings and providers through faxes, photocopies, printed reports, emails, texting and/or verbal discourse.  The point of this description is not to discount these methods en masse, as they are important, but to highlight those multi-media, hybrid approaches are creating situations of information loss, duplication, redundancies and inconsistencies (Coiera, 2015).  As part of a larger system of information exchange, the focus on data integrity related to missing and incorrect patient information in EHRs is a top patient safety concern for many organizations (ECRI Institute, 2015).  Nurses must be able to share reliable and accurate information regarding a patient’s wound care needs; standardized clinical terminologies, when integrated into EHRs with care and oversight, can assist clinicians in meeting this requirement.   1.2.5 Mapping Wound Care Parameters to Standardized Clinical Terminologies As researchers continue to map nursing wound care concepts to standardized terminologies, the choice of which one(s) to use must be tempered by what kind of concepts are to be mapped, how the standardized terminology is built, the organization’s reporting and 7  interoperability needs, and the vendor product’s capacity to use standardized clinical terminologies.   The following points were considered when deciding which terminology to use for mapping the BC nursing wound assessment, diagnosis, and outcome concepts.  Firstly, the types of data elements used in this study were very granular (i.e., slough, diffuse, and denuded).  After examining the literature, it was noted that other researchers have demonstrated success mapping similar data elements to LOINC and SNOMED-CT, thus making LOINC and SNOMED-CT possible contenders (Harris et al, 2015; Kim & Park, 2012).  Secondly, the data structure of standardized clinical terminologies often differ.  For example, ICNP has been built using Web Ontology Language (OWL), LOINC has been built in the communication protocol Health Level Seven International (HL7), and over time, SNOMED-CT has created its own guiding principles to structure it concepts (Kim & Matney, 2014).  Thus, the ability to map to multiple terminologies would require a deep understanding of the ontology structure for each terminology chosen.  Finally, two EHR products in use or soon to be used throughout BC (Pixalere™ and Cerner™) are SNOMED-CT enabled, thus the product of this thesis would be able to be used in the local context to share wound care information between disparate EHRs, thus making SNOMED-CT a logical choice.  Mapping wound care data elements to LOINC and ICNP were beyond the scope of this thesis and should be considered by future researchers or organizations to carry out.  1.3 Purpose Despite having nursing wound documentation standards in BC, there is no current method or process to share this information between disparate EHRs within or between BC 8  health authorities.  Further, there have been no known prior studies comparing the BC Standardized Nursing Wound Documentation standard to international standardized clinical terminologies.  The purpose of this study was to map the wound care data elements from the BC Standardized Nursing Wound Documentation standard to SNOMED-CT.  This was done to develop a knowledge representation model for the local context and to determine a rate of equivalence between this provincial standard and SNOMED-CT.  1.4 Significance Knowledge, implementation and spread of standardized terminologies into clinical information systems, EHRs, and databases remain limited and often poorly understood (Westra et al., 2015; Thoroddsen, Ehrenberg, Sermeus, & Saranto, 2012).  The care provided by nurses and their subsequent documentation in an EHR should be mapped to a standardized clinical terminology as an essential measure towards sematic interoperability, to ensure that clinical activities and judgements by nurses are accurately reflected in EHRs, and to facilitate future research and outcome analysis.  The outcome of this thesis project is a set of mapped nursing wound assessment, diagnosis, and outcome parameters to SNOMED-CT.  The thesis also contributes to the scientific field of nursing informatics though investigation of nursing knowledge representation of wound care in a standardized medical terminology.  The mapped list will be provided to the BC Provincial Nursing Skin and Wound Committee and other interested stakeholders with the goal to incorporate the coded data elements into existing and future EHRs.   9  1.5 Research Questions This thesis addresses three research questions: Research Question 1: What is the rate of equivalence between the pre-identified wound assessment, diagnosis and outcome data elements and SNOMED-CT?  Research Question 2: Are there any differences between the rates of equivalence between the groups of concepts (assessment, diagnosis, and outcome) and SNOMED-CT?  Research Question 3: Are there any differences between the rates of equivalence of the pre-identified wound assessment, diagnosis and outcome data elements and SNOMED-CT based on the methods of mapping?  1.6 Theoretical Framework Nursing theory can be used to facilitate the development of nursing knowledge while intersecting the health information age and technological advances (Matney, 2015).  As such, this thesis situates the use of standardized clinical terminologies in the context of nursing knowledge through the Theory of Wisdom in Action for Clinical Nursing (Figure 1.1) (Matney, 2015).  Matney (2015) suggests that the antecedents of nursing wisdom and knowledge are person-related factors and environment-related factors within the larger context of a nurse’s role in patient care.  Matney posits that Wisdom-in-Action is the use of knowledge for clinical judgment in context and care decisions.  Information system factors lie in the environment-related factors domain, noting that nursing knowledge requires data to be structured in context in order to be transformed into usable information and that electronic data that is out of context is not information.  For example, within or outside an EHR, the data element of “50” does not carry clinical meaning unless structured into context (e.g., Urine Output 50ml/hour; Diastolic Blood 10  Pressure 50mmHg; D50W IV solution).  Matney (2015) suggests that nurses can use the theory to understand what information is needed in specific clinical situations (judgement/action), and to use this to codify, organize, interpret and transform available information within EHRs to make it useful in practice (i.e., to make information an antecedent of wisdom).  By addressing and operationalizing the transformation of data into information through an advanced standardized clinical terminology, nurses can use patient care data collected in EHRs as a component of nursing knowledge and wisdom.    Figure 1.1 Theory of Wisdom in Action for Clinical Nursing  Image used with permission from Dr. Susan Matney.  11  1.6.1 Rationale A theoretical framework, especially in the field of health informatics where information sciences, computer sciences and health sciences merge, is important when analyzing and managing complex phenomena (Nelson & Staggers, 2014).  The theory of Wisdom in Action provides a conceptual space to examine the application of standardized clinical terminologies to transform data, which are collected and stored in an EHR, into information to support sematic interoperability, decision support tools, administrative needs and large-scale research analysis.  Thus, these codified data elements can be reused, shared and structured to display information, leading to the creation of nursing knowledge.   1.6.2 Theoretical Operationalization  This thesis project is a functional example of taking individual nursing data elements already implemented in an active EHR (Pixalere™), and mapping them to a structured standardized clinical terminology.  The theoretical operationalization of the Theory of Wisdom in Action for Clinical Nursing (Matney, 2015) provided the conceptual foundation to ground the argument for why nursing content representation in SNOMED-CT is important.  The benefits and considerations of using a standardized clinical terminology in an EHR are examined, highlighting the impact of semantic interoperability, data analysis and decision support on clinical practice.  As a measure to understand the concept orientation of the BC Standardized Nursing Wound Documentation standard, a conceptual model of knowledge representation was constructed.  Then, those same data elements were mapped to SNOMED-CT to determine the rate of equivalence.  Using Matney’s (2015) theory, the results of this study were critically analyzed for their impact on nursing knowledge and wisdom in practice.  Missing content 12  coverage in SNOMED-CT is specifically highlighted, postulating reasons why this may have occurred and the impact of the findings.     1.7 Summary This chapter sought to introduce the reader to the topic of standardized terminologies for use in nursing practice and the health care system.  The background described considerations and uses of standardized terminologies, relating it to the thesis purpose, significance, and subsequent research question.  The introduction of a nursing theory, Theory of Wisdom in Action for Clinical Nursing (Matney, 2015), was also highlighted, forming the lens which to situate this thesis work in the context of nursing care.  13  Chapter 2: Literature Review  2.1 Introduction  This chapter provides an introduction to the structure, meaning and current context of standardized clinical terminologies in health care.  Specifically, nursing and medical standardized terminologies are discussed and related to the challenge of achieving interoperability within and between health care organizations.  This chapter will also present current problems related to the uptake and operationalization of these standard terminologies, the clinical impact of using standardized terminologies in an EHR, and considerations once a standardized terminology has been implemented.  Finally, this chapter will conclude with a brief look at the use of standardized clinical terminologies in the Canadian health care context, as well as, the current work and knowledge available for using standardized terminologies in wound care clinical practice.   2.1.1  Identification of Literature The Cumulative Index to Nursing and Allied Health Literature (CINAHL) and Medline databases were searched for relevant and related information using the truncated search terms “standard* terminology”; “nurs* terminology”; and “terminology map*”.  After refining the search to show entries written in English, relevant journal articles were chosen for review based on the article title and abstract.  After reading the selected articles in full, their bibliographies of articles with similar topics and research methods were reviewed for additional relevant sources of research.  Additionally, recent medical and nursing informatics textbooks, along with seminal papers in knowledge representation in healthcare, were used to situate core and basic principles 14  of standard terminology history, structure, and current knowledge.  Authors whose publications were most closely related to this thesis topic were used to search in CINAHL by “author name” to find further related studies.  Finally, grey literature found through Canada Health Infoway, College of Registered Nurses of British Columbia, American Medical Informatics Association, Standard Terminology Organizations (i.e., SNOMED International, Regenstrief Institute, International Council of Nurses (ICN), World Health Organization, and Health Level 7 (HL7)) and the Health Information and Management Systems Society (HIMSS) were reviewed and used to add perspective from leading national and international organizations working with health terminologies in research and practice.   2.2 Clinical Interoperability  The importance of standardized clinical terminology structure and format is related to its function within the broader goal of interoperability.  As stated earlier, interoperability is the ability for related and disparate clinical information systems to share data within and across different organizational boundaries to support the delivery of patient care (HIMMS, 2005; Sensmeier & Murphy, 2014).  The goal of interoperability is a stated vision for the Pan-Canadian Clinical Interoperability Steering Committee, through Canada Health Infoway, as a way to “improve the quality of patient care through the effective sharing of clinical information among health care organizations, clinicians and their patients” (Infoway, 2015, May, p. 5).  There are three types of interoperability: 1) Foundational 2) Structural and 3) Semantic (HIMSS, 2013; National Committee on Vital and Health Statics (NCVHS), 2000).  Foundational interoperability allows data to be received in one information system, from another; however, it does not require the ability for that system to interpret the data (HIMSS, 2013; NCVHS, 2000).  15  Structural interoperability occurs when the syntax (i.e., structure) of the data exchange is defined, allowing for the uniform movement of health information between systems with the meaning of the data preserved (HIMSS, 2013; NCVHS, 2000).  Finally, semantic interoperability is the highest level of interoperability and is achieved when two or more clinical information systems are able to exchange information and can interpret and use that received information (HIMSS, 2013).  For semantic interoperability to occur, standards for data transport, content exchange, security protection and vocabulary management are required (Sensmeier & Murphy, 2014; Sensmeier, 2011).  Within the context of health care, these standards are agreed-upon ways to record and exchange information and can be created by interested groups, government sanctions, marketplace, and/or formal consensus by standards development organizations (Sensmeier, 2011; Westra et al., 2008).  This thesis focuses on the function of vocabulary management, through the use of standardized terminologies, as one measure to support semantic interoperability.   2.3 Clinical Standardized Terminologies  A standardized clinical terminology is a complex system of computer and human concept arrangement, merging to create a synergistic product that supports semantic information exchange.  Within human language, consider the words we combine to produce a concept; the meaning itself can stay constant, but the words we choose and the context to which we apply them can change and evolve (Cimino, 1998; Kim & Matney, 2014).  It is the concept itself, transcending the variety of linguistic representations, that is the focus of a standardized terminology.  For example, “orthostatic hypotension” has the same meaning as “postural hypotension”, which has the same meaning as “dizziness when standing up from a sitting or 16  supine position”.  These multiple expressions can be organized into one representative term and positioned into a hierarchy or network according to other concept relationships; the specification of these relationships allows for the creation of a domain specific terminology, which combined with standards created by regulatory organizations, becomes a standardized terminology (or nomenclature) (Hammond et al., 2014; Kim & Matney, 2014).  A standardized terminology becomes a classification when the concepts are organized according to their similarities rather than semantic meaning, whereas ontology refers to concepts that are formally specified and defined based on its multiple subclass relationships (Kim & Matney, 2014).  Generally, advanced standardized terminologies are created in an ontology format, allowing concept orientation to facilitate functionality in multiple situations, in many languages, and to be easily reviewed for quality (Rosenbloom, Miller, Johnson, Elkin, & Brown, 2006).  Further, in these advanced terminology systems, each concept is given a modeled definition, a position on a multi-hierarchy web with ontology class representation, a utility that allows granular dissection, and a non-sematic permanent code (Cimino, 1998; Hardiker et al., 2011). The idea of standardizing clinical concepts to support data exchange for the advancement of science is not a new idea.  In 1965 Gordon wrote: “The language of medicine is essential for the furtherance of scientific knowledge and the exchange of clinical data” (Gordon, 1965).  In the 1960s, as an early adopter of controlling clinical terminology nomenclature with technology, the American Medical Association and College of American Pathologists defined and coded clinical diseases with cardboard computer punch cards and electronic tape (Gordon, 1965). Nursing was quick to follow (Werley, Devine, Zorn, Ryan, & Westra, 1991) and for over 40 years have been involved in the development and maintenance of several nursing standardized terminologies (Westra et al., 2015).  Though there are several clinical standardized terminologies 17  available to clinicians worldwide, this thesis will focus on nursing content representation for wound care in SNOMED-CT.  As mentioned in Chapter 1, the reason for this is related to the type of granular nursing assessment data elements used, previous research in nursing, and the pragmatic fit with existing vendor products in the local context.  In the next section, greater detail and background is provided about the aforementioned standardized clinical terminologies of SNOMED-CT, ICNP and LOINC.  The purpose of this explanation is to provide greater context and anchor support related to the choice of SNOMED-CT, and also to highlight that nursing concept representation in standardized clinical terminologies is nuanced, duplicated, and disparate.     2.3.1 Systematized Nomenclature of Medicine – Clinical Terms The Systematized Nomenclature of Medicine – Clinical Term (SNOMED-CT) is the most comprehensive and multilingual standardized terminology system in the world (SNOMED International, 2014).  It is an ever-growing product which evolved from the historical Systemized Nomenclature of Pathology, the Systematized Nomenclature of Medicine, SNOMED-International, SNOMED- RT, and Clinical Terms Version 3, to become what it is known by today: SNOMED-CT (SNOMED International, n.d.a).  Each concept in SNOMED-CT is given a numeric (computable) code with a fully specified name, preferred term, synonym, and organized onto a hierarchal tree (with parent-child relationships) (Kim & Matney, 2014).   Since 2007, SNOMED International (previously the International Health Terminology Standards Development Organization (IHTSDO)) has owned and distributed SNOMED-CT (SNOMED International, n.d.a).  Canada Health Infoway is the formal liaison to SNOMED International for Canada and provides licensing, education, and support for Canadian clinicians 18  and vendors wishing to implement SNOMED-CT within an EHR (Infoway, 2015, March).  SNOMED-CT is listed as a pan-Canadian Standard to enable the electronic sharing of clinical information between disparate health care solutions (Infoway, n.d.).  Though the foundational structure remains medically driven, SNOMED International has made efforts to increase concept representation for other health disciplines.  For example, SNOMED International has created a nursing special interest group to advise the management board and act as key stakeholders in the development of SNOMED-CT (SNOMED International, n.d.b).  Already, researchers and collaborators have noted challenges mapping concepts related to the nurses’ clinical judgement on the health needs and assets of patient’s, families and communities (Kim, Hardiker, Coenen, 2014).  They indicate that the SNOMED-CT hierarchy structure restricted their mapping possibilities, posing a risk for lack of nursing content representation, and by extension, a risk for lack of interoperability and data analytic initiatives (Kim et al., 2014).  Supporting multi-disciplinary collaboration, as well as ongoing nursing research with SNOMED-CT, is important as SNOMED International continues to pursue international partnerships and global leadership in standardized clinical terminology development.    2.3.2 Logical Observation Identifiers Names and Codes (LOINC) In 1994, the Regenstrief Institute initiated the development of LOINC to support the computable exchange of laboratory values, patient measurements, assessment instruments, radiology exams and clinical observations (Hammond et al., 2014; Regenstrief Institute, n.d.).  In 2013, the Regenstrief Institute and IHTSDO (now SNOMED International) signed a cooperative agreement to begin laterally linking their terminology systems and aligning the representation of 19  laboratory and clinical measurement attributes, thus providing a common framework to use LOINC and SNOMED-CT together (IHTSDO & Regenstrief Institute, 2013).  As the content focus of LOINC has largely been dedicated to laboratory testing, the scope of nursing specific content is not extensive (Kim & Matney, 2014).  However, the structure of LOINC is useful in the context of scaled assessments, which are commonly used in nursing (e.g., pressure ulcer staging and the Braden Scale for Pressure Sore Risk).  As such, LOINC has been used by nurse researchers to extend nursing content representation.  For example, Harris et al. (2015) used LOINC to code nursing “questions” (i.e., wound drainage?) and SNOMED-CT to code nursing “answers” (i.e., serous exudate).  These researchers chose to include LOINC, partly because the structure has been harmonized to HL7 models, facilitating the use of Unified Modeling Language (UML) for computerized modeling (Harris et al., 2015).   Further, as a leader of health information and interoperability initiatives, Canada Health Infoway has customized LOINC to meet the needs of Canadian laboratory ordering and reporting (pan-Canadian LOINC Observation Code Database (pCLOCD)), and approved pCLOCD as a terminology standard to support Canadian EHR interoperability (Infoway, n.d.).  Although LOINC was outside the scope of this thesis, it is important that nurses continue to research and contribute to its development.  2.3.3 International Classification of Nursing Practice (ICNP) The International Council of Nurses (ICN) represents more than 130 nursing associations across the world and controls the ICNP formal terminology system (ICN, 2015; Kim & Matney, 2014).  The ICNP is a formal standardized nursing terminology that provides a framework into which nursing diagnosis, intervention, and outcome concepts are structured and coded for use in 20  an EHR (ICN, n.d.).  Due to its robustness and compliance to international standards, the CNA endorses ICNP as the recognized standard terminology to describe professional nursing practice in Canada (CNA & Infoway, 2008; ICN, 2014).  However, ICNP has not obtained the same recognition with Canada Health Infoway and it has not been added to the pan-Canadian Standards Inventory list (Infoway, n.d.; CNA & Infoway, 2008). In 2010, and again in 2013, the ICN and IHTSDO (now SNOMED International) signed a working collaborative agreement to develop and advance an equivalence table to cross-map ICNP concepts into SNOMED-CT to further a common understanding and support interoperability (ICN & IHTSDO, 2014).  In a large study mapping these two systems, it was found that 399/805 (55.7%) of the ICNP concepts had pre-coordinated matches to SNOMED-CT (Kim et al., 2014; ICN, 2014).  This study recommended future work to include: adding missing concepts to SNOMED-CT, to consider the use of post-coordinated terms, and to broaden the conversation at an international level to deliberate the potential of retaining each system’s domain independence while using a technology platform (such as the United Medical Language System (UMLS) Metathesaurus) to harmonize shared concepts (Kim et al., 2014).    2.4 Problems Facing the Operationalization of Standardized Terminologies  Identifying how and which terminology to use in an EHR can be complex and problematic.  The “terminology problem” can be characterized by several interrelated issues including: 1) the development of multiple standardized clinical terminologies, 2) inadequacies related to computers not understanding complex human language, and 3) minimal policy initiatives or federal incentives to drive adoption.  21  2.4.1 Multiple Standardized Clinical Terminologies  The first problem related to operationalizing standardized terminologies is associated with the simultaneous development of multiple, overlapping terminologies, created by numerous organizations and groups (Hardiker et al., 2011).  For example, the American Nursing Association recognizes ten different standardized terminologies, seven of which are nursing specific (Kim & Matney, 2014).  However, the paradox of quantity is that the proliferation of these terminology systems has not yet produced one singular terminology that is comprehensive enough to cover the extensive health care needs of its multidisciplinary users (Cimino, 1998; Hammond et al., 2014; Kim & Matney, 2014).  Another paradox is that the quantity and overlap of these terminologies perpetuates the problem of coded data which are not standardized.  As an effort to mediate this aspect of the “terminology problem”, international organizations, researchers, health IT professionals and clinicians are working to laterally cross-map and harmonize standardized terminology systems.  For example, SNOMED-CT has had the ICD standardized terminology and a number of nursing terminology systems (Clinical Care Classification; ICNP; North American Nursing Diagnosis Association International (NANDA-I); Nursing Interventions Classification; Nursing Outcomes Classification; Omaha System; Perioperative Nursing Data Set) cross-mapped to its comprehensive clinical terminology sets  (Kim & Matney, 2014).  This is an essential practice to ensure the meaning of exchanged information is preserved and the precise meaning is not lost when interpreted by different health care professionals (Kim & Matney, 2014).  22  2.4.2 Human-Computer Concept Interpretation  A second contributor to the “terminology problem” is the juxtaposition of human concept interpretation and computer concept interpretation within an EHR (Hardiker et al., 2011).  The problem being that humans create concepts based on natural language, full of diverse, semantic, contextual, and equivalent meanings; computers cannot readily understand these especially when clinical terms are intended to be re-used for interoperability, decision support and data extraction (Hardiker et al., 2011; Kim & Matney, 2014).  But herein lies the connection to the first problem: not enough concepts are available in one standardized terminology to describe everything for everyone in health care (Rosenbloom et al., 2006).  An idea to mediate this could be to keep adding concepts and recognize them as unique and distinct; this is the exercise of “pre-coordination” or “enumerative terminology” (Coiera, 2015; Rosenbloom et al., 2006).  Further, to aid functionality, a concept could be tagged with several synonyms and employed when developing the user interface for an EHR (Kim & Matney, 2014).  From the discussion above, a pre-coordinated concept could be “postural hypotension” and as such, would be given all the necessary attributes required in a standardized terminology.  However, if organizations continually add to their existing terminology system without oversight and knowledge of pre-existing concepts, the result can lead to an assortment of terms poorly organized with troublesome granularity, clarity, and redundancy (Cimino, 1998; Coiera, 2015).  When Kim, Hardiker, & Coenen (2014) mapped ICNP to SNOMED-CT, lack of clarity was cited between the two systems.  In one instance, they found that the SNOMED-CT concepts, “abnormal behaviour (finding)” and “problem behaviour (finding)” both could be mapped to the one ICNP concept of “negative behaviour”.  23  As an alternate idea, one could use existing concepts and combine them to produce a new clinical concept that is not available in a pre-coordinated format; this is the exercise of “post-coordination” or “compositional terminology” (Coiera, 2015; Rosenbloom et al., 2006).  Post-coordination occurs when complex concepts with varying levels of detail are described using more fundamental ones (also known as atomic concepts) (Rosenbloom et al., 2006).  Again from the discussion above, “postural hypotension” could be dynamically coded within an EHR using the separate concepts of “posture” and “hypotension.”  There is an inherent risk however, with this practice; Consequences of post-coordination include: i) difficulty restricting an EHR to medically meaningful concepts, ii) the potential for unrecognized duplicate entries, iii) inefficiency composing complex concepts from simpler ones, and iv) the burden to look up several unrelated concepts from distinct multiple sub-class lists (Hardiker et al., 2002; Lee, Cornet, Lau, & de Keizer, 2013; Rosenbloom et al., 2006). Keeping in mind the risks of pre- and post-coordination, both practices are fundamental when translating clinical need to computer need.  In his seminal paper, Cimino (1998) recognized the importance of terminology systems having the capacity to increase the number of concepts it included, but stressed the significance of formal, reproducible, and systematic methods of adding content.  These early realizations have become today’s standards, where the quality of health care terminology structure, content, mapping, and its life cycle have been addressed by the International Organization for Standardization (ISO) (Kim & Matney, 2014).  As one example of an ISO standard related to standardized terminology, ISO 18104:2014 defines the categorical structure methodology for nursing diagnosis and nursing action, which developers and clinical infomaticians use when adding and developing content for standardized nursing terminologies (International Standards Organization, 2014).  24  2.4.3 Limited Usage of Standardized Clinical Terminologies  A third challenge contributing to the “terminology problem” is the limited uptake in knowledge, clinical priority, health policy, and research available on the implementation of standardized clinical terminologies into existing EHRs (Westra et al., 2008).  Historically, items that have been systematically coded are related to concepts associated with billing and reimbursement and not specific to clinical usage (Kim & Matney, 2014).  Despite 40 years of defined, coded, and available nursing terminology systems, the wide integration of such standards into EHR systems available for purchase (i.e., vendor systems) has yet to be realized (Westra et al., 2015).  Associated with this challenge is the reality that the language nurses use to describe practice continues to be inconsistent across and within health care settings (Technology Informatics Guiding Education Reform, 2009).  This leaves many nursing-sensitive patient care indicators non-standardized, limiting the ability to capture nurses’ work and to share and compare information as patients move through the health care system (Chow et al., 2015).   Though there has been work to increase the availability of better data to create knowledge from aggregated EHRs, clinicians still do not fully realize how it can be used to support their practice and their organizations (Canadian Health Outcomes for Better Information and Care, 2015).  Often, clinicians view the incorporation of standards in documentation as a bureaucratic process, suggesting that on-going education and engagement are necessary to ensure the information produced is of value and utilized to inform practice (Canadian Health Outcomes for Better Information and Care, 2015).  In an integrative review of the literature, Stallinga et al. (2015) found that health care professionals were not concerned when ambiguous language was used to document patient care.  They suggest this could be the reason why the implementation of 25  standardized terminologies have failed and that these systems will only be successful when data can be meaningfully reused in areas such as decision-making and research.  To add another dimension, even for those organizations who have an agreed upon set of standardized nursing clinical charting parameters, and intend to operationalize a standardized clinical terminology to codify nursing care in their EHR, the methods which to map these existing parameters is diverse with no clear agreed-upon gold-standard (Monsen et al., 2016; Harris et al., 2015; Kim et al., 2014; Richesson, Andrews, & Krischer, 2006).  For example, manual mapping is often cited as a method for searching through a standardized clinical terminology; however, in one study, researchers have shown graduate prepared clinical nurses only had a 57.1% success rate when manually mapping a nursing clinical concept to SNOMED-CT (when compared to a pre-existing inter-terminology mapping list) (Monsen et al, 2016).  Further, in another study, when two researchers were mapping nursing clinical concepts from ICNP to SNOMED-CT, each individually manually mapped 436 concepts, and between the two had a kappa value of 0.45 (moderate agreement) when compared.  These few examples suggest that manual mapping may be inconsistent; however, necessary when exploring concept representation, not otherwise known.  Other studies have attempted to use automated or semi-automated measures to map existing clinical concepts to standardized clinical terminologies (also known as candidate mapping).  For example, inter-terminology mapping exercises might start with UMLS as a tool to investigate previous laterally mapped concepts as potential matches (Kim, 2016; Kim et al., 2014).  Yet, the generated result of these automated or semi-automated methods is often insufficient, requiring another method (e.g., manual mapping) to fill in the missing matches (Saitwal et al, 2012; Lau, Simkus, & Lee, 2008; Richesson et al., 2006).   26  Though these nuanced differences in mapping methods may seem trivial to some (after all, they do result in coded data elements matched to the desired standardized clinical terminology), the outcome may have a significant patient safety and quality care impact which needs to be addressed and clearly understood before the large scale implementation of standardized clinical terminologies are pushed out in an EHR.  Specifically, this relates to the actualized benefits of the codified standardization process itself; semantic interoperability, data aggregation and decision support.  These are threatened if dissociated and disparate mapping methods are employed without oversight by organizations tasked with integrating these standardized terminologies within existing EHRs.  To illustrate this, imagine if two different organizations decided to include the concept “dizziness when standing up” in their EHR.  One mapped this to the SNOMED-CT concept Orthostatic Hypotension (disorder) 28651003, the other mapped it to Dizzy Spells (finding) 315018008.  What would happen if the same patient was treated at different times for this condition, in each organization?  Would their documentation be coded in SNOMED-CT as Orthostatic Hypotension (disorder) or Dizzy Spells (finding)?  What would happen if we tried to employ semantic interoperability, data aggregation, or decision support networks on what we thought was the same concept, but coded differently?   The scope of this paper does not expand on the outcome of dissociated cross-organizational mapping; however, it would be an important consideration for future research. Another aspect to the uptake of standardized clinical terminologies relates to the engagement of vendors whose proprietary applications are (or are not) built with the capacity and standards adoption to support sematic interoperability (Harris et al., 2015).  For example, when free-text, natural language is the main form of clinical documentation, the content becomes largely inaccessible to manipulate and use with decision support tools or statistical research 27  (Hardiker et al., 2002).  Clinical information systems need to be designed in such a way that data are placed in a meaningful context and ready to be reused without the need for manual transformation and manipulation (Lenz, Beyer, & Kuhn, 2007).  Some of these challenges may be solved by the use of natural language processing in which free text is analyzed and mapped to a codified clinically equivalent concept (Topaz et al, 2016); however, this is yet to readily available through commercial clinical EHR vendors.  Another reflection on this limitation requires a view of current marketplace demands; currently, there are no Canadian federal governmental policies requiring health care organizations to choose EHRs with the built in standards and structure to facilitate sematic interoperability.  In BC, the  Ministry of Health (2014) released a cross sector policy discussion paper, Enabling Effective, Quality Population and Patient-Centered Care: A Provincial Strategy for Health Information Management and Technology, recognizing the need to improve data sharing throughout the health care system.  Though provisions and plans are underway to drive foundational information standards, the scope and scale of this work is still in the early phases of development in BC.   2.5 Clinical Impact of Using Standardized Terminologies  The implementation and maintenance of a standardized terminology in an EHR has many advantages, such as ensuring that the impact of nursing care on patient outcomes is represented in aggregated health care information (Hardiker et al., 2002).  Data aggregation for quality care monitoring can be harnessed through the use of standardized terminologies.  This can be accomplished by using equivalent coded concepts and compiling information to determine if an identified variable had an effect.  When health care organizations establish data quality standards (with terminology as one component), the result can be the production of high quality data to 28  facilitate knowledge generation important for hospital operations and patient care (Wager, Lee, & Glaser, 2013).  This utility also supports the requirement for nursing specific contributions to be represented in aggregated health care outcomes (Hardiker et al., 2002).  For example, this functionality would help analyze gaps in current practice, map patient outcomes associated with organizational guidelines, policy, or staffing, and follow medication error events after reconciliation (Kim & Matney, 2014).  As an outcome, when organizations use their own patient care data, adverse patient consequences decrease and new knowledge from the large accumulation of digitized records is accessible (Kim et al., 2014).  It is also suggested that data aggregation enables comparative effectiveness research as a product of using large quantities data retrieved from numerous EHRs (Kim & Matney, 2014).  Using a standardized terminology can also be operationalized to focus content utilized on a clinical interface (i.e., data elements on a pre-designed template within an EHR).  This can help an organization ensure that the terms used within an EHR are not redundant, numerous, or ambiguous (Kim & Matney, 2014).  Along with this noted precision, standardized terminologies can provide a rich set of flexible, user-friendly synonyms that increase usability for nursing documentation or when displaying computer-stored patient information (Rosenbloom et al., 2006).  When following standards set by the ISO for sematic interoperability, advanced terminologies support operative data retrieval and exchange of clinical information (Hardiker et al., 2011).  Yet, as many EHR systems are not designed with this functionality, custom point-to-point interfaces to share patient information have become commonplace.  But concern is mounting; as the complexity, volume, and risk connecting each system increases, this practice has become unsustainable (Hammond et al., 2014).  If the principle of sematic interoperability is 29  achieved, custom computer-to-computer interfaces would become obsolete as clinical information can be shared between disparate EHR’s, ensuring that the operational purpose and meaning of these data are maintained and unaltered (Technology Informatics Guiding Education Reform, 2009).  Essentially, this means sharing information that is translated as “apples to apples” and not “apples to car”. Finally, standardized terminologies can be leveraged as a scaffold to design and build logic for integrated decision support (Cimino, 1998).  This can be imagined as documenting an assessment on a screen that links to a clinical guideline or provides the clinician with an automated message alerting to a potential negative health outcome.  To meet this potential, terminologies need to be designed with enough depth and granularity to allow the system to trigger clinical guidelines to facilitate best practices (Moen, Henry, & Warren, 1999).  To this, the American Medical Informatics Association Board of Directors approved a White Paper that outlines a US national action plan to operationalize the full potential of clinical decision support (Osheroff et al., 2007).  It describes that if best knowledge is to be available, clinical knowledge and interventions need to be represented in a standardized format (interpretable to both humans and computers).  This interpretable format being, at least in part, standardized terminologies.  As a clinical example to represent nursing decision support, Zega et al. (2014) created a semi-structured assessment form utilizing standardized NANDA-I terminology, and built into it a decision support framework to suggest nursing diagnoses based on assessment findings.  The results of this study validated the assessment form used by the research team and the potential of using standardized terminology to support the structure needed to create a decision support mechanism.   30  2.6 Considerations Once Terminologies Have Been Implemented   As outlined above, there are genuine problems competing with highly desirable outcomes when considering standardized terminologies.  Truly, the use of a standardized terminology is not a decision to be taken lightly.  If a health care organization and vendor take on the development and implementation of a standardized terminology in an EHR, there are considerations that come along with its lifecycle adoption.  The first consideration is ‘what is the actual available content?’ for a chosen standardized terminology.  There has been a continual call for these terminologies to include more vocabulary and terms (Cimino, 1998; Hardiker et al., 2002).  Lee et al. (2013) found organizations using SNOMED-CT had difficulty with the available content, ambiguity of terms and different synonyms, syntactic consistency, and difficulty understanding hierarchal relationships.  To add complexity, often more than one standardized terminology is required in an EHR in order to achieve the desired content coverage (e.g., Pan-Canadian Standards Inventory of technical and clinical standards) (Infoway, n.d.; Hammond et al., 2014).  As a response to the continued call for more coded concepts, every organization controlling a standardized terminology must have an ongoing maintenance process, or else the terminology would become obsolete (Hammond et al., 2014; Kim & Matney, 2014).  As such, the organizations that control standardized terminologies continually add content based on clinical and clinician need.  A new version of ICNP is released every two years and includes elements to classify nursing diagnoses, actions, and outcomes (ICN, n.d.).  SNOMED International releases an updated version of SNOMED-CT every six months that include new concepts, preferred terms, concept status (i.e., inactive), supporting attributes, and hierarchical position changes (Lee, Cornet, & Lau, 2011).  31  Organizational processes to maintain these updates need to be implemented or the terminology system would quickly become outdated. The second consideration is the downstream effect from the first; organizations using standard terminologies need to anticipate, determine implications, and operationalize the release changes to their current EHR systems while maintaining and ensuring no unexpected consequences occur (Lee et al., 2011).  These types of standardized terminology updates also include the inactivation or re-description of existing concepts, as well as, concept model position changes (Lee et al., 2011).  For example, it could be that we structured our EHR with the post-coordinated concepts “posture” and “hypotension” to make up for the missing pre-coordinated concept “postural hypotension.”  Later, in a newer release of a chosen standardized terminology, “postural hypotension” could be added as a pre-coordinated concept.  The effect of terminology updates and modifications could incur changes to decision support networks, user interface displays, data extracts, and report generation.  Further, changes to a hierarchy position could make data retrieval difficult as the technical analyst would need to change their data query to obtain similar reportable data (Lee et al., 2013).  Knowledgeable staff, time, and resource allocation must be considered when planning the sustainment period of implementing a standardized terminology.  2.7 Canadian Nursing Perspective about Standardized Terminologies In Canada, much of the EHR development, investment, and outcome reporting has been driven by physicians, which has resulted in nursing sensitive clinical practice and outcomes being largely invisible (Hannah, White, Nagle, & Pringle, 2009).  Concerns are also mounting that health information systems are increasingly being built and used to make health system 32  decision-making, with expectations focused on billing, regulatory, and accreditation requirements, rather than clinical need (CIHI, 2013; Cusack et al., 2013).  Canada Health Infoway, which is sponsored by Health Canada, was created in 2001 to help transform the health care system by leading the national development of EHRs (Office of the Auditor General of Canada, 2009).  Canada Health Infoway works with provinces and territories to help ensure the collaborative key requirements and core components of this goal are met by providing financial support and planning guidelines (Office of the Auditor General of Canada, 2010).  The Infoway Pan-Canadian Clinical Interoperability Steering Committee creates, through stakeholder consensus, a pan-Canadian technical and clinical terminology standards list, chosen and evaluated to enable shareable clinical information (Infoway, n.d.).  This list includes SNOMED-CT; pCLOCD; several ISO technical standard requirements; diagnostic, pathology, medication, claims and non-clinical registry standards; and more.   Through funding from the CNA, provincial partners, and Canada Health Infoway, there has been a Canadian first step to use nursing standardized terminology to capture large-scale nursing sensitive outcomes within EHRs; this is the work of the Canadian Health Outcomes for Better Information and Care (C-HOBIC) project (Hannah et al., 2009).  This project (2007-2014) was designed to utilize structured language in nursing assessments to give nurses real-time information about their patient care, to measure patient outcomes through aggregation, and to facilitate data sharing (C-HOBIC, 2015).  The C-HOBIC data set has 24 standardized data elements and is arranged into four categories: functional status, symptom management, safety outcomes, and therapeutic self-care (C-HOBIC, 2015).  Though it is not widely distributed throughout Canada, the C-HOBIC content has been included in the National Nursing Quality Report in Canada; was designated as a Canadian Approved Standard; and has been formally 33  endorsed by the CNA and Canadian Nursing Informatics Association (C-HOBIC, n.d.).  Further, discussions are currently underway for Canadian Institute for Health Information to use this data set in the Discharge Abstract Database to inform clinical outcomes, health system use, and performance reporting (C-HOBIC, n.d.). In April, 2016, a National Nursing Data Standards meeting was hosted by the CNA, Canadian Institute of Health Information, Canada Health Infoway, Canadian Nursing Informatics Association, and the Lawrence S. Bloomberg Faculty of Nursing (Nagle & White, 2016).  Nursing leaders from across Canada were invited to unify and provide a strategy for the teaching, collection, and reporting of nursing care.  After this two-day meeting, the working groups concluded that the adoption of nursing data standards would serve to create visibility for nursing, bring credibility, inform care planning and quality improvement opportunities, support the delivery of safer patient care, and leverage and strengthen decision-making at a local, jurisdictional, and national level (Nagle & White, 2016).  An identified standardized clinical terminology was not recommended at the conclusion of this meeting; however, a new spark and interest has begun as clinical, policy, research and educational nurse leaders plan future opportunities to investigate and advocate for the use of a specific standard(s) to represent nursing care in Canada.  2.8 Current Wound Care Standardization Initiatives  As an overall population health matter, it has been estimated that complex wounds affect 1.47 per 1,000 of the population (Hall et al., 2014).  Further, in Canada, it has been estimated that 35.5% of individuals receiving community health services require wound care for skin ulcers (Woodbury & Houghton, 2005).  Wounds are a significant practice consideration for nurses as 34  nearly all health care areas have patients requiring wound care services (Bryant & Nix, 2007).  As a move to support the practice of nurses caring for patients with wounds, CRNBC recently updated the Scope of Practice For Registered Nurses (2016) to include a statement describing that registered nurses may carry out wound care without a physician or nurse practitioner’s order (CRNBC, 2016).  Globally, nurses have been actively pursuing collaborative projects aimed at standardizing different types of wound assessments, analyzing content coverage in standardized terminologies, and integrating the usage of wound assessments in decision support systems. Pressure ulcer prevention and care, as one type of complex wound, has been widely viewed as a nursing-sensitive quality care indicator (Wound Ostomy and Continence Nurses Society, 2009).  With many health care resources dedicated to pressure ulcer prevention and management, researchers have used this as an opportunity to study the functionality and benefits of mapping these pressure ulcer parameters to standardized clinical terminologies (Chow et al., 2015; Harris et al., 2015; Kim & Park, 2012).  Two large studies have mapped nursing pressure ulcer assessment data elements to SNOMED-CT and LOINC (Chow et al., 2015; Harris et al., 2015).  In another study, researchers mapped nursing pressure ulcer assessment data elements to SNOMED-CT and modeled the data elements using an entity-attribute-value format (Kim & Park, 2012).  In each of these three aforementioned studies, a set of mapped data elements were created and used to build further models to support sematic interoperability or enacted with a decision support tool.  Kim and Park (2012) used their coded data set in a structured electronic assessment form, where, based on the nurses surface injury characteristic charting, an integrated decision support system determined the pressure ulcer stage and provided the next assessment screen relevant to that pressure ulcer stage assessment requirements (i.e., wound size, wound 35  bed, and exudate).  As researchers begin to harness and use the functional benefits these structured terminologies provide, a new dialogue highlighting further challenges is beginning.  The authors in these studies suggest new questions and express real concerns including: What are the national and international plans to share coded data sets?, Who is responsible to update these lists?, Which standardized terminologies are used and in what context?, and How are the data stored and retrieved? (Chow et al., 2015; Harris et al., 2015; Kim & Park, 2012).  Rather than deterrents for clinicians to use these standardized terminologies, these questions help move the science forward and showcase the collaborative work already happening within the clinical informatics community and help elucidate the future work that is needed.   2.9 Summary In summary, standardized terminologies provide a mechanism for interoperability, an opportunity to design decision support networks, and extract semantically equivalent data to support research and patient care initiatives.  Yet, achieving consensus and standards for nursing terminologies is complex and not without a set its own set of challenges.  As a contribution to the complex solution, the purpose of this thesis is to examine the ability for existing standard terminologies to represent nursing wound care concepts used in BC, Canada.     36  Chapter 3: Methods  3.1 Introduction  In this chapter, the methods used to map the wound assessment, diagnosis, and outcome data elements to SNOMED-CT are outlined.  The study design type, sampling plan, procedure and analysis adhered to scientific protocol, with oversight from the thesis supervisor and committee members.  Finally, ethical considerations and study limitations were considered, ensuring governmental and university policy were enacted and a balanced perspective was presented when interpreting the final results.   3.2 Research Questions  Three research questions were addressed:  Research Question 1: What is the rate of equivalence between the pre-identified wound assessment, diagnosis and outcome data elements and SNOMED-CT?  Research Question 2: Are there any differences between the rates of equivalence between the groups of concepts (assessment, diagnosis, and outcome) and SNOMED-CT?  Research Question 3: Are there any differences between the rates of equivalence of the pre-identified wound assessment, diagnosis and outcome data elements and SNOMED-CT based on the methods of mapping?  3.3 Study Design  This study was conducted as a nonexperimental descriptive analysis, examining the rate of equivalence between the 107 pre-identified wound assessment, diagnosis and outcome data 37  elements and SNOMED-CT.  This design was chosen to match the study question: to describe the rate of equivalence between two variables (wound data elements and a standardized terminology), rather than to manipulate a variable and study its outcome or effect (Polit & Beck, 2012).  Structurally, the descriptive analysis was conducted in three phases; a) creation of a conceptual model using a Unified Modeling Language (UML) class diagram framework, b) mapping nursing wound care data elements to SNOMED-CT using three approaches, and c) clinician review of mapping results to inform an expert consensus list (Figure 3.1).   Figure 3.1 Overview of Study Methodology    38  Descriptive statistical analyses were carried out in phases 2 and 3 to determine a numerical rate of equivalence.  These multiple phases were purposefully implemented to provide the rigour required to answer the research questions, as well as, to support future research studies wishing to replicate nursing concept mapping activities.      3.4 Sampling Plan For the purpose of this study, 107 wound care data elements were composed of wound assessment (assessment), wound etiology (wound diagnosis) and goal of care (outcome) parameters, as per the BC Nursing Wound Documentation standard (Appendix A).  Concepts related to other assessment parameters (e.g., lower leg), planning (e.g., care planning), intervention (e.g., dressing change frequency) and evaluation (e.g., wound care clinician referrals) were not considered in this study.  The BC Nursing Wound Documentation standard was created in 2008 through a provincial nursing committee composed of nursing wound care clinicians from each of the six health authorities (Handfield, 2013).  This group is now known as the British Columbia Provincial Nursing Skin and Wound Committee.  Their collaboration has resulted in joint projects with CRNBC, the creation of dozens of clinical practice documents and guidelines, the development of a province-wide standardized nursing wound documentation tool, and more (Handfield, 2013).  The BC Nursing Wound Documentation standard has since been adapted to paper and electronic formats, and is currently used, or working towards being used, in all provincially funded, cross-continuum, clinical institutions.  For example, the BC Nursing Wound Documentation standard is integrated and used in the Vancouver Coastal Health (VCH) electronic wound documentation system (Pixalere™).  The procurement of these data elements for use in this thesis were obtained by formal request to the department manager of the Primary 39  and Community Solutions Team in VCH (Jamie Simpson).  As well, permission was sought and granted by the British Columbia Provincial Nursing Skin and Wound Committee (guardians and decision makers of clinical content utilized in the obtained BC Nursing Wound Documentation standard) (Committee Chair, Shannon Handfield).    3.5 Procedures and Data Collection This study was conducted in three phases (Figure 3.1).  The first phase was the creation of a conceptual model of knowledge representation for the BC Nursing Wound Care Documentation standard.  The purpose of this model was to guide the matching of wound assessment, diagnosis, and outcome data elements to SNOMED-CT.  Then, the second phase of the study used three mapping approaches  to search for concept matches in SNOMED-CT.  Specifically, these mapping methods included manual mapping, automated electronic mapping, and literature comparison mapping.  Manual mapping occurred first, followed by automated and literature comparison.  The intention to perform manual mapping first, was to decrease possible influence the other methods might have had on the researcher in concept selection.  It was also planned that in each mapping approach, the selection of a SNOMED-CT concept needed to meet the criteria of a pre-coordinated match.  Post-coordination was not considered in this study as it would have expanded the scope and scale of this thesis project beyond a feasible timeline.  In the third phase, all possible concept matches were reviewed by two expert clinicians, who created a final expert consensus list, using the criteria of match, one-to-many, and no match (Table 3.2).   40  3.6 Phase 1: Wound Care Conceptual Model of Knowledge Representation Due to the complex and granular representation of SNOMED-CT concepts, semantic representation of the 107 wound care data elements needed to be clearly symbolized.  As such, it was decided that a class diagram model would be created to visually display the knowledge represented in the BC Nursing Wound Care Documentation standard.  This decision was also influenced by previous mapping studies, in which researchers created data models to provide context and guide decisions regarding the conceptual meaning of each targeted data element (Harris et al., 2015; Chow et al., 2015; Kim & Park, 2012). The class diagram model created in this study merged three abstractions of hierarchy structuring: clinical meaning, UML class diagram design, and SNOMED-CT hierarchy types.  This model acted as a guide for the manual and literature mapping activities, as well as, the expert consensus mapping analysis for the expert consensus list.  This step was important because the mapping activities required more than lexical matching, rather, it required a deep understanding of the target concept ontological meaning and relationships (i.e., what is periwound erythema?).  The following sections describe these three influencing abstractions in detail.   3.6.1 Clinical Meaning The standardized nursing wound assessment, diagnosis, and outcome data elements were obtained from the presiding chair of the BC Provincial Nursing Skin and Wound Committee.   After obtaining the data elements, each wound definition and parameter heading was evaluated.  This provided the opportunity to group each data element into meaningful parts.  For example, the wound ‘etiologies’ (i.e., underlying causes) of Burn, Trauma, and Skin Tear could be 41  semantically clustered into the heading Traumatic Injury Etiology, thus creating a clinically driven, parent-child value set.   Clinical interpretation also facilitated the leveling of granularity and division into segments to represent the Nursing Process.  The Nursing Process can be described as the steps taken by nurses to deliver holistic, patient-focused care, and include 1) assessment, 2) diagnosis, 3) outcome/planning, 4) implementation (intervention), and 5) evaluation (American Nurses Association, n.d.).    The assessment parameters represented in this model describe wound assessment characteristics and are specific to the local body region associated with the skin injury.  Other supporting assessment parameters, such as intrinsic and extrinsic factors that affect wound healing, health assets, and body part involved, were not included in this mapping activity because it would not be feasible to carry out in the time and scope for a thesis project.  The nursing diagnosis parameters in this model represent the underlying cause (etiology) of the wound.  The nursing wound diagnosis is the application of the nurse’s clinical judgment about the patient’s skin injury assessment, and could include the accumulation of other assessment parameters (as noted above), investigations, and/or collaboration with other health care professionals (ANA, n.d.).  The outcome/planning concepts represented in this model relate to the goal of care for the management of the wound.  The nurse would consider the accumulated information related to the wound assessment and wound diagnosis, and together with the patient/family and other health care team members, would set a measurable and achievable goal of care for the management of their wound (ANA, n.d.). Though the BC Nursing Wound Documentation standard parameters had not been previously titled and divided into these categories, it became apparent that these pre-existing 42  concepts were the attributes observed in the nursing process.  Thus, the data elements obtained from the BC Nursing Skin and Wound Committee, and used in this research study, fit within the context of a nursing assessment, diagnosis, and outcome process for wound care.  Those nursing concepts related to the implementation (care planning, treatment) and evaluation of a patient’s wound were not included in this research study.  These nursing process concepts and concepts related to other aspects of a nursing assessment, diagnosis and outcome/planning for wound care, would be of value to map to SNOMED-CT in future research studies.   3.6.2 Unified Modeling Language The application of UML was applied to provide an opportunity to visually depict high-level interactions between the clinically meaningful parts (Medvidovic, Rosenblum, Redmiles, & Robbins, 2002).  UML has previously been used by researchers, such as Chow et al. (2015) and Harris et al. (2015), as they attempted to translate clinical knowledge into an unambiguous model, readable for multiple users.  In both studies, the researchers modeled skin, wound, and pressure ulcer assessment data elements using UML to represent value sets and linking attributes (Chow et al., 2015; Harris et al., 2015).   In the diagram generated for this thesis, the clinical finding of a wound is denoted as having the highest level of abstraction, with attributes of wound assessment, wound diagnosis, and goal of care.  The white diamond leading from the wound finding to the wound assessment was used to represent “aggregation”; the wound assessment is subordinate and is a part-of the wound finding (Table 3.1) (IBM, 2003; Medvidovic et al., 2002).  This subordinate behaviour may be temporary, such that if the wound finding relationship was removed, parts of the wound assessment may still be applicable (e.g., the periwound skin may still have a rash).   43   Table 3.1 UML Class Diagram Components and Definitions Component Name: Definition aggregation'has a' Aggregation; the wound assessment is subordinate and is a part-of the wound finding inheritance'is a' Inheritance; symbolizing a parent-child class structure    composition'has a' Composition; stronger form of aggregation    Relationship; represent the specific data elements   The solid line with the closed, unfilled arrowhead is used to represent an important “inheritance”, symbolizing a parent-child class structure (Table 3.1) (IBM, 2003; Medvidovic et al., 2002). The arrow leading from the wound diagnosis to wound finding represents an inheritance; thus, the wound diagnosis is a subclass of the clinical finding of a wound (i.e., the venous ulcer is a wound). The solid diamond leading from the wound finding to the goal of care was used to represent a “composition” and a stronger form of aggregation; the goal of care is subordinate to only the wound finding (IBM, 2003; Medvidovic et al., 2002).  For example, it is the wound (as part of the person) that has a goal of care.  This relationship is tied only to the wound finding, and once the wound finding is removed, the goal of care in this instance is also removed (i.e., the goal of care is not a venous ulcer).  Finally, the dashed lines extending from the wound assessment, wound diagnosis, and goal of care boxes to specific wound parameters were used to represent the specific data elements that are included in the BC Nursing Wound Documentation standard (see Table 3.1).   44  3.6.2.1 SNOMED-CT Hierarchy Types Finally, SNOMED-CT hierarchy types and relationship links were considered in the development of this conceptual model.  These were adopted after examining the structure of SNOMED-CT, the top level hierarchy definitions, and “is a” (concept)/“has a” (attribute) relationships (SNOMED International, 2014).  Prior to mapping, the definition of the hierarchy types in SNOMED-CT were evaluated and matched to the clinical meaning of the nursing parameters described in this study.  At the top of this class diagram, the concept of “wound” was added as it was clinically assumed, and an antecedent of, a wound assessment.  The SNOMED-CT hierarchy choice of Clinical Finding (finding) was used to describe the observation/judgement of this top clinical concept related to the patient having a wound (SNOMED International, 2016).  Within SNOMED-CT, the Clinical Finding hierarchy type has a related sub- hierarchy of Disorders, which was also used in this model to represents an abnormal clinical state/disease (wound diagnosis) (SNOMED International, 2016).  The visual representation of this in the conceptual model can be understood as “a wound diagnosis is a wound” or more specifically, “a neuropathic ulcer is a wound”.  Additionally, this SNOMED-CT sub-hierarchy relationship was represented by superimposing the addition of “is-a” to the inheritance UML line (SNOMED International, 2014). The SNOMED-CT hierarchy of Observable Entity was used to describe the clinical concepts related to wound assessment and goal of care.  In the SNOMED International (2016) editorial guide, this hierarchy type is used to group concepts representing a question which can produce an answer.  For example, we observe wound assessment parameters, such as “what is the exudate type”, and expect to answer with a finding or value, such as “serous” or “purulent”.  The concepts related to goal of care were harder to assign into the defined hierarchies of 45  SNOMED-CT.  However, based on the definitions provided by SNOMED International (2016), the goal of wound management could be considered a question requiring an answer, and was thus described as an Observable Entity.  The relationship links were added to the conceptual model to clinically represent attributes of the wound (finding).  This type of relationship link in SNOMED-CT is different than those relationship links describing parent-child, subtype meanings (SNOMED International, 2014).  The attribute relationships of a top source concept (i.e., wound finding) are meant to add further definition and greater semantic meaning.  In this case, a wound (finding) has wound assessment characteristics and a goal of care.  Put another way, these attributes occur because of the clinical state of a patient with a wound.  This “has-a” relationship was added to the UML lines for aggregation and composition because in both cases, these linkages represent a sub-ordinate state, reliant on the source concept (wound finding) to exist. The conceptual model also includes the specific data elements that are related to the parameter heading.  These were not conceptually described using SNOMED-CT hierarchies on this diagram; however, the related hierarchy which it was linked with was considered during the mapping process (i.e., the wound etiologies were successfully matched during mapping if the SNOMED-CT concept was part of the diagnosis hierarchy). The outcome of this work was the creation of a Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis and Outcome. Microsoft Visio was used to draw this model through its template for ULM modeling design.  46  3.7 Phase 2: Mapping for Cross-Coverage   3.7.1 Manual Mapping to SNOMED- CT The list of the 107 wound assessment, diagnosis, and outcome data elements were independently and manually mapped by Lori Block, using the open source SNOMED International, SNOMED-CT web browser (January 2016 International Edition).  Each data element was entered into the SNOMED-CT browser and matched to all possible pre-coordinated concept results.  The Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis and Outcome was used to guide the concept selection, supporting the retention of semantic meaning.  The matched concept(s) was/were then entered into an excel spread sheet with the SNOMED-CT fully specified name, the description concept, the concept identifier, the defined status, and the concept syntax.   3.7.2 Automated Electronic Mapping to SNOMED- CT Apelon TermWorks™ software (Apelon, Inc., 2016) January 2016 International Edition was utilized as an engine to machine automate the mapping of the wound care data elements to SNOMED-CT (Infoway, 2017, January).  This step was important to help increase the inference quality of possible matches because bias and possible human error can occur when working with the SNOMED-CT database of almost half a million concepts (Monsen et al., 2016; Polit & Beck, 2012).   Access to this software was gained through Standards Access membership with Canada Health Infoway.  Once the application plug-in was installed, the same list of wound care data elements were entered into an excel spread sheet and ‘cleaned’ of punctuation and grammatical characters, a requirement to increase matching accuracy.  The Apelon TermWorks™ settings 47  were modified to use the January 2016 International edition of SNOMED-CT, with matching return criteria of 90-100% accuracy.  This matching return setting allows the software to scan the entire SNOMED-CT database and display all possible matches (a few characters to the whole word).  The larger the accuracy spread, the greater volume of concept matches that will be displayed.  For example, if the accuracy was set at 70-100% a much larger set of possible concepts would be retrieved.   3.7.3 Literature Comparison of Wound Assessment, Diagnosis, and Outcome Mapping The wound care data elements in this study were matched to the published work of Harris et al. (2015) to find areas of overlap.  In the study conducted by Harris et al. (2015), the researchers used similar nursing wound care data elements (N=419) and mapped them to SNOMED-CT (January 2011) and LOINC.  Access to the detailed mapping results of Harris et al. (2015) was accessed through http://www.fhims.org/press_ulcer.html.  In a separate excel spread sheet, the BC nursing wound assessment, diagnosis, and outcome data elements were matched to equivalent wound care data elements compiled by Harris et al. (2015).  Consideration of the Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis and Outcome in British Columbia was used to accurately match the concept granularity and meaning.  Then, the SNOMED-CT concept identified by Harris et al. was added to the excel spread sheet of the BC Provincial Nursing Skin and Wound Committee, wound care data elements.  This component of comparison acted as a method to compare the results from this thesis to another similar study.  48  3.8 Phase 3: Expert Consensus Mapping  In a separate excel spread sheet, the accumulation of mapped results for the 107 data elements were listed in three parallel columns: 1) manual mapping, 2) automated mapping, and 3) literature comparison mapping results.  Using the Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis and Outcome as a guide, nursing subject matter expert consensus was reached to determine a selection for each data element.  Consensus was determined on the criteria of direct match, one-to-many, and no match (Kim et al., 2014) (see Table 3.2). Terms that were considered synonyms within SNOMED-CT were not counted as one-to-many matches.  The subject matter experts were two nursing wound care clinicians (Lori Block and Shannon Handfield).  Table 3.2 Matching criteria for mapping data elements to a standardized clinical terminology Criteria Definition Direct Match A data element matches the clinical concept available in SNOMED-CT. This is also known as a pre-coordination or enumeration.  One-to-many  A data element could be equivalently described using more than one concept in the same standardized terminology library (i.e., clinical term “sanguineous drainage” and SNOMED-CT concepts “sanguineous discharge from wound” and “sanguineous exudate from wound”) No match A data element does not have a direct match or one-to-many match in SNOMED-CT   3.9 Data Analysis Descriptive statistics were used to analyze the data.  Rates of equivalence of mapping between assessment, diagnosis and outcome concepts were compared.  Collective and grouped percentages were used to describe the numerical rate of equivalence between the BC Nursing Wound Assessment, Diagnosis, and Outcome data elements and SNOMED-CT.    49  3.10 Ethical Considerations No human subject participation or use of collected data was used in this study.  As per TCPS 2-Chapter 2 (Government of Canada, 2015), this research study did not qualify as requiring research ethics board approval, and as such, no formal ethics application was conducted.  Ethical consideration however, was exercised throughout the conduct of this study. The ethical practice of a nurse demands accountability in all actions taken in the care of a patient, including the sharing of knowledge (CNA & Infoway, 2008).  Confirmation was obtained from the Vancouver Coastal Health Research Institute and Data Steward that research ethics board approval was not needed for use of the data elements obtained from the Pixalere™.  Finally, the information generated from this study could be of use if plans to create an interoperable clinical patient chart with standardized terminologies are realized.  As such, the results will be published through the University of British Columbia, Faculty of Graduate Studies and shared with the BC Provincial Nursing Skin and Wound Committee.    3.11 Plans to Navigate Possible Difficulties  Advice and guidance was sought from my thesis supervisor and committee members throughout this studies progression.  I have also made contact with leaders in the field of terminology standards who were able to direct and suggest resources throughout this mapping activity (Manager of the Standards Collaborative within Canada Health Infoway, as well as, the Director of Terminology Standards in Alberta Health).  50  Chapter 4: Results  4.1 Introduction  The purpose of this study was to map the British Columbia (BC) Standardized Nursing Wound Assessment, Diagnosis, and Outcome parameters to SNOMED-CT and examine the rate of equivalence.  The method to obtain these results first involved creating a Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis and Outcome.  This guided the mapping process through three techniques to create a candidate list of concepts to support the production an expert consensus final list.  These mapping techniques included a process of manual mapping, followed by automated and literature comparison mapping.  Results of the expert consensus mapping process were evaluated using descriptive statistics.  The following chapter will outline the creation of the conceptual model and the rate of equivalence from the expert consensus mapping activity.  It will also highlight the compared results between each the manual, automated and literature comparison mapping to the final expert consensus list.  4.2 Phase 1: Conceptual Model Findings The creation of the Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis, and Outcome in British Columbia was informed by a) clinical meaning, b) unified modeling language (UML), and c) SNOMED-CT hierarchy types and relationship links.       51  4.2.1 Consensus of Model Development Figure 4.1 shows the resulting  conceptual model of knowledge representation for a nursing wound assessment, diagnosis, and outcome in BC.  This model acted as a foundational component of concept selection during manual mapping and consensus selection.  As described in section 3.5.1, this conceptual model only describes some of the nursing assessment, diagnosis, and outcome concepts related to wound care, but for the purpose of this study, only those 107 identified data elements were considered for mapping (see “*” in Figure 4.1, 4.2, 4.3, 4.4).  To ease readability, each section in Figure 4.1 has been portioned into subsections: Wound Assessment (Figure 4.2); Wound Diagnosis (Figure 4.3); Goal of care (Figure 4.4).      52  Figure 4.1 Conceptual Model of Knowledge Representation for Nursing Wound Care (overview)    53  Figure 4.2 Conceptual Model of Knowledge Representation for a Nursing Wound Assessment  Wound (finding)- Wound Diagnosis- Goal of CareWound Assessment (observable entity)- Date of Onsetaggregation'has a'- Recurrence- Wound Pain- Wound Measurement- Wound Bed- Periwound Skin- Wound Edge- Wound Exudate Type- Wound Odour-Wound Assessment-Wound Exudate AmountDate of Onset*Date FieldRecurrence*YesNoWound Pain*Value 0-10Wound Measurement*- Undermining*- Sinus Tract*Depth*Width*Length*Wound Edge*Diffuse*Demarcated*Callused*Attached*Epithelializing*Hypergranulation*Not Attached*Rolled*Scarred*Exudate Amount*None*Scant*Small*Moderate*Large/copious*Exudate Type*Nil*Serous*Sanguineous*Purulent*UnderminingClock face hourClock face hourLengthSinus TractClock face hourLengthWound Bed*- Non-Viable Tissue- Viable Tissue- Obscured TissueViable TissueBone*Epithelial islands*Fully epithelialized*Granulation*Superficial pnk, red*Tendon*Viable graft*Non-Viable TissueFriable*Non granulation*Hypergranulation*Blister*Fully callused*Fungating*Hematoma*Malignant*Weepy skin*Obscured TissueNot visable*Biochemical wound product*Periwound Skin*Callused*Bruised*Boggy*Blister*Dry*Edema*Erythema > 2cm*Erythema < 2cm*Excoriated/denuded*Fragile*Increased warmth*Indurated > 2 cm*Indurated < 2cm*Macerated*Intact*Weepy*Rash*Tape tear*Eschar - soft, boggy*Eschar - dry, soft*Scab*Slough*Wound Odour*Yes*No*Fistula*Assessment 54  Figure 4.3 Conceptual Model of Knowledge Representation for a Nursing Wound Diagnosis  Wound (finding)- Wound Diagnosis- Goal of CareWound Diagnosis (disorder)- Peripheral Vascular Disease Etiology- Pressure Injury Etiology- Surgical Etiology- Traumatic Injury Etiologyinheritance'is a'- Incontinence Associated Etiology-Wound AssessmentPressure Injury EtiologyPressure Ulcer - Stage 1*Pressure Ulcer - Stage 2*Pressure Ulcer - Stage 3*Pressure Ulcer - Stage 4*Pressure Ulcer - Stage X*Pressure Ulcer - SDTI*Traumatic Injury EtiologyBurn*Skin Tear*Trauma*Surgical EtiologyDonor Site*Graft Site*Surgery (Secondary Intent)*Other Wound EtiologyAbscess*Drug Reaction*Neuropathic EtiologyDiabetic/Neuropathic*Fistula - non stomatized*Fistula - stomatized*Infectious*Inflammatory Disease*Pilonidal Sinus*Skin Disease*- Other Wound Etiology Incontinence Associated EtiologyIncontience Associated Dermatitis*Lymphatic EtiologyLymphedema*- Neuropathic Etiology- Lymphatic EtiologyOncology EtiologyIrradiation*Malignant*- Oncology Etiology* Data elements used in mapping activityDiagnosis 55  Figure 4.4 Conceptual Model of Knowledge Representation for a Nursing Wound Outcome  Wound (finding)- Wound Diagnosis- Goal of CareGoal of Care (observable entity)-Goal of Wound Managementcomposition'has a'-Wound AssessmentGoal of Wound Management To Heal the wound*To Maintain the wound*To Monitor/Manage the non-healable wound*     4.3 Phase 2: Mapping Findings 4.3.1 Manual Mapping The manual mapping process was completed by Lori Block, using the open source SNOMED International, SNOMED-CT Web browser, with possible matches informed by the conceptual model of knowledge representation (Figure 4.1).  This Web browser was free for a Canadian to use, and required the user to agree to the SNOMED International, SNOMED-CT browser license agreement.  The January 2016 International edition was used, and all concepts were manually mapped using the words described in the conceptual model.  As this process was carried out by a clinical content expert, there were times when a search was conducted with a word that had the same semantic meaning, but was additional to the word in the conceptual map (e.g., exudate, discharge, and drainage were each used to search exudate type) (Appendix B).  The conceptual model also acted as a guide for the selection of the possible concept matches as it defined the clinical relationships and hierarchy types.  For example, when manually mapping SNOMED-CT, concepts for wound etiologies were only considered if they were part of the 56  wound diagnosis (disorder) hierarchy.  Further, given that the concepts related to the wound assessment and goal of care were within the observable entity hierarchy, concepts were matched only if the parameter and data elements matched the criteria of a “question” and “answer”.  Using the criteria for mapping data elements for manual mapping, a cumulative total of 62/107 (57.9%) had possible direct matches, 10/107 (9.4%) had possible one-to-many matches, and 35/107 (32.7%) had possible no matches (see Table 4.1 and Appendix B, C, and D).  When considering specific categories, wound assessment had 43/76 (56.6%) possible direct matches, 6/76 (7.9%) had possible one-to-many matches, and 27/76 (35.5%) had possible no matches.  Wound diagnosis had 19/28 (67.8%) possible direct matches, 4/28 (14.3%) possible one-to-many matches, and 5/28 (17.9%) possible no matches.  For goal of care, 0/3 (0%) had direct matched, 0/3 (0%) had one-to-many matches, and 3/3 (100%) had possible no matches (see Table 4.1).    Table 4.1 Manual Mapping of Nursing Wound Documentation Parameters to SNOMED-CT  Wound  Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n Wound Assessment 43 (56.6%) 6 (7.9%) 27 (35.5%) 76 Wound Diagnosis 19 (67.8%) 4 (14.3%) 5 (17.9%) 28 Goal of Care 0 0 3 (100%) 3 Total matches 62 (57.9%) 10 (9.4%) 35 (32.7%) 107 (100%)  4.3.2 Automated Mapping The automated mapping process was facilitated by Lori Block.  The list of data elements were entered into an excel spreadsheet and all grammatical characters were removed.  An Excel plug-in for TermWorks™ was then activated against these 107 data elements.  Results were 57  retuned with a matching rate of 90-100% accuracy, as per the software configuration.  The automated mapping results had 57/107 (53.3%) possible direct matches, 50/107 (46.7%) possible one-to-many matches, and 0/107 possible no matches.  Specifically for each category, wound assessment had 41/76 (53.9%) possible direct matches, 35/76 (46.1%) possible one-to-many matches, and 0/76 (0%) possible no matches.  Wound diagnosis had 14/28 (50%) possible direct matches, 14/28 (50%) possible one-to-many matches, and 0/28 (0%) no matches.  Goal of care had 2/3 (66.7%) possible direct matches, 1/3 (33.3%) possible one-to-many matches, 0/3 (0%) possible no matches (see Table 4.2).       Table 4.2 Automated Mapping of Nursing Wound Documentation Parameters to SNOMED-CT  Wound  Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n Wound Assessment 41 (54%) 35 (46%) 0 76  Wound Diagnosis 14 (50%) 14 (50%) 0  28  Goal of Care 2 (66%) 1 (34%) 0  3  Total matches 57 (53.3%) 50 (46.7%) 0% 107 (100%)   4.3.3 Literature Mapping The work of Harris et al. (2015) was used as a benchmark to match equivalent wound care concepts, already mapped to SNOMED-CT.  The researchers in that study took 419 wound care data elements related to skin, wound, and pressure ulcer prevention and mapped them to SNOMED-CT.   For this study, Lori Block took the 107 selected wound care data elements (those marked with an “*” in the conceptual model) and compared them against the 419 mapped concepts in the 58  Harris et al (2015) study.  These were matched, using the conceptual model (Figure 4.1) and concept definitions (Appendix A) to determine a match, no match, or one-to-many match (Table 3.2).  The outcome of this activity concluded that 62 of the BC nursing wound care data elements matched against the 419 Harris et al. (2015) wound care data elements. Of these 62 clinical matches, four were semantically close, but vague in SNOMED-CT meaning.  Such that they were added to the spread sheet as one-to-many match with the intent to consider further during the planned expert consensus matching phase (periwound skin: <2cm indurated, >2cm indurated, <2cm erythema, >2cm erythema).  These 62 semantically equivalent concepts from Harris et al. (2015) were then added to the outcome mapping Excel spread sheet.  Those parameters from this study that did not find a match in Harris et al. (2015) were not considered “No Match” as the clinical focus of each study was not equivalent.  However, it can be noted that between this study, and the Harris et al. (2015) study, 45 data elements were new.  Reasons for this overall match rate included differences in study design, concept scope, granularity, and SNOMED-CT version.    Table 4.3 Comparison of Mapped Nursing Wound Documentation Parameters to Harris et al.  Wound  Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n  Wound Assessment 47 (61.8%) 4 (5.3%) 25 (32.9%) 76 Wound Diagnosis 11 (39.3%) 0 17 (60.7%) 28 Goal of Care 0 0 3 (100%) 3 Total matches 58 (54.2%) 4 (3.7%) 45 (42.1%) 107 (100%)   59  4.4 Phase 3: Expert Consensus Mapping to find Rate of Equivalence  In this final step, the independent mapping results from the manual, automated, and literature comparison activities were gathered and listed in parallel, in an Excel spread sheet (Appendices B, C, D).  As a measure to decrease bias, two subject matter experts (Lori Block and Shannon Handfield) evaluated this compiled list and compared each data against the Conceptual Model of Nursing Knowledge Representation for a Nursing Wound Assessment, Diagnosis, and Outcome.  Shannon Handfield was not part of the previous manual, automated, or literature mapping activities, and thus provided an independent perspective when reviewing the accumulated results (Scichilone & Rihanek, 2015).  Using the description of match, no match, and one-to-many match (Table 3.2) they created, through summative evaluation and consensus, the final expert consensus list of concepts in SNOMED-CT (Figure 3.1).   It should be noted that the outcome of expert consensus mapping was different from the results of the manual mapping.  Though Lori Block was the researcher in both cases, after discussion with Shannon Handfield, interpretation of the SNOMED-CT hierarchies and relationships, and the re-review of the conceptual model, some of the initial manual mapping results were found to be too granular or the SNOMED-CT hierarchy placement did not match the semantic clinical meaning (i.e., blister (morphological abnormality)).    Thus, the expert consensus list is the outcome of this overall mapping research activity (i.e., the expert consensus mapping produced the final determination of the rate of equivalence between the BC standardized nursing wound assessment, diagnosis, and outcome and SNOMED-CT).   The rate of equivalence between the 107 wound care data elements and SNOMED-CT found that there were 43 (40.2%) direct matches, 2 (1.9%) one-to-many matches, and 62 (57.9%) no matches (Table 3.2).  60  Table 4.4 Expert Consensus Mapping of Wound Documentation Parameters to SNOMED-CT  Wound  Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n  Wound Assessment 27 (35.5%) 1 (1.3%) 48 (63.2%) 76 Wound Diagnosis 16 (57.1%) 1 (3.6%) 11 (39.3%) 28 Goal of Care 0 0 3 (100%) 3 Total 43 (40.2%) 2 (1.9%) 62 (57.9%) 107 (100%)   4.4.1 Rate of Equivalence between SNOMED-CT and Wound Assessment Considering these values at a lower level of granularity, the rate of equivalence between SNOMED-CT and the wound assessment data elements found 27/76 (35.5%) direct matches, 1/76 (1.3%) one-to-many match, and 48/76 (63.2%) no matches (Raw data in Appendix B).   For the specific instance of the one-to-many match, the SNOMED-CT concepts of Sanguineous exudate from wound (finding) and Sanguineous discharge from wound (finding) were manually mapped to the equivalent clinical concept of Sanguineous drainage.  This duplicate finding was only revealed in the manual mapping method.  As well, it was noted that some of the direct matches were found in only one single mapping method.  For example, the automated mapping for the SNOMED-CT concept Nil (qualifier value) was the only direct match for the clinical concept “nil” (drainage).  Both manual and literature mapping did not discover this SNOMED-CT concept.  As another example, the concepts Scanty (qualifier value), Exudate observable (observable entity), and Odour of Exudate (observable entity) were only discovered through the literature comparison mapping.      61  Table 4.5 Expert Consensus Mapping to SNOMED-CT for Wound Assessment Data Elements  Wound Assessment Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n  Date of Onset 0 0 1 (100%) 1 Recurrence 0 0 1 (100%) 1 Wound Pain 0 0 1 (100%) 1 Wound Measurement 5 (71.4%) 0 2 (28.6%) 7 Wound Bed 4 0 19 (82.6%) 23 Wound Exudate Type 4 (80%) 1 (20%) 0 5 Wound Exudate Amount 6 (100%) 0 0 6 Wound Odour 3 (100%) 0 0 3 Wound Edge 4 (40%) 0 6 (60%) 10 Periwound Skin 1 (5.3%) 0 18 (94.7%) 19 Total 27 (35.5%) 1 (1.3%) 48 (63.2%) 76 (100%)   4.4.2 Rate of Equivalence between SNOMED-CT and Wound Diagnosis The rate of equivalence between SNOMED-CT and the wound diagnosis data elements found 16/28 (57.1%) direct matches, 1/28 (3.6%) one-to-many match, and 11/28 (39.3%) no matches (Raw data in Appendix C).  The specific instance of the one-to-many match, the SNOMED-CT concepts of Irritant contact dermatitis caused by contact with urine and/or feces (disorder) and Irritant contact dermatitis due to incontinence (disorder) were manually mapped to the equivalent clinical concept of Incontinence Associated Dermatitis.  This duplicate finding was only revealed in the manual mapping method.       62  Table 4.6 Expert Consensus Mapping to SNOMED-CT for Wound Diagnosis Data Elements  Wound Diagnosis Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n  Peripheral Vascular Disease 3 (100%) 0 0 3 Neuropathic 1 (100%) 0 0 1 Pressure Injury 5 (83.3%) 0 1 (16.7%) 6 Surgical  1 (33.3%) 0 2 (66.7%) 3 Traumatic Injury 2 (66.7%) 0 1 (33.3%) 3 Incontinence Associated 0 1 (100%) 0 1 Lymphatic 1 (100%) 0 0 1 Oncology 1 (50%) 0 1 (50%) 2 Other Wound 2 (25%) 0 6 (75%) 8 Total 16 (57.1%) 1 (3.6%) 11 (39.3%) 28 (100%)  4.4.3 Rate of Equivalence between SNOMED-CT and Goal of Care The rate of equivalence between SNOMED-CT and the goal of care data elements found 0/3 (0%) direct matches, 0/3 (0%) one-to-many matches, and 3/3 (100%) no matches (Table 4.7) (Raw data in Appendix D).  This interpretation meaning, there were no equivalent matches in SNOMED-CT to represent the goal of care for wound management.   Table 4.7 Expert Consensus Mapping to SNOMED-CT for Goal of Care Data Elements  Goal of Care Direct Match n(%) One-to-Many n(%)  No Match n(%)  Total  n  To Heal the Wound 0 0 1 (100%) 1 To Maintain the Wound 0 0 1 (100%) 1 To Monitor/ Manage the Non-Healable Wound 0 0 1 (100%) 1 Total 0% 0% 3 (100%) 3 (100%)  63  4.4.4 Inter-Mapping Comparison against Expert Consensus List The three mapping methods (manual, automated and literature comparison) were compared against the final  expert consensus mapping results (i.e., data presented in Tables 4.1 – 4.4).  When reviewing the mapping results for each mapping method independently against the expert consensus list, selection agreement often occurred, but not exclusively (Raw data in Appendices B, C, D).  Table 4.8 shows the summary of this comparison.    Table 4.8 Differences between Expert Consensus Mapping and Manual, Automated and Literature Mapping  Method Direct Match n(%) One-to-Many n(%) No Match n(%) Expert Consensus Mapping 43 (40.2%) 2 (1.9%) 62 (57.9%)     Manual Mapping 62 (57.9%) 10 (9.4%) 35 (32.7%) Difference between manual and consensus mapping -19 (-17.7%) -8 (-7.5%) 27 (25.2%) Automated Mapping (Apelon™) 57 (53.3%) 50 (46.7%) 0% Difference between automated and consensus mapping -14 (-13.1%) -48 (-44.8%) -62 (-57.9%) Literature Comparison 58 (54.2%) 4 (3.7%) 45 (42.1%) Difference between literature and consensus mapping -15 (-14.0%) 2 (1.8%) 17 (15.8%)     The difference between manual and expert consensus mapping was -19 (-17.7%) for direct matches, -8 (-7.5%) for one-to-many, and 27 (25.2%) for no matched.  The difference between automated and expert consensus mapping was -14 (-13.1%) for direct matches, -48 (-44.8%) for one-to-many matches, and -62 (-57.9%) for no matches.  Finally, the difference between literature and expert consensus mapping was -15 (-14.0%), 2 (1.8%) for one-to-many matches, and 17 (15.8%) for no matches.   64  4.5 Summary In this chapter, the creation of the conceptual model and the rate of equivalence from the expert consensus mapping activity were analyzed and explained.  Mapping results showed that manual mapping found the highest proportion of direct matches (57.9%) followed by literature comparison mapping, automated mapping and comparison mapping with proportions of 54.2%, 53.3% and 40.2% respectively. Goal of care was only identified using automated mapping, but these were later considered non-matches when expert consensus mapping was completed.   65  Chapter 5: Discussion  5.1 Introduction  This thesis sought to use scientific inquiry to determine a rate of equivalence and concept representation between the BC Nursing Wound Documentation standard and SNOMED-CT.  The main findings include i) the creation and use of a Conceptual Model of Knowledge Representation for Nursing Wound Care, ii) a novel method to map nursing concepts in SNOMED-CT, and iii) the creation of an expert consensus list of mapped wound care data elements, with identification of direct, one-to-many, and no matches.   The following discussion presents these findings within the methodology of descriptive research using the theoretical lens of Matney’s (2015) Theory of Wisdom in Action for Clinical Nursing.  As a conclusion, this paper will posit the possibilities of using the lessons learned in this research for future standardized clinical terminology research activities.    5.2 Use of a Conceptual Model  Exploration of the phenomenon of knowledge representation for the BC nurses wound assessment, diagnosis and outcome documentation standard, occurred through development of a conceptual model.  As Fawcett and Gigliotti (2001) describe, conceptual models can provide discipline-specific frames of reference to guide and inform thinking.  Though this conceptual model was not intended for theory generation or theory testing, it did guide the mapping process and provide a format which to visualize the nursing “work” executed during a wound assessment, wound diagnosis and goal of care decision.  Further, in designing this model, computer architecture modeling (UML) and SNOMED-CT hierarchy features were included to 66  guide decision making in the ensuing mapping activities.  Within the science of clinical terminology mapping, clinical engagement with computer archetype models to guide mapping activities has been previously utilized (Harris et al., 2015; Kim & Park, 2012; Qamar, Kola, & Rector, 2007).  For example, Qamar, Kola, and Rector (2007) used existing pre-revision (ambiguous) and post-revision (unambiguous) archetype models to represent histology pap screening.  In these other studies, clinician concept mapping to SNOMED-CT occurred, using the models to guide mapping, and results were compared.  These researchers concluded that clinician mapping accuracy with SNOMED-CT improved when good modeling design unambiguously represented the intended clinical data element, making the concept clear to clinician interpretation.    5.3 Implications of the Conceptual Model of Knowledge Representation for Nursing Wound Care  This study suggests that a reference clinical model which integrates computer architecture design may be useful when mapping clinical concepts to a standardized clinical terminology.  Though its usage in this thesis to increase accuracy cannot be substantiated through statistical analysis (no group comparison), the creation and usage of the conceptual model was of value and may assist future mapping activities through clear representation of the intended target concept.    Additionally, to the best of this researcher’s knowledge, it is the first time that general and accumulated nursing wound assessment, diagnosis and outcome concepts have been represented in a conceptual model using UML class diagram design with SNOMED-CT hierarchy integration.  Previous work in this field has focused on skin and pressure ulcer 67  assessment data (Harris et al., 2015; Kim & Park, 2012); however, expansion of further nursing wound diagnosis and goal of care concepts appears to be new.  This visual account of nursing wound care “work”, in the context of computer architecture design, is important as we continue to find ways which to define nursing’s body of knowledge in clinical informatics (Ronquillo, Currie, & Rodney, 2016).  For example, when given to a computer architect, a UML class diagram can be used to create high-level interactions between decision support networks, or nodes to aggregate patient data (Medvidovic, Rosenblum, Redmiles, & Robbins, 2002).  In a quantifiable way, when placed within the context of the nursing process (ANA, n.d.), a BC nurse would consider these 107 parameters (and more), prior to performing a planned intervention. The model provides a computer infrastructure that permits documentation of nurses’ clinical judgement and underscores the idea for computer design (and beyond), that for nurses, changing a wound bandage is not a mutually exclusive task devoid of contextualized, critical thinking.    Overall, the desired intent behind the creation of a conceptual model through using a UML class diagram was met because the model framed the conceptual meaning of nursing wound assessment, diagnosis and outcome, and guided the researchers in concept mapping.  Future nursing mapping activities might benefit from developing a conceptual model of their sample clinical concepts using a UML class diagram in order to clarify clinical meaning, support mapping decisions, visualize nursing “work”, and aid in computer EHR design decisions.       5.4 Mapping Methods  Similar to other studies, mapping concepts to standardized clinical terminologies, manual, automated, and expert consensus methods were used in this research thesis (Monsen et al., 2016; Harris et al. 2015; Kim et al., 2014; Matney et al., 2012; Saitwal et al., 2012).  Further, 68  this study also incorporated a method to compare a previously mapped set of similar data elements: literature comparison mapping (Harris et al, 2015).    In the published literature considered in this paper, several researchers used an iterative approach to mapping, where an automated or semi-automated method would “cull” the concept list for initial matches, followed by review and analysis of the results (Kim et al., 2014; Saitwel et al., 2012; Lau et al., 2008; Richesson et al., 2006).  As an additional measure, some of these studies would then create a concept sub-list of vague or no matches to then manually map against the target terminology system (Kimet al., 2014; Saitwel et al., 2012).  For example, the methods used in the Richesson et al. (2006) two researchers independently used an automated mapping tool (Apelon TermWorks™) to create an initial list of possible matches. When matches were not found, the researchers then used knowledge of the SNOMED-CT hierarchy to manually find missing pre- or post-coordinated matches.  As well, Saitwal et al. (2012) employed automated and semi-automated mapping methods to map proprietary medication codes to SNOMED-CT.  These researchers noted that a further 1626/8447 (23%) needed to be manually mapped to SNOMED-CT.       The work in this thesis explored repetitive mapping without the iterative delineation of sequential matched concepts.  This resulted in each of the 107 data elements being individually mapped using three different methods (manual, automated, and literature comparison).  Though this was a labor intensive endeavour, it considered the weaknesses and strengths each method inherently possessed, and discussed in previous studies.  For example, 1) though manual mapping has known variability (Monsen et al., 2016; Kim et al., 2014; Kim et al., 2012), it allows the researchers to use their clinical expertise to search for alternative concept names (i.e., Pressure Ulcer is also known as Pressure Injury and Decubitus Ulcer); 2) though automated 69  mapping appears to be insufficient alone, requiring human interpretation and validation (Kim et al., 2014; Richesson et al, 2006), it can facilitate the quick processing and collation of several match possibilities, using integrated system rules outside human interpretation (and bias); 3) and though previously mapped lists might be older than the current terminology version (i.e., SNOMED-CT 2011 versus SNOMED-CT 2016) (Harris et al., 2015), mapping comparison facilitates peer review, maintenance and research utilization.    5.4.1 Manual Mapping  Previous research has demonstrated that clinical expertise is an important requirement for effective and efficient terminology mapping activities (Kim et al., 2012; Richesson et al., 2006).  Kim et al. (2012) identified medical diagnosis and operative concepts, which were then mapped to SNOMED-CT by two physicians (one neurosurgeon and one ophthalmologist) and one medical technician trained in clinical EHR coding.  Kim et al. found an increased accuracy in mapping between the two physicians, when compared to the medical technician.  Further, the neurosurgeon had greater mapping accuracy to neurosurgical concepts, and the ophthalmologist had greater mapping accuracy to ophthalmology concepts.  Kim et al. (2012) concluded that domain knowledge played a factor in mapping accuracy, and those who are involved in the generation of the actual clinical data, should be involved in the SNOMED-CT mapping activities.    Drawing upon this strength, this thesis incorporated manual mapping performed by the primary investigator: a wound care clinician (Lori Block).  It also utilized the domain expertise of two wound care clinicians (Lori Block and Shannon Handfield) to review all returned matched concepts, to create the expert consensus list.  Though group comparison to evaluate clinical 70  expertise in mapping accuracy was not possible in this study, and inter-rater agreement is not possible as the final expert consensus list was created only through consensus, clinical expertise did facilitate flexible concept searching and concept comprehension.  Expert consensus mapping identified a much smaller number of direct or one-to-many matches than the other forms of mapping.  This may be a result of having two wound care domain experts doing the expert consensus mapping rather than one person or an automated method.  Further, having two clinical experts may have been a contributing factor in finding two instances of one-to-many matches.  To explain this point, the finding of SNOMED-CT concepts Irritant contact dermatitis caused by contact with urine and/or feces (disorder) and Irritant contact dermatitis due to incontinence (disorder) were manually mapped to the clinical concept of Incontinence Associated Dermatitis.  This duplicate finding was only revealed in the manual mapping method.  These SNOMED-CT concepts were found by the primary researcher of this thesis, who recognized the different words, but equivalent meaning.  Clinically, these two concept classifications were considered equivalent because an ongoing, transient, or single “incontinence” episode includes skin “contact with urine and/or feces” (Doughty et al., 2012; Doughty, 2006).  This duplicate concept may have remained unknown if this study were designed to perform automated mapping first to “cull” initial matches.  This statement can be supported by reviewing the automated mapping results of Irritant dermatitis due to incontinence (disorder), which was also the same in the literature comparison.  The concept Irritant contact dermatitis caused by contact with urine and/or feces (disorder) did not display for either.    In a similar instance, mapping the clinical concept of sanguineous wound drainage revealed an instance of SNOMED-CT concept duplication.  Manual mapping revealed the two equivalent SNOMED-CT concepts of Sanguineous exudate from wound (finding) and 71  Sanguineous discharge from wound (finding).  Clinically, the difference between “exudate from wound” and “discharge from wound” is null within the context of a BC Nursing Wound Assessment.  Again, it is possible that using an automated method to “cull” initial matches may have missed this duplication as automated mapping results were considered non-matches and the literature comparison provided a single match to Sanguineous discharge from wound (finding).  The concept Sanguineous exudate from wound (finding) did not display for either.       Yet, clinical expertise advantage is only one consideration when working with a complex standardized clinical terminology, such as SNOMED-CT.  There were times when the researcher in this study manually mapped concepts to SNOMED-CT, which were later not included in the final expert consensus list.  This manual mapping variability may be attributed to the researcher’s inexperience using SNOMED-CT, possible selection bias, concept granularity and/or concept ambiguity.  As an example, the expert consensus list found 43 direct matches; however, the manual mapping found 62 (-17.7% difference).  This higher number may be anticipated or acceptable when creating a candidate match list; yet, if this were to be the only method of mapping to create a final match list, possible coding discrepancies may have resulted.    5.4.2 Automated Mapping   Another discussion point relates to automated mapping requiring human interpretation and evaluation.  In this, Apelon TermWorks™ was selected to use for automated mapping of data elements.  For the 107 study concepts, TermWorks™ returned 269 possible matches with a 90-100% restraint.  Of these possible matches, 17/43 (39.5%) were included as direct matches in the final expert consensus list.  Alternatively, this could be read as 17/269 (6.3%) possible returned matches could be used to codify these 107 concepts in an EHR.  Though the latter value 72  may seem quite low for mapping accuracy, the ease and speed at which these concepts were matched suggests that this form of mapping may provide a fast, low-barrier option, which researchers can use to create candidate mapping lists.  As mentioned above, other researchers have used this type of mapping successfully to “cull” larger lists clinical concepts (Kim et al., 2014; Saitwel et al., 2012; Lau et al., 2008; Richesson et al., 2006).  Further, this method identified one concept match which was not discovered in either the manual or literature comparison.  This can be demonstrated as the SNOMED-CT concept Nil (qualifier value) being the only direct match for the clinical concept “nil” (drainage).  Neither manual nor literature mapping discovered this SNOMED-CT concept.     5.4.3 Literature Comparison Mapping  The literature comparison also played a key role in concept matching.  Of those 62 concepts that were equivalent (direct match or one-to-many match) in both this thesis and the study by Harris et al. (2015), 30/43 (69.8%) mapped concepts from the Harris et al. study were included in the final expert consensus list for direct matches.  In other words, there were 62/107 candidate matches, and of those, 30 were included in the final expert consensus list.  Further, with the addition of mapping from the published literature, three additional concepts were included in the expert consensus final list, of which neither manual nor automated mapping found.  These concepts were: Scanty (qualifier value), Exudate observable (observable entity), and Odour of Exudate (observable entity).       73  5.4.4 Expert Consensus Mapping   Once the three above-mentioned mapping methods were completed by the author, the results of each were compiled in an excel spreadsheet (Appendices B, C, D) and reviewed by two domain experts.  As noted above, the purpose of the multiple methods was to discover as many mapping matches as possible.  Yet, to clarify, frequency of matches was only a consideration used during the final expert consensus list creation.  The process of concept selection also integrated the clinical expertise of the two expert nursing wound care clinicians and the Conceptual Model of Knowledge Representation for a Nursing Wound Assessment, Diagnosis, and Outcome, against the criteria of match, one-to-many, and no match.  For example, all three techniques matched the clinical concept of Pressure Ulcer- Stage 1 to the SNOMED-CT concept Pressure ulcer stage 1 (disorder).  This was considered by the two researchers and was selected as a direct match in the final expert consensus list.  However, in two mapping methods (manual and automated), the wound etiology of Trauma was matched to SNOMED-CT concept of Traumatic injury (disorder).  This appears to have concept convergence, and a positive match.  However, when the two reviewers considered this concept against the conceptual model as an entity of traumatic injury causing a wound, the SNOMED-CT concept was not specific enough and became a non-match in the final expert consensus list.  5.5 Implications of Mapping Methods  The individual mapping methods discussed in this thesis highlight the point that one mapping method alone does not provide the rigour required to provide a comprehensive mapped data set.  Each phase was designed to guide the researcher towards a reportable outcome, which was well-balanced and attempted to decrease selection bias.  It also provides an alternative 74  approach to traditional mapping methods, where concurrent mapping methods, versus iterative mapping methods, can be used to create candidate mapping options, and reviewed by content experts to create an expert consensus list.  Though the approach used in this study was time consuming, and perhaps not a pragmatic choice for local organizations who will be coding large clinical data sets, it may be useful for exploratory or descriptive research studies whose purpose is to closely examine domain content coverage, discover concept duplication, or to peer review previously mapped value sets.  For example, Harris et al. (2015) discuss the expectation that clinicians would want to build and expand upon their work, however were challenged with a standard process to facilitate this.  Ideally, the products of such mapping activities would be shared across and between organizations, with subgroups providing updates at regular intervals.  Clinically, it is also worth noting that the six American study sites included in the Harris et al. (2015) paper, had 62/107 (57.9%) equivalent (direct match or one-to-many match) wound care concepts to the BC Nursing Wound Care Documentation standard.  Again, though the scope of what was included as sample wound care concepts differed, there does exist an opportunity to share and compare clinical documentation standards, beyond the codified lists.  Future researchers may look to reuse and collate several smaller mapping studies, with the purpose to find similar or shared clinical standards, which would ultimately help organizations looking to share disparate EHR information and improve semantic interoperability outcomes.  This type of collaborative evaluation could advance the practice of nursing, whose profession has historically struggled with coming to clinical consensus regarding patient care documentation standards.      75  5.6 Missing Content Coverage  This study was executed in three phases, examining nursing wound care concept representation in SNOMED-CT.  The main findings demonstrated that the rate of equivalence between the BC nursing wound assessment, diagnosis, and outcome data elements and SNOMED-CT found 43/107 (40.2%) had pre-coordinated, direct matches, 2/107 (1.9%) had one-to-many matches, and 62/107 (57.9%) had non matches. When this rate of equivalence was abstracted for each of the nursing process categories, Wound Assessment had 27/76 (35.5%) direct matches, 1/76 (1.3%) one-to-many match, and 48/76 (63.2%) no matches in SNOMED-CT; Wound Diagnosis had 16/28 (57.1%) direct matches, 1/28 (3.6%) one-to-many match, and 11/28 (39.3%) no matches in SNOMED-CT; and Goal of Care had 3/3 (100%) no matches in SNOMED-CT.    These values appear to hold similarities to other SNOMED-CT mapping studies.  For example, in one of the largest activities to map nursing domain content to SNOMED-CT, Kim, Hardiker, and Coenen (2014) mapped ICNP concepts to SNOMED-CT.  These researchers found that 436/783 (55.7%) of the ICNP nursing diagnostic and outcome concepts had pre-coordinated matches in SNOMED-CT.  Reasons cited by the authors for this relatively low equivalence rate include difference in content coverage, lack of pre-coordinated matches in SNOMED-CT, and semantic differences with the (medical) hierarchy structure.   In contrast, the benchmark US study by Harris et al. (2015) included nursing skin and wound care assessment and diagnostic parameters from six large health care organizations.  Working with experts from SNOMED-CT and LOINC, the researchers were able to map 261/320 (82%) of these concepts to SNOMED-CT.  A possible reason for this higher rate of equivalence than what was identified in this thesis could be explained by Harris et al.’s decision 76  to use both the international core SNOMED-CT extension and the US extension of SNOMED-CT for mapping.  This effectually increased the quantity of concepts available for mapping and provided a greater opportunity to find SNOMED-CT matches.  Further, from their larger data set of identified skin and wound assessment parameters (N=419), researchers approached mapping with the intention of using LOINC and SNOMED-CT together.  The outcome of this included some assessment parameter headings, such as “periwound skin”, being excluded from the SNOMED-CT mapping and instead, only mapped to a LOINC code.   Finally, in an earlier medical study by Richesson et al. (2006), researchers analyzed case report forms used in clinical research for vasculitis studies and found 616 unique assessment parameters.  The intent of the study by Richesson et al. was to estimate content coverage for these data elements in SNOMED-CT.  The final result found an estimated 23% of the concepts had pre-coordinated matches to SNOMED-CT (Richesson et al., 2006).  Though researchers were exploring SNOMED-CT to only estimate content coverage, they noticed that their mapping data elements were generally patient specific clinical “questions,” and that the more complex a clinical concept was, the greater difficulty they had in matching it to SNOMED-CT.  As the above discussion demonstrates, the nuances of mapping clinical parameters to SNOMED-CT are variable, not mutually exclusive, and concept-type dependent.  Difficulties, such as concept granularity, hierarchy placements, and missing coverage are used by authors to explain their mapping results.  Similar considerations can be posited for this thesis study as well.  For example, the mapping result for the concepts, To Heal Wound, To Maintain the Wound, and To Monitor/Manage the Wound, scored 0/3.  The context, for which these concepts exist, is within the nursing process of outcome planning.  Though SNOMED-CT does have concepts related to wound healing, such as Wound healed (finding), O/E wound healing delayed (finding), 77  and Wound healing status (observable entity), they do not represent the same thing.  Clinically, ‘goal of care’ concepts are set by the nurse, with consultation from the patient/family and other health care professionals, to direct the intervention and evaluation measures for the wound (BC Provincial Nursing Skin & Wound Committee, 2015).  Specifically, the goal of care (i.e., To Heal the Wound) represents a clinical judgment, which directs an intervention (i.e., Wound debridement (procedure)) which would then be evaluated (i.e., Wound healed (finding) or O/E wound healing delayed (finding)).  Noting this, placement in the current hierarchy structure of SNOMED-CT is difficult and unclear, and requires further exploration. Further, a question and consideration could be posed within this study regarding the differences between Wound Assessment (35.5% direct match) and Wound Diagnosis (57.1% direct match) concept representation.  Historically, it appears that there has been previous work to map nursing diagnoses, interventions and outcomes to SNOMED-CT (Kim et al., 2014; Kim & Matney, 2014), with perhaps less focus on mapping granular nursing assessment parameters.  It is possible that the SNOMED-CT rate of equivalence has been impacted by these mapping priorities.  It may also be related to the local clinical culture driving the BC Nursing Wound assessment, diagnosis and outcome standard.  It is possible that BC (or Canadian) nursing wound care approaches documentation differently than other countries, and as this study is the first of its (known) kind in BC; there has been limited opportunity to add these concepts to SNOMED-CT.  For example, when compared to the Harris et al. (2015) skin and wound parameters, only 62/107 concepts were equivalent.  Though the clinical parameter focus had some differences (i.e., BC: all wounds, Harris et al.: skin and pressure wounds), those which were related had granular and semantic differences (i.e., BC: “soft, boggy eschar” and “dry, stable eschar” versus Harris et al.: “eschar”).  In this example, one could argue that “soft, boggy” and “dry, stable” could be post-78  coordinated; however, the clinical meaning of these two presentations are very different.  This then reinforces the need for nursing, as a profession, to continue to work towards a common patient care language, of which, standardized clinical terminologies can help identify commonalities and redundancies (Kim & Matney, 2014). As well, working towards a common nursing patient care clinical language would help standardized clinical terminology implementation initiatives and cross-organizational (and ultimately cross-border) semantic interoperability and data aggregation projects.   5.7 Implications of Missing Content  The overall published mapping results used to compare against this thesis, range from 23%-82%; however, if you were to examine each of these papers closely, you would quickly discover that the types of concepts, the types of mapping methods, and the terminology system year/version are different.  Yet consistently, challenges related to concept hierarchy placement, semantic clarity, and availably of pre-coordinated matches exist. So what does this mean?  Does it mean that these studies’ comparative results are good or bad?  The findings from this thesis study are that BC nursing wound care concept representation in SNOMED-CT is 40.2%.  This finding is within the rage of mapping results considered and compared in this thesis.  However, the pragmatic response is perhaps not so much about the answer related to “good” or “bad”, but what can be done about it.  The contribution of these findings relate to the identification of which wound care concepts now need to be added to SNOMED-CT.  As well, this study has demonstrated a measure of quality control through identification of two, one-to-many, duplicate concepts.  Alternatively, it may provide a further discussion point regarding increasing nursing domain coverage in another standardized clinical terminology (e.g., ICNP) versus movement to 79  increase domain coverage in SNOMED-CT.  Regardless, the next step of this research is to share these findings with Canada Health Infoway, who can liaise with SNOMED International, to consider missing and one-to-many concepts.  5.8 Relationship to Theoretical Model  Matney’s (2015) Theory of Wisdom in Action for Clinical Nursing can be used to think about the integration of structured data in EHRs to create information (i.e., data aggregation, decision support alerts, and integration with disparate health information systems).  Matney (2015) describes the accumulation of person-related factors and environmental-related factors contributing to the creation of knowledge, which when applied in a clinical situation, facilitates clinical judgement (and thus, wisdom in action).  It is within the environmental-related factors section, that her model describes the process of data, structured in context, in creating information.  If that information has logic or rules applied, it can be then used to support the creation of knowledge though contributions to lifespan contextualism, rich procedural knowledge, and rich factual knowledge.    So, by way of this model design, it can be posited that if the nurses’ wound measurements were structured in a graph, against previous ones, it could support the development of information.  If that information were to be given some logic or rules, such as an automated alert when the wound did not decrease in size by 30%, by the third week, that it could help in the creation of clinical knowledge.  This, with other accumulated antecedents, could support the nurse’s clinical judgment and care decisions regarding client’s wound care management needs. 80   Noting this, standardized clinical terminologies can help in this process through the realized benefits of semantic interoperability (i.e., wound care provided in different health settings and documented in disparate charts), data aggregation (i.e., combining three weeks of wound measurements), and decision support (i.e., using the codified data elements to a system “rule” to flag the chart).  As well, the findings in this thesis highlight the concern that if nursing knowledge is missing, or not represented in a standardized clinical terminology, that it may limit those possible benefits. However, it is important to consider that this model does not suggest that data must be structured by computers, or through the use of standardized terminologies, but instead, provides that lens which to envision and understand where these standardized terminologies can be leveraged and used to support clinical practice.  This model might also be thought of as a map, which to plan the integration of a standardized clinical terminology.  For example, before you design a system-wide “rule” to alert a nurse about a slow-healing wound, you might be better to start with the assurance that those wound measurements are structured in the EHR in such a way that they can be extracted and compared against previous ones.  By way of this model, the value of standardized clinical terminologies can be envisioned, supporting the ongoing development and advocacy of this science.       5.9 Limitations of Study   This study has several limitations.  For those related to the study methods, it is possible that because the researcher conducted the manual mapping, and was involved in the expert consensus mapping, that some of the results may have been influenced by her knowledge of both processes.  As well, this study may have been strengthened if more than two researchers were involved in the creation of the final expert consensus list or if two researchers independently 81  performed manual mapping and inter-rater reliability scores were considered.  Further, the decision to use only pre-coordinated matches in the International Edition SNOMED-CT would have impacted the rate of equivalence results.   Another limitation of this study relates that the sample data elements were only related to nursing wound assessment, diagnosis, and outcome parameters examined in BC, Canada.  Its applicability to other domains of practice, provinces, and countries may be limited.  Finally, there remains the reality that with most mapping activities, there is limited pragmatic use to replicate the exact mapping methods to validate findings.  This relates to the frequent SNOMED-CT updates, changes to clinical practice guidelines, and differing reporting priorities.  For example, it was decided that SNOMED-CT January 2016 would be used in this study, and already, the SNOMED-CT July 2016 edition may have placed these findings out-of-date.  This will likely remain a limitation to all clinical mapping research activities, and a consideration for those who wish to conduct research in this field.    5.10 Implications and Recommendations for Nursing Practice, Policy and Research  The science of standardized clinical terminologies has been a topic in nursing literature for over 40 years, and as this science matures, researchers are developing methods to increase concept representation, harmonize with ontologies outside our domain, and ensure that the essence of nursing is depicted in data representation as the technology paradigm shift occurs in health care.  Researchers such as Harris et al. (2015), forums such as the Canadian National Nursing Data Standards meeting, and organizations such as the CNA and CNIA, have brought forward a renewed interest to ensure that the praxis of standardized clinical terminologies 82  continues to meet the demands of our profession and most importantly, to support the patients we serve.    As this evolution unfolds, the methods of concept mapping must be scrutinized and understood to ensure clarity for those who wish to operationalize these technologies, and to ensure the desired benefits of these technologies are met with careful oversight and consideration for patient safety and quality control (Scichilone & Rihanek, 2015).  Similarly, the work to ensure the knowledge of nursing is represented in standardized terminologies must continue, or else threats to our professional representation in semantic interoperability, data aggregation, and decision support, might occur.  Through a desire to address the “terminology problem”, many researchers and organizations are working and publishing inter-terminology mapping activities to harmonization disparate nomenclatures.  Though it appears that less has been recently published regarding “bottom-up” mapping activities, researchers are continuing to map nursing concepts between existing EHRs and domains of knowledge to standardized clinical terminologies (Harris et al., 2015; Kim & Park, 2012).    As one outcome of this thesis project, a list of codified wound care concepts has now been compiled.  After completion of this thesis work, it is planned that this list will be shared with the BC Provincial Nursing Skin and Wound Committee who will confirm mapping results, as well, to review those data elements that did not have matches to ensure each is clear and non-redundant.  Another planned step, post-completion, will be to share this list with Canada Health Infoway, and advocate for the inclusion of missing data elements to SNOMED-CT (Infoway as the SNOMED International liaison).   Finally, as researchers continue to publish, it is now upon us working within health care originations to start planning for the next shift: standardized clinical terminology implementation 83  with nursing content representation.  It may still remain unclear if the methods of either inter-terminology mapping or “bottom-up” mapping will work well for local organizations wishing to move forward in adoption; however, the opportunity to start these conversations and to embrace collaboration is now.  This is especially significant as groups such as the CNA and Canada Health Infoway have taken measures to envision a sharable pan-Canadian health record with the inclusion of standards.  5.11 Summary    This thesis sought to map the BC standardized wound assessment, diagnosis, and outcome parameters to SNOMED-CT.  One of the main outcomes is a knowledge representation model that can be integrated into clinical information systems and extended to capture additional nursing wound care data elements.  As far as this author knows, it is the first study of its kind in Canada, in which a “bottom-up” approach was taken to map nursing wound care concepts to SNOMED-CT.  The outcome of this mapping activity was a list of 107 data elements that have been analyzed as having a SNOMED-CT match, one-to-many match, or a no match.  It also provided an opportunity to refine another approach to clinical mapping: conceptual model development, repetitive mapping (manual, automated, and, literature comparison), and descriptive analysis (consensus and statistical) to determine a rate of equivalence.    The final work presented in this research, now has the opportunity to be shared and used by clinical groups wishing to implement SNOMED-CT wound care concept codes in a SNOMED-CT enabled EHR.  As well, identification of those concepts missing or duplicate can be shared with Canada Health Infoway and SNOMED International, with the hope that future SNOMED-CT iterations may incorporate the findings.   Finally, as one small steps towards 84  interoperability, data aggregation, and decision support integration, this study has added to the body of scientific knowledge for mapping nursing knowledge in a standardized clinical terminology.             85  References American Nurses Association (n.d.). The nursing process. Retrieved from http://www.nursingworld.org/EspeciallyForYou/What-is-Nursing/Tools-You-Need/Thenursingprocess.html. Accessed 5 Nov. 2016. Apelon Inc. (2016). Termworks [Computer software]. Retrieved from http://www.apelon.com/ British Columbian Ministry of Health. (2014). Setting priorities for the B.C. health system. Retrieved from http://www.health.gov.bc.ca/library/publications/year/2014/Setting-priorities-BC-Health-Feb14.pdf. Accessed 5 Nov. 2016. British Columbia Nursing Skin and Wound Committee (2015). Guideline: Wound bed preparation for healable and non-healable wounds in adults & children. Retrieved from https://www.clwk.ca/buddydrive/file/guideline-wound-bed-preparation/. Accessed 9 Oct. 2015. Bryant, R. A., & Nix, D. (2007). Acute & chronic wounds: Current management concepts (3rd ed.). St. Louis: Mosby. Canadian Institute for Health Information (2013). Better information for improved health: A vision for health system use of data in Canada. Ottawa, ON: CIHI. Retrieved from https://www.cihi.ca/en/hsu_vision_report_en.pdf. Accessed 5 Nov. 2016.   Canada Health Infoway. (2014). Accelerating Clinical Interoperability in Canada The Path Forward. Retrieved from https://www.infoway-inforoute.ca/en/component/edocman/2329-accelerating-clinical-interoperability-in-canada-the-path-forward-2/view-document?Itemid=101. Accessed 6 Aug. 2016. Canada Health Infoway. (2015, May). Clinical interoperability action plan. Retrieved from https://ic.infoway-inforoute.ca/en/resources/docs/254-clinical-interoperability-action-plan-notes-may-2015/view-document. Accessed 10 Jan. 2016. Canada Health Infoway. (2015, March). SNOMED- CT Canadian National release Licence Agreement. Retrieved from https://infocentral.infoway-inforoute.ca/@api/deki/files/9245/=SCT_Ca_National_License_Agreement_20151006.pdf. Accessed 5 Nov. 2016. Canada Health Infoway. (2017, January). Apelon TermWorks.   Retrieved from https://ic.infoway-inforoute.ca/en/tools/standards-tools/termworks. Accessed 5 Jan. 2017. Canada Health Infoway. (n.d.). Standards inventory. Retrieved from https://ic.infoway-inforoute.ca/en/standards/inventory. Accessed 5 Jan. 2017. Canadian Health Outcomes for Better Information and Care. (2015). C-HOBIC Phase 2 Final Report. Retrieved from http://c-hobic.cna-aiic.ca/documents/pdf/Canadian-Health-Outcomes-for-Better-Information-and-Care_C-HOBIC-Phase-2_Final-Report_January-2015.pdf.  Accessed 9 Oct. 2015. Canadian Health Outcomes for Better Information and Care. (n.d.). Inclusion of nursing-related patient outcomes in electronic health records.   Retrieved from http://c-hobic.cna-aiic.ca/about/default_e.aspx. Accessed 24 Mar. 2015. Canadian Nurses Association. (2006). Position statement nursing information and knowledge management. Retrieved from http://www.cna-aiic.ca/~/media/cna/page-content/pdf-en/nursing-information-and-knowledge-management_position-statement.pdf?la=en. Accessed 24 Mar. 2015. 86  Canadian Nurses Association, & Canada Health Infoway. (2008). Mapping Canadian clinical outcomes in ICNP. Retrieved from http://c-hobic.cna-aiic.ca/documents/pdf/ICNP_Mapping_2008_e.pdf. Accessed 24 Mar. 2015. Chow, M., Beene, M., O'Brien, A., Greim, P., Cromwell, T., DuLong, D., & Bedecarré, D. (2015). A nursing information model process for interoperability. Journal of the American Medical Informatics Association, 22(3), 608-614 607p. doi:10.1093/jamia/ocu026 Cimino, J. J. (1998). Desiderata for controlled medical vocabularies in the twenty-first century. Methods of Information in Medicine, 37(4-5), 394-403.  Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3415631/. Accessed 24 Mar. 2015. Coiera, E. (2015). Guide to health informatics (3rd ed.). Boca Raton: Taylor & Francis Group. College of Registered Nurses of British Columbia. (2016). Scope of practice for registered nurses. Retrieved from https://www.crnbc.ca/Standards/Lists/StandardResources/433ScopeforRegisteredNurses.pdf. Accessed 5 Nov. 2016. College of Registered Nurses of British Columbia. (2013). Nursing Documentation. Retrieved from https://crnbc.ca/Standards/Lists/StandardResources/151NursingDocumentation.pdf. Accessed 24 Mar. 2015. Cusack, C. M., Hripcsak, G., Bloomrosen, M., Rosenbloom, S. T., Weaver, C. A., Wright, A., . . . Mamykina, L. (2013). The future state of clinical data capture and documentation: A report from AMIA's 2011 policy meeting. Journal of the American Medical Informatics Association, 20(1), 134-140. doi:doi:10.1136/amiajnl-2012-001093 Doughty, D. B. (2006). Urinary & fecal incontinence: Current management concepts (3rd ed.). St. Louis, Mo: Mosby Elesevier.  Doughty, D., Junkin, J., Kurz, P., Selekof, J., Gray, M., Fader, M., Bliss, D., Beeckman, D., Logan, S. (2012). Incontinence-associated dermatitis: Consensus statements, evidence-based guidelines for prevention and treatment, and current challenges. Journal of Wound, Ostomy, Continuce Nursing, 39(3), 303-315. ECRI Institute (2015). Top 10 patient safety concerns for healthcare organizations. Retrieved from https://www.ecri.org/EmailResources/PSRQ/Top10/2015_Patient_Safety_Top10.pdf. Accessed 5 Nov. 2016. Fawcett, J. & Gigliotti, E. (2001). Using conceptual models of nursing to guide nursing research: The case of the neuman systems model. Nursing Science Quarterly, 14(4), 339-345. Gordon, B. L. (1965). Standard medical terminology. Journal of the American Medical Association, 191(4), 311-313.  doi:10.1001/jama.191.4.311 Government of Canada. (2015). TCPS 2 (2014) - the latest of Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Retrieved from http://www.pre.ethics.gc.ca/eng/policy-politique/initiatives/tcps2-eptc2/Default/. Accessed 9 Oct. 2015. Hall, J., Buckley, H. L., Lamb, K. A., Stubbs, N., Saramago, P., Dumville, J. C., & Cullum, N. A. (2014). Point prevalence of complex wounds in a defined United Kingdom population. Wound Repair and Regeneration, 22(6), 694-700. doi:10.1111/wrr.12230 Hammond, W. E., Jaffe, C., Cimino, J. J., & Huff, S. M. (2014). Standards in biomedical informatics In E. H. Shortliffe & J. J. Cimino (Eds.), Biomedical Informatics Computer 87  Applications in Health Care and Biomedicine (4th ed., pp. 211-253). London: Springer-Verlag. Handfield, S. (2013). Formation of a provincial nursing skin and wound committee. Journal of Wound, Ostomy, and Continence Nursing : Official Publication of the Wound, Ostomy and Continence Nurses Society / WOCN, 40(6), 568. Hannah, K. J., White, P. A., Nagle, L. M., & Pringle, D. M. (2009). Standardizing Nursing Information in Canada for Inclusion in Electronic Health Records: C-HOBIC. Journal of the American Medical Informatics Association, 16(4), 524-530 527p. doi:10.1197/jamia.M2974 Hardiker, N. R., Bakken, S., Casey, A., & Hoy, D. (2002). Formal nursing terminology systems: a means to an end. Journal of Biomedical Informatics, 35(5-6), 298-305.  doi:10.1016/S1532-0464(03)00002-9 Hardiker, N. R., Bakken, S., & Kim, T. Y. (2011). Advanced terminological approaches in nursing. In V. K. Saba & K. A. McCormick (Eds.), Essentials of nursing informatics (5th ed.). New York: McGraw-Hill Medical. Harris, M. R., Langford, L. H., Miller, H., Hook, M., Dykes, P. C., & Matney, S. A. (2015). Harmonizing and extending standards from a domain-specific and bottom-up approach: an example from development through use in clinical applications. Journal of the American Medical Informatics Association, 22(3), 545-552 548p. doi:10.1093/jamia/ocu020 Häyrinen, K., Lammintakanen, J., & Saranto, K. (2010). Evaluation of electronic nursing documentation--nursing process model and standardized terminologies as keys to visible and transparent nursing. International Journal of Medical Informatics, 79(8), 554-564 doi:10.1016/j.ijmedinf.2010.05.002 Healthcare Information and Management Systems Society. (2005). Interoperability Definition and Background. Retrieved from http://s3.amazonaws.com/rdcms-himss/files/production/public/HIMSSorg/Content/files/AUXILIOHIMSSInteroperabilityDefined.pdf. Accessed 10 Jan. 2016. Healthcare Information and Management Systems Society. (2013). Definition of interoperability.   Retrieved from http://s3.amazonaws.com/rdcms-himss/files/production/public/FileDownloads/HIMSS%20Interoperability%20Definition%20FINAL.pdf. Accessed 10 Jan. 2016. Hurd, T., & Posnett, J. (2009). Point prevalence of wounds in a sample of acute hospitals in canada. International Wound Journal, 6(4), 287-293. doi:10.1111/j.1742-481X.2009.00615.x IBM (2003). An introduction to the unified modeling language. Retreived from http://www.ibm.com/developerworks/rational/library/769.html. Accessed 5 Nov. 2016. International Council of Nurses. (2014). ICN and IHTSDO extent collaboration to advance harmonisation of health care terminology. Retrieved from Geneva, Switzerland: http://www.snomed.org/news-articles/icn-and-ihtsdo-extend-collaboration-to-advance-harmonisation-of-health-terminology. Accessed 5 Jan. 2017. International Council of Nurses. (2015). ICNP Definition.   Retrieved from http://www.icn.ch/what-we-do/definition-a-elements-of-icnpr/. Accessed 24 Feb. 2015. International Council of Nurses. (n.d.). International classification for nursing practice (ICNP) information sheet.  Retrieved from 88  http://www.icn.ch/images/stories/documents/pillars/Practice/icnp/ICNP_FAQs.pdf. Accessed 9 Oct. 2015. International Council of Nurses, & International Health Terminology Standards Development Organization. (2014). ICN and IHTSDO extend collaboration to advance harmonisation of health terminology [Press release]. Retrieved from http://www.icn.ch/images/stories/documents/news/press_releases/2014_PR_17_ICN-IHTSDO.pdf. Accessed 24 Feb. 2015. International Health Terminology Standards Development Organization, & Regenstrief Institute. (2013). New Regenstrief and IHTSDO agreement to make EMRs more effective at improving health care [Press release]. Retrieved from http://www.ihtsdo.org/about-ihtsdo/partnerships/loinc. Accessed 1 Feb. 2016. International Standards Organization. (2014). ISO 18104:2014 Health informatics – Categorical structures for representation of nursing diagnoses and nursing actions in terminological systems.   Retrieved from http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=59431. Accessed 1 Feb. 2016. Kinnunen, U., Saranto, K., Ensio, A., Iivanainen, A., & Dykes, P. (2012). Developing the standardized wound care documentation model: A delphi study to improve the quality of patient care documentation. Journal of Wound, Ostomy, and Continence Nursing : Official Publication of the Wound, Ostomy and Continence Nurses Society / WOCN, 39(4), 397. Kim, H. Y., & Park, H. A. (2012). Development and evaluation of data entry templates based on the entity-attribute-value model for clinical decision support of pressure ulcer wound management. International Journal of Medical Informatics, 81(7), 485-492. doi:10.1016/j.ijmedinf.2011.10.008  Kim, S. Y., Kim, H. H., Shin, K. H., Kim, H. S., Lee, J. I., Choi, B. K. (2012). Comparison of knolwedge levels required for SNOMED CT coding of diagnosis and operation names in clinilca records. Healthcare Informatics Research, 18(3), pp. 186-190. doi: 10.4258/hir.2012.18.3.186 Kim, T. Y., Hardiker, N. R., & Coenen, A. (2014). Inter-terminology mapping of nursing problems. Journal of Biomedical Informatics, 49, 213-220 218p. doi:10.1016/j.jbi.2014.03.001 Kim, T. Y., & Matney, S. A. (2014). Standards. In R. Nelson & N. Staggers (Eds.), Health informatics: An interprofessional approach (pp. 351-369). St. Louis: Elsevier Mosby. Lau, F., Simkus, R., & Lee, D. (2008). A methology for encoding problem lists within SNOMED CT in general practice. Proceedings of the 3rd Interational Conference on Knolwedge Representation in Medicine (KR-MED 2008).  Lee, D., Cornet, R., & Lau, F. (2011). Implications of SNOMED- CT versioning. International Journal of Medical Informatics, 80(6), 442-453 412p. doi:10.1016/j.ijmedinf.2011.02.006 Lee, D., Cornet, R., Lau, F., & de Keizer, N. (2013). A survey of SNOMED- CT implementations. Journal of Biomedical Informatics, 46(1), 87-96 10p. doi:10.1016/j.jbi.2012.09.006 Lenz, R., Beyer, M., & Kuhn, K. A. (2007). Semantic integration in healthcare networks. Studies in Health Technology & Informatics, 116, 385-390. doi:10.1016/j.ijmedinf.2006.05.008 89  Matney, S. A. (2015). Development of the theory of wisdon in action for clinical nursing. (Doctor of Philosophy), The University of Utah. Retrieved from https://www.researchgate.net/publication/282353776_MatneyWisdomInActionDissertationFinal08172015. Accessed 6 Aug. 2016.  Medvidovic, N., Rosenblum, D., Redmiles, D., Robbins, J. (2002). Modeling software architechtures in the unified modeling language. ACM Transactions on Software Engineering and Methology, 11(1), 2-57. Moen, A., Henry, S. B., & Warren, J. J. (1999). Representing nursing judgements in the electronic health record. J Adv Nurs, 30(4), 990-997.  Monsen, K., Finn, R., Fleming, T., Garner, A., LaValla, A., Riemer, J. (2016). Rigor in electronic health record knowledge representation: Lessons leaned from SNOMED CT clinical content encoding exercise. Informatics for Health and Social Care, 41, 97-111.  Nagle, L. & White, P. (2016). National nursing data standards symposium proceedings. Retrieved from  https://cnia.ca/national-nursing-data-standards-symposium-proceedings/. Accessed 5 Nov. 2016. National Committee on Vital and Health Statistics (NCVHS) Report on Uniform Data Standards for Patient Medical Record Information, July 6, 2000, 21-22. Retrieved from http://www.ncvhs.hhs.gov/wp-content/uploads/2014/08/hipaa000706.pdf. Accessed 28 Nov. 2016. Nelson, R., & Staggers, N. (2014). Theoretical foundations of health informatics. In R. Nelson & N. Staggers (Eds.), Health Informatics An Interprofessional Approach (pp. 18-39). St. Louis: Elsevier Mosby. Office of the Auditor General of Canada. (2009). 2009 fall report of the Auditor General of Canada. Retrieved from http://www.oag-bvg.gc.ca/internet/English/parl_oag_200911_04_e_33205.html. Accessed 24 Feb. 2015 Office of the Auditor General of Canada. (2010). Electronic health records in Canada. An overview of federal and provincial audit reports April 2010. Retrieved from http://www.oag-bvg.gc.ca/internet/docs/parl_oag_201004_07_e.pdf. Accessed 24 Feb. 2015. Osheroff, J. A., Teich, J. M., Middleton, B., Steen, E. B., Wright, A., & Detmer, D. E. (2007). A roadmap for national action on clinical decision support. Journal of the American Medical Informatics Association : JAMIA, 14(2), 141-145. doi:10.1197/jamia.M2334 Polit, D. F., & Beck, C. T. (2012). Nursing research: Generating and assessing evidence for nursing practice (9th ed.). Philadelphia: Wolters Kluwer Health/Lippincott Williams & Wilkins. Qamar, R., Kola, J., & Rector, A. (2007). Unambiguous data modeing to ensure higher accuracy term binding to clinical terminologies. AMIA 2007 Symposium Proceedings (pp. 608-613) Regenstrief Institute. (n.d.). About LOINC.   Retrieved from https://loinc.org/background. Accessed 5 Nov. 2016. Ronquillo, C., Currie, L. M., & Rodney, P. (2016). The evolution of data-information-knowledge-wisdom in nursing informatics. Advances in Nursing Science, 39(1), E1-E18. doi:10.1097/ANS.0000000000000107 Rosenbloom, S. T., Miller, R. A., Johnson, K. B., Elkin, P. L., & Brown, S. H. (2006). Interface terminologies: facilitating direct entry of clinical data into electronic health record 90  systems. Journal of the American Medical Informatics Association, 13(3), 277-288. doi:10.1197/jamia.M1957 Scichilone, R. & Rihanek, T.(2015). Data mapping of code sets, clinical terminologies and classifications systems. In K. Giannangelo (Ed.), Health code sets, clinical terminologies, and classification systems (3rd ed.). Chicago, Il: AHIMA.   Sensmeier, J. (2011). Health data standards: Development, harmonization, and interoperability. In V. K. In Saba & K. A. McCormick (Eds.), Essentials of nursing informatics (5th ed., pp. 191-201). New York: McGraw-Hill Medical. Sensmeier, J., & Murphy, W. J. (2014). Health policy and informatics. In R. Nelson & N. Staggers (Eds.), Health informatics an interprofessional approach (pp. 397-409). St. Louis: Elsevier Mosby. SNOMED International. (2016). SNOMED- CT editorial guide. Retrieved from https://confluence.ihtsdotools.org/display/DOCEG/Previous+Versions. Accessed 5 Jan. 2017.  SNOMED International. (2014). SNOMED- CT starter guide. Retrieved from http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwir-YH566zRAhUP5GMKHYaPArsQFggdMAA&url=http%3A%2F%2Fwww.ihtsdo.org%2Fresource%2Fresource%2F41&usg=AFQjCNGvbiaeA_MHmAI3vj5StNaV2tyQRw&sig2=XvKjk3FNVhnTxDp9YUYyjg. Accessed 5 Jan. 2017.  SNOMED International. (n.d.a). History of SNOMED- CT.   Retrieved from http://www.snomed.org/snomed-ct/what-is-snomed-ct/history-of-snomed-ct. Accessed 5 Jan. 2017. SNOMED International. (n.d.b). Special interest group.   Retrieved from http://www.ihtsdo.org/participate/special-interest-groups. Accessed 5 Jan. 2017. Stallinga, H. A., Napel, H., Jansen, G. J., Geertzen, J. H., Vries Robbé, P. F., & Roodbol, P. F. (2015). Does language ambiguity in clinical practice justify the introduction of standard terminology? An integrative review. Journal of Clinical Nursing, 24(3-4), 344-352. doi:10.1111/jocn.12624 Technology Informatics Guiding Education Reform. (2009). Collaborating to integrate evidence and informatics into nursing practice and education: An executive summary. Retrieved from http://s3.amazonaws.com/rdcms-himss/files/production/public/FileDownloads/tiger-report-executive-summary.pdf. Accessed 24 Feb. 2015. Thoroddsen, A., Ehrenberg, A., Sermeus, W., & Saranto, K. (2012). A survey of nursing documentation, terminologies and standards in European countries. NI 2012: Proceedings of the 11th International Congress on Nursing Informatics, 2012, 406. Topaz, M., Lai, K., Dowding, D., Lei, V. J., Zisberg, A., Bowles, K. H., & Zhou, L. (2016). Automated identification of wound information in clinical notes of patients with heart diseases: Developing and validating a natural language processing application. International Journal of Nursing Studies, 64, 25-31. doi:10.1016/j.ijnurstu.2016.09.013 Wager, K. A., Lee, F. W., & Glaser, J. P. (2013). Health care information systems: A practical approach for health care management. San Francisco: Jossey-Bass. Werley, H. H., Devine, E. C., Zorn, C. R., Ryan, P., & Westra, B. L. (1991). The nursing minimum data set: abstraction tool for standardized, comparable, essential data. American Journal of Public Health, 81(4), 421-426 426p. doi:10.2105/AJPH.81.4.421 91  Westra, B. L., Delaney, C. W., Konicek, D., & Keenan, G. (2008). Nursing standards to support the electronic health record. Nursing Outlook, 56(5), 258-266.e251. doi:10.1016/j.outlook.2008.06.005 Westra, B. L., Latimer, G. E., Matney, S. A., Park, J. I., Sensmeier, J., Simpson, R. L., . . . Delaney, C. W. (2015). A national action plan for sharable and comparable nursing data to support practice and translational research for transforming health care. Journal of the American Medical Informatics Association, 22(3), 600-607 608p. doi:10.1093/jamia/ocu011 Woodbury, M. G., & Houghton, P. E. (2005). The extent of chronic wounds in Canada: What we know and what we don’t know. Wound Care Canada, 3(1).  Retrieved from http://cawc.net/images/uploads/wcc/3-1/woodbury-houghton.pdf. Accessed 24 Feb. 2015. Wound Ostomy and Continence Nurses Society. (2009). Position statement avoidable versus unavoidable pressure ulcers. Retrieved from http://c.ymcdn.com/sites/www.wocn.org/resource/resmgr/Publications/PS_Avoidable_vs_Unavoidable_.pdf. Accessed 6 Aug. 2016. Zega, M., D'Agostino, F., Bowles, K. H., De Marinis, M. G., Rocco, G., Vellone, E., & Alvaro, R. (2014). Development and Validation of a Computerized Assessment Form to Support Nursing Diagnosis. International Journal of Nursing Knowledge, 25(1), 22-29 doi:10.1111/2047-3095.12008 92  Appendices  Appendix A- Wound Care Parameters This section presents the concepts and definitions, which were used as sample data elements.  These concepts and definitions are represented in the BC Nursing Wound Documentation standard, and provided to the researcher by the BC Provincial Nursing Skin and Wound Committee Chair, Shannon Handfield.    1. Parameters Related to Wound Assessment  Wound Assessment Parameters Clinical Definition >Wound Bed    Biochemical wound product Residual/remaining biochemical wound care product in wound bed Blister Elevation or separation of the epidermis containing fluid Bone Hard, rigid white connective tissue Epithelial islands Within an open wound bed, islands (small areas) of epithelial tissue proliferating and migrating from the center to the edge of the wound Eschar - dry, stable Firm, dry necrotic tissue with an absence of drainage, edema, erythema or fluctuance. It is black or brown in color and is attached to the wound edges and wound base Eschar - soft, boggy Soft necrotic tissue which is black, brown, grey, or tan in color. It may be firmly or loosely attached to the wound edges and wound base; fluctuance and drainage may be present. Foreign body Objects such as mesh, hardware, suture(s) Friable Fragile tissue that may bleed easily when manipulated (i.e., palpated, probed, irrigated) Fully callused Wound bed that is 100% covered with a callus. This choice often used on areas that are being monitored. Do not use this choice for any open wound bed with a callused edge. The area must be completely covered with a callus. Fully epithelialized Covered completely with new epithelial tissue Fungating tissue Cancerous or non cancerous tissue, rapidly growing tissue; appears cauliflower-like Granulation tissue Firm, red, moist, pebbled healthy tissue. Hematoma Localized collection of blood Hypergranulation tissue Red, moist tissue raised above the level of the skin (proud flesh) Malignant tissue Cancerous tissue Non-granulation tissue Moist, red (pale to bright) non-pebbled (smooth) tissue that may be unhealthy Not visible A portion or all of the open wound bed that cannot be visualized Scab Superficial, dry crust Slough Dry or wet, loose or firmly attached, yellow to brown dead tissue Superficial pink, red Clean, open pink/red area with non-measurable depth   93  1. (con’t) Parameters Related to Wound Assessment   Wound Assessment Parameters Clinical Definition Tendon Shiny white cord of fibrous connective tissue that connects muscle to bone. Viable graft Grafted area of skin that has “taken” and is living  >Wound Measurement   Length From edge to edge, the longest measurement of the wound Width From edge to edge, the widest measurement of the wound at right angles to the length Depth The deepest vertical measurement from the base of the wound to the level of the skin Sinus Tract A channel that extends from any part of the wound and tracks into deeper tissue. Undermining A destruction of tissue that occurs underneath the intact skin of the wound perimeter parallel to the skin surface Fistula An abnormal track connecting an organ to the skin surface, wound bed or to another organ >Exudate Amount   None   Scant Wound drainage amount considered in relationship to the size of the wound. Small Wound drainage amount considered in relationship to the size of the wound. Moderate Wound drainage amount considered in relationship to the size of the wound. Large/copious Wound drainage amount considered in relationship to the size of the wound. >Exudate Type   Nil  no exudate Purulent Thick, cloudy Sanguineous Bloody Serous Thin, clear, yellow >Wound Odour   Yes Unpleasant smell noted from wound after cleansing No No noted odour from wound after cleansing >Wound Edge   Attached Edge appears flush with wound bed or as a “sloping” edge Callused Hyperkeratosis, thickened layer of epidermis Demarcated Well defined, distinct, easy to clearly define wound outline Diffuse Not well defined, indistinct, difficult to clearly define wound outline Epithelializing New, pink to purple, shiny migrating tissue Hypergranulation Tissue Wound edge that is red, moist tissue raised above the level of the skin (proud flesh) Not Attached Edge appears as a “cliff” Rolled  Epithelial wound edge of a cavity wound which rolls under Scarred Fibrotic regenerated tissue following wound healing >Periwound Skin   Blister Elevation or separation of the periwound epidermis containing fluid Boggy Soft, spongy tissue around the wound 94  1. (con’t) Parameters Related to Wound Assessment   Wound Assessment Parameters Clinical Definition Bruised Dark red, purplish, blue, tissue that fades to yellow, green, grey, depending on the skin colour Callused Hyperkeratosis, thickened layer of epidermis Dry Flaky or scaly skin Edema Interstitial collect of fluid to the periwound skin Erythema <2cm Redness of the skin; may be intense bright red to dark red, that is limited within 2cm from wound edge to the periwound skin Erythema >2cm Redness of the skin; may be intense bright red to dark red, that extends beyond 2cm from the wound edge to the periwound skin Excoriated/denuded Superficial loss of tissue Fragile Skin that is at risk for breakdown Increased warmth Increased warmth when compared to skin in adjacent area Indurated <2 cm Abnormal firmness of the tissues with palpable margins, that is limited within 2cm from the wound edge to the periwound skin Indurated >2cm Abnormal firmness of the tissues with palpable margins, that extends beyond 2cm from the wound edge to the periwound skin Intact  Unbroken skin around the wound Macerated Wet, white looking skin – caused by excessive moisture Rash Temporary eruption on the skin-often raised, red, sometimes itchy Tape tear Superficial skin loss due to tape removal Weepy Moist, draining areas >Wound Pain Quantified on the Visual Analogue Scale where 0 = no pain and 10 = excruciating as described by the patient/client/resident >Date of Onset Best known date of when the wound occurred >Recurrence When a wound had previously occurred in the same area, usually with the same underlying etiology           95  2. Parameters Related to Wound Diagnosis Wound Diagnosis Parameter Clinical Definition Abscess A local collection of purulent and/or sanguineous drainage that may be incised and drained Arterial Caused by a disruption to arterial blood flow leading to moderate / severe tissue ischemia Burn Tissue loss/damage as a result from thermal, chemical, electrical, and radiation sources. Diabetic/Neuropathic  A result of neuropathy, structural foot changes, and/or altered blood flow related to diabetes mellitus (or alcoholism, renal failure, HIV, late stage syphilis, trauma and surgery). Donor Site An area of the patient’s skin (with or without deeper tissue structures) that is surgically removed and placed elsewhere on the body. Drug Reaction A wound or skin reaction where the cause is related to the intake/absorption of a drug/medication. Fistula- Non Stomatized A fistula where the nurse cannot visualize a specific area that has connected with the skin. There may be several diffuse areas open on the skin (or in tunnel/undermining) Fistula-Stomatized A fistula with a budded opening between the skin and the associated structure/space/organ (eg: small bowel). Graft Site An area of the patient’s body where skin (with or without deeper tissue structures) is surgically grafted from another area of the body. Incontinence Associated Dermatitis Skin damage associated to urine and/or fecal incontinence.  Infectious Skin rash or ulcer from an infectious process and can be categorize according to the organism: fungus, bacteria, virus, or arthropod.  Inflammatory Disease Present as atypical ulcers and can be misidentified as venous or arterial ulcers. (i.e., Pyoderma Granulosum, Vasculitis, Drug-Induced Vasculitis).  Irradiation Complications from radiation therapy that can present as erythema, dry desquamination, moist desquamination, and/or permanent skin changes. Lymphedema Lymphatic obstruction / blockage of the lymph vessels causing swelling. It is caused by damage to the lymph system, congenital defect to the lymph system, or parasitic infection (filariasis).  Malignant Cancerous lesion/wound that may present as painful, fungating and/or friable wounds. Pilonidal Sinus An abnormal tract/sinus extending from the skin, usually near the cleft of the buttock. Pressure Ulcer- Stage 1 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 1– Intact skin with localized nonblanchable erythema; darkly pigmented skin may not show visible blanching but will appear different than the colour of surrounding skin. Pressure Ulcer- Stage 2 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 2– Partial thickness wound presenting as a shallow open ulcer with a red / pink wound bed, slough may be present but does not obscure the depth of tissue loss; may also present as an intact or open/ruptured serum-filled or serosanguineous filled blister. Pressure Ulcer- Stage 3 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 3– Full thickness wound; subcutaneous tissue may be visible but bone, tendon and muscle are not exposed; may include undermining and/or sinus tracks; slough or eschar may be present but does not obscure the depth of tissue loss. 96  2. (con’t) Parameters Related to Wound Diagnosis    Wound Diagnosis Parameter Clinical Definition Pressure Ulcer- Stage 4 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 4- Full thickness wound with exposed bone, tendon or muscle; often includes undermining and/or sinus tracks; slough or eschar may be present on some parts of the wound bed but does not obscure the depth of tissue loss. Pressure Ulcer- SDTI Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Suspected Deep Tissue Injury (SDTI) – A localized purple or maroon area of intact skin or a blood filled blister that occur when underlying soft tissue is damaged from friction or shear. Pressure Ulcer- X Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Unstageable, Stage X – A wound in which the wound bed is covered by sufficient slough and / or eschar to preclude staging. Skin Disease Wounds develop as a manifestation of a disease process or complication associated with the treatment of a disease.(i.e., Epidermolysis Bullosa, Calciphylaxis, Graft-versus-Host Disease). Skin Tear Skin tears are the result of trauma caused by shearing, friction, or blunt force to the skin. Consider risk factors (i.e., advanced age, immobility, inadequate hydration/nutrition, falls)  Surgery (Secondary Intent) Incision has dehisced and surgical closure is not possible; wounds must heal by granulation  Trauma Wounds caused by traumatic injury and expected to heal with measureable, objective signs of improvement after three weeks of care (eg: Puncture, Abrasion, Stab, Bite, Gunshot, Laceration). Venous  Caused by venous insufficiency due to valve dysfunction, complete or partial blockage of the deep veins, and / or failure of the calf muscle pump Venous/Arterial  Caused by both venous insufficiency and disrupted arterial blood flow.  3. Parameters Related to Goal of Care  Goal of Care Parameters Clinical Definition To Heal the Wound Wound healing is anticipated to occur according to a predictable trajectory.  To Maintain the Wound Maintenance wounds have the potential to heal but are impacted by client, wound and/or system factors that cannot be mitigated resulting in wound healing that is slow or stalled. The factors that are barriers to healing may change over time and periodic re-evaluation is indicated with maintenance wounds. To Monitor/Manage the Non-healable Wound Non-healable wounds are sometimes called palliative wounds because their inability to heal may be due to an untreatable cause such as cancer. These wounds may be stable or may deteriorate over time. 97  Appendix B – Wound Assessment: Terms and Matches by Different Mapping Methods  This section describes the specific instances of concept, concept matches and mapping method.  Wound Assessment Matches by Different Mapping Methods  Parameter   Manual Mapping  Apelon Autofill 90-100% match Harris et al Consensus Mapping Wound Assessment  Clinical Definition SNOMED 2016 Jan SNOMED 2016 Jan SNOMED 2011 Jan Consensus Results >Wound Bed     Fibrinous wound bed (finding)   no match Biochemical wound product Residual/remaining biochemical wound care product in wound bed   Biochemical (qualifier value)   no match Blister Elevation or separation of the epidermis containing fluid Blister (morphological abnormality) Blister (morphologic abnormality) blistering eruption (disorder) no match Blistering eruption (disorder) Blister of skin AND/OR mucosa (disorder) Blister - unit of product usage (qualifier value) Bone Hard, rigid white connective tissue Skin loss exposing bone (finding) Entire bone (organ) (body structure) bone (tissue) structure (body structure) no match Entire bony skeleton (body structure) Epithelial islands Within an open wound bed, islands (small areas) of epithelial tissue proliferating and migrating from the center to the edge of the wound   Island (environment)   no match Eschar – dry, stable Firm, dry necrotic tissue with an absence of drainage, edema, erythema. It is black or brown in color and is attached to the wound edges and wound base   Crust (morphologic abnormality) skin eschar (disorder) no match Eschar (morphologic abnormality) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  98  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter   Manual  Apelon Autofill 90-100% match Harris et al Consensus  Wound Assessment  Clinical Definition SNOMED 2016 Jan SNOMED 2016 Jan SNOMED 2011 Jan Consensus Results Eschar – soft, boggy Soft necrotic tissue which is black, brown, grey, or tan in color. It may be firmly or loosely attached to the wound edges and wound base; fluctuance and drainage may be present.   Crust (morphologic abnormality) skin eschar (disorder) no match Eschar (morphologic abnormality) Foreign body Objects such as mesh, hardware, suture(s) Foreign body in wound (morphological abnormality) Foreign body (disorder)   no match Foreign body (morphologic abnormality) Friable Fragile tissue that may bleed easily when manipulated (i.e.: palpated, probed, irrigated)   Friability (finding) fragile skin (finding) no match Fully callused Wound bed that is 100% covered with a callus. This choice often used on areas that are being monitored. Do not use this choice for any open wound bed with a callused edge. The area must be completely covered with a callus. Hyperkeratosis (morphological abnormality) Callosity (disorder) callosity (disorder) no match Fully epithelialized Covered completely with new epithelial tissue Wound epithelialization (finding) Epithelialization (morphologic abnormality) wound epithelialization (finding) direct match Epithelial metaplasia (morphologic abnormality) Fungating tissue Cancerous or non cancerous tissue, rapidly growing tissue; appears cauliflower-like Fungating tumor (finding) Fungating tumor (finding)   no match    Granulation of tissue (finding)   Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  99  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter   Manual  Apelon Autofill 90-100% match Harris et al Consensus  Wound Assessment  Clinical Definition SNOMED 2016 Jan SNOMED 2016 Jan SNOMED 2011 Jan Consensus Results Granulation tissue Firm, red, moist, pebbled healthy tissue. Granulation of tissue (finding) Granulation tissue (morphologic abnormality) granulation of tissue (finding) direct match Hematoma Localized collection of blood Wound hematoma (finding) Hematoma (disorder) hematoma of skin (disorder) direct match Hematoma (morphologic abnormality) Hematoma (morphologic abnormality) Hypergranulation tissue Red, moist tissue raised above the level of the skin (proud flesh) Hypertrophic granulation tissue (morphologic abnormality) Bone (tissue) structure (body structure)   no match Blood (substance) Myelin (substance) Fascial (qualifier value) Mucosal (qualifier value) Ependyma (body structure) Syngraft (substance) All bone (tissue) (body structure) Subserosa (body structure) Brown fat (body structure) Tooth bud (body structure) Endosteum (body structure) Free flap (substance) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus    100  Wound Assessment Matches by Different Mapping Methods (con’t)   Parameter   Manual  Apelon Autofill 90-100% match Harris et al Consensus  Malignant tissue Cancerous tissue Malignant neoplasm, primary (morphologic abnormality) Clear cell sarcoma (morphologic abnormality)   no match Malignant neoplasm of soft tissue (disorder) Malignant tumor of fibrous tissue (disorder) Malignant melanoma of soft tissues (disorder) Malignant tumor of mesothelial tissue (disorder) Malignant infiltration of soft tissue (disorder) Malignant neoplasm of retrocecal tissue (disorder) Malignant neoplasm of connective tissue (disorder) Malignant neoplasm of perinephric tissue (disorder) Primary malignant neoplasm of perirenal tissue (disorder) Malignant tumor of mesothelial and soft tissue (navigational concept) Primary malignant neoplasm of periadrenal tissue (disorder) Sarcoma, no International Classification of Diseases for Oncology subtype (morphologic abnormality) Non-granulation tissue Moist, red (pale to bright) non-pebbled (smooth) tissue that may be unhealthy   Granulation tissue (morphologic abnormality)   no match Granulation of tissue (finding) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 101  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Not visible A portion or all of the open wound bed that cannot be visualized   No view of vocal cords (finding)   no match Epiglottis not visible (finding) Pressure ulcer not visible (disorder) Tympanic membrane not visible (finding) Scab Superficial, dry crust Crust (morphologic abnormality) Crust (morphologic abnormality)   no match Infestation caused by Psoroptes (disorder) Slough Dry or wet, loose or firmly attached, yellow to brown dead tissue Necrotic debris (morphologic abnormality) Necrotic debris (morphologic abnormality) wound slough (finding) direct match Wound slough (finding) Superficial pink, red Clean, open pink/red area with non-measurable depth   Pink color (qualifier value)   no match Pink color (finding) Tendon Shiny white cord of fibrous connective tissue that connects muscle to bone. Tendon finding (finding) Entire tendon (body structure) tendon structure (body structure) no match Tendon structure (body structure) Viable graft Grafted area of skin that has “taken” and is living    Viable (qualifier value)   no match >Wound Measurement     Wound measure (physical object)   no match Length From edge to edge, the longest measurement of the wound Length of wound (observable entity) Long (qualifier value) Length of wound (observable entity) direct match Length (attribute) Length property (qualifier value) Width From edge to edge, the widest measurement of the wound at right angles to the length Width of wound (observable entity) Width (qualifier value) Width of wound (observable entity) direct match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 102  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Depth The deepest vertical measurement from the base of the wound to the level of the skin Depth of wound (observable entity) Deep (qualifier value) Depth of wound (observable entity) direct match Depth (qualifier value) Depth (attribute) Sinus Tract A channel that extends from any part of the wound and tracks into deeper tissue. Wound sinus (finding) Fistula (morphologic abnormality) wound Sinus (finding) direct match Acquired fistula (morphologic abnormality) Inflammatory fistula (morphologic abnormality) Enlargement of sinus tract of skin (procedure) Periapical abscess with sinus tract (disorder) Diagnostic injection of sinus tract (procedure) Therapeutic injection of sinus tract (procedure) Incision and exploration of sinus tract (procedure) Undermining A destruction of tissue that occurs underneath the intact skin of the wound perimeter parallel to the skin surface Wound tissue undermining (finding) Wound tissue undermining (finding) wound tissue undermining (finding) direct match Fistula An abnormal track connecting an organ to the skin surface, wound bed or to another organ   Fistula (morphologic abnormality) Fistula (disorder)   no match Fistula (morphologic abnormality) Fistula route (qualifier value) >Exudate Amount   Amount of exudate (observable entity) Amount of exudate (observable entity) Amount of exudate (observable entity) direct match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 103  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  None   Absence of wound discharge (situation) None (qualifier value) none (qualifier value) direct match Scant Wound drainage amount considered in relationship to the size of the wound.   Scant hemorrhage (morphologic abnormality) scanty (qualifier value) direct match Small Wound drainage amount considered in relationship to the size of the wound. Small (qualifier value) Small (qualifier value) small (qualifier value) direct match Moderate Wound drainage amount considered in relationship to the size of the wound. Moderate (severity modifier) (qualifier value) Moderate (severity modifier) (qualifier value) moderate (qualifier value) direct match Large/copious Wound drainage amount considered in relationship to the size of the wound. Large (quantifier value) Copious sputum (finding) large (qualifier value) direct match >Exudate Type         Exudate (substance) exudate observable (observable entity) direct match Exudate (morphologic abnormality) Exudation, function (observable entity) Nil  no exudate Absence of wound discharge (situation) Nil (qualifier value)   direct match Purulent Thick, cloudy Purulent discharge from wound (finding) Purulent (morphologic abnormality) purulent discharge from wound (finding) direct match Sanguineous Bloody Sanguineous exudate from wound (finding) Nevus sanguineous (disorder) sanguineous discharge from wound (finding) one-to-many Sanguineous discharge from wound (finding) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 104  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Serous Thin, clear, yellow Serous discharge from wound (finding) Serous (qualifier value) serous discharge from wound (finding) direct match >Wound Odour     Wound odor (finding) odor of exudate (observable entity) direct match Yes Unpleasant smell noted from wound after cleansing Wound odor (finding) Wound odor (finding) offensive wound odor (finding) direct match Offensive wound odor (finding) No No noted odour from wound after cleansing  Absence of wound odor (situation) Absence of wound odor (situation)   direct match >Wound Edge                   Wound edge sharp (finding)   no match Rolled wound edge (finding) Wound edge finding (finding) Wound edge diffuse (finding) Scab of wound edge (finding) Wound edge necrosis (finding) Wound edges attached (finding) Wound edges approximated (finding) Attached Edge appears flush with wound bed or as a “sloping” edge Wound edges attached (finding) Entire attached surface of gingiva (body structure) wound edges attached (finding) direct match Callused Hyperkeratosis, thickened layer of epidermis   Callosity (disorder) callosity (disorder) no match Demarcated Well defined, distinct, easy to clearly define wound outline Wound edge sharp (finding) Retinopathy of prematurity stage 1 - demarcation line (disorder) wound edge sharp (finding) direct match Diffuse Not well defined, indistinct, difficult to clearly define wound outline  Wound edge diffuse (finding) Diffuse (qualifier value) wound edge diffuse (finding) direct match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 105  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Epithelializing New, pink to purple, shiny migrating tissue Wound epithelialization (finding) Epithelialization (morphologic abnormality) wound epithelialization (finding) no match Epithelial metaplasia (morphologic abnormality) Hypergranulation Tissue Wound edge that is red, moist tissue raised above the level of the skin (proud flesh)   Bone (tissue) structure (body structure)   no match Blood (substance) Myelin (substance) Fascial (qualifier value) Mucosal (qualifier value) Ependyma (body structure) Syngraft (substance) All bone (tissue) (body structure) Subserosa (body structure) Brown fat (body structure) Tooth bud (body structure) Endosteum (body structure) Free flap (substance) Not Attached Edge appears as a “cliff”   Entire attached surface of gingiva (body structure)   no match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus      106  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Rolled  Epithelial wound edge of a cavity wound which rolls under Rolled wound edge (finding) Log roll (procedure) rolled wound edge (finding) direct match Does roll (finding) Swiss roll (substance) Arctic roll (substance) Skin rolling (finding) Pill rolling (finding) Chicken roll (substance) Pancake roll (substance) Sausage roll (substance) Able to roll (finding) Skin rolling (procedure) Tongue-rolling (finding) Unable to roll (finding) Does roll over (finding) Rolling of eyes (finding) Ability to roll (observable entity) Pericardial roll (substance) Difficulty rolling (finding) Does roll on to side (finding) Rolling of conjunctiva (procedure) Brush rolling, function (observable entity) Inferior rolling of breast (procedure) Superior rolling of breast (procedure) Scarred Fibrotic regenerated tissue following wound healing     Scar (disorder) scar (disorder) no match Scar (morphologic abnormality) Healing scar (morphologic abnormality) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  107  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  >Periwound Skin               Boggy periwound skin (finding)   no match Finding of periwound skin (finding) Rupture of periwound skin (finding) Swelling of periwound skin (finding) Hemorrhage of periwound skin (finding) Inflammation of periwound skin (finding) Blister Elevation or separation of the periwound epidermis containing fluid Blister (morphological abnormality) Blister (morphologic abnormality) blister of skin AND/OR mucosa (finding) no match Blistering eruption (disorder) Blister of skin AND/OR mucosa (disorder) Blister - unit of product usage (qualifier value) Boggy Soft, spongy tissue around the wound Boggy periwound skin (finding) Uterus boggy (finding) Boggy periwound skin (finding) direct match Boggy prostate (finding) Tissue bogginess (finding) Bruised Dark red, purplish, blue, tissue that fades to yellow, green, grey, depending on the skin colour Contusion - lesion (morphologic abnormality) Contusion (disorder) contusion (disorder) no match Ecchymosis (morphologic abnormality) Ecchymosis (morphologic abnormality) Contusion - lesion (morphologic abnormality) Callused Hyperkeratosis, thickened layer of epidermis Hyperkeratosis (morphological abnormality) Callosity (disorder) callosity (disorder) no match Dry Flaky or scaly skin Dry skin (finding) Dry (qualifier value) dry skin (finding) no match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 108  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Edema Interstitial collect of fluid to the periwound skin Edema (morphologic abnormality) Edema (observable entity) edematous skin (disorder) no match Edema (finding) Edema (finding) Edema (morphologic abnormality) Erythema <2cm Redness of the skin; may be intense bright red to dark red, that is limited within 2cm from wound edge to the periwound skin   Vestigial cyst (morphologic abnormality) wound erythema (finding) no match Injury of trunk (disorder) erythema of skin (finding) Does not move leg (finding) Unable to move arm (finding) Insect bite granuloma (morphologic abnormality) Cervical dilatation, 2cm (finding) Injury of nervous system (disorder) Hides lit cigarettes in pockets (finding) Erythema >2cm Redness of the skin; may be intense bright red to dark red, that extends beyond 2cm from the wound edge to the periwound skin   Vestigial cyst (morphologic abnormality) wound erythema (finding) no match Injury of trunk (disorder) erythema of skin (finding) Does not move leg (finding)   Unable to move arm (finding)   Insect bite granuloma (morphologic abnormality)   Cervical dilatation, 2cm (finding)   Injury of nervous system (disorder)   Hides lit cigarettes in pockets (finding)   Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 109  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Excoriated/denuded Superficial loss of tissue Excoriation (morphologic abnormality) Excoriation (morphologic abnormality) excoriation of skin (disorder) no match Fragile Skin that is at risk for breakdown Fragile skin (finding) Fragility, function (observable entity)   no match Increased warmth Increased warmth when compared to skin in adjacent area   Increased (qualifier value) increased skin temperature (finding) no match Indurated <2 cm Abnormal firmness of the tissues with palpable margins, that is limited within 2cm from the wound edge to the periwound skin   square centimeter (qualifier value) induration of skin (disorder) no match Induration of wound (finding) Indurated >2cm Abnormal firmness of the tissues with palpable margins, that extends beyond 2cm from the wound edge to the periwound skin   Vestigial cyst (morphologic abnormality) induration of skin (disorder) no match Injury of trunk (disorder) Induration of wound (finding) Does not move leg (finding)   Unable to move arm (finding)   Insect bite granuloma (morphologic abnormality)   Cervical dilatation, 2cm (finding)   Injury of nervous system (disorder)   Hides lit cigarettes in pockets (finding)   Intact  Unbroken skin around the wound Intact skin (finding) Intact (qualifier value) intact skin (finding) no match Macerated Wet, white looking skin – caused by excessive moisture Macerated skin (finding) Macerated skin (finding) macerated skin (finding) no match Macerated fetus (disorder) Macerated stillbirth (finding) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 110  Wound Assessment Matches by Different Mapping Methods (con’t)  Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus  Rash Temporary eruption on the skin-often raised, red, sometimes itchy Cutaneous eruption (morphologic abnormality) Eruption of skin (disorder) eruption of skin (disorder) no match Cutaneous eruption (morphologic abnormality) Tape tear Superficial skin loss due to tape removal   Strapping procedure (procedure)   no match Weepy Moist, draining areas   Crying associated with mood (finding) moist skin (finding) no match >Wound Pain Quantified on the Visual Analogue Scale where 0 = no pain and 10 = excruciating as described by the patient/client/resident Pain score (observable entity) Wound pain (finding)    no match >Date of Onset Best known date of when the wound occurred Date of onset (observable entity) Date of onset (observable entity)   no match >Recurrence When a wound had previously occurred in the same area, usually with the same underlying etiology Recurrence of problem (finding) Recurrence (qualifier value)   no match Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus     111  Appendix C – Wound Diagnosis: Terms and Matches by Different Mapping Methods  This section describes the specific instances of concept, concept matches and mapping method.  Wound Diagnosis Matches by Different Mapping Methods Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus   Parameter   Manual  Apelon Autofill 90-100% match Harris et al Consensus Wound Diagnosis  Clinical Definition SNOMED 2016 Jan SNOMED 2016 Jan SNOMED 2011 Jan Consensus Results Abscess A local collection of purulent and/or sanguineous drainage that may be incised and drained Wound abscess (disorder) Abscess (disorder)   direct match Abscess (morphologic abnormality) Arterial Caused by a disruption to arterial blood flow leading to moderate / severe tissue ischemia Ischemic ulcer (disorder) Ischemic ulcer (disorder)   direct match Burn Tissue loss/damage as a result from thermal, chemical, electrical, and radiation sources. Burn (disorder) Burn (disorder) burn of skin (disorder) direct match Burn of skin (disorder) Burn injury (morphologic abnormality) Burning sensation quality (qualifier value) Diabetic/ Neuropathic  A result of neuropathy, structural foot changes, and/or altered blood flow related to diabetes mellitus (or alcoholism, renal failure, HIV, late stage syphilis, trauma and surgery). Skin ulcer associated with diabetes mellitus (disorder) Neuropathic diabetic ulcer - foot (disorder)   direct match Neuropathic ulcer (disorder) 112  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus   Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Donor Site An area of the patient’s skin (with or without deeper tissue structures) that is surgically removed and placed elsewhere on the body.   Donor site (body structure) donor site (body structure) no match Donor site (attribute) Drug Reaction A wound or skin reaction where the cause is related to the intake/absorption of a drug/medication. Adverse reaction caused by a drug (disorder) Drug reaction (qualifier value)   no match Adverse reaction caused by drug (disorder) Fistula- Non Stomatized A fistula where the nurse cannot visualize a specific area that has connected with the skin. There may be several diffuse areas open on the skin (or in tunnel/undermining)   Non-puerperal fistula of nipple (disorder)   no match Fistula-Stomatized A fistula with a budded opening between the skin and the associated structure/space/organ (eg: small bowel).   Fistula (disorder)   no match Fistula (morphologic abnormality) Fistula route (qualifier value) Graft Site An area of the patient’s body where skin (with or without deeper tissue structures) is surgically grafted from another area of the body.   Skin care: graft site (regime/therapy) recipient site (body structure) no match Site of fistula/graft (attribute) Site of origin of graft (attribute) 113  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Incontinence Associated Dermatitis Skin damage associated to urine and/or fecal incontinence.  Irritant contact dermatitis caused by contact with urine and/or feces (disroder) Irritant contact dermatitis due to incontinence (disorder) irritant contact dermatitis due to incontinence (disorder) one-to-many Irritant contact dermatitis due to incontinence (disorder) Chronic papillomatous dermatitis due to incontinence (disorder) Conjunctivitis associated with dermatitis herpetiformis (disorder) Infectious Skin rash or ulcer from an infectious process and can be categorize according to the organism: fungus, bacteria, virus, or arthropod.  Infectious disease (disorder) Wind (event)   no match Wound (morphologic abnormality) Wound (disorder) Inflammatory Disease Present as atypical ulcers and can be misidentified as venous or arterial ulcers. (i.e.: Pyoderma Granulosum, Vasculitis, Drug-Induced Vasculitis).  Inflammatory disorder (disorder) Inflammatory disorder (disorder)   no match Irradiation Complications from radiation therapy that can present as erythema, dry desquamination, moist desquamination, and/or permanent skin changes. Dermatosis caused by therapeutic ionizing irradiation (disorder) Irradiation (physical force)   direct match 114  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Lymphedema Lymphatic obstruction / blockage of the lymph vessels causing swelling. It is caused by damage to the lymph system, congenital defect to the lymph system, or parasitic infection (filariasis).  Lymphedema (disorder) Lymphedema (disorder)   direct match Lymphatic edema (morphologic abnormality) Malignant Cancerous lesion/wound that may present as painful, fungating and/or friable wounds. Malignant neoplastic disease (disorder) Malignant (qualifier value)   no match Malignant neoplasm, primary (morphologic abnormality) Pilonidal Sinus An abnormal tract/sinus extending from the skin, usually near the cleft of the buttock. Cyst - pilonidal (disorder) Cyst - pilonidal (disorder)   direct match Pilonidal sinus of natal cleft (disorder) Pressure Ulcer- Stage 1 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 1– Intact skin with localized nonblanchable erythema; darkly pigmented skin may not show visible blanching but will appear different than the colour of surrounding skin. Pressure ulcer stage 1 (disorder) Pressure ulcer stage 1 (disorder) pressure ulcer stage 1 (disorder) direct match 115  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus  Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Pressure Ulcer- Stage 2 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 2– Partial thickness wound presenting as a shallow open ulcer with a red / pink wound bed, slough may be present but does not obscure the depth of tissue loss; may also present as an intact or open/ruptured serum-filled or serosanguineous filled blister. Pressure ulcer stage 2 (disorder) Pressure ulcer stage 2 (disorder) pressure ulcer stage 2 (disorder) direct match Pressure Ulcer- Stage 3 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 3– Full thickness wound; subcutaneous tissue may be visible but bone, tendon and muscle are not exposed; may include undermining and/or sinus tracks; slough or eschar may be present but does not obscure the depth of tissue loss. Pressure ulcer stage 3 (disorder) Pressure ulcer stage 3 (disorder) pressure ulcer stage 3 (disorder) direct match 116  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus    Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Pressure Ulcer- Stage 4 Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Stage 4- Full thickness wound with exposed bone, tendon or muscle; often includes undermining and/or sinus tracks; slough or eschar may be present on some parts of the wound bed but does not obscure the depth of tissue loss. Pressure ulcer stage 4 (disorder) Pressure ulcer stage 4 (disorder) pressure ulcer stage 4 (disorder) direct match Pressure Ulcer- SDTI Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Suspected Deep Tissue Injury (SDTI) – A localized purple or maroon area of intact skin or a blood filled blister that occur when underlying soft tissue is damaged from friction or shear.   Pressure ulcer (morphologic abnormality)   no match Pressure ulcer (disorder) 117  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus   Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Pressure Ulcer- X Caused as a result of pressure, friction, shearing and often (though not always) located over a bony prominence. Unstageable, Stage X – A wound in which the wound bed is covered by sufficient slough and / or eschar to preclude staging. Nonstageable pressure ulcer (disorder) Pressure ulcer (morphologic abnormality) Nonstageable pressure ulcer (disorder) direct match Pressure ulcer (disorder) Skin Disease Wounds develop as a manifestation of a disease process or complication associated with the treatment of a disease.(i.e.: Epidermolysis Bullosa, Calciphylaxis, Graft-versus-Host Disease). Disorder of skin (disorder) Disorder of skin (disorder)   no match Skin Tear Skin tears are the result of trauma caused by shearing, friction, or blunt force to the skin. Consider risk factors (i.e.: advanced age, immobility, inadequate hydration/nutrition, falls)  Traumatic tear of skin (disorder) Traumatic tear of skin (disorder) traumatic tear of skin (disorder) direct match Surgery (Secondary Intent) Incision has dehisced and surgical closure is not possible; wounds must heal by granulation  Dehiscence of surgical wound (disorder) Wound healing status: secondary intention (observable entity) dehiscence of wound of skin (disorder) direct match 118  Wound Diagnosis Matches by Different Mapping Methods (con’t) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus     Parameter   Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Trauma Wounds caused by traumatic injury and expected to heal with measureable, objective signs of improvement after three weeks of care (eg: Puncture, Abrasion, Stab, Bite, Gunshot, Laceration). Traumatic injury (disorder) Traumatic injury (disorder)   no match Traumatic abnormality (morphologic abnormality) Venous  Caused by venous insufficiency due to valve dysfunction, complete or partial blockage of the deep veins, and / or failure of the calf muscle pump Stasis ulcer (disorder) Stasis ulcer (disorder)   direct match Stasis ulcer (morphologic abnormality) Venous/Arterial  Caused by both venous insufficiency and disrupted arterial blood flow. Mixed arteiovenous leg ulcer (disorder) Mixed arteriovenous leg ulcer (disorder)   direct match 119  Appendix D – Goal of Care: Terms and Matches by Different Mapping Methods  This section describes the specific instances of concept, concept matches and mapping method.  Goal of Care Matches by Different Mapping Methods Parameter  Clinical Definition Manual  Apelon Autofill 90-100% match Harris et al Consensus Goal of Care Parameters Clinical Definition SNOMED 2016 Jan SNOMED 2016 Jan SNOMED 2011 Jan Consensus Results To Heal the Wound Wound healing is anticipated to occur according to a predictable trajectory.    Wound healed (finding)   no match To Maintain the Wound Maintenance wounds have the potential to heal but are impacted by client, wound and/or system factors that cannot be mitigated resulting in wound healing that is slow or stalled. The factors that are barriers to healing may change over time and periodic re-evaluation is indicated with maintenance wounds.   Maintenance of wound drain (procedure)   no match To Monitor/ Manage the Non-healable Wound Non-healable wounds are sometimes called palliative wounds because their inability to heal may be due to an untreatable cause such as cancer. These wounds may be stable or may deteriorate over time.   Non-invasive blood pressure monitor (physical object)   no match   Airway pressure monitor, non-powered (physical object) Key to colour coding: Blue = concept and definition; Beige = manual mapping; Green = automated mapping; Grey = literature mapping; White = consensus mapping; Yellow = direct match for consensus; Red = one-to-many for consensus 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0340679/manifest

Comment

Related Items