UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Infection prevention and control effectiveness and safety : validation of a survey for long term care… Schall, Valerie 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_schall_valerie.pdf [ 9.15MB ]
Metadata
JSON: 24-1.0066441.json
JSON-LD: 24-1.0066441-ld.json
RDF/XML (Pretty): 24-1.0066441-rdf.xml
RDF/JSON: 24-1.0066441-rdf.json
Turtle: 24-1.0066441-turtle.txt
N-Triples: 24-1.0066441-rdf-ntriples.txt
Original Record: 24-1.0066441-source.json
Full Text
24-1.0066441-fulltext.txt
Citation
24-1.0066441.ris

Full Text

Infection prevention and control effectiveness and safety: validation of a survey for long term care facilities by Valerie Schall BScN, University of British Columbia, 2002  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Health Care and Epidemiology) THE UNIVERSITY OF BRITISH COLUMBIA June 2008 © Valerie Schall, 2008  ABSTRACT Objectives To develop and validate a survey that can be used to measure key infection prevention and control (IP&C) structures and processes in LTC facilities. Methods This study was designed using a three-phase methodology. In Phase I, six structural and process composite indices were developed based on the 2004 PHAC recommendations for IP&C in LTC and other literature. During the second phase of the study, a group of 7 experts in LTC IP&C used the Delphi methodology to validate and further develop the survey based on group consensus. Five Safety Principles published by the Institute of Medicine were also provided to the experts so they could be used to complement and further develop the concepts covered by the survey. The Delphi phase began in April and ended in October 2007; 114 worksheets were sent to experts to support the consensus-reaching process. Once the validity of a survey draft had been established based on expert-group consensus, it was pilot-tested in Phase III using 20 randomly selected LTC facilities in Fraser Health. Conclusions The three-phase methodology used in this study was very useful and innovative way to further develop and validate the literature-based survey developed in Phase I for IP&C in long term care. In addition, by merging two bodies of knowledge and thought into the process, concepts and components that are not explicitly described in IP&C literature yet were felt to be key in program success, were incorporated into the measurement tool. Using Delphi, the experts expressed a need for IP&C professionals working in LTC to increase their knowledge, understanding and use of safety theory and strategies. They also felt that interdisciplinary work, the development of a culture of safety, and the development clear and simple IP&C systems are key ways in which infections can be prevented and outbreaks quickly controlled. In Phase III, the pilot-study analysis demonstrated the utility, validity and reliability of the survey. In addition, the analysis showed that there is a tendency for facilities to have lower levels of components within the Leadership Index and the ICP Index.  ii  TABLE OF CONTENTS ABSTRACT ................................................................................................................................... ii
 TABLE OF CONTENTS ............................................................................................................. iii
 LIST OF TABLES ......................................................................................................................... v
 LIST OF FIGURES ..................................................................................................................... vi
 ACKNOWLEDGEMENTS ......................................................................................................... vii
 DEDICATION............................................................................................................................. viii
 1
 LITERATURE REVIEW AND STUDY RATIONAL.......................................................... 1
 1.1.
 Introduction..................................................................................................................... 1
 1.2.
 Why focus on long term care (LTC)? ............................................................................. 2
 1.3.
 Why is infection prevention and control an issue in long term care?............................. 3
 1.4.
 Evidence of infection prevention and control program effectiveness ............................ 8
 1.5.
 Infection prevention and control and patient safety........................................................ 9
 1.6.
 Gaps in the literature..................................................................................................... 15
 1.7.
 Purpose and significance............................................................................................... 17
 1.8.
 Overarching goal: knowledge translation ..................................................................... 18
 2.
 DESIGN AND METHODS.................................................................................................. 20
 2.1.
 Phase I: Planning Phase ................................................................................................ 21
 2.1.1.
 Step 1: Conceptual model ..................................................................................... 21
 2.1.2
 Step 2: Concept definition and operationalization................................................ 23
 2.1.3
 Step 3: Question development using QFD matrices ................................................. 33
 2.2
 Phase II: Expert Group Validity Testing Phase using Delphi....................................... 35
 2.3
 Phase III: Pilot Testing Phase ....................................................................................... 42
 2.3.1
 Face validity, content validity, and feasibility testing .............................................. 46
 3.
 RESULTS ............................................................................................................................. 50
 3.1.
 Phase II: Expert group validity testing phase using Delphi.......................................... 50
 3.1.1
 Individual expert response time for questions and worksheets ................................ 53
 3.1.2
 Results for Question 1 – Infection control concepts and components...................... 53
 3.1.3
 Results for Questions 2 – Safety concepts and components..................................... 61
 3.1.4
 Results for Questions 3 – Safety concepts and components..................................... 62
 3.1.5
 Results for Questions 4 – Operationalization of components................................... 76
 3.1.6
 Results for Questions 5 – Question validation and development ............................. 80
 3.2.
 Phase III: Pilot Testing Phase ....................................................................................... 82
 3.2.1.
 Sample participation ................................................................................................. 82
 3.2.2.
 Descriptive statistics ................................................................................................. 83
 3.2.3.
 Validity and feasibility test #1: Question Relevance................................................ 85
 3.2.4.
 Validity and feasibility test #2: Data Completeness ................................................. 88
 3.2.5.
 Validity and feasibility test #3: Score Distribution................................................... 89
 4.
 DISCUSSION ....................................................................................................................... 92
 4.1
 Study summary ............................................................................................................. 92
 4.2
 Phase II – Delphi expert-group validation .................................................................... 94
 4.2.1
 Question 1 – Key IP&C concepts and components .................................................. 94
 4.2.2
 Question 2 & 3 – Merging safety and IP&C theory ................................................. 98
 iii  4.2.3
 Question 5 – Question design and survey content validation ................................. 106
 4.2.4
 The Delphi methodology ........................................................................................ 107
 4.2.5
 Moderator effect...................................................................................................... 112
 4.3
 Phase III: Pilot-testing phase ...................................................................................... 113
 4.3.1
 Findings from validity & feasibility tests ............................................................... 114
 4.3.2
 Facility ‘Not-for-profit’ versus ‘For-profit’ status.................................................. 120
 4.3.3
 Potential biases in the pilot study............................................................................ 123
 4.4
 External validity.......................................................................................................... 124
 5.
 CONCLUSIONS & FUTURE RESEARCH ...................................................................... 126
 5.1
 Lessons learned: Merging IP&C and safety ............................................................... 127
 5.2
 Lessons learned: The Delphi methodology................................................................. 129
 5.3
 Knowledge translation and future research................................................................. 131
 REFERENCES ......................................................................................................................... 133
 APPENDIX A: Draft 1 of survey (After literature review) ................................................... 142
 APPENDIX B: BREB certificate of approval – Minimal Risk ............................................. 148
 APPENDIX C: Draft 2 of survey (After Delphi phase) ........................................................ 150
 APPENDIX D: Invitation letter for pilot study (Phase 3)..................................................... 156
 APPENDIX E: Cover letter for pilot study (Phase 3) .......................................................... 157
 APPENDIX F: Question 1 – Reformatted Worksheet 1 ..................................................... 158
 APPENDIX G: Question 2 – Safety Principle 1: Provide leadership................................ 167
 APPENDIX H: Question 2 – Safety Principle 2: Respect human limits in process design ..................................................................................................................................................... 168
 APPENDIX I: Question 2 – Safety Principle 3: Promote effective team functioning ..... 169
 APPENDIX J: Question 2 – Safety Principle 4: Anticipate the unexpected.................... 170
 APPENDIX K: Question 2 – Safety Principle 5: Create a learning environment ........... 171
 APPENDIX L: Indices after Question 4 ................................................................................ 172
 APPENDIX M: Matrix of scatterplots – Correlation between number of beds and staffing ..................................................................................................................................................... 174
 APPENDIX N: Draft 3 of survey (After pilot study) ............................................................. 176
  iv  LIST OF TABLES Table 1.1: Patient safety principles matched by general theme to infection prevention and control (IP&C) structures and processes recommended for long term care (LTC) facilities. ............................................................................................................................... 12
 Table 2.1: Structural composite indices.................................................................................... 24
 Table 2.2: Process composite indices......................................................................................... 24
 Table 2.3: Top four rows of the QFD matrix for the Leadership Index................................ 27
 Table 2.4: Top four rows of the QFD matrix for the Infection Control Practitioner (ICP) Index..................................................................................................................................... 28
 Table 2.5: Top four rows of the QFD matrix for the Policy & Strategy Index..................... 29
 Table 2.6: Top four rows of the QFD matrix for the Partnership & Resources Index ........ 30
 Table 2.7: Top four rows of the QFD matrix for the Surveillance Index .............................. 31
 Table 2.8: Top four rows of the QFD matrix for the Control Index...................................... 32
 Table 2.9: Sample QFD matrix.................................................................................................. 35
 Table 2.10: Questions posed to expert group and supporting documents & worksheets provided ............................................................................................................................... 41
 Table 3.1: Timeline of group membership and of documents sent to experts during Phase II ............................................................................................................................................... 52
 Table 3.2: Descriptive statistics for individual expert response time (in days)..................... 53
 Table 3.3: Leadership Index after modifications based on Question 1 consensus decisions 56
 Table 3.4: ICP Index after modifications based on Question 1 consensus decisions............ 57
 Table 3.5: Unmodified Policy & Strategy Index after Question 1.......................................... 58
 Table 3.6: Partnership & Resources Index after modifications based on Question 1 consensus decisions ............................................................................................................. 59
 Table 3.7: Surveillance Index after modifications based on Question 1 consensus decisions ............................................................................................................................................... 60
 Table 3.8: Control Index after modifications based on Question 1 consensus decisions ..... 60
 Table 3.9: The six indices after modifications based on Question 3 consensus decisions. ... 63
 Table 3.10: Participation of invited facilities in the pilot study.............................................. 83
 Table 3.11: Descriptive statistics of front line staff within sample facilities.......................... 84
 Table 3.12: Mean and Standard Deviation (SD) for percentage of missing data in the survey. .................................................................................................................................. 89
 Table 3.13: Descriptive statistics of standardized scores for entire surevey and for individual indices ................................................................................................................ 90
  v  LIST OF FIGURES Figure 2.1: Study design using a modified QFD methodology ............................................... 20
 Figure 2.2: Conceptual model for infection prevention and control...................................... 21
 Figure 2.3: Delphi methodology used with the expert group.................................................. 36
 Figure 2.4 Sample size calculation............................................................................................. 45
 Figure 2.5 Survey relevance questions ...................................................................................... 47
 Figure 3.1: Distribution of time taken for Question 1 comments and votes.......................... 55
 Figure 3.2: Distribution of time taken for Questions 2 & 3 comments and votes................. 65
 Figure 3.3: Distribution of time taken for Question 4 comments & votes............................. 77
 Figure 3.4: Distribution of time taken for Question 5 votes ................................................... 81
 Figure 3.5: Distribution of facility size based on number of beds .......................................... 84
 Figure 3.6: Boxplots describing distribution of standardized total scores and index scores. ............................................................................................................................................... 90
  vi  ACKNOWLEDGEMENTS I wish to thank each member of the faculty in the department of Health Care and Epidemiology for helping me to learn and understand epidemiology, statistics, health services research, and health policy. Their support and patience helped to make learning a positive and stimulating experience. I feel very blessed to have had Sam Sheps as my thesis supervisor and I can never thank him enough for his guidance and support throughout my Masters program. Sam believed in my ability to do well in this program even when I doubted myself. He was always available to help, and he guided me with respect, kindness and patience. I am truly indebted to Bonnie Henry and David Matheson who agreed to be part of my thesis committee, and who provided me with insightful feedback and continuous support. Even with their hectic professional schedules, they generously shared their large breadth of knowledge and expertise with me whenever needed. Their wisdom and guidance helped me to design and write a thesis with which I can be proud. I would like to acknowledge the contribution of the Provincial Infection Control Network (PICNet) who helped me to develop a thesis question that would be useful to the infection prevention and control community of practice. I am also very grateful towards the experts and facility senior managers who participated in this study and generously volunteered their precious time and expertise to make this study a success. Finally, I would not have been able to this project without the support and funding of the Western Regional Training Centre in Health Services Research (WRTC), which is funded by Canadian Health Services Research Foundation (CHSRF), Alberta Heritage Foundation for Medical Research (AHFMR) and Canadian Institutes of Health Research (CIHR). The training they provided for me greatly enriched my studies with new experiences in leadership and research that I would otherwise have never had in a Masters program.  vii  DEDICATION  To my loving husband and children who never failed to give me their support and understanding as I endeavored to continue my education. Je vous remercie du fond du cœur mes chers amours.  viii  1 LITERATURE REVIEW AND STUDY RATIONAL 1.1. Introduction As the Canadian population ages and the proportion of elderly Canadians increases, the need for safe, high quality long term care (LTC) has become more urgent. The Canadian Institute for Health Information (CIHI) defines LTC as “Health care provided by provincially licensed or approved institutions to persons who reside in the institution” 1  . Given the vulnerability of the elderly population living in LTC facilities to a variety of  infectious diseases, the cornerstone to ensuring safe care is provided in what are in fact the residents’ home, is the implementation of effective programs to prevent infections and to control the spread of pathogenic organisms. Moreover, at a time when new residential care facilities need to be built or older ones expanded, it is particularly important to establish the effectiveness of infection prevention and control (IP&C) in achieving this goal. Currently, the ability to do this is limited by a lack of validated tools to measure IP&C programs in the LTC context. The purpose of this study is to develop and validate a survey instrument that will enable the accurate and reliable measurement of IP&C structures and processes in the LTC setting. Such data could then be correlated with key outcome measures so that researchers may better understand the effectiveness of these programs in preventing infections and controlling outbreaks, thereby enhancing the quality of life, and reducing the likelihood of unnecessary suffering and premature death among LTC residents..  1  1.2. Why focus on long term care (LTC)? Over thirteen percent of Canadians and fourteen percent of British Columbians are now 65 years of age or older 2. In 2011, the baby boom generation will begin to enter retirement age. After that, the proportion of Canadians in this age group is projected to rapidly increase to approximately 25 percent by 2031 3. According to Statistics Canada, the proportion of seniors requiring care has, in fact, been slowly declining possibly because many are living more years without serious health conditions, but also because increasing numbers are choosing to receive care in their homes, and governments are finding it more cost effective to fund home support rather than institutional care 4. Nevertheless, the continuing increase in the total number of elderly Canadians means that there will be a substantially increased need for various forms of LTC in the future.  Two key predictors of an individual’s need for LTC are increasing age and more complex co-morbidities 4, 5. The National Population Health Survey (NPHS) found that between 1994/95 and 1996/97, over half of those who moved from their homes to a long term care facility were 80 years or older 6. In comparison, only 16 percent of seniors who lived at home were in that age group. In addition, 65 percent of seniors who moved to a long term care facility had recently experienced the onset or exacerbation of a chronic health condition that significantly affected their ability to care for themselves. As a result of their heightened vulnerability, the NPHS found a 3.5 fold increase in the proportion of deaths in seniors living in long term care facilities compared to seniors living at home, after adjusting for age and sex.  2  An additional consideration is the increasing cost of care for the elderly. Although less than two percent of Canadians are living in long term care facilities 6, over nine percent of total health expenditures – from all governmental and non-governmental sources are on this form of care 7. In addition, approximately half of provincial acute care hospital expenditures were for the care of seniors, with most of that amount aimed at caring for those between 70 and 84 years old 7; an age group comprising those much more likely to require some form of LTC prior to and/or after hospitalization.  1.3. Why is infection prevention and control an issue in long term care? Recent studies have raised major concerns regarding the quality and safety of the care provided in LTC facilities in Canada and the United States 8-12. This research has not only demonstrated that LTC facilities score much more poorly on key quality and safety indicators than acute care hospitals, but also that the gap between for-profit (FP) and not-for-profit (NP) is much wider in LTC facilities than in acute care. For example when studying LTC facilities in British Columbia, McGregor et al (2006) found that “Adjusted hospitalization rates for all outcomes were significantly higher at FP facilities compared with NP facilities attached to a hospital, and in many cases risk ratios were more than 2fold higher” 12 (p.942).  In 1985, the US Centers for Disease Control and Prevention (CDC) estimated that facility-associated infections occurred at a mean rate of one per resident per year, or 1.5 million facility-acquired infections per year in LTC facilities 13. The specific definition for  3  ‘Facility-associated’ infections varies based on the epidemiology of each causative organism, but in general these are defined as infections occurring during a stay in a long term care facility with an onset 48 to 72 hours after admission. The exclusion of infections occurring early after admission is a generic means of excluding infections acquired from the community or from acute care hospitals. Recent estimates show that pneumonia, the infection that causes the most deaths in LTC, was found to have an approximately ten times greater incidence in LTC compared to rates in communitydwelling seniors 14. This high rate is not surprising given that predictors of the need for LTC are also risk factors for infection.  As part of the normal aging process, physical and biochemical changes make elderly people more vulnerable to infections 13, 15. Urinary retention, decreased pulmonary elasticity, decrease mucociliary clearance of secretions from the lungs, depressed cellmediated immunity, altered inflammatory response, and decreased secretion of gastric acid are known risk factors for infection and occur with normal aging 13, 16. In addition, chronic diseases such as diabetes mellitus, chronic obstructive pulmonary disease, cancer and congestive heart failure, as well as chronic disorders such as dementia, urinary or bowel incontinence, and neurological disorders are all highly associated with increased risk of infection 13, 16, 17. All of these pathophysiological mechanisms result in increased susceptibility to infection.  The average acuity of LTC facility residents has also significantly increased over the past few years. This is partly due to increasingly early discharge from acute care, and  4  partly due to increased use of invasive devices in LTC 13, 18. Early discharge from acute care means that patients are in LTC facilities at a time when they are particularly vulnerable to re-current infections or surgical site infections. According to Sheps et al 19, the number of acute care days per 1000 population declined by 30.0% and the average length of stay dropped by 12.9% in British Columbia between 1991-92 and 1996-97. This study also showed that between 1986-88 and 1993-95, the age-adjusted death rate increased by 7-8% for full-time residents of long term care facilities. Long term use of invasive devices provides pathogens with a direct portal of entry into the tissues, and their use is highly associated with an increased risk of infection 13. In most studies, infections most commonly occur in the urinary tract, especially in residents with indwelling urinary catheters and in those who suffer from urinary or bowel incontinence. Infections also frequently occur in the respiratory tract, in skin and soft tissues, and in the gastro-intestinal tract 18, 20-23. Septicemia is particularly common in residents with end-stage renal disease who require long-term hemodialysis 24.  Infections occur when pathogenic organisms are transmitted from their reservoirs to a susceptible host. LTC facilities can be a highly efficient environment for such transmissions. LTC facilities are populated with contiguous, highly susceptible hosts, and these hosts are increasingly subjected to the risk of antibiotic-resistant organisms from other colonized or infected people within their own facility or while spending time in acute care. To exacerbate this situation, detection is particularly difficult because elderly people often exhibit depressed or altered signs and symptoms of infection due to the aging process and concomitant diseases or disorders 25 (e.g. stroke), they are seldom  5  routinely assessed by medically-trained professionals, and they have decreased access to diagnostic testing compared to patients in acute care facilities 13, 18, 26. Even when a diagnosis of infection is made, control of spread to other residents is difficult. Attempts to isolate residents, even for a short period, may place the elderly person at high risk for physical or psychological trauma 27-29. Furthermore, some control measures, such as long term isolation for a decubitus ulcer infected with an antibiotic-resistant organism, may be seen as inappropriate given that the LTC facility is the person’s home. Moreover, such isolation policies may be difficult to enforce because of cognitive impairment in the resident (i.e. the resident may not remain in his or her room as instructed). Finally, many studies have shown that hand hygiene, the cornerstone of IP&C, is not consistently performed by health care providers in health care settings 30-34, and residents may be unable to perform hand hygiene on their own due to impaired motor or neurological function. Thus, control measures instituted in LTC should go well beyond repeated calls to improve hand hygiene.  Although infections acquired while living in a LTC facility may not postpone a pending discharge home as they do in acute care, they are often the catalyst for a cascade of events that increase the financial costs of care for the public health care system 35, 36. Infected residents require more hours of care from staff, increased use of medical equipment (some of which may be invasive and place the resident at further risk of infection), as well as treatment with increasingly costly antibiotics 37. Since elderly residents often develop more severe infections due to their age and poor health status, they often require transfer to a hospital for treatment 13, 20. During hospitalizations of less  6  than 30 days, vacated LTC beds are held for a pending return. Therefore, one acute as well as one long term care bed may be unavailable for 30 days after a transfer to an acute care hospital, and when the hospital stay lasts longer than 30 days, the resident may be unable to return to their home and may be obliged to remain in hospital until a new LTC bed becomes available. Infections caused by antibiotic resistant organisms are of particular concern because they not only require more costly antibiotics for longer treatment periods, they also have significantly higher relapse rates 38-40. Since antibiotic usage can lead to the development of increasingly resistant pathogens, infections create a cycle of transmission and re-infection within LTC facilities with organisms that are increasingly difficult and costly to treat 20. These infections negatively impact the quality of life of the residents, and in some cases lead to the premature death.  Moreover, infections are known to have long-term detrimental affects on the functional status of seniors. Infections accelerate the aging process and the development of ageassociated diseases such as atherosclerosis, chronic obstructive pulmonary disease, and dementia 41. In addition, studies have found that the case fatality rate associated with an infection is as high as 7.1% in LTC 20. Although many of these infections and deaths may not be preventable, studies have shown that key IP&C structures and processes can be very effective at reducing rates (mainly in acute care - see Section 1.4), although the degree to which infections can be prevented in the LTC setting using specific IP&C structures and processes remains to be evaluated.  7  1.4. Evidence of infection prevention and control program effectiveness IP&C programs have been called the “Premier Quality Assessment Program” 42. This is partly due to findings from a very large 5-year study called “The Study on the Efficacy of Nosocomial Infection Control (SENIC)” conducted by the United States Centers for Disease Control (CDC) and published in 1985. The primary reason for this study was to establish the validity and cost-effectiveness of CDC recommendations so IP&C programs would not succumb to fiscal pressures. This study provided strong evidence of IP&C program effectiveness in acute care hospitals after adjusting for a large number of confounders 43. Thirty two percent of all hospital-associated infections were found to be preventable by establishing certain key infection surveillance and control structures and processes. The authors calculated that this could result in the prevention of 150 hospital-associated infections and three to five avoidable deaths per year in a typical 250-bed US hospital 44.  Numerous smaller studies from around the world have also shown an association between implementation of infection surveillance and control activities, such as isolation, use of personal protective equipment, or cohorting of sick patients or residents in one room or in a set of rooms with care provided by dedicated staff, and lower rates of hospital-associated infections 45, 46. For instance, research has shown major decreases in rates of infection or colonization from antibiotic-resistant organisms such as methicillin-resistant Staphylococcus aureus (MRSA) 47, 48, Vancomycin-resistant enterococcus (VRE) 49, and Clostridium difficile-associated diarrhea (CDAD) 50. Other  8  studies have shown the value of specific processes in reducing the rate of urinary tract infections 51, respiratory infections 52, 53, and surgical site infections 54.  1.5. Infection prevention and control and patient safety The Agency for Healthcare Research and Quality defines patient safety as “Freedom from accidental injury,” or “avoiding injuries or harm to patients from care that is intended to help them.” 55. Based on these definitions, indicators used by researchers to help measure key aspects of patient safety have commonly included healthcareassociated infections 56-59. In 1991, the Harvard Medical Practice Study II on patient safety published results showing that wound infections were the second most common type of healthcare-associated adverse events 60. Patient safety literature describes preventable healthcare-associated infections as adverse events related to or caused by errors embedded in institutional or clinical processes 60. According to theory, “Ensuring patient safety involves the establishment of operational systems and processes that minimize the likelihood of errors and maximizes the likelihood of intercepting them when they occur.” 55. Therefore, the authors of the Harvard Medical Practice Study II concluded that IP&C improvements were a critical method of reducing risk of infectionrelated adverse events.  On the other hand, IP&C experts have not historically made the connection between patient safety and IP&C, nor have they directly used patient safety theory to complement IP&C theory. Hospital IP&C programs were developed in the 1950s and  9  1960s, therefore decades prior to the development of safety theory, and primarily in response to staphylococcal outbreaks 43. By the 1970s, the US CDC formulated specific recommendations regarding the necessary structures and processes of acute care hospital IP&C programs. In 2004, Health Canada (now the Public Health Agency of Canada – PHAC) published as set of recommendations for key IP&C structures and processes specific to long term care facilities 61. These guidelines were developed in 1997-98 and based on consensus decisions by the Infection Control Alliance, a comprehensive group of 27 IP&C experts from across Canada. Although the birth of IP&C programs pre-date by three decades public concerns regarding safety in health care, the theoretical underpinnings of these programs seem to follow principles that are very similar to those of safety system design endorsed in 2000 by the Institute of Medicine (IOM), and described in the report “To Err Is Human” 62.  Table 1.1 shows the similar general themes that can be noted between the five safety principles endorsed by the IOM and the IP&C structures and processes recommended by Health Canada / PHAC for long term care facilities 61. From this table it can be seen that, similar to the recommendations made in Safety Principle 1, one of the main objectives of an IP&C program is to provide leadership and expertise within the facility to set a common focus on the prevention and control of facility-associated infections. Principle 2 is exemplified by the development and use of simple standardized definitions for infections as well as screening and immunization protocols. These are some of the common methods used in IP&C to simplify processes so that human limits of recall are appreciated and compensated for. Providing IP&C education to entire clinical teams on  10  a regular basis actualizes the promotion of effective health care team functioning recommended in Principle 3. Continuous active and passive surveillance for infection is a common and effective mechanism used in IP&C to identify, investigate and change unsafe practices, and enable IP&C personnel to anticipate the unexpected as recommended in Safety Principle 4. Finally, Principle 5 is often fulfilled by regular and timely dissemination of surveillance data to physicians and other health care workers. This creates a feedback loop and promotes a learning environment. Consultation and liaison with other experts also helps facilities to learn from past errors and make changes to improve care.  11  Table 1.1: Patient safety principles matched by general theme to infection prevention and control (IP&C) structures and processes recommended for long term care (LTC) facilities. Safety Principles and their subRecommended IP&C structures & processes for components LTC (Institute of Medicine “To Err is Human” 2000) Principle 1: Provide Leadership • • • • •  Make patient safety a priority corporate objective Make patient safety everyone’s responsibility Make clear assignments for and expectation of safety oversight Provide human and financial resources for error analysis and systems redesign Develop effective mechanisms for identifying and dealing with unsafe practitioners  Principle 2: Respect Human Limits in Process Design • • • • • •  Design jobs for safety Avoid reliance on memory Use constraints and forcing functions Avoid reliance on vigilance Simplify key processes Standardize work processes  (Health Canada / PHAC, 2004) The following program leadership is recommended: • An ICP with: o 1.6 FTE per LTC facility with up to 250 beds o 1 FTE per facility with up to 150 beds for resident groups requiring a high level of care • An Infectious Disease Physician The following resources are recommended: • Epidemiologist support • Secretarial Support • Personal computer with internet access appropriate software • Laboratory resources with serology, virology & molecular epidemiology The following activities are recommended: • Surveillance with the use of simple, effective systems using standard definitions. Process should be standardized by basing it on the Canadian consensus surveillance criteria • An occupational health program which includes mandatory Tuberculosis screening and work restrictions during outbreaks. There should also be an immunization programs which includes yearly influenza vaccine. • A resident health program with includes admission tuberculosis screening and an influenza and pneumococcal vaccine program  Principle 3: Promote Effective Team Functioning • Train in teams those who are expected to work in teams • Include the patient in safety design and the process of care  Education should be provided to the following stakeholders: • Staff • Families • Visitors  Principle 4: Anticipate the Unexpected  The following recommendations are also made: • Effective infection prevention and control policies and procedures should be development and implemented • Surveillance should be used to quickly identify potential outbreaks and it should focus on high morbidity/mortality infections • When issues are identified using surveillance, the IP&C program should work at finding ways to resolve these quickly • The facility should respond quickly to potential outbreaks by using control measures (e.g. antivirals, cohorting, etc.) • The facility should use proven preventative strategies such as limiting unnecessary antimicrobial use, and using effective personal protective equipment and cleaning products  •  • •  Adopt a proactive approach: examine processes of care for threats to safety, and redesign them before accidents occur Design for recovery Improve access to accurate, timely information  12  Safety Principles and their subcomponents  Recommended IP&C structures & processes for LTC  (Institute of Medicine “To Err is Human” 2000)  (Health Canada / PHAC, 2004) Additional recommendations: • Consultants in IP&C and epidemiology should be used to help in the development and implementation of a surveillance program • Surveillance findings should be used for continuous quality improvement • The expertise and guidance of Public Health professionals should be used to find ways to prevent infections and control outbreaks • The ICP and staff should be provided with adequate training in IP&C  Principle 5: Create a Learning Environment • Use simulations whenever possible • Encourage reporting (of errors and hazardous conditions) • Ensure no reprisals for reporting of errors • Develop a working culture in which communication flows freely regardless of authority gradient • Implement mechanisms of feedback and learning from error  Nevertheless, many important components of the five safety principles have not explicitly been part of IP&C theory and practice. In particular, the safety principles stress the importance of developing and maintaining a culture of safety within which everyone works together to prevent adverse events, and “communication flows freely regardless of authority gradient” (i.e. there is a flat hierarchy with regard to communication). “Organizations with a positive safety culture are characterized by communications founded on mutual trust, by shared perceptions of the importance of safety, and by confidence in the efficacy of preventive measures” 63. In addition, the safety principles place great emphasis on adverse event analysis. This concept involves early detection of adverse events using methods similar to surveillance used in IP&C. Adverse event analysis then goes beyond activities involved in outbreak investigations by actively seeking to understand not simply the event or outbreak itself, but also the dynamics within organizations that give rise to adverse events, and changing system designs to remove actual or potential “threats to safety”.  13  Thus, patient safety theory may not only enhance IP&C processes to improve prevention and control of infections, but the use of safety language may also improve communication between IP&C professionals and decision-makers within the organization and at all levels of government when lobbying for IP&C resources. Thanks to governmental reports such as the Kirby 64 and Romanow 65 reports on the health of Canadians in 2002, as well as reports from the Canadian Institute for Health Information 66 and the Canadian Centre for Policy Alternatives 67, studies such as ‘The Canadian Adverse Events Study’ in 2004 58, and initiatives like ‘Safer Health Care Now’, many Canadian health care administrators, senior managers and governmental decision-makers, particularly those working in acute care settings, should be well versed in healthcare safety theory and language. For instance, the Calgary Health Region is currently updating a policy entitled ‘Just and Trusting Culture’ 68. The new version will clearly state that all senior managers have many safety-related responsibilities such as doing safety analyses which require advanced knowledge in systems safety theory. Making the link between IP&C and safety languages may therefore create a harmony and a more effective partnership between IP&C professionals and decision makers, administrators and senior managers. It may facilitate and enhance overall communication and assist in reaching the common goal of improved safety in health care.  14  1.6. Gaps in the literature Although seniors living in LTC facilities are highly vulnerable to infection, research specifically related to infection prevention and control (IP&C) in LTC has been slow to progress. While there are now recommendations from expert advisory groups describing IP&C program organizational structures and activities that should be in place in LTC 13, 69, 70, these are largely based on extrapolations from structures and processes shown to be effective in acute care. This has occurred even though all expert groups agree that IP&C programs and processes in LTC and acute care need to and should be fundamentally different 18. Surveillance definitions and methods need to be population appropriate, particularly in an environment in which confirmed acute diagnoses are more difficult to establish. In addition, control measures need to maintain a fine balance between resident safety and quality of life within their home. Because of the challenges of diagnosing infections in LTC facilities due to physiological changes stemming from the aging process and from co-morbidities, and because of infrequent assessments by trained professionals and decreased access to diagnostic testing, a validated set of definitions has already been developed specifically for the surveillance of facilityassociated infections 71. The recommendations published by Health Canada / PHAC in 2004 were also a major milestone for Canadian IP&C programs because they more clearly defined the differences and similarities in the structures and processes required by three fundamentally different types of health care environments: acute care facilities, long term care facilities, and community and home care settings 61. In addition, a few studies have attempted to describe the state of infection prevention and control in LTC,  15  but this research has generally used poorly validated surveys derived from instruments adopted from acute care programs 72-74.  In Canada, a survey of IP&C structures and processes was sent to a large number of the long-term care facilities in 2005 for the purpose of describing the state of IP&C within each province, but the results have yet to be published 75. The main weakness of this study is that the survey was 15 pages long and only 23 percent (38 out of 165 facilities) of the facilities in British Columbia were reported to have responded. In 1992, the Community and Hospital Infection Control Association of Canada (CHICA-Canada) sent a survey to its members working in LTC facilities across Canada, to identify the salient needs of this special group in the IP&C community of practice 73. Survey respondents identified quality assurance and risk management, surveillance systems, basic IP&C, and policy and procedure development as their most urgent needs.  Only one study attempted to evaluate the effectiveness of infection surveillance and control in LTC facilities 76. Investigators attempted to replicate the SENIC design on a smaller scale by surveying all LTC facilities in the state of Maryland using a modified and lengthened SENIC-derived survey. They evaluated effectiveness by comparing surveillance and control indices to self-reported rates of infection. Although there was an attempt to use the design of the SENIC study to establish the effectiveness of IP&C structures and processes in LTC, this study had several weaknesses. First, instead of designing a survey specific to long term care based on national expert body recommendations, the SENIC study survey, designed and validated for acute care, was  16  used as the foundational document and modified using recommendations from local infection control practitioners with expertise in long term care. Secondly, investigators for the SENIC study had distilled their survey to a maximum of six pages because of the low response rates associated with long surveys 77. However, in order to make the SENIC survey more applicable to the long term care setting, certain questions were removed and many were added thereby making the survey longer (exact number of pages was not reported). However, using this lengthy survey, the authors reported a 57 percent response rate, thereby leaving a possibility that the findings do not apply to all LTC facilities in their study population; it is possible that those who participated in the study after being invited may systematically be different from those who did not respond. Thirdly, only content and face validity assessments were performed. Reliability, internal consistency, criterion and construct validity testing were not performed. Finally, self-reported crude rates of infection have a large potential for surveillance bias. This type of bias is likely to occur because facilities expending more effort and resources to identify infections seem to have higher rates than those that do not seek them out.  1.7. Purpose and significance The objectives of this study are the following: 1.To develop a short survey based on the 2004 Health Canada / PHAC recommendations for key IP&C structures and processes in long term care facilities and the five IOM safety principles.  17  2.To establish the validity and feasibility of this survey when used to describe key infection prevention and control structures and processes in LTC facilities within the context of the five safety principles listed above.  The validated survey is being developed in order to evaluate the effectiveness of IP&C structures and processes in British Columbian LTC facilities. This study proposes to improve on previous attempts at developing a validated survey tool for LTC in four major ways. First, the researcher will develop a survey tool that is based on nationally recommended IP&C structures and processes specific to long term care and published by Health Canada / PHAC in 2004. Secondly, the researcher will use the validation process to limit the final survey to include only questions that are deemed highly relevant, valid and reliable for measuring key IP&C structures and processes. Keeping the survey to the shortest possible length will minimize survey non-response when large numbers of facilities are surveyed using a self-administered method. Thirdly, the researcher will assess the feasibility as well as face and content validity of the survey using both a panel of experts and a sample of facilities. Such information will provide guidance to decision-makers when planning IP&C and safety initiatives in LTC facilities. Finally, the researcher will develop a method of evaluating IP&C structures and processes within the context of the five safety principles described above.  1.8. Overarching goal: knowledge translation A fundamental and overarching goal is to ensure that the findings of this study are translated into action by providing decision-makers with an evaluation tool that enables  18  them to assess and improve the IP&C services provided to residents of LTC facilities. To achieve this, the researcher engaged stakeholders by forming a partnership with experts and decision-makers in IP&C through the Provincial Infection Control Network (PICNet) of British Columbia. Knowledge gaps were identified by PICNet and the study question was jointly developed to not only enable us to meet our goal of developing a useful tool to help improve infection prevention and control in LTC facilities, but also to help fulfill PICNet’s mandate in providing guidance to the broader IP&C community of practice, and in making recommendations to the Ministry of Health. This study also involved PICNet stakeholders with expertise in LTC infection prevention and control to participate in the expert group validation testing phase.  19  2. DESIGN AND METHODS In order to design a survey that accurately and reliably measures the numerous key aspects of LTC infection prevention and control (IP&C) programs, the study used a survey development strategy called Quality Function Deployment (QFD) 78. This methodology was altered to imbed validity testing into the different phases of the survey development process (See Figure 2.1).  Figure 2.1: Study design using a modified QFD methodology Phase I: Planning phase Step 1: Conceptual model development Step 2: Concept definition and operationalization Step 3: Question development using a QFD matrix Phase II: Expert group validity testing phase using Delphi Phase III: Pilot testing phase with feasibility and validity testing  A three-step planning phase involved the use of a QFD matrix to develop a survey based on Health Canada recommendations for IP&C in long term care facilities. In the second phase of the study, a small group of PICNet stakeholders with expertise in LTC infection prevention and control and public health were recruited. This group performed a set of validity tests using the Delphi methodology, and the findings were used to refine the survey. During the third phase, a group of 20 LTC facilities in the Fraser Health region of British Columbia were randomly selected for survey pilot-testing and assessment of validity and feasibility.  20  2.1. Phase I: Planning Phase 2.1.1. Step 1: Conceptual model In the planning phase of the QFD methodology, the crucial first step of developing a conceptual model of IP&C effectiveness (see Figure 2.2) was undertaken. This model provides a theoretical framework needed for developing the key concepts to be measured by the survey. Since this survey is being developed and validated for the purpose of future work to evaluate the effectiveness of key IP&C structures and processes across British Columbia in decreasing the rate of facility-associated infections, this study was conceptualized using a combination of two models.  Figure 2.2: Conceptual model for infection prevention and control  21  Donabedian’s three-component model is useful for assessing the quality of health care 79  . The ‘structure’ component describes both human and material resources specifically  needed to reduce preventable infections and enhance patient safety in LTC facilities. The ‘process’ component denotes the specific activities needed to identify and control infections and outbreaks within facilities. Finally, the ‘outcome’ category encompasses all of the positive and negative effects of IP&C structural and process components. Donabedian theorized that when these components are closely interrelated, the structural components drive care processes, which in turn affect proximal and distal patient outcomes. This interrelatedness enables researchers to link and evaluate the effectiveness of key structural or process components of care.  A slightly modified European Foundation for Quality Management (EFQM) model was added to complement and expand upon the Donabedian Structure-Process-Outcome model 80, 81. In our EFQM modified model, six structural and process dimensions (denoted by four boxes under STRUCTURE, and two boxes under PROCESS) are seen as comprehensively describing all key activities and stakeholders that/who ‘enable’ excellence in IP&C. These components are linked to highlight the necessity for teamwork and partnership building, and the larger leadership box pictorially represents the need for strong leadership and constancy of purpose. With regard to this study, the process dimension was subdivided to represent two main processes involved in IP&C: 1) ‘Surveillance’ aimed at detecting infections and outbreaks, and 2) ‘Control’ processes aimed at preventing and managing infections and outbreaks. The four outcome dimensions describe indicators related to all stakeholders when IP&C is enhanced in  22  LTC facilities. According to the model, measurable results can be expected in key performance indicators, and these can be used to evaluate the effectiveness of IP&C structures and processes. These indicators include acute care admissions or readmissions for serious infections, and mortality associated with an infection, thus individual-level data for the key indicators can be obtained using the BC Linked Health Database 82.The outcome dimensions are linked to show the interconnectedness of effects on all sectors of this population. Similar to Deming’s plan-do-check-act cycle of continuous quality improvement, the model also demonstrates the iterative nature of quality improvement, with a looping back mechanism that emphasizes continuous innovation and learning. This model not only makes it possible to develop a tool for measuring key IP&C structures and processes as is being done in this study, but using regression analysis, it can also be used to compare the relative effectiveness of the various structures and processes with key outcome indicators.  2.1.2 Step 2: Concept definition and operationalization The second step of the planning phase involved the use of the four structural and two process dimensions of the conceptual model to create the following six composite indices of IP&C (see Tables 2.1 and 2.2): Leadership Index, Infection Control Practitioner Index, Policy & Strategy Index, Partnership & Resources Index, Surveillance Index and Control Index.  23  Table 2.1: Structural composite indices  Table 2.2: Process composite indices  24  In order to better define each index in a clear and measurable way, the key concepts and components of the six indices were then used to build six Quality Function Deployment (QFD) matrices 78, 83. As shown in Table 2.3, Table 2.4, Table 2.5, Table 2.6, Table 2.7, and Table 2.8, Row 1 of each index is the index name. The 2004 Health Canada / PHAC recommendations for IP&C in long term care settings 61 were used to identified key concepts within each of the six indices, and these were entered into Row 2 of each index. Numerous sources of information were then used (as described below) to better understand each concept so these could be comprehensively and systematically broken down into their key components. These components were then entered into Row 3 of each matrix. Finally, each component was operationalized by identifying ‘how’ each component would be measured in an accurate and valid way given the resources available to long term care facilities. Although the development of the exact weighting calculation for each index is beyond the scope of this study, in order to be used in an analytic study each index will need to be calculated with each operationalized subcomponent weighted to reflect the proportion of content coverage for that particular index.  Leadership Index According to Health Canada / PHAC, an Infectious Disease physician and an Epidemiologist who dedicate specified hours to a facility are necessary components of an IP&C program in long term care settings 61. This requirement is generally supported by the SENIC study which found that having a physician or microbiologist leader “interested enough in IP&C to have taken a course on the subject” (p. 194) 43 was  25  associated with a significant decrease in hospital-associated infections (Note: Although the SENIC study provided no recommendations regarding the specified number of hours needed to achieve this, Health Canada / PHAC has recommended that the Infection Control Physician spend four hours per month on IP&C for a 250 bed LTC facility). Nevertheless, the Society for Healthcare Epidemiology of America (SHEA) Long Term Care Committee and the American Professionals in Infection Control and Epidemiology (APIC) state that, due to fiscal constraints, long term care facilities are often unable to have such clinical leadership for IP&C 13. As an alternative, they recommend the use of consultants with expertise in infectious diseases and IP&C. Based on this information, the researcher felt that the physician’s role in the program should be weighted by the type of relationship (i.e. more weight given to a consistent leadership role versus an “as needed” consultant role), and expertise in IP&C related specialties, and the time devoted to IP&C within the facility. Although not specifically recommended by Health Canada / PHAC, LTC facilities may obtain consultative support through local or regional long term care IP&C committees. Local IP&C committees may have representation from facility administrative staff who work collaboratively with and support IP&C. This direct organizational leadership involvement in LTC infection prevention and control programs is strongly recommended by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) 84. According to JCAHO, “Only with the ongoing attention and direction of organizational leadership can the appropriate scope of the program be determined and adequately resourced” (P.2). Accordingly the presence of an IP&C committee was added to the leadership index. The presence of a  26  committee was weighted by its perceived ability to assist with local IP&C issues (i.e. whether it is a local versus regional committee).  Table 2.3: Top four rows of the QFD matrix for the Leadership Index Row 1  Index Name  Row 2  Key concepts  Row 3  Concept components  Row 4  Operationalized components  Leadership Index Physician  IC Committee  Role  Time  Education  Leader vs consultant  Hours per month  One of (micro, epi, ID, IC)  Degree of leadership provided  Presence (Y/N)  Local vs regional  Infection Control Practitioner (ICP) Index All Canadian and American IP&C advisory groups strongly recommend that IP&C programs in any setting have a specially trained infection control practitioner available to perform key IP&C activities 13, 43, 61, 84, 85. According to the ICP professional standards developed by the Community and Hospital Infection Control Association of Canada (CHICA-Canada), ICPs must have “knowledge and experience in areas of patient care practices, microbiology, asepsis, disinfection/sterilization, adult education, infectious diseases, communication, program administration, and epidemiology.” (P.3) 86. However, in the LTC setting, the ICP position is often covered by a staff member or facility administrator with several other roles within the institution 18, 26, 74. A CHICACanada survey of long term care ICPs found that less than one third of the designated ICP’s time was spent on infection prevention and control activities 73. In British Columbia, a group of facilities may also share the time of an ICP devoted to LTC 27  facilities only, or managing activities for a mixture of acute and long term care facilities. The ICP Index was therefore weighted by the time devoted to the facility, the years of experience and the degree of educational expertise in infection prevention and control.  Table 2.4: Top four rows of the QFD matrix for the Infection Control Practitioner (ICP) Index Row 1  Index Name  ICP Index  Row 2  Key concepts  Infection Control Practitioner (ICP)  Row 3  Concept components  Row 4  Operationalized components  Time  Education  Experience  Hours per 100 LTRC beds per week  Courses in IC, epi (0, 1, >1)  In years (large > 2 yrs)  Policy & Strategy Index This index was developed using the elements of Health Canada / PHAC recommendations related to policies and procedures. The presence of policies and procedures was weighted by ICP or expert consultant participation in the development process as a proxy for document quality. Also, the presence of monitoring systems for staff compliance was used as a proxy for implementation as was done in the SENIC study 77. ICP participation in the selection of products related to the prevention and control of infections (e.g. personal protective equipment, cleaning and sterilization products and equipment, and nursing care products, equipment and devices) were also used to weigh the effectiveness of the facility IP&C policies and procedures.  28  Table 2.5: Top four rows of the QFD matrix for the Policy & Strategy Index Row 1  Index Name  Policy & Strategy Index  Row 2  Key concepts  Policies & Procedures  Row 3  Concept components Development  Row 4  Operationalized components Written policy (Y/N)  Quality (ICP or consultant involvement) (Y/N)  Product selection quality  Implementation  Staff education (Y/N)  Estimate of compliance (proportion of time staff are compliant)  Monitoring system for compliance (Y/N)  ICP participation (Y/N)  Partnership & Resources Index The components for calculating this index were:   the frequency of contacts with Public Health or IP&C services from other LTC or acute care facilities for decision making support or educational materials    the availability of information technology such as computers, internet service, software appropriate for surveillance and surveillance data entry support    the degree of access to laboratory services during each day of the week to quickly diagnose infections    access to educational materials, either in hard copy or on the internet. These materials include IP&C journals, and books, as well as access to provincial outbreak reports and national guidelines to assist in decision making and guideline development. Except for the Public Health component and specifics regarding the needed educational material, all of these components are explicitly recommended by Health 29  Canada / PHAC for IP&C programs in long term care facilities. In British Columbia, Public Health must be contacted when an outbreak of respiratory infection or gastroenteritis occurs. Once contacted, Public Health provides outbreak investigation and containment support, as well as educational resources to LTC facilities. This component was therefore added to the index calculation.  Table 2.6: Top four rows of the QFD matrix for the Partnership & Resources Index  Books Access to any book on IC or epi  Hand hygiene (Y/N)  Journals Access to any IC journal (Y/N)  National guidelines Routine & additional precautions (Y/N)  Bacteriology Access (days per week)  BCCDC outbreak reports  Virology Access (days per week)  One of GI or influenza regular reports  Secretarial support Access (Y/N)  Educational material  Access (Y/N)  Internet Access (Y/N)  Lab services  Appropriate software for surveillance  Computer Access (Y/N)  IT for ICP  Liaison with PH  Row 4  Operationalized components  Row 3  Partnershi p support  Frequency  Key conce pts  Frequency  Row 2  Partnership & Resources Index  Support from IC program in other facility  Index Name  Concept component s  Row 1  Surveillance and Control Indices As in the SENIC study, both a surveillance and a control index were developed for measuring each of these component’s individual effectiveness in decreasing infections in LTC settings. Since the needs and priorities of LTC are different from those of acute care, Health Canada / PHAC recommended elements that were used to define each of the three processes involved in surveillance 86: methods used and frequency of data collection, data analysis for the purpose of identifying unusual patterns of infection, and 30  reporting of significant findings to those with the ability to affect the process of care directly (i.e. facility staff and physicians) or indirectly (i.e. facility administration and IP&C committee if available) . Similarly, the four control elements and weighting variables were specifically recommended by Health Canada / PHAC and other expert advisory bodies 18, 84 for long term care facilities: rapid outbreak control (i.e. control of the outbreak before a threshold number of cases have occurred), education of stakeholders (i.e. including paid educational time, materials and opportunity for ICPs and front-line staff, and the availability of educational material for visitors and family), and infection prevention measures directed at both residents (i.e. admission screening and immunization program) and staff (i.e. new staff screening, influenza immunization program, and policy for work restrictions during influenza outbreaks).  Table 2.7: Top four rows of the QFD matrix for the Surveillance Index Row 1  Index Name  Row 2  Key concepts  Surveillance Index Data analysis  Data collection Row 3  Row 4  Concept components  Simple written standardized definitions for infections  Operationalized components  Use (Y/N)  Surveillance for high M&M infections Case finding using one of: Charts, Kardex, vitals record, lab, rounds, staff report  Types of infections: pneumonia, UTI, MRO, RI, viral gastro, CDAD, flu, skin infections  Monitoring of Antibiotic use  Early detection of outbreaks  Done (Y/N)  Number of unusual infections or outbreaks detected after 1 case or 2 cases  Reporting  Report prepared  Report access by stakeholders  Use of report for CQI  At least one per year (Y/N)  Access of: Staff, physicians, administratio n, IC committee  Report review and improveme nts made (Y/N)  31  Table 2.8: Top four rows of the QFD matrix for the Control Index  Potential confounders When the survey is used to measure the effectiveness of IP&C structures and processes, potential confounders will need to be measured concurrently so that their effects may be controlled. One of the factors with the potential to affect infection outcomes is facility size; the more beds a facility has, the more likely it is that a resident, staff or visitor in the facility will have an infection with the potential to begin an outbreak. Facility size may also increase the likelihood that a facility will have an ICP and written IP&C policies and procedures. It also increases the likelihood that the facility has a larger ‘high-risk’ population. Admission policies relating to living wills or ‘Do Not Resuscitate” status may affect whether a resident is aggressively treated for an infection or sent to acute care for treatment. Since the outcome of interest for an effectiveness study is projected to be acute care admissions for infections, it is very important to control for this confounder. The staff-to-resident ratio 87 and the level of education of the front line staff have also been found to affect quality outcomes in many studies 18, 26,  32  Yearly staff flu immunization program  Influenza prevention  Work restrictions during outbreaks (Y/N)  New staff screening  TB skin test (Y/N)  Immunization  Occupational transmission prevention  Immunization status (Y/N)  Admission screening  One time pneumococcal (Y/N)  Staff  Yearly flu (Y/N)  ICP  Visitor s& family  TB (Y/N)  Rapid control  Preventative care for residents  Education  Availability of material  Row 4  Concept components  Operationalized components  Row 3  Outbreak prevention & managemen t  Access (hours per year)  Key concepts  Paid education time (0, 1 day, >1 day)  Row 2  Control Index  Access to IC courses or conf. (Y/N)  Index Name  Number of outbreaks detected after 2 cases  Row 1  46  . Resident case mix has been found to have a major impact on the facility’s ability to  detect and control infections 26. Nevertheless, this composite variable will not be evaluated by the survey because this tool is to be complemented by individual-level administrative data on LTC residents (i.e. From the BC Linked Health Database if these data are made available to researchers within a reasonable time-frame) 82. Such data permit a more complete and reliable calculation of case mix than would a survey answered by one staff member within each facility 88. Finally, the length of time that a formal IP&C program has been in existence also needs to be taken into account 77 for its effects on the rate of infection and as a measure of the extent to which the IP&C program is embedded into the culture of the facility. Two specific events were selected as indicators of the beginning of an infection prevention and control program in a LTC facility: the establishment of a hand-hygiene policy, and the creation of an ICP position. The starting date of the infection prevention and control program was defined as the earliest year during which either of these two events occurred.  2.1.3 Step 3: Question development using QFD matrices The final step in the planning phase was to use the six QFD matrices (one matrix for each index) to translate the operationalized concepts (written in the top row of each matrix) into a set of clear questions that cover the full definition range of each concept. The set of questions specific to each matrix and generated through this process was then written in the first left hand column of the appropriate matrix (See  33  Table 2.9 for a sample matrix). As an internal validation of the close relationships between the questions generated and the matching operationalized components of the matrix, the research team then assessed the six matrices to explicitly identify their perception of each relationship. Three degrees of relationship proximity were used: 1) A strong relationship between a question and an operationalized component was identified as an ‘S’, 2) A moderate relationship was identified as an ‘M’, 3) A weak relationship was identified as a ‘W’. The main benefit of this methodology is that it enables easy visualization of the perceived strength of each relationship so that questions with moderate relationships can be strengthened, and questions with weak relationships can be redesigned or deleted. This method also helps to minimize the length of the survey by enabling the investigators to delete redundant questions they believe cover the same concepts more than once. Since the SENIC study questions and question formats have been shown to yield valid and reliable information in numerous similar studies, these were used whenever they were found to strongly correlate with the operationalized concepts of this study 89 (p.177). Questions were modified, added or deleted until each operationalized component was found to have one strong relationship to a question. The six final sets of questions were then used to prepare the first survey draft (See Appendix A).  34  Table 2.9: Sample QFD matrix Index Name  Leadership Index  Key concepts Concept components Operationalized components Does your facility have a physician or epidemiologist serving as leader or consultant for infection control? Leader (Y/N), Consultant (Y/N)  Physician Role  Time  Leader vs consultant  Hours per month  Education  Degree of leadership provided  One of (micro, Presence epi, ID, IC) (Y/N)  Local vs regional  S  Please circle any specialty training that your physician leader of consultant has had: a)Infection control b)Epidemiology c)Microbiology d)Infectious Diseases On average, how many hours per month does this physician or epidemiologist spend on infection control activities for your facility?  IC Committee  S S  Is there an infection control committee that acts as an infection control resource for your facility?  S S  Is this committee within your facility or for the region?  2.2 Phase II: Expert Group Validity Testing Phase using Delphi The next phase of this process involved the formation of a small criterion sample 90 of PICNet experts to improve and validate the key concepts, components and operationalized components of the six QFD matrices, to develop a method of evaluating IP&C structures and processes in terms of the five safety principles, to improve the survey questions for each of the six indices, and to assess the face and content validity of the survey before it is pilot tested in the next phase 89-92. Once ethics approval had been obtained for Phase II from the University of British Columbia Behavioural Research Ethics Board on March 6th, 2007 (See Appendix B), an e-mail soliciting members was sent to all stakeholders from the Provincial Infection Control Network of 35  British Columbia (PICNet), and volunteers were selected according to the following inclusion criteria. The experts were required to have a minimum of five years in Infection prevention and control or Public Health, and they needed to have experience working with LTC facilities. They were asked to be able to commit one hour per week for active electronic communication with the research team for a projected minimum of four to six weeks, and they could not be from Fraser Health as this was the health authority chosen for the pilot project. Four to eight experts were sought with a minimum of two in Public Health and two in Infection prevention and control 93, 94.  Figure 2.3: Delphi methodology used with the expert group  36  The Delphi methodology was used during this phase of the validation process (See Figure 2.3) 90, 91, 94. The Delphi is an iterative process developed to enable controlled and anonymous communication between experts so that they may reach consensus on questions posed. Anonymity is maintained throughout the process to minimize the possible biasing effect of professional status and degree of reputation as an expert. In order to maintain anonymity, experts were only aware of the identity of the moderator and only communicated with this person. Each question was sent to the experts via email with the names and addresses of group members hidden. The response rates were tracked and responses from each participant were collected and used to prepare worksheets used by each group member to comment and vote on every change suggested by any of the group members for that particular question. New worksheets were prepared with a quantitative summary of the results from the previous worksheet as well as a collated set of comments made by all group members regarding the suggested change in question.  Throughout the process, the moderator was not permitted to vote on any suggested modifications. The role of the moderator was limited to sending out questions, as well as preparing and sending the worksheets to all group members. Communication from the moderator was kept transparent when responding to expert group member comments or concerns by explicitly identifying herself. In addition, when group members requested supplemental information to more fully understand a question or a worksheet, the moderator communicated with that group member by e-mail or by phone to provide the needed clarification. Detailed communication logs were kept of all individual  37  communications and the same clarifications were sent to all group members in case they had the same unspoken concerns. Every change suggested by any member of the group was included, unaltered, in a worksheet (i.e. Worksheet 1) and sent to all group members for voting and or comments so that an exchange of ideas and opinions could take place and a group consensus decision could be reached. In an analytical study of the validity and reliability of the Delphi methodology, Norman Dalkey reported that the most valid and reliable group consensus is not defined as all group members providing the same response 94. Instead, his research shows that the most reliable response is the median of the final responses from the group after an iterative discussion has taken place. For the purpose of this study where the responses are dichotomous (i.e. yes or no), the median response cannot be calculated. Therefore, based on the size of the group, ‘consensus’ was defined as a minimum of 67% (i.e. two thirds) concordance of group member opinion.  Once completed by each group member, worksheets were returned to the moderator for analysis. The moderator first calculated the percentage of group members who had voted for or against each suggested change, or had not responded to the request for a vote. All comments made by group members regarding a suggested change were then anonymized and triaged using the following procedure: 1. Suggested changes that had achieved a concordance of votes greater or equal to 67% without comments from group members were considered final decisions (for or against the suggested change) and entered into a Final Results document for that worksheet and sent to the group for review. If the  38  suggested change was supported in the final decision, the moderator made the required change to the index or survey question as appropriate. 2. Suggested changes that had achieved a concordance of votes less than 67% were revised according to the comments made by group members and entered into a subsequent worksheet (i.e. Worksheet 2). This process was repeated (i.e. a Worksheet 3, then 4 was created) until consensus was reached on whether or not to make the change to the index or survey question. 3. Suggested changes with comments achieving a concordance of votes of greater or equal to 67% were triaged in the following way: (a) If the comments were found to bring to light new information that the moderator felt may have changed the votes of other group members (had they seen the comment prior to voting), the suggested change was entered into a subsequent worksheet for a re-vote. (b) If the comments did not seem to bring forth new information, the consensus decision was considered to be a final decision (for or against the suggested change) and entered into a Final Results document for that worksheet and sent to the group for review. If the suggested change was supported in the final decision, the required change was made to the index or survey question as appropriate.  Once consensus was reached for a particular suggested change, the results were sent to the group for review, and the changes made to either the indices or survey questions (depending on the question posed to the group), the next question was sent to the  39  group of experts and the entire process was repeated. For any given question, experts were sent a copy of all documents or literature needed as background knowledge. Please see Table 2.10 for the list of questions posed to expert group members, the supporting documents provided by the moderator, and the worksheets developed to assist group members in the consensus-reaching process. At the end of the Delphi process with the expert group a second draft of the survey was created for validation by a sample of the target population in the pilot testing phase (see Appendix C).  40  Table 2.10: Questions posed to expert group and supporting documents & worksheets provided Questions 1  nd  rd  Looking at the 2 and 3 rows of each matrix, are all key infection control structural and process concepts and components listed for each index?  Supporting documents provided • Study proposal  Worksheets provided •  Worksheets 1 to 4  • Health Canada / PHAC 61 recommendations for IP&C in LTC . • Survey draft (See Appendix A) • Top four rows of each index (See Tables 2.3 to 2.8)  2  Please reflect on the infection control structures and processes already in each index: What specific infection control structures and processes are embodied in, subsumed by or map onto the 5 principles?  • Paper: “To Err is Human” pages 165 to 62 182 .  •  Question 2 and 3 Worksheet  3  What other measurable infection control structures and processes should be added to the indices to describe the five safety principles more fully?  • Paper: “To Err is Human” pages 165 to 62 182 .  • • • • • •  Question 2 and 3 Worksheet Safety Principle 1 - Worksheets 1 and 2 Safety Principle 2 - Worksheet 1 Safety Principle 3 - Worksheets 1 and 2 Safety Principle 4 - Worksheet 1 Safety Principle 5 - Worksheets 1 and 2  4  Please look at the operationalized components written in row 4 of each index: Do these describe each concept component (listed in row 3) appropriately and adequately?  • Paper: “Applying the Toyota production system: Using a patient safety alert 59 system to reduce error”  •  Worksheets 1 to 3  5  Please describe the strength of the relationship between the questions and their matching operationalized component for each index using one of the following descriptors:  • Updated survey draft (See Appendix C)  •  Worksheets 1 and 2       • Paper: “Hospital survey on patient 63 safety culture”  “S” for strong relationship “M” for moderate relationship (please explain) “W” for weak relationship (please explain).  41  2.3 Phase III: Pilot Testing Phase Once the survey was refined and validated by experts, we began the crucial phase of pilot testing and validating the tool in the field with a subsample of those who may be included in a larger survey. The first step was to transform the survey into an electronic format that had the following characteristics: 1. The format used for the survey needed to block the respondent from being able to modify either the format or the content of the survey. 2. The format needed to allow respondents to enter their answers easily electronically or manually (by printing the survey) onto the form. 3. The format needed to minimize the possibility of electronically sending responses to the wrong recipient (i.e. someone other than the research team), and it also needed to enable respondents to easily send their responses electronically to minimize the time required to do this. 4. Finally, the format needed to be compact enough to be easily sent by most e-mail servers (i.e. Less than 10 MB).  Based on these requirements, Adobe Acrobat Professional was used to transform the survey questions validated by the expert group in the previous phase of the study into a Portable Document Format (PDF) with fillable spaces. Depending on the question, respondents were able to answer with either unlimited text, numerical answers, or a selection of pre-written answers using either a drop-down box format, a radio button format (for single answer questions) or a check-box format (for multiple answer questions).  42  Ethics approvals for Phase III were obtained from the University of British Columbia Behavioural Research Ethics Board on March 6th, 2007, and from the Fraser Health Research Ethics Board on February 6th, 2007. A random sample of 20 LTC facilities stratified by facility ownership and governance structure was then selected out of the 86 LTC facilities in Fraser Health. In order to obtain a sample representative of the diversity of ownership and governance categories in Fraser Health, five facilities from each of the following four broad categories were selected: 1) Not-for profit (NP) facilities owned and operated by Fraser Health (including facilities with beds attached to an acute care hospital), 2) NP facilities that are owned and operated by an independent non-profit community or religious society and contracted by Fraser Health, 3) For profit (FP) facilities that are contracted by Fraser Health, and 4) FP facilities that are not contracted by Fraser Health. Only two facilities in Fraser Health were found to belong to a fifth category: NP facilities that are owned and operated by an independent non-profit community or religious society but not contracted by Fraser Health. These two facilities were therefore excluded resulting in 84 facilities within our sampling frame. The ownership and governance status of facilities seemed at first to be information that would be simple to locate using common directory resources such as The Red Book Directory of Services in the Lower Mainland 95. When this task was undertaken, this information was actually found to be very elusive and difficult to obtain. Directories such as the Red Book, the 2002 edition of the Canadian Health Facilities Directory 96, and the Seniors Resource Directory 95, 97-100 each only contained this type of information about some of the facilities. The licensing office was then contacted to obtain the missing information, but this office stated that this type of information was ‘confidential’ and could not be divulged to the general public. When facility websites also did not provide  43  the needed information and in order to verify the accuracy of the information found in the directories, facilities were called and asked the two following questions: “Is your facility is a Non-Profit society?” and “Is your facility contracted by Fraser Health?”. Although Not-forprofit (NP) societies were quick to answer these questions, many facilities which had been identified as For-profit (FP) using other sources refused to answer. The ownership status of the remaining facilities was confirmed by calling the Fraser Health corporate office.  The sample size was calculated based on the following literature, assumptions and sample size calculation (see Figure 2.4): •  It is unlikely that facilities from each particular ownership category have a broad variation in their IP&C structures and processes. Facilities that are owned and operated by Fraser Health have the support of the acute care IP&C programs. On the other hand, most NP facilities are contracted by Fraser Health and share a small group of LTC Infection Control Practitioners whose services are provided by Fraser Health. Finally, FP facilities are usually not contracted by Fraser Health and are unlikely to have the services of a specialized Infection Control Practitioner. A maximum variation of 30% was therefore estimated within groups.  •  According to a recent BC study by McGregor et al 12, there is an incrementally significant association between ownership and governance structures and rates of acute care admissions for pneumonia and urinary tract infections. After adjusting for a large number of confounders, the hazard ratio (HR) of pneumonia for FP ownership compared to NP facilities attached to a hospital was 2.01 (95% Confidence Interval (CI) = 1.75,2.31). This hazard ratio remained significant but  44  decreased as the FP facilities were compared to NP facilities amalgamated with a regional health authority (HR = 1.13; CI = 1.01,1.27) and NP facilities with multiple sites (HR = 1.20; CI = 1.03,1.38). The hazard ratio for urinary tract infections showed a similar trend. Since ownership and governance structures associated with lower rates of infections are also more likely to have access to a broader range of services, diagnostic tests and health care professionals, the study investigators hypothesized that larger groupings of NP facilities may provide greater capacity for the development of infection control infrastructure. In order to have a 95 percent probability of detecting a minimum difference of 25 percent between the four ownership categories, and allowing for a 20 percent probability of making a type II error (i.e. power = .80), the required sample size was calculated as 12. This was increased to 20 to account for refusals to participate and to provide a margin of error in our estimates.  Figure 2.4 Sample size calculation  45  In order to obtain a list of all Fraser Health facilities for each category, the facilities were first stratified by NP or FP status using SPLUS. The NP facilities were then subdivided into those owned and operated by Fraser Health, those contracted by Fraser Health, and those not contracted. Similarly, the NP facilities were subdivided into those contracted and those not contracted by Fraser Health. Each of the four random samples of five facilities was obtained using the program Research Randomizer obtained on the internet at www.randomizer.org/form.htm. All 20 facilities obtained for the pilot sample were amalgamated into one list and ordered by city. Each facility was then invited to participate in the pilot study via an electronic invitation letter from the investigators (See Appendix D). A web-based survey in fillable PDF format and a cover letter (See Appendix E) describing the study were then e-mailed to each facility in which the senior manager agreed to participate by completing the survey. The responses were electronically entered in the survey by the senior manager, and these were sent to the research team automatically by clicking on the ‘Submit by Email’ icon located on the first page of the electronic survey. Respondents were also given the option to print their responses so that they could keep a record for themselves and/or send their responses to the research team by regular mail. Phoned or mailed reminders were used to enhance response rates.  2.3.1 Face validity, content validity, and feasibility testing Although consensual face and content validity will were previously established with infection prevention and control experts using the Delphi process, the face validity, content validity, and the feasibility of the survey within the target population were measured in three ways. The first method 89 measures the relevance of the survey questions by using the act  46  of responding to the survey as a trigger to gather information that will assist in evaluating it. Immediately after completing the survey, when first impressions are still fresh, all respondents were asked to note their general impressions regarding the content and format of the questions (see Figure 2.5). In addition, respondents were asked the following questions to help them to articulate any concerns they had while responding to specific survey questions. The questions were created to assist respondents to categorize their concerns based on the modification they feel would best improve the survey: (1) A question deletion, (2) A question addition, and (3) A question modification.  Figure 2.5 Survey relevance questions 1)  Please give us your general impression of the content and format of the survey you have just completed.  2)  Please describe any survey questions that you feel are not relevant in evaluating the structures and processes that assist your facility in preventing and controlling infections? Please explain.  3)  Please describe any additional questions that you feel should have been added to more fully evaluate the structures and processes that assist your facility to prevent and control infections? Please explain.  4)  Please describe any questions that you feel lacked clarity and should be reworded? Please explain.  The second method of evaluating face and content validity as well as feasibility is through assessment of data completeness. As Safran writes, a good indicator of whether a survey is able to provide valuable information is in the extent to which respondents have the ability to answer the survey questions 101. The proportion of survey questions left unanswered or being answered as “I don’t know” with mean and standard deviation for all of the facilities 47  was therefore calculated. Low levels of data completeness would indicate issues with either the level of knowledge of survey respondents, perhaps suggesting that the survey should be completed by another member of the facility staff or administration, or issues with the survey content, format, relevance or clarity.  Finally, since the survey has been developed to examine the association between IP&C structures and processes and serious infections in LTC facilities, we also calculated the score distribution for the entire survey and for each index 101. This provides an estimate of the variability in the scores obtained across facilities which if reasonably variable will allow for such an association to be detected. In order to evaluate score distribution the following calculation was used:  Score calculation: A maximum of one point could be obtained for each component within a particular index. A partial point could be obtained where questions had numerous answer choices. The partial point was calculated by dividing the point by the number of answer choices with the exception of “no”, “never or rarely”, or “I don’t know”. The total raw score was then calculated by adding all of the whole or partial points. A standardized score for the index was calculated by dividing the raw score by the total number of components, and multiplying by 100. The total score was calculated by adding all of the raw scores, dividing by the total number of components in all six indices and multiplying by 100.  48  The following characteristics of score distribution were calculated: mean, standard deviation, and Tukey’s five-number summary (lowest score, 25th percentile, median, 75th percentile and highest score). A floor effect with a large proportion of respondents in the lower scores, and a ceiling effect with most in the higher scores, may show that IP&C processes are consistently either very good or very poor, but may also indicate a mismatch between the items addressed by the questions and the respondent’s ability to answer 102.  49  3. RESULTS 3.1. Phase II: Expert group validity testing phase using Delphi Recruitment of experts began on March 16, 2007 and continued until May 16, 2007 when the final member joined. Seven experts were found to meet the inclusion criteria and were enrolled into the group: •  Four Infection Prevention and Control Practitioners working with long term care facilities of four different health authorities in British Columbia  •  One Public Health Infection Control Educator  •  One Medical Health Officer working with Public Health  •  One Medical Microbiologist and Professor with extensive expertise in long term care IP&C.  As can be seen, the expert panel was appropriately multidisciplinary. All group members had worked directly with long term care facilities of all sizes to resolve real day-to-day IP&C issues. None were exclusively academic experts removed from the realities of infection prevention and control in this type of setting.  Table 3.1 shows that the Delphi methodology began on April 2, 2007 when the expert group consisted of four members (the minimum number required). Experts continued to join until May 16th, and each new member was sent all of the e-mails and supporting documents that had already been sent to the group. Since the Delphi methodology is a process to facilitate on-going discussion and consensus-reaching, new members were sent the worksheet in progress at the time of their entrance into the group. Therefore, new members were given the opportunity to vote on the recommended changes currently being discussed, and were 50  also able to make comments on any decisions made on worksheets completed prior to their entrance into the group. Since one expert dropped out of the group at approximately the same time that our seventh member joined, the expert group was comprised of six members throughout the entire voting and consensus-reaching process (which began April 18th and ended October 29th 2007). Table 3.1 also shows that Expert #1 was obliged to temporarily stop participating in the group discussion but returned on May 2nd. With these changes in membership all taking place early on in the process, Table 3.2 shows that the six experts who remained in the group throughout the process were given the opportunity to respond to a minimum of seventeen of the twenty worksheets. With the exception of Expert #7 who did not join in time to vote on the reformatted Question 1 – Worksheet 1, they were also all given the opportunity to respond to every worksheet requesting votes.  51  52  Experts #3 & 6 finish  End of Delphi 26 to 31 Expert #5 finishes  21 to 25  September  16 to 20 Expert #7 finishes  Experts #1 & 2 finish  Expert #7 joins  1 to 5  SP1W2, SP3W2, SP5W2 sent for votes 26 to 30  21 to 25  16 to 20  11 to 15  SP2W1, SP3W1, SP4W1, SP5W1 sent for votes 6 to 10  Q1W4 & SP1W1 sent for votes  26 to 31  Q1W3 sent for votes 21 to 25 Expert #4 drops out  Q2&3 W1 sent for comments 16 to 20  11 to 15  Q1W2 sent for votes 6 to 10  1 to 5  26 to 30  21 to 25  Reformatted Q1W1 sent for votes 16 to 20  May  Q5W2 sent for votes 11 to 15  6 to 10  1 to 5  26 to 30  21 to 25  16 to 20  August  11 to 15  Expert #5 joins  6 to 10  Expert #6 joins Expert #1 restarts  Temporary stop for Expert #1  Expert #4 joins  1 to 5  Q1W1 sent for comments 11 to 15  Start of Delphi - Q1 sent for comments  26 to 31  21 to 25  April  6 to 10  Q5W1 sent for votes 1 to 5  26 to 31  21 to 25  Q4W3 sent for votes 16 to 20  July  11 to 15  6 to 10  6 to 10 11 to 15  Timeline for group membership  Invitation letter sent 16 to 20 Experts #1, 2, 3 join  UBC ethics approval  Timeline for documents Day sent to experts interval  March  1 to 5  Q4W2 sent for votes 26 to 31  21 to 25  16 to 20  Month:  11 to 15  Q4W1 sent for comments 6 to 10  1 to 5  Timeline for documents Day Timeline for group sent to experts interval membership  Month:  Table 3.1: Timeline of group membership and of documents sent to experts during Phase II June  Q = Question W = Worksheet SP = Safety Principle  October  3.1.1  Individual expert response time for questions and worksheets  The time taken by individual experts to respond to worksheets from the time it was sent to them electronically is summarized in Table 3.2. Only summary data are presented in order to protect the anonymity of individual experts. Four out of the 114 worksheets were not responded to, but otherwise all group members were given the time they required to respond to each worksheet. The mean response time varied from 2.6 days (standard deviation of 3.1 days) to 12.1 days (standard deviation of 9.1 days) depending on the expert (see Table 3.2). Although the minimum response time varied minimally between experts (i.e. from 0 to 4 days), the maximum response time varied between from 7 to 40 days.  Table 3.2: Descriptive statistics for individual expert response time (in days) Expert:  #1  #2  #3  #4  #5  #6  #7  Minimum time:  0  0  1  4  1  0  0  1st quartile:  1  5  6.75  5  6.5  1  1  Mean:  8.1  11.4  8.3  5.7  12.1  3.5  2.6  Median:  6  8.5  8  6  13.5  1  1  3rd quartile:  12  20.5  11  6.5  16  2  3  Maximum time: Standard Deviation  25 7.3  23 8.0  15 3.5  7 1.5  40 9.1  18 5.8  10 3.1  Total responses  17  18  20  3  18  17  17  Maximum possible responses based on start and stop date in the group  18  20  20  3  18  18  17  3.1.2  Results for Question 1 – Infection control concepts and components  In Question 1, experts were asked to discuss the following question: ‘Looking at the 2nd and 3rd rows of each matrix, are all key infection control structural and process concepts and components listed for each index?’. This question yielded fourteen recommended changes 53  to the second and third rows of the indices. Recommended changes were therefore entered into Worksheet 1 and sent to the group for further discussion and feedback. Unfortunately, Worksheet 1 yielded very little on-going discussion. Most respondents stated that they were unclear about the process and mostly restated the same changes they had originally recommended. Based on these responses the worksheet was reformatted to better assist experts to come to a consensus regarding each suggested change (See Appendix F). All worksheets that followed throughout the process continued with this new format. The boxplots in Figure 3.1 show the distribution of time taken by experts to respond to Question 1. From this figure, we can see that the median expert response time slowly decreased with every completed worksheet submitted for this question. Experts took between 0 and 20 days to respond, often with one or two taking approximately twice as long as the others to respond.  Boxplot Key: • • • • •  The lower edge of each blue box is located at the first quartile The upper edge is at the third quartile The black horizontal line is at the median. The whiskers extending from the box reach out to the most extreme values up to 1.5 times the inter-quartile range. Any outliers beyond the whiskers would be represented by dots.  54  Figure 3.1: Distribution of time taken for Question 1 comments and votes  Leadership Index Based on Question 1, five changes were recommended for the Leadership Index. Using the iterative Delphi methodology described in Chapter 2, the following consensus decisions were made by the group and modification were made to the Leadership Index (See Table 3.3): 1.  83% voted in favor of adding ‘Senior management representation’ as a key component of ‘Infection Control Committee’. Subsequently, 83% voted in favor of changing ‘Senior management representation’ to ‘Active participation of senior management’ as a component of ‘Infection Control Committee’. 55  2.  83% voted against adding ‘Pharmacy’ as a key concept of the Leadership Index.  3.  83% voted against making modifications to the ‘Education’ component of ‘Physician’.  4.  67% voted in favor of adding ‘Infection Control reporting structure’ as a separate concept of the Leadership Index, and for adding ‘Reporting to administration’ as a component for this new concept. In a separate worksheet, 67% voted in favor of changing ‘Infection Control reporting structure’ to ‘Infection Control leadership structure’ and to add ‘Financial support’ as a key component of this re-named concept.  5.  No consensus could be reached on whether or not to add ‘Epidemiologist’ as a separate concept for this index as the group was evenly divided on this issue (i.e. 50% in favor; 50% against). One expert felt that this concept was better suited in the Partnership & Resources Index. Another stated “SENIC and others suggest a physician with expertise in infection control and epidemiology is associated with fewer infections. I know of no evidence that an epidemiologist without physician involvement is effective”. Since we are already asking whether the physician has training in epidemiology, two experts felt that the ‘Education’ component of ‘Physician’ adequately covered this concept. Since no consensus could be achieved, the decision was made to resume the discussion under the Partnership & Resources Index.  Table 3.3: Leadership Index after modifications based on Question 1 consensus decisions Leadership Index Physician Role Leader vs consultant  Time  IC leadership structure Education  Hours per One of (micro, epi, month ID, IC)  Reporting to administration  Financial support Program financed by Facility Admin &/or Health Authority  IC Committee Active participation of senior management  Degree of leadership provided Local vs regional  56  ICP Index For the ICP Index, the experts suggested that four modifications be made based on Question 1. After the discussion and voting, modifications were made to this index (See Table 3.4) based on the following consensus decisions: 1.  83% felt that ‘Certification’ in infection prevention and control should be added as an operationalized component of ‘Education’.  2.  100% of experts thought that the survey should include limited acronyms, and the acronyms kept in the survey (e.g. ICP) should be defined at the beginning of the survey.  3.  No consensus could be reached on whether or not to change ICP (i.e. Infection Control Practitioner) to the new terminology IP&CP (i.e. Infection Prevention and Control Practitioner) as the group was evenly divided on this issue (i.e. 50% in favor; 50% against). One expert felt that the survey should adhere to the new name endorsed by the Community and Hospital Infection Control Association of Canada (CHICA-Canada). Another disagreed stating that survey respondents will be senior managers who are likely to be unfamiliar with the new terminology. Since no consensus could be achieved, the decision was to make no change.  4.  67% voted to add ‘Additional duties’ as a key component of the ‘ICP’ concept.  Table 3.4: ICP Index after modifications based on Question 1 consensus decisions ICP Index Infection Control Practitioner Time Hours per 100 LTRC beds per week  Education Additional duties Experience Courses in IC, epi In years (large > 2 Certification (0, 1, >1) yrs)  57  Policy & Strategy Index No modifications were suggested for the Policy & Strategy Index based on Question 1 (See Table 3.5). Table 3.5: Unmodified Policy & Strategy Index after Question 1 Policy & Strategy Index Policies & Procedures  Development  Quality (ICP or consultant Written policy involvement) (Y/N) (Y/N)  Implementation  Product selection quality  Estimate of compliance (proportion Monitoring system of time staff are for compliance Staff education ICP participation (Y/N) (Y/N) compliant) (Y/N)  Partnership & Resources Index In response to Question 1, the expert group suggested that three modifications be made the Partnership & Resources Index. Table 3.6 shows the updated index after modifications were made based on the following consensus decisions: 1.  All experts agreed to add ‘Liaison with Occupational Health and Safety’ as a separate key component of ‘Partnership support’.  2.  67% felt that ‘Partnership with pharmacy’ should be added as a key component of ‘Partnership support’.  3.  83% agreed to add ‘Epidemiologist consultant’ as a separate key component of ‘Partnership support’.  58  Table 3.6: Partnership & Resources Index after modifications based on Question 1 consensus decisions Partnership & Resources Index Partnership support Support from IC program in Partnership Liaison other Epi. with with facility pharmacy Consultant OHS  Frequency  IT for ICP  Liaison with PH  Frequency  Computer  Access (Y/N)  Internet  Access (Y/N)  Appropriate software for surveillance  Access (Y/N)  Lab services  Secretarial support Virology  Access (Y/N)  Bacterio.  Educational material  Journals  Books  BCCDC outbreak reports  National guidelines  One of GI Access to Access or Routine & Access Access any IC to any influenza additional (days per (days per journal book on regular precautions week) week) (Y/N) IC or epi reports (Y/N)  Hand hygiene (Y/N)  Surveillance Index Two modifications were suggested and approved for the Surveillance index based on Question 1 (See Table 3.7). 1.  83% felt that ‘Trend review and corrective action’ should be added as a key component of ‘Data analysis’.  2.  83% agreed to change the component ‘Report access by stakeholders’ to ‘Infection Control report provided to stakeholders’.  59  Table 3.7: Surveillance Index after modifications based on Question 1 consensus decisions Surveillance Index Data collection  Data analysis  Simple written standardized definitions for infections  Survellance for high M&M infections  Use (Y/N)  Case finding using Types of infections: one of: Charts, pneumonia, uti, Kardex, vitals record, MRO, RI, viral lab, rounds, staff gastro, CDAD, flu, report skin inf  Monitoring of Abx use  Reporting  Trend review and Early detection of corrective action outbreaks  Report prepared  Number of unusual infections or outbreaks detected after 1 case or 2 At least one cases per year (Y/N)  Done (Y/N)  IC report provided to stakeholders Use of report for CQI given to: Staff, physicians, administration, IC committee  Report review and improvements made (Y/N)  Control Index Only one modification was suggested for the Control Index based on Question 1. All experts (100%) agreed that ‘Rapid staff reporting of suspected outbreak to Infection Control’ should be added as a component of the ‘Outbreak prevention and management’ concept. Table 3.8 shows the new index after the modification was made.  Table 3.8: Control Index after modifications based on Question 1 consensus decisions Control Index Outbreak prev. & man. Rapid staff reporting of suspected outbreak to IC Rapid control # of outbreaks detected after 2 cases  Education  Preventative care for residents  ICP  Staff  Visitors & family  Access to IC Paid educ. courses or time (0, 1 conf. (Y/N) day, >1 day)  Access (hours per year)  Avail. of material  Admission screening  TB (Y/N)  Immunization  Yearly flu (Y/N)  One time pneumo (Y/N)  Occupational transmission prevention  New staff screening  Immun. status (Y/N)  TB skin test (Y/N)  Influenza prevention Work restrictions during Yearly staff outbreaks flu vac. (Y/N) program  60  3.1.3  Results for Questions 2 – Safety concepts and components  After reading an excerpt from “To Err is Human” 62. (Pages 165 to 182) and using Question 2, experts were asked to reflect on the infection control structures and processes already in each of the six indices, and formulate a common understanding of the infection control structures and processes they felt are embodied in, subsumed by or map onto the five safety principles. The responses from experts were summarized in five tables (See Appendices G to K), each highlighting the strongest correlating components of the six indices with one of the five safety principles.  Overall, experts felt that Safety Principle 1 ‘Provide leadership’ correlated most strongly with approximately half of the components of the Leadership Index and the ICP Index, all of the Policy & Strategy Index and the Surveillance Index, and most of the Control Index. They felt that Safety Principle 2 ‘Respect human limits in process design’ currently correlated only with the ‘information technology (IT)’ concept of the Partnership & Resources Index, some of the Policy & Strategy Index, and some of the ‘data collection’ concepts of the Surveillance Index. The expert group felt that there was very little correlation between Safety Principle 3 ‘Promote effective team functioning’ and the six IP&C indices. The only correlation they identified with this safety principle was with the Policy & Strategy Index and with the ‘partnership support’ concept of the Partnership & Resources Index. Most of the indices were thought to already have concepts that strongly correlated with Safety Principle 4 ‘Anticipate the unexpected’. The specific indices that mapped well onto this principle were as follows: the ‘Infection Control committee’ concept of the Leadership Index, the ‘Information Technology’ and  61  ‘laboratory services’ of the Partnership & Resources Index, the ‘outbreak prevention and management’ and ‘preventative care for residents’ concepts of the Control Index, the Policy & Strategy Index and Surveillance Index. Finally, the expert group found that, with the exception of the ICP Index, Safety Principle 5 ‘Create a learning environment’ strongly correlated with approximately half of the key concepts of the ICP Index, the Policy & Strategy Index, the Surveillance Index, and the Control Index.  3.1.4  Results for Questions 3 – Safety concepts and components  Once the expert group had developed a common conceptualization of how the five safety principles are embodied in the infection control structures and processes described in the six indices, the group was given the following Question #3 to ponder: ‘What other measurable infection control structures and processes should be added to the indices to describe the five safety principles more fully?’. This question yielded fifty recommended changes to the indices so that key safety concepts could be incorporated. Table 3.9 shows the six indices after modifications had been made based on the consensus decisions for this question.  62  Table 3.9: The six indices after modifications based on Question 3 consensus decisions. Red boxes represent added, moved or modified concepts (Row 2), components (Row 3), or operationalized components (Row 4). Leadership Index Physician  Role  Time  IC leadership structure Reporting to administration  Education  Clear assignment of IC oversight  One of In res Leader vs Hours per (micro, epi, safety consultant month ID, IC) initiatives  IC Committee  Financial support  Active participation Knowledge & Degree of of senior understanding of leadership management res safety initiatives provided  Program financed by Facility Admin &/or Health Authority  Local vs regional  ICP Index ICP Time Hours per 100 LTRC beds per week  Education  Certification  Within facility partnership Additional duties  Courses in IC, epi (0, 1, >1)  Experience  Housekeeping  Liaison with OHS  Resident / family partnership Nursing  Resident / family / staff meetings  In years (large > 2 yrs)  Policy & Strategy Index Policies & Procedures  Emergency PreparednesStrategy  Safety Culture  Development  Implementation  Product selection quality  Quality (ICP Written Brief, clear, Regularly or consultant policy concise scheduled involvement) Staff (Y/N) language Checklists participation reviews (Y/N)  Readily available in work area  ICP participation (Y/N)  IC Free promotion communication by Admin flow within facility Initiative supported by Admin  Blameless reporting of errors  Clear IC G&O  Plan for surge capacity  Pandemic / Emergency plan  Initiative supported by Admin  63  Partnership & Resources Index External Partnership support  IT for ICP  Lab services  Support from Partnership Appropriate IC program in with software for Secretarial Epi. Partnership other facility pharmacy Consultant with PH Computer Internet surveillance support Virology  Frequency of contact  Frequency  Access (Y/N)  Access (Y/N) Access (Y/N)  Access (Y/N)  Bacterio.  Access Access (days per (days per week) week)  Journals  Educational material  Budget  BCCDC outbreak reports  Dedicated to IC  Books  National guidelines  Access to Access to One of GI or Routine & any IC any book influenza additional Hand journal on IC or regular precautions hygiene (Y/N) epi reports (Y/N) (Y/N)  Surveillance Index Data collection Simple written standardized definitions for infections  Use (Y/N)  Data analysis  Surveillance for high M&M infections Types of Case finding infections: using one of: pneumonia, Charts, Kardex, UTI, MRO, RI, vitals record, viral gastro, lab, rounds, CDAD, flu, skin staff report inf  Quick & easy access to accurate Monitoring of data Abx use  Trend review and corrective action  Early detection of outbreaks  Reporting Report prepared  IC report provided to stakeholders  Use of report for CQI  Number of unusual infections or Given to: Staff, outbreaks At least one physicians, Report review and detected after 1 per year administration, IC improvements case or 2 cases (Y/N) made (Y/N) committee  Done (Y/N)  Control Index Outbreak prev. & man. Rapid staff reporting of suspected outbreak to IC  Rapid control # of outbreaks detected after 2 cases  Education  ICP  Staff  Preventative care for residents  Visitors Admission & family screening  Access Paid ICP to IC educ. involvement Materials Access courses time (0, 1 in new staff Access used are (hours or conf. day, >1 IC to Avail. of approved per (Y/N) day) orientation materials by IC year) material TB (Y/N)  Immunization  Occupational transmission prevention  New staff screening  Error analysis  Monitoring Estimate of system for Regular Influenza prevention compliance compliance IC audits  Work restrictions Yearly One Yearly time Immun. TB skin during staff flu flu test outbreaks pneumo status vac. (Y/N) (Y/N) (Y/N) (Y/N) (Y/N) program  64  In order to maintain a focus on each safety principle individually, separate worksheets were created each covering recommended changes for all six of the indices based on one of the five safety principles. The box plots in Figure 3.2 show that experts took approximately 10 days to respond to the first set of safety principle worksheets and only four days for the second set of worksheets for safety principles 1, 2, and 5. Consensus was reached with the first worksheets on all recommended changes for safety principles 2 and 4, therefore no second worksheets were sent. Experts took between zero and 23 days to respond to the safety principle worksheets for Question 3.  Figure 3.2: Distribution of time taken for Questions 2 & 3 comments and votes  65  Safety Principle 1: Provide leadership The expert group recommended that 18 modifications be made to the six indices to ensure that the following key safety concepts were being measured adequately in the survey: • Make patient safety a priority corporate objective • Make patient safety everyone’s responsibility • Make clear assignments for and expectation of safety oversight • Provide human and financial resources for error analysis and systems redesign • Develop effective mechanisms for identifying and dealing with unsafe practitioners  Leadership Index 1. 83% voted in favor of adding ‘Clear assignment of infection control oversight’ to the concept of ‘Infection Control leadership structure’. 2. 83% voted in favor of adding ‘Knowledge and understanding of resident safety initiatives’ to the concept of ‘Infection Control committee’. The expert who rejected this addition explained that they are not aware of how this knowledge could help with the work of this type of committee. 3. 67% voted in favor of adding education ‘in resident safety initiatives’ to the physician ‘Education’ component.  ICP Index 4. 83% voted in favor of adding ‘Budget’ as a key concept. 5. 83% voted in favor of adding ‘Dedicated to the Infection Control program’ as a key component of ‘Budget’ 66  Policy & Strategy Index 6. 83% agreed that policy reviews were necessary, but they did not all feel that these have to be done on a yearly basis. ‘Regularly scheduled reviews’ was therefore added to the policy and procedure ‘Development’ component. In addition, all experts voted to move ‘Staff education’, ‘Estimate of compliance’ and ‘Monitoring system for compliance’ from the ‘Implementation’ component to the Control Index. 7. 67% felt that we should add ‘Safety culture’ as an additional key concept of this index. One vote against was based on the concern that such a concept was difficult to measure and lacked evidence of effectiveness. 8. 67% voted to add ‘Infection Control promotion by administration’ as a component of the new ‘Safety culture’ concept. Two experts who rejected this addition were concerned about measurability; they felt that budget and membership in the infection control committee were the only valid ways to measure this component, and these items had already been included in the indices. 9. 83% agreed to add ‘Clear infection control goals and objectives’ as a component of ‘Safety culture’.  Partnership & Resources Index 10. 83% voted in favor of changing the ‘Liaison with Public Health’ component to ‘Partnership with Public Health’.  67  Control Index 11. The group was evenly split on whether or not to add ‘Infection prevention and control as a staff priority’ as a component of ‘Occupational transmission prevention’. Those against the addition of this component argued that it was already covered by the ‘Safety culture’ concept in the Policy & Strategy Index. They further stated that this concept was ‘nebulous’, difficult to define and even more difficult to measure. Since no consensus could be reached, this component was dropped. 12. The group was evenly split on whether or not to add ‘Error prevention’ as a key concept of this index. One expert rejected the addition stating that error prevention had not been shown to be effective in the long term care setting to prevent and control the spread of infection. Since no consensus could be reached and other similar concepts had been proposed and agreed upon for other safety principles, this concept was dropped. 13. , 14. & 15. The group was evenly split on all three of the following additions to the component of ‘Error prevention’: 1) ‘Immunization record-keeping errors’, 2) ‘Error analysis’, and 3) ‘Plan of action for non-compliance’. Experts who rejected these explained that they were concerned with the lack of staff to implement these initiatives, and with the lack of evidence to support their effectiveness. They also felt that the development of a plan for non-compliance would be seen as punitive by staff. Since the concept ‘Error prevention’ had already been dropped, all three of these suggestions were also dropped.  68  16. All group members rejected the addition of ‘Identification system for unsafe practitioners’ to the (rejected) concept ‘Error prevention’. They feared that there would not be enough staff to implement this, and felt that this would be a quality improvement initiative rather than an IP&C initiative. 17. 83% voted to add ‘Error analysis’ as a key concept of this index. The expert who rejected this addition felt that this level of monitoring was not applicable to the long term care setting. 18. All agreed to add ‘Regular infection control audits’ as a key component of ‘Error analysis’.  Safety Principle 2: Respect human limits in process design The expert group recommended that four modifications be made to the six indices to ensure that the following key safety concepts were being measured adequately in the survey: •  Design jobs for safety  •  Avoid reliance on memory  •  Use constraints and forcing functions  •  Avoid reliance on vigilance  •  Simplify key processes  •  Standardize work processes  69  Policy & Strategy Index 1. All group members agreed to operationalize the policy and procedure ‘Development’ component by using the following terms ‘Brief, clear, concise language’. 2. All group members agreed to also operationalize the policy and procedure ‘Development’ component by the use of ‘Checklists’. 3. All group members agreed to operationalize the policy and procedure ‘Implementation’ by using the term ‘Readily available in work area’.  Control Index 4. 83% rejected to add ‘Regular safety audits’ (as opposed to infection control audits proposed and approved for Safety Principle 1) as an operationalization of the component ‘Identification system for unsafe practitioners’ (this component was rejected in the Safety Principle 1 worksheets). Experts felt that this type of audit was more related to a safety program rather than an infection control program.  Safety Principle 3: Promote effective team functioning The expert group recommended that 14 modifications be made to the six indices to ensure that the following key safety concepts were being measured adequately in the survey:  70  •  Train in teams those who are expected to work in teams  •  Include the patient in safety design and the process of care  Leadership Index 1. 67% rejected ‘Resident / family representation’ as a component of ‘Infection Control committee’. One expert was concerned with the lack of evidence to support that this would decrease rates of infection. Another felt that it would not be appropriate to include families in this type of committee, but that infection control issues could be discussed at family meetings. 2. The group was evenly split on whether to add the component ‘Team-building activities’ to the ‘Infection Control committee’ concept. Those who rejected this addition felt that although team-building may have benefits, it would not be a key activity of the committee.  ICP Index 3. 67% felt that we should move the ‘Budget’ concept to the Partnership & Resources Index. 4. 83% voted in favor of adding ‘Within facility partnership’ as a key concept of the ICP Index. This concept would then contain the transdisciplinary partnerships that must be maintained within a facility to prevent and control infections. 5. All experts agreed that ‘Housekeeping’ must be included as a component of the new ‘Within-facility partnership’ concept.  71  6. 83% felt that ‘Nursing’ must be included as a component of the new ‘Within facility partnership’ concept. 7. The group was evenly split on including ‘Ward-level infection control champions’ as an operationalized component of the ‘Nursing’ component. One expert felt that there was no evidence to support the inclusion of this component. Another stated that this was a volunteer role for nursing staff and could not be included because it is not a remunerated role. 8. 67% voted in favor of adding ‘Resident / family partnership’ as a third key concept of this index. 9. & 10. 67% felt that we should have ‘Resident / family / staff meetings’ as a key component of the new ‘Resident / family partnership’ concept, but the group was evenly split on stating that these meetings should have a ‘Standing infection prevention and control agenda’. 11. 67% felt that the component ‘Liaison with Occupational Health and Safety’ should be moved from the Partnership & Resources Index to the ICP Index. 12. The group was evenly split on various attempts to change the name of this index to one that would highlight the inclusion of all stakeholders within the facility who work together to prevent and control infections (e.g. People Index or WithinFacility People Index).  Policy & Strategy Index 13. All group members agreed that ‘Staff participation’ was an important way to operationalize the policy and procedure ‘Development’ component.  72  Partnership & Resources Index 14. Since within-facility partnerships had been moved to the ICP Index, 67% voted in favor of changing the concept ‘Partnership support’ to ‘External partnership support’.  Safety Principle 4: Anticipate the unexpected The expert group recommended that six modifications be made to the six indices to ensure that the following key safety concepts were being measured adequately in the survey: • Adopt a proactive approach: examine processes of care for threats to safety, and redesign them before accidents occur • Design for recovery • Improve access to accurate, timely information  Policy & Strategy Index 1. All experts felt that ‘Emergency preparedness strategy’ was a key concept for this index. 2. 67% voted to include ‘Plan for surge capacity’ as a component of the new ‘Emergency preparedness strategy’ concept. 3. 83% felt that ‘Pandemic / Emergency plan’ should also be included as a key component of the new ‘Emergency preparedness strategy’ concept. One expert felt that emergency planning should not be included. 73  Surveillance Index 4. 67% wanted to add ‘Quick and easy access to accurate data’ as a key component of ‘Data collection’. Those who voted against this addition were concerned about feasibility for most facilities and measurability in the survey.  Control Index 5. & 6. 67% voted to add “Orientation of new staff to IC” as a key concept of “Occupational transmission prevention”, and experts unanimously agreed that ‘ICP involvement’ was necessary in the orientation process.  Safety Principle 5: Create a learning environment The expert group recommended that eight modifications be made to the six indices to ensure that the following key safety concepts were being measured adequately in the survey:  •  Use simulations whenever possible  •  Encourage reporting (of errors and hazardous conditions)  •  Ensure no reprisals for reporting of errors  •  Develop a working culture in which communication flows freely regardless of authority gradient  •  Implement mechanisms of feedback and learning from error  •  Standardize work processes  74  Policy & Strategy Index 1. 83% agreed that ‘Free communication flow within the facility’ should be added to the key concept ‘Safety cultures’. 2. 67% felt that the new component ‘Free communication flow within the facility’ should be operationalized by ‘Initiative supported by administration’. One expert voted against because of the difficulty in measuring this component. 3. 83% agreed to add ‘Blameless reporting of errors’ as a component of ‘Safety cultures’. 4. 67% felt that the new component ‘Blameless reporting of errors’ should also be operationalized by ‘Initiative supported by administration’. 5. The group was evenly split on whether or not to add ‘Error management policies or algorithms’ to operationalize the component ‘Blameless reporting of errors’. One expert rejected this inclusion stating that this issue was exclusively related to a ‘safety program’ and not an infection control program.  Partnership & Resources Index 6. The group was evenly split on whether or not to add facility ‘License’ as a component of either this index or the Leadership Index. Experts could not form an agreement on how being licensed would affect infection prevention and control within facilities.  75  Control Index 7. 83% voted to include ‘Access to educational material’ to operationalize ‘Staff’ education. 8. 67% voted to include ‘Educational material approved by infection control’ to operationalize ‘Staff’ education. One expert who voted against stated that facilities who do not have trained ICPs would not have the option of obtaining this approval. Another felt that the term ‘approved’ was too restrictive; they felt that infection control should participate in the development of the material rather than just approve it.  3.1.5  Results for Questions 4 – Operationalization of components  When Question 4 was introduced, the following tasks had already been completed: (Question 1) Consensus had been reached on all of the key concepts in the second rows and the components in the third rows of the six indices, and (Questions 2 and 3) patient safety concepts had been integrated into the six indices. Question 4 thus asked experts to look at the operationalized components written in the fourth row of each index, and to decide whether each component was described appropriately and adequately. This question yielded 18 recommended modifications to the fourth row of the indices. Once a consensus had been reached on each of the suggested modifications, changes were made to the indices. The revised fourth row of each index can be seen in Appendix L. Figure 3.3 shows a different response time pattern for this question when compared to previous questions. Instead of taking less time with each  76  worksheet, the mean response time increased from 4.2 days with the first worksheet, to 6.6 days with the second, and finally 7.7 days with the third worksheet.  Figure 3.3: Distribution of time taken for Question 4 comments & votes  High morbidity and mortality infections (Surveillance Index) Question 4 ignited a debate on the types of infections that should be included in the Surveillance Index as high morbidity and mortality infections for which experts felt there should be a system of surveillance in long term care facilities. The infections presented to the group for discussion was an exhaustive list of the eight types of infections that 77  commonly cause morbidity in long term care facilities 14, 16, 18, 21, 22, 24, 26, 41, 103. Experts were asked to state which infections they felt caused high morbidity as well as mortality in long term care, and should therefore be used in the survey when asking questions about surveillance. The following votes were obtained: •  67% selected urinary tract infections  •  67% selected pneumonia  •  100% selected influenza  •  67% selected skin and soft tissue infections  •  100% selected Clostridium difficile-associated diarrhea  •  67% selected viral gastroenteritis  •  83% selected antibiotic resistant organisms  •  50% selected tuberculosis  All agreed that each type of infection is known to commonly cause morbidity in the long term care setting. However, some group members debated that urinary tract infections and skin and soft tissue infections do not cause high levels of mortality. Given these results, and in order to only include key representative infections that cause high morbidity and mortality in long term care, only infections selected by at least 83% of experts were included.  78  Types of outbreaks (Surveillance Index and Control Index) There was also a long discussion on the best method of obtaining information on whether or not facilities are achieving ‘Early detection of outbreaks’ (measured in the Surveillance Index), and ‘Rapid control of outbreaks’ (measured in the Control Index). The following critical points led to the final selection of the type of outbreak to use: •  All facilities are required to report both influenza and gastroenteritis outbreaks to Public Health. Either of these two categories of infections were therefore ideal choices because all facilities must use methods of surveillance to detect these two type of outbreaks.  •  Gastroenteritis and influenza have varying incubation periods and different definitions for outbreaks. Combining numerous types of outbreaks into one question regarding rapid detection would not be valid; only one kind of infection needed to be selected. Influenza was therefore selected.  •  Outbreaks of respiratory infection (a.k.a. influenza-like-illnesses), are often reported before influenza is identified as the causative organism based on the definition for respiratory outbreak. Although more control measures are used to quickly halt the spread of an influenza outbreak, all respiratory outbreaks can cause high levels of morbidity and mortality, and they are usually reported and stopped in a similar fashion. The decision was therefore to include all respiratory outbreaks instead of just influenza.  79  Measurement issues for outbreaks (Surveillance Index and Control Index) After experts had agreed to use respiratory outbreaks to gather information on whether or not facilities are achieving ‘Early detection of outbreaks’, and ‘Rapid control of outbreaks’, other concerns became evident. The group felt it was unlikely that facilities would readily be able to provide information on respiratory outbreaks their facility has had over the past two years in any valid way. Based on this feedback, a decision was made that survey users will have to request and hopefully obtain this reportable information from regional health authorities where these data are collected.  3.1.6  Results for Questions 5 – Question validation and development  Once all of the concept components had been operationalized in Question 4, the moderator ensured that each of the operationalized components had a matching question in the far left-hand column of the index matrices exactly as it had been done in Phase I (See Table 2.9 in Chapter 2). For the first worksheet of Question 5, experts were asked to use the six index matrices sent to them in Excel format to describe the strength of the relationship between the questions and their matching operationalized components using ‘S’ for a strong relationship, ‘M’ for a moderate relationship, and ‘W’ for a weak relationship. Experts were also asked to suggest how moderate or weak relationships could be maximized. Figure 3.4 shows that Question 5 - Worksheet 1 required more time than all other worksheets used during Phase II. It shows that experts required between 10 and 40 days to respond with an average of 20.8 days. One expert did not respond explaining she did not understand how to use the Excel worksheet and  80  did not have time to learn. Two other experts needed to be individually helped by the moderator to understand both how they needed to respond and how to use the Excel worksheet. Figure 3.4: Distribution of time taken for Question 5 votes  Although questions were only accepted into the survey after at least 67% of experts felt that the question had a ‘Strong’ relationship with the operationalized component, all expert comments were taken into consideration to improve the questions. Experts felt that nine of the 61 questions needed to be improved (i.e. The questions scored less than 67% ‘S’). These questions were entered into a second worksheet, each with a proposed new question developed based on the feedback. The new round of votes 81  resulted in all questions being accepted by consensus, with four questions obtaining 100% of votes as ‘Strong’, two obtaining 83% of votes as ‘Strong’, and three obtaining 67% of votes as ‘Strong’.  3.2. Phase III: Pilot Testing Phase 3.2.1. Sample participation On November 19th, 2007, an invitation letter (See Appendix D) was sent to 20 facilities which had been randomly selected from the four broad ownership and governance strata of 84 Fraser Health facilities. Out of the 20 facilities, only two FP facilities not contracted by Fraser Health (Stratum #4) immediately refused to participate without asking any further questions (See Table 3.10). One did not provide a reason for the refusal and the other explained that they did not have time to complete a survey. Of the 18 facilities that agreed to participate, one facility from each of the four categories subsequently withdrew from the study resulting in a response rate of 70%. Facility senior managers who refused explained that they either did not have the time to complete the survey due to high workloads and multiple roles within the facility, or they had reviewed the survey with their supervisors and had decided that they no longer wished to participate. No further discussion was pursued regarding these explanations as doing so may have been interpreted as pressure for participation.  82  Table 3.10: Participation of invited facilities in the pilot study Stratum #1  Stratum #2  Stratum #3  Stratum #4  Total  # Fraser Health facilities  14  27  32  11  84  # invited facilities  5  5  5  5  20  # accepted invitations  5  5  5  3  18  # completed surveys  4  4  4  2  14  Total refusals & drop-outs  1  1  1  3  6  3.2.2. Descriptive statistics In order to maintain the anonymity of the facilities participating in the study, only summary data are reported. While the invited facilities were located in 12 out of the 15 cities/municipalities within Fraser Health, the 14 participating facilities were located in only nine. All surveys were completed by one of the following categories of senior managers: 1) Resident Care Managers - 5 facilities, 2) Coordinators - 3 facilities, or 3) Directors - 6 facilities. Only two facility senior managers requested the assistance of another person to complete the survey, and both of these requested assistance from the Fraser Health ICP who works with their facility.  Figure 3.5 shows that there was a broad distribution of facility size within the portion of the sample that responded to the survey. The number of beds ranged from 11 to 300, with a mean of 108 beds (Standard Deviation = 65.2).  83  Figure 3.5: Distribution of facility size based on number of beds  Facilities within the sample also varied broadly on the number and type of front-line staff working within their facility (See Table 3.11). Further analysis using scatterplots showed that there seems to be only a weaker than expected correlation between the number of beds and the number of Registered Nurses (RN), Licensed Practical Nurses (LPN), Care Aides, RNs plus LPNs, and all front-line staff within the facilities (See Appendix M).  Table 3.11: Descriptive statistics of front line staff within sample facilities Number of Registered Nurses  Number of Licensed Practical Nurses  Number of Care Aides  Minimum  0  0  3  1st Quartile  5.0  2.0  24.0  Mean  9.4  7.0  38.2  Median  10.0  6.8  31.0  3rd Quartile  12.0  13.3  50.0  Maximum  18  15  123  Standard Deviation  5.2  5.7  29.5  84  Only one of the facilities within the sample had an admission policy requiring residents to have a ‘Do Not Resuscitate’ order for admission. Thus, this facility may have a larger ‘high-risk’ population. This type of requirement could also affect whether residents from this facility are aggressively treated or transferred to an acute care hospital for infections. Finally, the length of time that a formal IP&C program has been in existence was defined as the earliest of either of the two following events: the year a handhygiene policy was established or the year an ICP position was created. Two facility senior managers did not report that either indicators that an IP&C program had been established within their facility. The earliest date reported was 1990 and the latest (reported by three facilities) was 2007. The mean year for the establishment of an IP&C program was 2001 (Standard Deviation = 6 years).  3.2.3. Validity and feasibility test #1: Question Relevance In order to evaluate the relevance of the survey questions for facility senior managers, the following four questions were posed at the end of the survey. Following each question is a summary of the feedback provided by respondents.  1)  Please give us your general impression of the content and format of the survey you have just completed. Nine respondents sent feedback for this question. Eight of these described the survey using one of the following terms: “easy to complete”, “generally good”,  85  “easy to follow”, “very good”, “comprehensive and relevant”, “thorough”, “pretty clear and easy to follow”. A few suggestions were also made to enhance the clarity of particular questions. Words such as “staff” and “educational time” were found to be difficult to interpret. In addition, one respondent found it frustrating that the number of direct-care staff did not include part-timers. Another senior manager stated that she found it difficult to complete the survey as the role of the ICP is shared by the Director, Manager and LPN of their small facility. One respondent thought that the regional ICP would be a better choice for responding to the survey as their assistance was required. Finally, two were frustrated that the survey questions did not allow for “not applicable” or “no answer” as answer choices, but rather only provided the choice to respond “I don’t know” if none of the other answers were appropriate.  2)  Please describe any survey questions that you feel are not relevant in evaluating the structures and processes that assist your facility in preventing and controlling infections? Please explain. Five responses were sent for this question. Two facility senior managers stated that all questions were relevant and important. Two respondents felt that some questions were inapplicable to them either because of their small facility size or because they do not have an ICP, and another felt that some of the questions should be answered by Occupational Health and Safety (but they did not ask for any assistance to answer the questions and simply responded “I don’t know”).  86  Although one respondent was the designated ICP, she stated that she did not feel knowledgeable enough to answer many of the questions.  3)  Please describe any additional questions that you feel should have been added to more fully evaluate the structures and processes that assist your facility to prevent and control infections? Please explain. Recommended additions from six senior managers were as follows: •  Request information on hours of coverage for ICPs. Facilities covered by regional ICPs do not have many hours of coverage per month.  •  Develop separate surveys for large, medium and small facilities  •  Request information on last revision for hand hygiene policy to ensure that facilities are remaining current with best practice  •  Ask about past and present infection rates.  •  Request information about N95 fit-testing.  •  Ask information about outbreak management.  •  Investigate the facilities’ multidisciplinary approach in dealing with infection.  •  Ask about room set up on the unit for staff to take their break during quarantine periods (Meaning not explained).  •  Ask questions about Microbiologist coverage.  87  4)  Please describe any questions that you feel lacked clarity and should be reworded? Please explain. Seven respondents requested the following clarifications: •  How do you answer section 2 when the ICP is the Care Director?  •  How do you answer section 2 when the ICP role is shared by multiple people and the IC physician is fulfilled by the General Practitioners caring for the residents of the facility?  •  In section 2, are the questions referring to a person within the facility or from the health authority?  •  Question 10 stated that if you answer "no" to skip to question 14, however it should have said to skip to question 16 as questions 14 and 15 still referred to the IC physician and ICP.  •  In section 3, is the precaution sign on the door considered as education material already or the education material needs to be in a pamphlet form?  3.2.4. Validity and feasibility test #2: Data Completeness The second method of evaluating face and content validity as well as feasibility was through an assessment of data completeness. The survey is only able to provide investigators with valuable information if respondents are able to answer the questions of the survey. Table 3.12 shows that, on average, fewer than nine percent of survey questions either had no response of received the answer “I don’t know”. 88  Table 3.12: Mean and Standard Deviation (SD) for percentage of missing data in the survey. No response provided  “I don’t know” response selected  Total missing data  Mean  2.9 %  5.8 %  8.7 %  SD  5.8 %  7.9 %  9.0 %  3.2.5. Validity and feasibility test #3: Score Distribution This survey is being developed and validated so that it can be used on a large scale to examine the association between IP&C structures and processes as measured by the six indices and serious infections in LTC facilities. In order to be able to evaluate such an association (or lack of), the survey needs to be able to detect the variability in IP&C structures and processes between facilities. Table 3.13 describes the distribution of the standardized total scores as well as the standardized individual scores for all survey respondents. Figure 3.6 provides a visual depiction of Tukey’s five number summary using boxplots. Based on the table and figure, we can see that the survey was able to detect variability in all scores. The smallest amount of variability was found in Control Index score, which had a standard deviation of 15.7%. The survey was also able to detect lower levels of variability in the Surveillance Index (Inter-quartile range (IQR) = 17.8%), and Policy & Strategy Index (IQR = 17.6%), and more variability in the ICP Index (IQR = 44.9%), and Partnership & Resources Index (IQR = 39.9%).  89  Table 3.13: Descriptive statistics of standardized scores for entire surevey and for individual indices Total score (%)  Leadership Index (%)  ICP Index (%)  Partnership & Resources Index (%) 59.7  Surveillance Index (%)  Control Index (%)  45.5  Policy & Strategy Index (%) 71.0  Mean  63.1  47.4  73.4  69.7  Standard Deviation (SD) Minimum score st 1 Quartile  16.1  22.5  29.7  16.4  21.5  16.8  15.7  38.4  17.5  0.0  35.8  30.0  31.7  49.6  50.7  30.0  19.0  62.2  36.6  66.2  56.5  Median  60.0  45.3  45.4  69.8  68.9  73.8  67.9  3 Quartile  78.3  56.3  63.9  79.8  76.5  84.0  79.5  Maximum score  87.1  90.0  90.8  97.1  85.0  98.3  97.9  Inter-quartile range (IQR)  27.6  26.3  44.9  17.6  39.9  17.8  23.0  rd  Figure 3.6: Boxplots describing distribution of standardized total scores and index scores.  90  Boxplot Key: • • • • •  The lower edge of each blue box is located at the first quartile The upper edge is at the third quartile The black horizontal line is at the median. The whiskers extending from the box reach out to the most extreme values up to 1.5 times the inter-quartile range. Any outliers beyond the whiskers would be represented by dots.  91  4. DISCUSSION 4.1 Study summary Long term care facilities are often perceived as either small sub-acute hospitals where elderly people go to live when they require assistance to care for themselves, or a place to which elderly people move when they are approaching death. These assumptions have made the importance of IP&C unclear in this setting, and have, perhaps subconsciously, led decision-makers to focus IP&C efforts mostly in acute care hospitals of British Columbia. In reality, those living in LTC are some of our oldest and most vulnerable to infections, and they expect and should receive safe, high quality care within their facility. The problem decision-makers now face is how to enhance IP&C in long term care when so little research has been done to elucidate what is already in place and what works using research tools specifically designed for this setting.  This study used an innovative three-phase methodology to develop a valid tool that can be used to measure IP&C structures and processes in the LTC setting so that their effectiveness can then be evaluated based on key IP&C outcomes. Phase I of this study involved designing a new conceptual model for IP&C that highlighted the close interplay between IP&C structures that drive care processes, and in turn affect resident outcomes. It also reflects the iterative process of innovation and learning that is necessary to maintaining an effective program that continuously improves over time. Using the 2004 Health Canada / PHAC recommendations for IP&C in LTC, and after drawing together the information found in current scientific literature and other expert body advisory reports, this new IP&C model was then used to outline the key concepts  92  and components for the four structural and two process composite indices. These six indices were then used to develop a literature-based survey draft.  In Phase II, the Delphi methodology was effectively used to discuss, further develop and validate the six indices and survey questions by a group of experts in LTC infection prevention and control. The Delphi enabled the twenty iterations of discussion to take place in a controlled and anonymous setting to minimize the biasing effects of perceived status, strong personalities, and the fear of loosing face or appearing foolish. By responding to 110 out of a possible 114 worksheets for discussion and voting, this process generated 82 suggested modifications to the six indices. Phase II of this study therefore shows how the Delphi methodology can be successfully used to develop and validate a survey based on expert opinion. Because of the paucity of research in the field of infection prevention and control in the LTC setting, this method of inquiry was particularly beneficial as it enabled the researcher to enhance and validate the literature-based survey developed in Phase I using experts in the field. Moreover, since the incorporation of safety principles in IP&C research has never been explicitly done in the past, the survey developed in this study could become the initiating component in the development of a new understanding of how infections may be prevented and outbreaks may be controlled in health care facilities.  In Phase III of this study the expert validated survey was pilot tested using a random sample of facilities in Fraser Health. The pilot study obtained a response-rate of 70% and facility Senior Managers were able to complete at least 80% of survey questions. 93  This final step in the validation process thus demonstrated that large-scale use of the survey would be feasible. Further analyses also showed that the data collected are reliable, and validly measure the key IP&C concepts and components contained in the six indices. The end result is a validated survey to assess IP&C in the LTC setting. The following discussion describes observations, potential advantages and pitfalls noted during the survey development and validation process.  4.2 Phase II – Delphi expert-group validation 4.2.1 Question 1 – Key IP&C concepts and components The Delphi methodology used in Phase II of the study was an effective way to further develop and validate a survey based on expert consensus. The first question sent to the group asked them to draw upon their expertise in LTC infection prevention and control to ensure that all IP&C concepts and components were described in the six indices. All experts in the group felt that the relationship of the administrative or senior management staff to those responsible for IP&C is crucial in determining the effectiveness of the program, and must therefore be clearly measured in the survey. An IP&C program without adequate administrative support is likely more or less powerless in dealing with IP&C issues, and thus is perhaps more symbolic than effective. This is concern supported by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) 84 which stated that “Only with the ongoing attention and direction of organizational leadership can the appropriate scope of the program be determined and adequately resourced” (P.2). This requirement is echoed by the Institute of Medicine in their report “To Err is Human” 62. Under the first safety principle (i.e. Provide leadership 94  – See Table 1.1), they explain that governing boards, management and clinical leadership must not only explicitly state their safety-related organizational goals, but also provide “…regular, close oversight of the safety of the institution they shepherd” (p.166). Group members felt that three components were key in measuring the commitment of facility administrators and senior managers to IP&C. There should be direct reporting of issues to facility administrators; echoing the above statement from JCAHO, experts also felt that a strong commitment to IP&C initiatives should translate into financial support from facility administration; they felt that the active participation of senior management in the infection control committees would make them much more effective in dealing with issues that arise and proactively preventing problems from occurring.  Question 1 also began a long discussion and heated debate on whether an Epidemiologist was needed as a distinct component of the Leadership Index. Those in favor based their opinion on the fact that many physician leaders do not have the skills needed to design a surveillance program, analyze surveillance data and report findings in an effective way, thus greater expertise was needed. Those against reasoned that the SENIC study 43 (the only major study designed to evaluate the effectiveness of IP&C structures and processes, and now 20 years old) provided sufficient evidence that a physician leader with IP&C related education had a major impact on program effectiveness. In the end, the group came to the consensus that support from an Epidemiologist consultant was important, interestingly not as part of the leadership structure of the IP&C program where the role may have had the most impact, but as an  95  important partnership in the Partnership & Resources Index where the impact may be diminished.  For the Infection Control Practitioner (ICP) Index, the group spent considerable time debating the need to include professional certification as a key component of ICP education. The Certification in Infection Control (CIC) is obtained through testing by the Certification Board of Infection Control and Epidemiology (Inc.) in the United States. It is an internationally recognized certification for Registered Nurses and Physicians requiring two years (minimum 800 hours) of experience working in an IP&C program within the most current five-year period 104. Since years of experience and education in IP&C were already being measured in the survey, the moderator voiced a concern that this component did not seem to measure any new aspects of ICP competence. Because of the requirements for certification, this component indirectly measures both ICP knowledge in infection prevention and control and experience in this field. Since education and experience in infection prevention and control were already being measured directly in the ICP Index, the inclusion of certification seemed redundant. However, since the support for the inclusion of this component was very strong from the expert group (i.e. 83%), it was added to the ICP Index.  Two additional interdisciplinary relationships (in the Partnership & Resources Index) were felt to be crucial to having an effective IP&C program in any facility: a liaison with the person in charge of Occupational Health and Safety (OHS), and a partnership with pharmacy. While the mandate of the IP&C program is to prevent and control  96  transmission of infectious organisms within the resident population of the facility, the objective of the OHS program is to prevent transmission within the employee population. Since there is strong evidence that outbreaks often result from transmission between health care professionals and facility residents 105, 106, experts felt that it is crucial to have clear communication between those responsible for these programs to coordinate their efforts effectively. The experts also felt that a close partnership with pharmacy can also be critical. They stated that the pharmacist is often key in alerting the ICP of a sudden increase in the usage of antimicrobials. When this information is conveyed quickly, it can help the ICP to rapidly identify, investigate and control an outbreak of infection. In addition, the experts felt that the pharmacist can alert the ICP of potential misuses of antimicrobials 107, 108. Overuse of antibiotics has no benefits, and may be harmful to residents, and also enhance the development of antibiotic resistant organisms 109.  Finally, the experts wanted a more clearly defined outline of the data analysis and reporting phases of surveillance. John Last defines surveillance as “Continuous analysis, interpretation, and feedback of systematically collected data... By observing trends in time, place, and persons, changes can be observed or anticipated and appropriate action, including investigative or control measures, can be taken” 110. Separate data analysis components were therefore added for trend review and corrective actions. In addition, they felt that stakeholders should not simply have access to surveillance reports, they should actively be provided with a copy of the report. Even though the discussion for Question 1 took place prior to the introduction of the five  97  safety principles, this small but significant change follows the recommendation made under safety principle 5 – Creating a learning environment. Under this principle, health care organizations are encouraged to foster a “…working culture in which communication flows freely regardless of authority gradient”, and to “Implement mechanisms of feedback and learning from error”62. These mechanisms should be detailed, clear, timely, and should lead to process improvements.  4.2.2 Question 2 & 3 – Merging safety and IP&C theory The Delphi methodology was also an innovative way to incorporate two bodies of thought into the survey in a way that has never previously been done. At the completion of Question 1, it was assumed that all key concepts and components stemming from infection prevention and control theory had been incorporated into the six indices. The primary purpose of Question 2 was therefore to introduce the expert group to safety theory by asking them to reflect on how the IP&C structures and processes in the six indices overlap with the five safety principles outlined in the report “To err is Human”62. While identifying the areas of overlap as per Question 2, it was expected that potential gaps in the six indices would become evident. Therefore Question 3 asked experts: “What other measurable IP&C structures and processes should be added to the indices to describe the five safety principles more fully?”. The utility of this two-step process was evidenced by the fifty modifications recommended for the six IP&C indices based on the five safety principles.  98  Nevertheless, the sense of concern that immediately resulted from this question seemed to potentially threaten the continuation of the study. Phone calls to five group members were required to discuss the purpose of introducing safety theory to an IP&C survey. Even after numerous calls and e-mails, one group member dropped out of the study stating that she could not understand the safety principles being introduced. Although the introduction of safety theory proved very beneficial in complementing IP&C theory to ensure the survey measures all key IP&C structures and processes, some experts seemed to find it challenging to understand that IP&C is an integral and fundamental component of resident safety. One expert expressed her concern by repeatedly pointing out that there is no evidence that strategies designed to prevent adverse events can effectively prevent infections, even though infections are clearly one of the types of adverse events described in safety theory. Unfortunately, there is also little evidence that most IP&C strategies effectively prevent infections in the LTC setting. Indeed, this lack of evidence further highlights the importance of developing a tool that will enable researchers to study the effectiveness of all of these strategies.  Based on the responses to Questions 2 and 3, the experts felt that the first Safety Principle – Provide Leadership, touched upon key concepts contained all of the six IP&C indices. In their minds, the provision of leadership for the purpose of maximizing resident safety is not simply about assigning someone to be responsible for infection prevention and control. Instead, leadership requires that a whole range of structures and processes be in place to effectively support and evaluate IP&C activities. In particular, they felt that foundational structures for the provision of leadership are  99  knowledgeable ICPs, physician leaders and actively participating IP&C committee members. Because of this, they felt that IP&C program leaders should not only be educated in infection prevention and control, they should also be committed enough to have knowledge of safety theory and currently used safety initiatives for LTC. These components were therefore added to the Leadership Index.  In addition, experts felt that the administrative commitment to resident safety (and therefore to IP&C) described in Safety Principle 1 should translate into the development of a culture of safety within a facility with trust and open communication as the central components. The expert group brought forth the same concern about including safety culture in the indices when discussing Safety Principle 5. This principle states that adverse events can only be effectively reduced or eliminated when dealt with in a trusting environment in which adverse events are used as a learning and improvement mechanism. The Health Quality Council of Alberta review of the IP&C issues in the East Central Health Region 111 clearly stated that a lack of safety culture played a large role in allowing the spread of Methacillin-Resistant Staphylococcus aureus within health care facilities. This was reflected in poor “patient safety advocacy” from health care professionals, lack of use of “patient safety criteria for making decisions”, “lack of alignment of processes and structures to support patient safety”, low levels of funding from the Board to support patient safety programs, and lack of involvement of physicians in patient safety initiatives. They stated that the lack of a safety culture “…did not allow for optimal identification of key safety issues that required immediate action at the senior administration and Board levels…”. The expert group felt that the culture of  100  safety so clearly described in Safety Principles 1 and 5 only incidentally correlated with some of the components of the Surveillance and Control Indices. The following components were noted as important gaps in the indices and were therefore added to the Policy & Strategy Index based on the consensus decisions: the promotion of IP&C by administration, clearly defined IP&C goals and objectives, free communication flow within the facility and blameless reporting of errors. In addition, a budget dedicated to IP&C was added as a key component of the Partnership & Resources Index. Interestingly, although the group felt strongly about adding components of safety culture within the indices, half of the group did not feel that ward-level infection control champions should be added. One expert explained that such a role would likely require additional remuneration for the additional responsibility. Nevertheless, since outbreaks of infection create a much larger workload for front-line staff, some may see other benefits to becoming a volunteer IP&C champion.  Safety Principle 1 also initiated a discussion on the concept of error analysis. Error analysis is one of the fundamental activities of health care organizations that have a culture of safety. When safety is a priority objective, systems need to be in place to facilitate the reporting and rapid identification of problems, and the analysis of the human, technological and system dynamics which allow error trajectories to develop. Some experts felt that IP&C audits and other monitoring systems for compliance would provide data necessary for error analysis. But experts remained reluctant to include any error analysis components that may be regarded as punitive or negative by health care professionals. The components that were rejected included the development of a plan  101  of action for non-compliance, and an identification system for unsafe practitioners. The experts were well aware of the IP&C breaches commonly caused by non-compliers in health care facilities. These were defined by one expert as “you know who I mean – those people who will either not wear the appropriate gear or those who wear gloves and contaminate the world”. According to Safety Principle 1, mechanisms need to be developed to “…identify, retrain, remove, or redirect physicians, nurses, pharmacists, or others who cannot competently perform their responsibilities” (p.169). Nevertheless, the experts rejected these components based on a mixture of concerns: from fear that staffing is inadequate to develop these systems, to a feeling that this should not be an IP&C issue but rather a safety or a quality issue. The concerns of some of the experts may stem from an unstated recognition that most errors result from inherent weaknesses within a particular system rather than individual incompetence. They may have felt that the development of a plan of action for non-compliance and an identification system for unsafe practitioners were at odds with the development of a culture of safety that includes blameless reporting of errors. Moreover, all but one of the experts in the group are members of BC’s Provincial Infection Control Network (PICNet). The overarching strategic objective of PICNet is to optimize sustainable capacity within the infection control community of practice. This is achieved by providing strategic advice for pressing infection control issues, by providing and sharing knowledge to guide practice and research, by supporting and coordinating communication, education and research, by sharing information and by advocating for the community of practice. The expert group may therefore feel discussing monitoring systems for unsafe practice is at odds with PICNet’s mandate. Perhaps these concepts  102  simply needed to be reframed in more supportive terms, for instance in the identification of educational needs and the development of a plan for professional improvement. In addition, the experts may have been unsure how ICP practices align with other safety activities. Yet, the development of a culture of safety was identified as a necessary precursor to effective IP&C, and facilities with a culture of safety need to ensure that their staff is well trained, and that their training is reflected in their professionalism, accountability, and ability to provide effective infection control within the LTC context.  Experts felt that Safety Principle 2, which calls for health care leaders to respect human limits of process designs, mapped well on to the IP&C requirement to provide Infection Control Practitioners with information technology and other resources to effectively do their work. They also felt that it reflected the requirements for clear IP&C policies and procedures, and simple surveillance definitions. This safety principle also alerted experts that they needed to further define the components of effective policies and procedures for IP&C so as to better describe the requirements needed to make them effective in IP&C. For instance, they agreed that IP&C policies and procedures must have clear and concise language and be reviewed regularly. They must easily be accessible by direct-care staff, and quick reminders such as checklists need to be used in the work area. Although regular audits were acknowledged as a very useful way to evaluate compliance with IP&C policies and procedures, the group almost unanimously rejected to call these ‘Safety Audits’. Performing activities such as audits in conjunction with the safety program may save funds, reduce overlap and redundancy, and improve collaboration between professionals. Unfortunately, experts may fear that by  103  acknowledging that IP&C activities are also resident safety activities, IP&C programs may loose their independence and identity. Indeed, such collaboration may entice administrators to simply integrate IP&C into a more general safety program, and this may lead to a net reduction in funds and control for IP&C experts.  Principle 3 urges facilities to promote effective team functioning by training staff in teams and by including the resident in their process of care. Experts felt that this principle correlated well with the internal (i.e. Occupation Health and Safety) and external (i.e. Pharmacy, Epidemiology consultant, Public Health) partnerships described in the indices. Nevertheless, there was less consensus as to which stakeholders constitute the health care ‘team’. According to this principle, “A team includes the practitioners, patients, and technologies used for the care…” (p.174). Based on this principle experts agreed that Housekeeping and Nursing should be added as key interdisciplinary team members with whom IP&C leaders need to work effectively to prevent and control infections. In order to increase the effectiveness and implementation of policies and procedures, they felt that it was crucial to involve directcare staff in the process, thereby acknowledging their value as team members. Evidence shows that care aide involvement in interdisciplinary care meetings is significantly associated with lower staff turnover rates in LTC 112. The resulting increased staff satisfaction and continuity of care may also lead to reduced infection rates. The role that residents and their families could play in the health care team for the purpose of IP&C seemed more difficult for the experts to define. They felt that more evidence was needed to support any potential benefits of including them in infection  104  control committee meetings. Nevertheless, regular meetings of staff, residents and family members to discuss any concerns (including IP&C concerns) could be very important in fostering an open and trusting culture of safety.  Safety Principle 4 recommends that systems be proactive in preventing adverse events by expecting them to occur, fixing problems before adverse events occur, and setting up information systems to quickly identify issues once they have occurred. Experts felt that this principle fit very well with the all of the components required for surveillance. It was further described within the information technology and laboratory resources included in the indices. The main gap noted in the indices based on this principle was the lack of concepts outlining emergency strategies, such as those needed in case of an influenza pandemic. An influenza pandemic is seen as an inevitable and devastating threat to the Canadian population 113. Only through careful planning can facilities hope to use their resources in the most effective and efficient way possible to minimize influenza transmissions and fatalities in residents and staff during such an event 114. Based on the 2006 pandemic influenza plan from the Public Health Agency of Canada, an effective pandemic plan for facilities must not only include control measures for limiting transmissions (e.g. activity restrictions), but it must also plan for the surge in the number and acuity of the residents within the facility. As acute care hospitals become overwhelmed with very sick patients, stable patients will likely need to be transferred to non-acute facilities. Experts therefore decided to add pandemic planning and surge capacity planning to the Policy and Strategy Index.  105  4.2.3 Question 5 – Question design and survey content validation When Question 5 was introduced, all of the concepts, components and operationalized components in each index had been discussed at length, and had been selected by consensus. The purpose of Question 5 was therefore not to re-discuss whether concepts or components needed to remain in the indices, but instead to evaluate the way in which the individual questions measured a particular component in a valid way. Although 23 written comments were made in relation to Question 5, only eight of these actually critiqued either the wording of the question or the relationship between the components and questions. The other fifteen comments and attached votes of ‘Moderate’ or ‘Weak’ relationships actually questioned the necessity to include the question in the survey at all. Some of the experts therefore used Question 5 as a means to re-initiate the discussion on concepts and components agreed upon by consensus. In all 15 cases, the other experts who responded described the relationships as ‘Strong’, therefore making the lower-scored response an outlier that did not affect the ability to achieve a consensus. However, the comments questioning the inclusion of certain components after these had already been agreed upon by consensus may potentially betray a certain degree of misunderstanding of the purpose of Question 5 and of the process used in the question development and validation process. It could be a sign that experts within the group had difficulty recalling the consensus decisions made throughout this lengthy process.  106  4.2.4 The Delphi methodology In order to make the Delphi phase of the study possible an adequate number of experts in LTC infection prevention and control needed to volunteer to participate. In British Columbia, health authority owned and operated LTC facilities share the support of a few ICPs and physicians. Because of recent budgetary increases, many of those working in this setting are recent hires and some have little experience. It was therefore feared that finding experts from various health authorities with expertise in a variety of long term care settings (i.e. large versus small, rural versus urban, attached to hospitals or contracted) would be difficult. However, with the endorsement of BC’s Provincial Infection Control Network (PICNet), many of BC’s foremost experts in LTC infection prevention and control volunteered to participate in this study. As soon as the minimum number of experts (i.e. four) was reached the Delphi methodology was initiated but members were allowed to join. The following assurances were taken to ensure the broadest response during this process: •  When the Delphi process was initiated with Question 1, only comments for ongoing discussion and recommended changes were requested from experts. No voting on particular recommended changes was requested in an attempt to reach consensus until six experts had volunteered to participate in the group.  •  When new members began participating in the group, all previously sent documents required as background information, e-mails sent by the moderator and Question 1 comments and recommended changes by group members were sent to them. New group members were also asked for their own comments and  107  feedback regarding Question 1 and related to past comments and recommended changes made by other group members.  When one expert unexpectedly dropped out of the group after only responding to the Reformatted Question 1 – Worksheet 1 requesting votes on recommended changes, there was a real concern regarding the potentially biasing effects of this change to the group. Although this group member had years of experience in LTC infection prevention and control, she explained that she “really struggled to understand” the concepts in the excerpt from the document To Err is Human which had just been sent to group members in preparation for Question 2. She stated that she is “not familiar with the (safety) models” and found them all “very abstract”. Since a new, highly qualified expert with considerable expertise in LTC infection prevention and control volunteered to participate in the group at exactly the same time, no change in total number of group members occurred when this expert decided to drop out. The new group member was provided with all previously sent background documents, e-mails and worksheet results, and asked for comments or recommended changes related to Question 1 and comments related to the results of the vote for the Reformatted Question 1 – Worksheet 1. These new recommended changes were then added to Question 1 – Worksheet 2 as new recommended changes. Previous recommended changes for which a consensus had been reached in Worksheet 1 were also re-entered in Worksheet 2 for a re-vote when a potentially vote-altering comment was made by any group member (including Expert #7). Thus, the new group member was given ample opportunity to participate in the current as well as any previous rounds of discussions and voting.  108  Following the initial changes with regard to group membership that ended May 23, experts were given an unlimited amount of time to respond to the Worksheets for comments and/or votes. Because of this flexible time allowance, only four of the 114 Worksheets were left unanswered, thereby limiting the potentially biasing effects of missing responses. All consensus decisions still required that 67% of all group members, including the non-respondent, agree with the decision. The data collected on variations in response time within this small group could be useful when planning future research using Delphi. From Figure 3.1, Figure 3.2, Figure 3.3 and Figure 3.4, we can see that the response pattern remained unchanged whether one, three or five Worksheets were sent on the same day (See Table 3.1). Experts consistently returned all Worksheets of a grouped-set within a 24 hour period. Their method seemed to be one of setting aside a period of time for the work of this study, and completing all of the work at hand at one time regardless of the amount of work given. In addition, it was falsely assumed that summer vacations and vacation coverage would leave little time for experts to participate in the project. Yet, experts responded to Worksheets sent during the months of July and August 2007 were responded to just as quickly as those sent during previous months. It would seem that the decrease in projects and meetings within their individual work places provided experts with the time needed to quickly answer the Worksheets. An additional observation is that two experts within the group were consistently quicker in answering Worksheets while two others were usually much slower (See Table 3.2). Although many reasons may explain why a longer response time was inevitable for some experts, the long wait between Worksheets may have  109  sapped some of the enthusiasm of fast respondents and may also have made it more difficult to recall previously made consensus decisions.  The first few weeks of this phase of the study focused on introducing the Delphi methodology to experts, explaining the purpose of the study and survey, and biding the time for consensus-reaching until recruitment was completed. Most importantly, this period was used for developing a tool that would facilitate the consensus-reaching Delphi process. Previous studies using the Delphi methodology first developed a survey containing a set of specific questions that were sent to group members to initiate the ‘discussion’ process 91, 115. A new survey was then sent to the group with a qualitative and quantitative summary of the previous survey’s responses so these may be taken into consideration when responding to the new survey. Although using a survey would have provided the expert group in this study with a clear and simple method to begin the discussion process, it would also have greatly limited the scope of responses they could make, and therefore the scope of discussion regarding modifications that could be made to the survey. Rather than immediately leading the discussion towards prespecified portions of the survey, the moderator began the Delphi process by asking experts to consider the key IP&C structural and process concepts and components within the survey. These concepts and components were provided to them, clearly outlined in the top 2nd and 3rd rows of the six index matrices (See Table 2.3, Table 2.4, Table 2.5, Table 2.6, Table 2.7, and Table 2.8). This open-ended method of initiating discussion enabled group members to provide a rich variety of feedback on the survey without being limited to discuss survey concepts and components pre-determined by  110  the researcher. Moreover, the use of the index matrices also focused their attention on the foundational concepts underlying the survey rather than on the details of the survey questions. For example, focusing on the wording of questions early on would have slowed the process because questions may need to be deleted or radically modified based on changes in the foundational concepts and components. Questions 1 to 5 systematically focused the discussion, first on the general underlying concepts and components, and slowly allowed the discussion to progress to the specific survey questions.  This method of initiating the discussion was very effective at enabling group members to provide a large variety of comments and recommended changes for the survey, and therefore directing the focus of the discussion. Nevertheless, using this open format rather than an initiating survey did not provide a clear method of forming consensus decisions in order to make modifications to the indices, and subsequently to survey questions. This was demonstrated by the responses (or lack of responses) to the original Question 1 – Worksheet 1, which was an anonymous collation of all comments and recommended modifications from group members under the various index headings. Instead of considering other expert’s comments and continuing the discussion, the comments stated “My comments to consider remain the same”, “Perhaps I don’t understand what I’m supposed to be adding here”, and “I do not have any response to the comment”. It seemed that group members needed a more clearly defined and concrete process for considering the opinions of other group members and coming to a consensus on the recommended changes. The Reformatted Question 1 –  111  Worksheet 1 (See Appendix F – Reformatted Worksheet 1) was therefore developed by the moderator to provide the necessary structure to support ongoing discussion. The reformatted Worksheet was very well accepted by group members, as clearly expressed by one expert: “Should’ve known you would make it easy for us…”. This study therefore showed how important it is for the moderator to spend the necessary time supporting group members as they learn the consensus-reaching process using Delphi.  4.2.5 Moderator effect Many steps were taken to prevent moderator bias from influencing the results of the discussion and consensus process. Moderator bias arises from a moderator consciously or subconsciously influencing the responses provided by respondents 116. Not only was the moderator in this phase of the study not permitted to vote on any recommended changes, but she was also required to identify herself prior to making any comments on the Worksheets. In addition, the moderator did not summarize comments sent by group members. These were entered into Worksheets unaltered to minimize the possibility of this bias. Although the effect of this type of bias was minimized, it is impossible to eliminate completely. When no consensus was reached on a recommended change, the moderator needed to use her own judgment in deciding whether to continue discussions based on a new comment that should be considered by everyone, or whether to drop the discussion for the time being until the topic reemerged from the discussion.  112  In Question 4, the moderator was also required to make a critical decision. In this question, an exhaustive list of eight infections that commonly cause high levels of morbidity in the long term care setting was provided to the experts. Experts were asked to state which infections they felt caused high morbidity as well as mortality in long term care. Since facilities are more likely to use their most intensive methods of surveillance for infections causing the highest level of morbidity and mortality in LTC, these infections would be used to collect data on surveillance practices used within the facility. The results showed that, although there were strong conflicting opinions regarding some of the types of infections, at least 67% of experts felt that seven out of the eight types of infection caused high morbidity and mortality. The moderator decided to increase the level of required consensus to for this particular issue to a minimum of 83% for the following reasons: 1) To take into consideration the strong opinions of experts, and 2) To limit the length of the survey by including only a subset of the exhaustive list which could provide data on surveillance practices. Since the moderator did not influence the choices made by experts, this decision was unlikely to have a biasing effect on the results.  4.3 Phase III: Pilot-testing phase The pilot study performed in Phase III of the study was a critical last step in the validation process. It enabled the researcher to establish the validaty the survey questions from the perspective of the target population, and to ensure the feasibility of a larger study. Based on the responses and feedback of facility Senior Managers, improvements were made to questions to improve their clarity and utility in measuring  113  IP&C structures and processes. The final draft (See Appendix N – Survey Draft 3) was then prepared using the revised questions. The following discussion describes observations and findings of the pilot-testing phase of this study.  4.3.1 Findings from validity & feasibility tests In general, facility Senior Managers stated that they approve of the survey format and of most of the survey questions. Nevertheless, important recommendations were made to improve their clarity and utility of certain questions. The following describes the modifications made to the survey based on their feedback.  Definition for “staff” and “educational time”: The word “staff” was reported as difficult to interpret in two general areas of the survey. In Section 1, Question 2 asks how many “full-time direct-care staff” within the categories of RN, LPN and Care Aide work within the facility. Numerous Senior Managers reported that part-time staff should be included in this question as most of their staff work parttime and equally contribute to IP&C within their facility. Part-time staff was excluded because they may work varying amounts of time within the facility and therefore their number may not accurately reflect the coverage they provide. The problem associated with this question was also evident from the survey responses provided by Senior Managers. Logically, there should be some association between the number of beds within a facility and the number of RNs, LPNs and Care Aides. Analysis of the data reported by Senior Managers showed little association between the number of staff and the number of beds in facilities. It is possible that some respondents decided to 114  somehow count and report their part-time staff while others strictly followed the directions and did not. In order to improve the utility of this question, the wording was changed to ask for the total number of Full Time Equivalents for all of the full-time and part-time direct-care staff working in the facility.  Senior Managers also felt that the word “staff” used throughout the document should be defined to clearly identify which staff the questions are referring to. For the most part, staff will therefore be preceded by the adjective “direct-care”, and will include RNs, LPNs and Care Aides working directly with residents. The term “educational time” as also found to be problematic as much of the education is provided to direct-care staff as short in-services or workshops. The question was therefore modified to clearly include this type of informal education.  Definition for ICP: One of the facility Senior Managers reported being confused by the person to identify as the ICP in Section 2 – Question 1. Her primary concern was related her small facility size and the fact that all three supervisory and management staff within the facility share this role. The clear assignment of IP&C oversight to a particular individual within the facility was identified as a key component of the Leadership Index by the expert group in the Delphi phase of the study based on recommendations from Health Canada / PHAC, the Institute of Medicine, the Joint commission for Accreditation of Healthcare Organizations, and many other sources 13, 61, 62, 84, 85. The question was therefore not modified to include management teams with shared IP&C responsibilities. Nevertheless,  115  other issues emerged from the data related to the clarity of the definition of ICP for respondents. First of all, Question 1 did not make it clear whether the ICP could be someone working for the health authority rather than the facility. The question was therefore changed to make it clear that the ICP could work from either within the facility or from the regional IP&C program. In addition, some respondents may have thought all of their nursing education and experience could be counted as education and experience in infection control. One respondent answered Questions 2 (How many college or university courses has your ICP taken in infection control or epidemiology?) and 3 (How many years of experience does your ICP have in infection control?) with the answer: “Nursing since 1971”. Since these questions aimed to measure IP&C-specific education and experience, these questions were clarified.  Selection of best respondent: Based on the feedback and responses, it became evident that facility Senior Managers may not have all of the information required to answer all of the survey questions. Furthermore, only two respondents obtained assistance from another member of the health care team to provide the missing information so the survey could be accurately complete. Since completing the survey requires inside information regarding each specific facility, a regional ICP doing IP&C for a large number of facilities seems like an inadequate choice of respondent. Even if such an ICP were able to provide accurate information regarding all of their facilities, they would not have the time required to do so for a large-scale study covering all facilities. Table 3.12 shows that Senior Managers, most knowledgeable about their facility IP&C activities, were able to answer at least  116  80% of the questions on the survey. It is therefore likely that they are the best choice of respondent to complete the survey. Nevertheless, Senior Managers felt that it was difficult to answer questions regarding their regional ICP’s training, experience and number of Full Time Equivalents (FTE), or about Occupational Health & Safety (OHS) issues such as immunization programs. The best way to accurately and reliably obtain such specific information would likely be to remove these specific portions of the survey and to complete them manually by contacting the personnel most knowledgeable with the required information. For example, regional ICPs could be contacted for information on their training, experience and FTEs, and this information could be used for all facilities with which the ICP works. This method may unfortunately be more time consuming and costly and could potentially reduce the ability to do a large-scale study.  Recommended additions: Senior Managers identified many important additions for the survey. Some, such as ICP and Microbiologist coverage, past and present infection rates, outbreak management, and multidisciplinary approach to IP&C were identified as key components of the indices and already covered by the survey or through data which could most validly and reliably be obtained by other means (i.e. from the health authorities or the BCCDC). All of the recommendations were used to make minor modifications to survey questions to enhance clarity where confusion seemed to be evident. However, some of the recommended additions had not been identified as key components by the expert group in the Delphi phase (e.g. Break room set-up during quarantine periods and N95 respirator fit-testing). These were therefore not added to the survey.  117  Indicators for IP&C program duration: The length of time a formal IP&C program has been established in a facility was also thought to potentially have an impact on the rate of preventable facility-associated infections. A long-standing, well established program may be an indictor of how firmly the program initiatives are embedded in the culture of the facility. On the other hand, when safety programs (such as IP&C programs) have been established for a long time, it is possible for complacency to begin to develop when no issues or outbreaks have been identified over a long period of time. Sidney Dekker calls this phenomenon ‘Drift into failure’ 117, and states that it relates “…not so much about breakdowns or malfunctioning of components, but about an organization not adapting effectively to the complexity of its structure and environment”. Since program duration may have an impact on infection rates, the survey attempted to measure this factor using a combination of two indicators: the establishment of a hand-hygiene policy and the creation of an ICP position for the facility. The duration of the program was defined using the most distant of these two dates (if both were provided), and was measured in years. Although all respondents stated that they had a written hand-hygiene policy, only twelve out of the fourteen facility Senior Managers entered a date for its establishment. In addition, eleven of the twelve stated that their hand-hygiene policy was first established within the past seven years (i.e. 2000 and later), with seven of these established either in 2006 or 2007. The respondent who answered that their policy was established in 2000 (i.e. the second oldest date entered) also commented that the last revision date for the policy should be requested to ensure that facilities are maintaining  118  best-practice over time. It is therefore plausible that some respondents misinterpreted the question as requesting the date of the latest version of the hand-hygiene policy rather than the date it was first established. This question was therefore revised to highlight the exact date sought. The second date used as an indicator for the start of the IP&C program of the facility also proved to be problematic. Although nine respondents stated that they had a nurse or person dedicated to IP&C in their facility, only six of them were able to state the year during which the position was created. Nevertheless, by using the older of the two dates, a marker for the start of the IP&C program was established for all but two of the facilities.  Survey and index score distribution: In order to determine whether this survey could be useful in determining whether there is an association between the survey or individual index scores and the number of serious infections developed by residents living in LTC, the distribution of scores needs to have an adequate amount of variance. The pilot study results showed that there is at least a 17% difference between the first quartile score and the third quartile score for all indices and for the total score (See Table 3.13). No floor effect, where a large proportion of respondents score very low, or ceiling effect, where most score high, were noted. Therefore the score distribution did not highlight any mismatch between the items addressed by the questions and the Senior Managers’ ability to answer.  When one point was allocated to each component within each index, the resulting mean scores for the six indices showed that there is a tendency for facilities to have lower  119  levels of components within the Leadership Index and the ICP Index. If confirmed in a larger survey, this would be a major concern as the SENIC study on the effectiveness of IP&C programs in acute care hospitals found these components to be associated with a significant decrease in hospital-associated infections 43. In fact all expert advisory bodies for both IP&C and safety state that a strong, dedicated and knowledgeable leadership is the key to program effectiveness 13, 61, 62, 84-86, 118. Although there were very low and very high scores in all indices, the mean scores reflected that, on average, over 60% of the index components were being met by facilities. A larger study would be able to confirm this finding and enable IP&C professionals and facility management to direct their efforts at improving the IP&C structures and processes that require increased funding and/or attention.  4.3.2 Facility ‘Not-for-profit’ versus ‘For-profit’ status Recent studies have highlighted significant differences in quality of care and safety indicators between FP and NP facilities 9, 11, 12, 119, one of these finding a more than 2 fold increased risk of requiring hospitalization for pneumonia and urinary tract infections for residents dwelling in FP facilities12. These data have lead to the hypothesis that fewer IP&C resources may be available in FP facilities. In order to ensure that all four major categories of ownership and governance were equally well represented in the final sample of 20 facilities, stratified random sampling was used, as described above, in the study design. However, an analysis of the differences in IP&C structures and processes between ownership and governance strata using the data from this pilot study would not have yielded valid results for the following reasons. First, Stratum #4  120  represented FP facilities not contracted by Fraser Health. It is therefore the only stratum without regional ICP coverage, and thus is the stratum of interest in the NP versus FP analysis. Unfortunately, three of the five facility Senior Managers invited to participate from this stratum either refused to participate or dropped out of the study. The data are therefore inadequate for even a preliminary comparative analysis. In addition, one of the major issues identified by the Senior Manager respondents was their difficulty in answering questions concerning their regional ICP’s training, education and number of full-time equivalents. The ICP Index data, when fully gathered at the facility level, may therefore be biased against facilities covered only by regional ICPs. Triangulation should therefore be used, whereby both Senior Managers and regional ICPs are asked to provide data specific to the regional ICP coverage and training, so the data may be compared in the analysis. This would enable researchers to identify issues related to the reliability of Senior Manager’s responses to these questions.  The pilot study revealed major difficulties when attempting to elucidate the ownership and governance status of facilities (See Methods section 2.3). The inaccessibility of information regarding ownership and governance by the general public reduces the ability for residents and families to make an informed consumer decision regarding providers of long term care. Proponents of the market model of service provision argue that, when individuals are free to make choices, exchanges of goods, services and money will always benefit both of the parties making the exchange 120. Unfortunately, barriers exist that make it excessively difficult for elderly residents of long term care facilities to exercise their freedom of choice or make informed decisions when selecting  121  their long term care provider. Residents and their families are primarily interested in purchasing long term care services that have high levels of quality and safety. Yet, quality and safety are abstract terms that are very difficult to measure. In our quest to improve quality, we continuously evaluate a set of measurable indicators, plan and implement changes that may improve the system, and compare the key quality indicators before and after changes are made to evaluate whether the changes improved quality. On the other hand, safety is an emergent process that requires constant vigilance for latent flaws and surveillance for measurable key safety indicators in systems to be well constructed 121. There exist many barriers that limit the ability of residents and their families to have access to meaningful information regarding quality and safety. First of all, since quality and safety indicators need to be readily measurable, they may not include many aspects of quality and safety that are very relevant for residents (e.g. cleaning technique and products used by housekeeping staff, frequency and appropriate timing of hand hygiene, staff level of knowledge of infection prevention and control procedures ...). Secondly, the continuous collection of data for each indicator requires time and personnel, and therefore extra front-end costs. In addition, findings resulting from the analysis of these data may not be flattering for facilities and facility owners may be concerned that their dissemination to residents and families may damage their reputation. Also, the findings may uncover major flaws that would necessitate potentially expensive changes. Facility owners may therefore prefer not to collect these types of data unless they are compelled to do so by a governing body. Even if residents and their families had access to the necessary data, they may not have the necessary 122  understanding of quality and safety issues, and may therefore not understand the relative importance of certain measures. Moreover, once an elderly person has been admitted to a long term care facility, there are no formalized ways for them to make their voices heard so they may affect change if they become concerned with the quality or safety of their care. Therefore facility ownership and governance can be key pieces of information to which residents and families should have easy access to enhance their knowledge and decision-making power during the key time when they are selecting a long term care facility. Open access to this information would also help researchers further investigate the association between facility ownership and governance and key IP&C outcome indicators.  4.3.3 Potential biases in the pilot study The pilot phase of the study helped us to evaluate the survey clarity and make broad estimates of its reliability, validity, and feasibility. By inviting 20 facilities to participate in the study, it remained adequately powered even with a 70% response rate (see Figure 2.4). Nevertheless, it is possible that those who volunteered to participate in the study after being invited are systematically different from those who refused to participate 116;. For example, facilities that reported being unable to participate due to a lack of time may be too busy working on important IP&C initiatives. On the other hand, they may be more poorly staffed than participating facilities and may therefore not have the resources to implement IP&C initiatives. It is interesting to note that three of the six facilities that refused to participate were FP facilities not contracted by Fraser Health. Senior Managers from this category of facilities may therefore feel more disengaged  123  from Regional or Provincial initiatives and research than Managers from the other three ownership and governance categories.  It is also possible for bias to arise from other sources 116, also potentially diminishing the pilot study’s ability to detect a true difference in IP&C structures and processes between the facilities. Acquiescence response bias refers to the fact that survey respondents are more likely to answer ‘yes’ than ‘no’ where these dichotomous responses are required. Evaluation apprehension bias and the Hawthorne effect may also be affecting the results as respondents may have felt particularly anxious about reporting a lack of certain key infection prevention and control processes. This may have been one of the factors involved in non-response, and it may have biased respondents to provide answers they feel place their facility in the best possible light. Respondents may also have become more interested in their infection prevention and control processes as they became involved in the study, and this may have lead them to make improvements.  4.4 External validity This study used a three-phase methodology to develop a valid tool that can be used to measure IP&C structures and processes in the LTC setting so their relative effectiveness can be evaluated based on key IP&C outcomes. Because of the paucity of research in the field of IP&C in the long term care setting, this method of inquiry was particularly beneficial as it enabled the researcher to enhance and validate the literature-based survey developed in Phase I using experts in the field from across 124  British Columbia. Moreover, since the incorporation of safety principles in IP&C research has never been explicitly undertaken in the past, the survey developed in this study could become the initiating component in the development of a new understanding of how infections may be prevented and outbreaks may be controlled in health care facilities.  The pilot study completed in Phase III of the study provided valuable data for the validation process. These data guided modifications that will enhance the clarity and utility of survey questions and the feasibility of a larger study across British Columbia (BC). Although the survey was only piloted in one of the six health authorities in British Columbia, the broad representation of experts in Phase II enabled concerns from other authorities to be voiced. In addition, concerns brought forth by Senior Managers regarding particular survey questions would likely be similar to those described by Senior Managers from the other health authorities. Those who responded to the survey worked in facilities of all sizes and the stratification method used ensured that facilities in the sample had access to the broad spectrum of IP&C resources found in BC. When the large study is performed using all facilities across BC further analysis can be performed to confirm the external validity of the survey.  125  5. CONCLUSIONS & FUTURE RESEARCH As Dekker explains, “organizations that involve safety-critical work are essentially trying to reconcile irreconcilable goals (staying safe and staying in business)” 117. If effective systems are not in place to collect, analyze, report and prevent IP&C issues, organizations such as LTC facilities may be tempted to make incremental changes that improve the bottom line, yet take the organization to the boundaries of safe practice and beyond. LTC facilities are particularly vulnerable to this kind of ‘drift into failure’ because many adverse events such as infections (even those leading to death) are often rationalized as being unavoidable in the highly vulnerable residents who dwell in them. This notion must be rejected and replaced with an urgency to transform all LTC facilities into organizations that provide safe, high quality care.  According to Jim Collins 118, the first and most critical step in achieving positive transformation is to have the strength to face the brutal facts. He explains that “only in a climate where truth is heard can positive change occur”. The primary objective of this study was to put this first step into motion. This was achieved by recognizing, and addressing important gaps in the literature. First of all, the literature was used to develop a conceptual framework that could frame studies designed to measure IP&C structures and processes in the LTC setting, and to evaluate their effectiveness as had been done in the acute care setting. Secondly, an innovative 3-phase process was used to build and validate a tool that would enable researchers to measure key IP&C structures and processes in the LTC setting in a feasible, reliable and valid manner. Finally, although safety theory had taken on increasing importance in making positive 126  change in the acute care setting over the last decade, it had not yet been used in the LTC setting. In order to facilitate the emergence of safety culture in this setting, an initial step was taken to merge IP&C and safety theory.  5.1 Lessons learned: Merging IP&C and safety The work done by the expert group in Phase II provided some very important insight regarding the critical role of safety theory for IP&C professionals in the long-term care setting. Although experts seemed ambivalent regarding the introduction of safety theory to complement IP&C theory, 50 out of the 82 suggested modifications to the six indices were made in relation to the five Safety Principles. On the one hand, experts rejected many concepts and components strictly because the word ‘safety’ was included (i.e. Safety audits rather than IP&C audits); on the other hand experts described at length the numerous overlaps between the five Safety Principles and IP&C structures and processes. Moreover, experts were able to use the Safety Principles to describe many additional IP&C structures and processes that are not commonly described in IP&C recommendations, yet are highly relevant in preventing infections and controlling outbreaks. The following recommendations for the IP&C community of practice summarize the 32 consensus decisions that experts made based on the five Safety Principles: •  Increase knowledge, understanding and use of safety theory and strategies: This study clearly showed the complimentary value of safety theory for IP&C. Experts found that concepts such as a culture of safety and error analysis really helped them to articulate necessary components of IP&C that were not explicitly 127  described in IP&C literature. As IP&C professionals become more knowledgeable about safety theory, they can begin to utilize even more complex safety strategies to further improve IP&C in the LTC setting. For example the concept of vigilance was not included in the indices, yet it could be used by IP&C professionals to go beyond simply doing surveillance for the purpose of rapidly responding to cases of infection or outbreaks. Vigilance requires a cognitive understanding of safety dynamics so that adverse events such as preventable infections are prevented by actively seeking out weaknesses in the system before these lead to infections. •  Enhance interdisciplinary work: IP&C professionals have always understood the importance of working closely with Public Health, Occupational Health and Safety, Pharmacy, and Housekeeping. Nevertheless, safety theory helped the experts to describe the crucial importance of also involving administration, senior management and direct care staff in IP&C processes to develop a sense of teamwork towards common and clearly defined objectives. Working closely with all disciplines within the facility also helps to develop multiple channels of communication that can help the Infection Control Practitioner to rapidly identify problems.  •  Place priority on developing a culture of safety: Experts felt that the development of a culture of safety was a key ingredient for an effective IP&C program. Clearly, staff needs to feel free to voice any concerns they may have to any of their superiors. Concerns need to be taken seriously, investigated quickly, and staff  128  needs to be notified of the results of this investigation (i.e. any actions taken to prevent further problems from occurring). •  Develop IP&C systems that are clear, simple and visible: Safety theory also helped experts to identify the importance of having clearly articulated IP&C goals and objectives from facility administrators. They also described the importance of developing IP&C policies, procedures and surveillance mechanisms that are clear and simple. In order to maximize their utility in the clinical setting, these need to be developed in conjunction with front-line staff and be strategically placed within the clinical setting. Memory aids such as checklists and reminders also need to be used consistently to assist recall.  5.2 Lessons learned: The Delphi methodology In Phase II of this study, the Delphi methodology was very useful in enabling a group of experts to make consensus decisions for the purpose of developing and validating a survey. Although this methodology was developed in the 1960’s, it had never been formally used in health care survey development and validation. Based on this experience, a few recommendations can be gleaned for the future use of the Delphi in survey development and validation. • Provide adequate time to respond: The workloads of experts varied broadly throughout the process so adequate time to respond must be provided. Most experts were able to respond within a two to three week period, so perhaps setting a consistent deadline within that range would maximize the responserate and minimize recall bias and worksheet fatigue. 129  • Don’t be afraid to begin or continue group work during the summer months: Experts responded to worksheets more quickly during July and August than at any other time of the year. • Begin discussion with an open-ended question: The questions used in this process sequentially focused the attention of the experts first on the most general concepts and slowly towards the specific questions in the survey. Their open-ended nature ensured that experts directed the discussion and felt at liberty to voice all of their concerns and recommendations rather than only those preselected by the moderator. • Use worksheets to frame the discussion and consensus-reaching process: The experts clearly voiced a need to have a clear framework within which to discuss and reach decisions on recommended changes to the survey. • Send multiple worksheets at the same time whenever possible: Experts took the same amount of time to respond to one worksheet as they did to respond to five sent together. Once experts had had a chance to achieve a certain level of comfort with the process being used, sending multiple worksheets clearly reduced response time for all experts. • Restrict and clearly define the role of the moderator: The role of the moderator must be transparent and clearly defined for all experts within the group. This reduces moderator bias and helps to develop and maintain an open trusting and respectful environment in which the discussion and consensus-reaching process can flourish.  130  5.3 Knowledge translation and future research The fundamental and overarching goal of this study was to ensure that the findings filled a knowledge gap within the IP&C community of practice. In order to achieve this goal, a partnership was formed with BC’s Provincial Infection Control Network (PICNet) at the very beginning of the process. The research question developed within this partnership was the first step in filling an urgent need to further evaluate and enhance IP&C in LTC facilities. In February 2008, the PICNet Steering Committee approved a project to use the survey in all LTC facilities throughout British Columbia. This process should provide PICNet with a rich source of data about IP&C in LTC that will permit them to evaluate the relative effectiveness of the various structures and processes currently being used in all BC facilities. Rigorous methods such as regression analysis can then be used to evaluate the relative impact of these structures and processes on key outcome indicators (e.g. mortality or admissions to acute care hospitals for serious infections). In addition, the large group size will also provide a valuable opportunity to further validate the survey and to determine the best method of calculating the composite index scores.  With the work completed in this study, the first step has been taken towards ensuring that preventable infections are effectively avoided and outbreaks are quickly controlled within all LTC facilities. Relentless drive for evaluation and research will now be required in order to achieve the momentum needed for positive change. The creative process used in this study enabled the development of a feasible, reliable and valid way by which researchers can measure key IP&C structures and processes. In addition, by merging two bodies of knowledge and thought into the process, concepts and 131  components that are not explicitly described in IP&C literature yet were felt to be key in program success, were incorporated into the measurement tool. This study has also facilitated the emergence of safety culture in the LTC setting by providing IP&C experts and facility Senior Managers with an introduction to safety theory. Use of the survey in future research will continue this process and will help to forge a common language and understanding within the worlds of IP&C, safety and LTC.  132  REFERENCES 1. Hicks V, Fortin G, Button I. Long term residential care in national health expenditures feasibility study. Ottawa, Ontario: Canadian Institute for Health Information (CIHI); 2002;ISBN 1-55392-051-1 (PDF):i-iii-1-15. 2. Statistics Canada. Table: Population by sex and age group, by province and territory 2006;CANISM Table 051-0001. Available from: http://www40.statcan.ca/l01/cst01/demo31a.htm?sdi=british%20columbia%20age. 3. Statistics Canada. Population projections 2005 to 2031. The Daily. 2005;December 15. 4. Cranswick K, Thomas D. Elder care and the complexities of social networks. Canadian Social Trends. 2005;77:10-15. 5. Tully P, Mohl C. Older residents of health care institutions. Health Reports. 1995;7:27-30. 6. Statistics Canada. National population health survey, cycle 2 - Residents of health care institutions. The Daily. 1999. 7. Canadian Institute for Health Information (CIHI). National health expenditure trends. 2005:1150. 8. Deber RB. Delivering health care services: Public, not-for-profit, or private? Saskatoon: Commission on the Future of Health Care in Canada; 2002;No.17:233-296. 9. Handler SM, Castle NG, Studenski SA, et al. Patient safety culture assessment in the nursing home. Qual Saf Health Care. 2006;15:400-404. 10. US nursing homes less safe than hospitals. BMJ. 2007;334:68-a. 11. Castle NG, Sonon KE. A culture of patient safety in nursing homes. Qual Saf Health Care. 2006;15:405-408. 12. McGregor. Care outcomes in long-term care facilities in British Columbia, Canada: Does ownership matter? Medical care [serial online]. 2006;44:929. 13. Smith PW, Rusnak PG. SPECIAL COMMUNICATION - Infection prevention and control in the long-term–care facility. Am J Infect Control. 1997;25:488-512. 14. Richards C. Infections in residents of long-term care facilities: An agenda for research. Report of an expert panel. J Am Geriatr Soc. 2002;50:570-576. 15. Norris S, Williams T. Healthy aging: Adding life to years and years to life. Ottawa, Ontario: Government of Canada; 2000;PRB 00-23E. 16. Feldman C. Pneumonia in the elderly. Med Clin North Am. 2001;85:1441-1459.  133  17. Golliot F, Astagneau P, Cassou B, Okra N, Rothan-Tondeur M, Brucker G. Nosocomial infections in geriatric long-term-care and rehabilitation facilities: Exploration in the development of a risk index for epidemiological surveillance. Infection Control & Hospital Epidemiology. 2001;22:746-753. 18. Nicolle LE. Infection control in long-term care facilities. Clin Infect Dis. 2000;31:752-756. 19. Sheps SB, Reid RJ, Barer ML, et al. Hospital downsizing and trends in health care use among elderly people in British Columbia. Can Med Assoc J. 2000;163:397-401. 20. Strausbaugh LJ, Joseph CL. The burden of infection in long-term care. Infection Control & Hospital Epidemiology. 2000;21:674-679. 21. Stevenson KB. Regional data set of infection rates for long-term care facilities: Description of a valuable benchmarking tool.[see comment]. Am J Infect Control. 1999;27:20-26. 22. Brusaferro S, Regattin L, Silvestro A, Vidotto L. Incidence of hospital-acquired infections in Italian long-term-care facilities: A prospective six-month surveillance. J Hosp Infect. 2006;63:211-215. 23. Alvarez S, Shell CG, Woolley TW, Berk SL, Smith JK. Nosocomial infections in long-term facilities. J Gerontol. 1988;43:M9-17. 24. Berman SJ, Johnson EW, Nakatsu C, Alkan M, Chen R, LeDuc J. Burden of infection in patients with end-stage renal disease requiring long-term dialysis. Clinical Infectious Diseases. 2004;39:1747-1753. 25. Bentley DW, Bradley S, High K, Schoenbaum S, Taler G, Yoshikawa TT. Practice guideline for evaluation of fever and infection in long-term care facilities. Clinical Infectious Diseases. 2000;31:640-653. 26. Bradley SE. Double, double, toil and trouble: Infections still spreading in long-term-care facilities. Infect Control Hosp Epidemiol. 2005;26:227-230. 27. Day K, Carreon D, Stump C. The therapeutic design of environments for people with dementia: A review of the empirical research. Gerontologist. 2000;40:397-416. 28. Stelfox HT, Bates DW, Redelmeier DA. Safety of patients isolated for infection control. JAMA. 2003;290:1899-1905. 29. Gammon J. The psychological consequences of source isolation: A review of the literature. J Clin Nurs. 1999;8:13-21. 30. Institute for Healthcare Improvement. How-to guide: Improving hand hygiene. A guide for improving practices among health care workers. 2006:1-32. 31. Harbarth S, Pittet D, Grady L, Goldmann DA. Compliance with hand hygiene practice in pediatric intensive care. Pediatr Crit Care Med. 2001;2:311-314. 134  32. Pittet D, Boyce JM. Hand hygiene and patient care: Pursuing the Semmelweis legacy. Lancet. 2001;1:9-20. 33. Pittet D. Compliance with hand disinfection and its impact on hospital-acquired infections. J Hosp Infect. 2001;48 Suppl A:S40-6. 34. Pittet D, Hugonnet S, Harbarth S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. The Lancet. 2000;356:1307-1312. 35. Pinner RW, Haley RW, Blumenstein BA, Schaberg DR, Von Allmen SD, McGowan JE,Jr. High cost nosocomial infections. Infect Control. 1982;3:143-149. 36. Coello R, Glenister H, Fereres J, et al. The cost of infection in surgical patients: A casecontrol study. J Hosp Infect. 1993;25:239-250. 37. Andersen BM, Rasch M. Hospital-acquired infections in Norwegian long-term-care institutions. A three-year survey of hospital-acquired infections and antibiotic treatment in nursing/residential homes, including 4500 residents in oslo. J Hosp Infect. 2000;46:288-296. 38. Capitano B, Leshem OA, Nightingale CH, Nicolau DP. Cost effect of managing methicillinresistant Staphylococcus aureus in a long-term care facility. J Am Geriatr Soc. 2003;51:10-16. 39. Rubin RJ, Harrington CA, Poon A, Dietrich K, Greene JA, Moiduddin A. The economic impact of Staphylococcus aureus infection in New York city hospitals. Emerg Infect Dis. 1999;5:9-17. 40. Moise-Broder PA, Sakoulas G, Eliopoulos GM, Schentag JJ, Forrest A, Moellering RC,Jr. Accessory gene regulator group II polymorphism in methicillin-resistant Staphylococcus aureus is predictive of failure of vancomycin therapy. Clin Infect Dis. 2004;38:1700-1705. 41. Gavazzi G, Krause KH. Ageing and infection. Lancet Infect Dis. 2002;2:659-666. 42. Wenzel RP, Pfaller MA. Infection control: The premier quality assessment program in United States hospitals. Am J Med. 1991;91:27S-31S. 43. Haley RW, Culver DH, White JW, et al. The efficacy of infection surveillance and control programs in preventing nosocomial infections in US hospitals. Am J Epidemiol. 1985;121:182205. 44. Haley RW. Managing hospital infection control for cost-effectiveness: A strategy for reducing infectious complications. Chicago: American Hospital Association. 1986. 45. Verdier R, Parer S, Jean-Pierre H, Dujols P, Picot MC. Impact of an infection control program in an intensive care unit in France. Infect Control Hosp Epidemiol. 2006;27:60-66. 46. Stevenson K, Loeb M. Performance improvement in the long-term-care setting: Building on the foundation of infection control. Infection control and hospital epidemiology [serial online]. 2004;25:72. 135  47. Karchmer TB, Durbin LJ, Simonton BM, Farr BM. Cost-effectiveness of active surveillance cultures and contact/droplet precautions for control of methicillin-resistant Staphylococcus aureus. J Hosp Infect. 2002;51:126-132. 48. Zoutman DE, Ford BD. The relationship between hospital infection surveillance and control activities and antibiotic-resistant pathogen rates. Am J Infect Control. 2005;33:1-5. 49. Montecalvo MA, Jarvis WR, Uman J, et al. Infection-control measures reduce transmission of vancomycin-resistant enterococci in an endemic setting. Ann Intern Med. 1999;131:269-272. 50. Zafar AB, Gaydos LA, Furlong WB, Nguyen MH, Mennonna PA. Effectiveness of infection control program in controlling nosocomial Clostridium difficile. Am J Infect Control. 1998;26:588-593. 51. French GL, Cheng AF, Wong SL, Donnan S. Repeated prevalence surveys for monitoring effectiveness of hospital infection control. Lancet. 1989;2:1021-1023. 52. Seto W, Tsang D, Yung R, et al. Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). The Lancet. 2003;361:1519-1520. 53. Macartney KK, Gorelick MH, Manning ML, Hodinka RL, Bell LM. Nosocomial repiratory syncytial virus infections: The cost-effectiveness and cost-benefit of infection control. Pediatrics. 2000;106:520. 54. McConkey SJ, L'Ecuyer PB, Murphy DM, Leet TL, Sundt TM, Fraser VJ. Results of a comprehensive infection control program for reducing surgical-site infections in coronary artery bypass surgery. Infect Control Hosp Epidemiol. 1999;20:533-538. 55. Agency for Healthcare Research and Quality. Guide to patient safety indicators. . 2003. 56. Cooper JB, Gaba DM, Liang B, Woods D, Blum LN. The national patient safety foundation agenda for research and development in patient safety. MedGenMed. 2000;2:E38. 57. Agency for Healthcare Research and Quality. Patient Safety Indicators Overview. Available at: http://www.qualityindicators.ahrq.gov/psi_overview.htm. Accessed 3/27/2007, 2007. 58. Baker GR, Norton PG, Flintoft V, et al. The Canadian adverse events study: The incidence of adverse events among hospital patients in Canada. Can Med Assoc J. 2004;170:1678-1686. 59. Furman C, Caplan R. Applying the Toyota production system: Using a patient safety alert system to reduce error. Joint Commission Journal on Quality and Patient Safety. 2007;33:376386. 60. Leape LL, Brennan TA, Laird N, et al. The nature of adverse events in hospitalized patients. Results of the Harvard medical practice study II. N Engl J Med. 1991;324:377-384.  136  61. Morrison J. Development of a resource model for infection prevention and control programs in acute, long term, and home care settings: Conference proceedings of the infection prevention and control alliance. American journal of infection control [serial online]. 2004;32:2. 62. Kohn LT. To Err is Human. Washington, DC: National Acad. Press; 2000. 63. Sorra JS, Nieva VF. Hospital survey on patient safety culture. Rockville, Maryland: Agency for Healthcare Research and Quality. 2004. 64. Kirby M, Standing Senate Committee on Social Affairs, Science, and Technology. The health of canadians. Ottawa: Senate of Canada; 2002. 65. Romanow RJ, Commission on the future of health care in Canada. Building on values: The future of health care in Canada. Ottawa, Canada: Health Canada; 2002:i-xxxiv-1-357. 66. Canadian Institute for Health Information (CIHI). Waiting for health care in Canada: What we know and what we don't know. 2006:1-69. 67. Rachlis M. Public solutions to health care wait lists. Ottawa, Canada: Canadian Centre for Policy Alternatives; 2005:1-38. 68. Calgary Health Region Executive Management. Safety Policy - Just and Trusting Culture. Available at: http://www.calgaryhealthregion.ca/policydb/ShowPolicy?policy_id=1628. Accessed 2/19/2008, 2008. 69. Joint Commission for Accreditation of Healthcare Organizations. Joint Commission Resources: New Standard IC.4.15. Available at: http://www.jcrinc.com/subscribers/perspectives.asp?durki=12882&site=10&return=13517. Accessed Nov. 8, 2006. 70. Jenkinson H, Wright D, Jones M, et al. Prevention and control of infection in non-acute healthcare settings. Nurs Stand. 2006;20:56-63; quiz 66. 71. McGeer A. Definitions of Infection for Surveillance in Long-Term Care Facilities. Dept. of Microbiology and Infectious Disease, Mount Sinai Hospital; 1999. 72. Campbell B. Surveillance and control of infections in long-term care: The Canadian experience. Am J Med. 1991;91:286S-288S. 73. Piaskowski P, Augustin A, McArthur M, Simor A. Results of CHICA-Canada survey of long term care infection control practitioners. Can J Infect Control. 1992;7:116-119. 74. Roup BJ, Roche JC, Pass M. Infection control program disparities between acute and longterm care facilities in Maryland. Am J Infect Control. 2006;34:122-127. 75. Ford B. Personal communication regarding future publication of a study on the state of infection surveillance and control in Canadian long term care facilities. Nov. 2006.  137  76. Goldrick B, Larson E. Assessment of infection control programs in Maryland skilled-nursing long-term care facilities. Am J Infect Control. 1994;22:83-89. 77. Haley RW. Appendix B - design of the preliminary screening questionnaire and specifications for computing indexes of surveillance and control. J of Epi. 1980;111:613-621. 78. Glushkovsky E, Florescu R, Hershkovits A, Sipper D. Avoid a flop: Use QFD with questionnaires. Quality Progress. 1995;28:57-60. 79. Donabedian A. The quality of care. how can it be assessed? JAMA. 1988;260:1743-1748. 80. Nabitz U, Klazinga N, Walburg J. The EFQM excellence model: European and Dutch experiences with the EFQM approach in health care. European foundation for quality management. Int J Qual Health Care. 2000;12:191-201. 81. Russell S. ISO 9000: 2000 and the EFQM excellence model: Competition or co-operation? Total Qual Manage. 2000;11:657-665. 82. Chamberlayne R, Green B, Barer ML, Hertzman C, Lawrence WJ, Sheps SB. Creating a population-based linked health database: A new resource for health services research. Can J Public Health. 1998;89:270-273. 83. Ermer DS. Using QFD becomes an educational experience for students and faculty. Qual Prog. 1995;28:131-136. 84. Joint Commission for Accreditation of Healthcare Organizations. 2005 pre-publication edition infection control chapter for LTC - surveillance, prevention, and control of infection. 2005. 85. Scheckler WE, Brimhall D, Buck AS, et al. Requirements for infrastructure and essential activities of infection control and epidemiology in hospitals: A consensus panel report. society for healthcare epidemiology of America. Infect Control Hosp Epidemiol. 1998;19:114-124. 86. Horan-Murphy E, Barnard B, Chenoweth C, et al. APIC/CHICA-Canada infection control and epidemiology: Professional and practice standards. association for professionals in infection control and epidemiology, inc, and the community and hospital infection control associationCanada. Am J Infect Control. 1999;27:47-51. 87. Spector WD, Takada HA. Characteristics of nursing homes that affect resident outcomes. J Aging Health. 1991;3:427. 88. Palin J. Measuring health care utilization: A comparison between the Canadian community health survey and the BC linked health database. HCEP rounds presentation:2006. 89. Abramson JH. Survey Methods in Community Medicine. 4th ed. Broadway, New York: Churchill Livingstone; 1990.  138  90. Hasson F, Keeney S, McKenna H. Research guidelines for the Delphi survey technique. J Adv Nurs. 2000;32:1008-1015. 91. Brown BB. Delphi Process: A Methodology used for the Elicitation of Opinions of Experts. Rand Corp; 1968. 92. Jones J, Hunter D. Consensus methods for medical and health services research. BMJ. 1995;311:376-380. 93. Dalkey N, Helmer O. An experimental application of the Delphi method to the use of experts. Management Science. 1963;9:458-467. 94. Dalkey NC. THE DELPHI METHOD: AN EXPERIMENTAL STUDY OF GROUP OPINION. . 1969. 95. Information Services Vancouver. Residential Care Facilities - The Red Book Online electronic version of The Red Book: Directory of Services for the Lower Mainland. Available at: http://www2.vpl.vancouver.bc.ca/DBs/RedBook/orgPgs/3/3248.html2007. 96. Gardiner C, Christopher R, Bajaj A, de Jong D, Jozic J, eds. Canadian Health Facilities Directory. 4th ed. North York, Ontario: Business Information Group; 2002Canadian Cataloguing in Publication Data. 97. SHIP Seniors Housing Information Program in Vancouver, BC, Canada. Available at: http://www.seniorshousing.bc.ca/default.asp. Accessed 03/06, 2007. 98. Canadian Health Facilities Directory. Don Mills, Ontario: Southam Information Products; 2004. 99. Fraser Valley Seniors Resource Centre Society. Seniors Resource Directory. Available at: http://www.seniorsresourcecentre.ca/resources_facilities.html2007. 100. Speirs D. Senior Housing and Care Facilities - Greater Vancouver. Available at: http://members.shaw.ca/bcseniors/housing.html2007. 101. Safran DG, Kosinski M, Tarlov AR, et al. The primary care assessment survey: Tests of data quality and measurement performance. Med Care. 1998;36:728-739. 102. McHorney CA. Health status assessment methods for adults: Past accomplishments and future challenges. Annu Rev Public Health. 1999;20:309-335. 103. Sax H, Hugonnet S, Harbarth S, Herrault P, Pittet D. Variation in nosocomial infection prevalence according to patient care setting: A hospital-wide survey. J Hosp Infect. 2001;48:2732. 104. The Certification Board of Infection Control and Epidemiology, Inc. Available at: http://www.cbic.org/candidatehandbook.asp. Accessed 2/22/2008, 2008.  139  105. Potter J, Stott D, Roberts M, et al. Influenza vaccination of health care workers in longterm-care hospitals reduces the mortality of elderly patients. J Infect Dis. 1997;175:1-6. 106. National Advisory Committee on Immunization (NACI). Statement on influenza immunization for the 2006-2007 season. Public Health Agency of Canada; 2006;Canada Communicable Disease Report Volume 32 • ACS-7. 107. Nicolle LE, Bentley D, Garibaldi R, Neuhaus E, Smith P. Antimicrobial use in long-termcare facilities. Infect Control Hosp Epidemiol. 1996;17:119-128. 108. Gonzales R, Sauaia A, Corbett KK, et al. Antibiotic treatment of acute respiratory tract infections in the elderly: Effect of a multidimensional educational intervention. J Am Geriatr Soc. 2004;52:39-45. 109. Monroe S, Polk R. Antimicrobial use and bacterial resistance. Curr Opin Microbiol. 2000;3:496-501. 110. Last JM. A dictionary of epidemiology. Int J Epidemiol. 2002;15:277-277. 111. Health Quality Council of Alberta. Summary of recommendations - Review of the IPC and CSR sterilization issues in ECHR Available at: http://www.hqca.ca/index.php?id=87. Accessed 2/26/2008, 2008. 112. Banaszak-Holl J, Hines MA. Factors associated with nursing home staff turnover. Gerontologist. 1996;36:512-517. 113. Stohr K. Avian influenza and pandemics--research needs and opportunities. N Engl J Med. 2005;352:405. 114. Public Health Agency of Canada. The Canadian Pandemic Influenza Plan for the Health Sector. Available at: http://www.phac-aspc.gc.ca/cpip-pclcpi/index-eng.php. Accessed 2/27/2008, 2008. 115. Peck S. The role of a medical coordinator in extended and long term care facilities in British Columbia - A Delphi study. [Masters of Science]. University of British Columbia; 1980. 116. Bowling A. Research Methods in Health. Open University Press Philadelphia, PA; 2002. 117. Dekker S. Why we Need New Accident Models. School of Aviation, Lund Univ. 118. Collins JC. Good to Great: Why some Companies make the Leap... and Others Don't. HarperCollins; 2001. 119. McGregor MJ, Cohen M, McGrail K, et al. Staffing levels in not-for-profit and for-profit long-term care facilities: Does type of ownership matter? Can Med Assoc J. 2005;172:645-649. 120. Stone DA. Policy Paradox: The Art of Political Decision Making. New York, NY: Norton and Company; 2002. 140  121. Sheps S. Reflections on safety and interprofessional care: Some conceptual approaches. Journal of interprofessional care. Abington, Oxfordshire, OX, U.K.: Carfax Pub. Co.; 2006;20:545.  141  APPENDIX A: D R A F T 1  O F S U RV E Y ( After literature review )  Instructions: The following survey asks important questions regarding the infection surveillance and control resources available to, and activities of, your facility. We request that the senior manager most knowledgeable with your facility infection control activities complete the survey. Name and role of the person completing the survey: _______________________________ Section 1: Facility specifications 1)  2)  Which of the following categories does your facility ownership fall into? (Please circle one only) a.  Not-for-profit facility owned and operated by a regional health authority  b.  Not-for-profit facility Not owned and operated by a regional health authority  c.  For-profit facility  How many residential care beds does your facility have? ________beds  3)  Does your facility have a policy restricting admission to people who do not have a particular code status? !No !Yes, please describe the policy below: _______________________________________________________________________ ______________________________________________________________________ 4)  How many of the following categories of full-time direct-care staff work in your facility: a.  Registered Nurses: ____________  b.  Licensed Practical Nurses: ____________  c.  Care aides: ____________  d.  Other: __________ Section 2: Infection Control Resources  1)  Does your facility have a nurse or other person in charge of infection control activities in your facility? (From now on, this person will be called ICP) ! I don’t know  !Yes !No  (If no, please skip to question 9)  2)  In what year did your facility first create a position with specific infection control responsibilities? _________  3)  How many courses has your ICP taken in infection control or epidemiology? A) No courses B) One course C) More than one course D) I don’t know 142  4)  How many years of experience does your ICP have in infection control? __________years  5)  On average, how many hours per week does the ICP spend on residential care infection control activities in your facility? ___________hours per week  6)  How many residential care beds is the ICP responsible for in any facility? (Please only count beds for which the ICP has infection control responsibilities) _______beds  7)  Does your ICP participate in the following activities for your facility: a. Developing infection control policies and procedures ! I don’t know !Yes !No b. Teaching infection control policies and practices to staff ! I don’t know !Yes !No c. Providing expertise related to product selection (e.g. Hand-hygiene products, germicidal agents, wound care products) ! I don’t know !Yes !No  8)  Does your ICP have access to the following resources? a. Computer: !Yes !No b. Internet: !Yes !No  ! I don’t know ! I don’t know  c. Appropriate software for surveillance (e.g. Excel): !Yes !No ! I don’t know d. Secretarial support: !Yes !No ! I don’t know e. Journals on infection control: !Yes !No  ! I don’t know  f. BC Centre for Disease Control influenza and enteric reports: !Yes !No ! I don’t know g. Resource books on infections and infection control: !Yes !No ! I don’t know h. Health Canada guidelines for routine practices (AKA Standard Precautions) and additional precautions: !Yes N ! o ! I don’t know i. Health Canada guidelines for hand-hygiene, cleaning, disinfection and sterilization: !Yes !No ! I don’t know j. Influenza immunization recommendations from the National Advisory Committee on Immunization (NACI): !Yes !No ! I don’t know 9)  10)  Does your facility have a physician or epidemiologist serving as leader or consultant for infection control? Leader: !Yes !No ! I don’t know Consultant: !Yes !No ! I don’t know Please circle any specialty training that your physician leader of consultant has had: a) Infection control b) Epidemiology c) Microbiology or Infectious Diseases d) Not applicable e) I don’t know 143  11)  On average, how many hours per month does this physician or epidemiologist spend on infection control activities for your facility? _________hours per month  12)  Is there an infection control committee that acts as an infection control resource for your facility? !Yes !No ! I don’t know  13)  Is this committee within your facility or for the region? ! Local  ! Regional  14)  How frequently is Public Health contacted for infection control resources or for consultative advice? A) At least monthly B) Every few months C) Rarely or never D) I don’t know  15)  How frequently is the infection control program team from another acute or long term care facility contacted for infection control resources or for consultative advice? A) At least monthly B) Every few months C) Rarely or never D) I don’t know  16)  How many days per week can you obtain the following lab tests? a. Bacterial culture: ________days per week b. Viral test for influenza or gastroenteritis: __________days per week  Section 3: Infection surveillance activities Please use the letters corresponding to each of the following common infections to answer questions 1 and 2: a. Urinary tract infections b. Pneumonia c. Influenza d. Skin and soft tissue infections e. C. difficile-associated diarrhea f. Viral gastroenteritits g. Infections or colonization from antibiotic-resistant bacteria (e.g. Methicillin Resistant Staph aureus (MRSA) or Vancomycin Resistant Enterococcus (VRE)) 1)  For which infections does your facility have a written set of simple standardized definitions useful for quickly determining the presence of the infection in residents based on limited information? ________________(Enter letters from list of infections above)  2)  For which infections does your facility use a method to routinely gather information on their occurrence in residents? _____________(Enter letters from list of infections above)  3)  At the present time, how often are any of the following case-finding methods used to identify new cases of infection in your facility?  144  Method  Daily  Weekly  Monthly  Every few months  Rarely or never  A) Charts reviews by ICP B) Kardex or vital sign charts reviewed by ICP C) ICP participation in facility rounds D) Review of laboratory culture results E) Reporting of infections to ICP by staff and physicians  4)  How many unusual infection problems (e.g. outbreaks, unusual clusters, unusual or new pathogens) were identified and investigated in the past 12 months? _________  5)  How many of these problems were identified before a second resident had become infected?______. Before a third resident?_______  6)  In how many of these was there further spread of the infection to other residents after the second person had become infected? ________  7)  Is your ICP responsible for monitoring the use of antimicrobials in your facility to ensure they are appropriately used? !Yes !No ! I don’t know  8)  Does your ICP prepare at least one yearly report regarding infections in your facility? !Yes !No ! I don’t know  9)  If yes, to whom are these reports given? (Please circle all that apply) a. Facility administration b. Infection Control committee c. Facility physicians d. Facility staff e. I don’t know  10)  Are these reports reviewed and used for quality improvement in your facility? A) Always B) Sometimes C) Rarely or never D) I don’t know  145  Section 4: Infection control activities 1)  Approximately in what year was your Hand-hygiene policy established? ___________  2)  How many paid educational days per year does your ICP have? A) Zero days B) One day C) More than one day D) I don’t know  3)  Does your ICP have access to financially subsidized infection control courses or conferences? Yes No  I don’t know  4)  Approximately how many hours per year of infection control education do your patient-care staff members receive? __________ hours per year  5)  For each of the infection control policies/guidelines/programs in column A, please answer either ‘yes’, ‘no’, or ‘Don’t know’ for each question in columns B to G:  A Infection control policies, guidelines, and programs  B Do you have a written policy?  C Did the ICP participate the policy development?  D  E  F  G  Is there a system in place to teach this to staff?  Do you monitor compliance of staff with the policy?  Do you think that that staff follow this policy correctly at least 80% of the time?  Do you give visitors written educational material on this topic?  1) Hand-hygiene 2) Isolation of infectious residents 3) Cleaning and sterilization  N/A  4) Respiratory outbreak prevention & control 5) Staff influenza vaccination  N/A  6) Staff exclusion policy for influenza outbreaks  N/A  7) Resident influenza & pneumococcal vaccination  N/A  8) Pneumonia prevention  N/A  9) Urinary tract infection prevention  N/A  10) Skin and soft tissue infection prevention  N/A  146  11) Antibiotic-resistant bacterial infection prevention and control 12) C. difficile-associated diarrhea prevention and control 13) Gastroenteritis outbreak prevention & control  14) Screening of new staff (e.g. TB screening, vaccines)  N/A  6) Did you require support to answer any of the questions? !Yes !No If so, please enter the person’s name and role: _________________________________  147  APPENDIX B: BREB certificate of approval – Minimal Risk The University of British Columbia Office of Research Services Behavioural Research Ethics Board Suite 102, 6190 Agronomy Road, Vancouver, B.C. V6T 1Z3  CERTIFICATE OF APPROVAL - MINIMAL RISK PRINCIPAL INVESTIGATOR:  INSTITUTION / DEPARTMENT: UBC/Medicine, Faculty of/Health Care & Samuel B. Sheps Epidemiology INSTITUTION(S) WHERE RESEARCH WILL BE CARRIED OUT:  UBC BREB NUMBER: H06-03980  Institution  Site  BC Centre for Disease Control  BC Centre for Disease Control  Other locations where the research will be conducted:  A sample of long term residential care facilities in Fraser Health CO-INVESTIGATOR(S): Valerie Schall Bonnie Henry SPONSORING AGENCIES: N/A PROJECT TITLE: Infection Control Effectiveness and Safety: Validation of a Survey for Long Term Residential Care Facilities CERTIFICATE EXPIRY DATE: March 8, 2008 DOCUMENTS INCLUDED IN THIS APPROVAL:  DATE APPROVED: March 8, 2007  Document Name  Protocol: Research Proposal Questionnaire, Questionnaire Cover Letter, Tests: Survey draft Cover letter Cover letter Letter of Initial Contact: Invitation letter for pilot Invitation letter for pilot Invitation letter for experts Invitation letter for experts Other Documents: Fraser Health ethics board e-mailed notice of approval  Version  Date  N/A  January 18, 2007  Draft 2 Draft 2 N/A  January 18, 2007 February 10, 2007 January 18, 2007  Draft 3 Draft 2 Draft 2 Draft 3  February 10, 2007 January 18, 2007 January 18, 2007 February 10, 2007  N/A  February 6, 2007  The application for ethical review and the document(s) listed above have been reviewed and the procedures were found to be acceptable on ethical grounds for research involving human subjects. Approval is issued on behalf of the Behavioural Research Ethics Board and signed electronically by one of the following:  Dr. Peter Suedfeld, Chair 148  Dr. Jim Rupert, Associate Chair Dr. Arminee Kazanjian, Associate Chair Dr. M. Judith Lynam, Associate Chair  149  APPENDIX C: Draft 2 of survey ( After Delphi phase ) Instructions: The following survey asks important questions regarding the infection surveillance and control resources available to, and activities of, your facility. We request that the senior manager most knowledgeable with your facility infection prevention and control activities complete this survey.    Name of the primary person completing the survey: _____________________________ Role: ____________________________     Name of any person providing assistance with the survey (if applicable): ______________________ Role: ____________________________  Section 1: General Facility Information 1)  How many residential care beds does your facility have? ________beds  2)  How many of the following categories of full-time direct-care staff work in your facility: # of staff:  Registered Nurses _________  Licensed Practical Nurses __________  Care Aides ________  Other:_________ _________  Section 2: Infection Control Leadership & Resources 1)  Does your facility have a nurse or other person dedicated to infection prevention and control in your facility? (Please circle the best answer and fill any blanks for the answer selected) Yes (year position was created:______)  No (Please skip to question 9)  I don’t know  From now on, this person will be called ICP  For each question below, please circle the best answer to the right and fill any blanks: 2) How many college or university courses has your ICP taken in infection control or epidemiology? 3) How many years of experience does your ICP have in infection control? 4) Is your ICP certified in infection control? (i.e. CIC) 5) How many Full Time Equivalents (FTE) are designated for infection prevention & control in your facility? 6) To whom does your ICP report?  0 courses  1 course  >1 course  I don’t know  2 yrs or less  > 2 years  I don’t know  Yes  No  I don’t know  # of FTE:_______  Name:_____________________________________  I don’t know I don’t know  Role:_______________________________________  150  7)  Please circle all other duties your ICP has within or outside your facility? Occupational Health & Safety Charge Nurse  8)  Resident Safety  Facility Senior management  Staff Educator  ICP for other long term care facilities (# of facilities:____)  ICP for acute care facilities (# of facilities:____)  I don’t know  Does your ICP have access to the following resources at your facility? (For each resource, please select one of the answers to the right with an ‘x’) Resources  Yes  No  I don’t know  a. b. c. d. e. f. g. h.  Computer Internet Appropriate software for surveillance (e.g. Excel or EpiInfo) Secretarial support Journals on infection control Influenza and / or enteric outbreak reports for BC Resource books on infection control or epidemiology Health Canada guidelines for routine practices (AKA Standard Precautions) and additional precautions i. Health Canada guidelines for hand-hygiene, cleaning, disinfection and sterilization j. A budget dedicated to the infection control program k. Subsidized educational experiences in infection control (e.g. Workshops or conferences) 9)  How many days per week can you obtain the following lab tests? c. Bacterial culture: ________days per week d. Viral test for influenza or gastroenteritis: __________days per week e. Please describe any concerns with specimen turn-around time (i.e. the time between specimen collection and the availability of the results from the laboratory): ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________  10)  Does your facility have an infection control physician serving the following roles? (Select the best answer with an ‘x’ in one of the boxes to the right) (a) Infection Control Leader: (b) Infection Control Consultant:  Yes  No  I don’t know  Yes  No  I don’t know  If you answered ‘No’ to 10a and 10b, please skip to question 14  151  11)  Please circle every type of specialty training that your infection control physician has had: Infection control  12)  30 min to <1 hr/month  Infectious Diseases  I don’t know  1 to 4hrs/month  >4hrs/month  I don’t know  On average, your infection control physician makes reference to regional, national or international resident safety initiatives: (circle the best answer) Rarely or never  14)  Microbiology  On average, circle the number of hours per month that your infection control physician spends on infection prevention and control activities for your facility? < 30 min/month  13)  Epidemiology  Sometimes  Often or always  I don’t know  For each of the three following activities, please select the answer that best describes the frequency of participation of your ICP or infection control physician by placing an ‘x’ in one of the boxes to the right: Activities  Never  Sometimes  Usually  Always  I don’t know  (a) Development of infection control educational material for staff (b) New staff orientation (c) Providing expertise related to infection control product selection (e.g. Handhygiene products, germicidal agents, wound care products)  15)  Please select the best description of the type of communication the ICP or infection control physician usually has with the following departments in relation to infection control: None or rare Only outbreak Outbreak control & infection I don’t Departments control  prevention and control initiatives  know  (a) Housekeeping: (b) Occupational Health & Safety: (c) Nursing:  16)  Please select the best estimates of the accessibility and amount of support received from each of the following types of external infection control resources when these have been needed in the past: Never or rarely Sometimes Often I don’t Never External resources know  (a) Pharmacy  needed  Accessible Supportive  (b) Epidemiologist consultant  Accessible Supportive  (c) Public Health Unit or Medical Health Officer  Accessible Supportive 152  External resources (d) Infection control program from another facility or hospital  17)  Never or rarely  Sometimes  Often  I don’t know  Never needed  Accessible Supportive  I estimate that our infection control program is financed _________% by our facility administration, and _________% by the health authority.  Please answer each of the following questions by selecting with an ‘x’ one of the boxes to the right: Yes  No  I don’t know  18) Does your facility have a policy of admitting only residents who have a Do Not Resuscitate order? 19) Does your facility have clearly written infection control goals and objectives? 20) Are your infection control goals and objectives regularly reviewed and updated? 21) Does your facility keep records of staff immunization status? 22) Does your facility have a written plan that can readily be implemented in case of emergency (e.g. Influenza pandemic)? 23) Does this plan contain a method of increasing the number of staff and/or the number of resident beds if needed during an emergency such as an Influenza pandemic? 24) Are infection control audits routinely done in your facility to identify potential issues? 25) If yes to Question 24, how frequently are the audits done?  26)  Approximately every _______months  I don’t know  Is there an infection control committee that acts as an infection control resource for your facility?  Yes, a local committee for Yes, a regional Yes, both a local and regional our facility committee committee If there is no committee, please skip to Question 29  No committee  I don’t know  Please answer the following questions by selecting one of the boxes to the right with an ‘x’: Rarely or never  Sometimes  Often or always  I don’t know  27) On average, your infection control committee members make reference to regional, national or international resident safety initiatives: 28) Is the facility senior management actively involved in this committee? 29) Does your facility administration provide financial support for short-term infection prevention and control initiatives? 30) Does your facility administration demonstrate their support of infection prevention and control initiatives in other ways (e.g. Written support in newsletters or other correspondence)? 31) In your opinion, do staff members feel free to speak up to senior management if they see something that may negatively impact residents? 32) Are staff given feedback about changes put into place based on infection control or other safety related reports? 33) In your opinion, does senior management seriously consider staff suggestions for improving resident care?  153  Section 3: Infection Surveillance Activities For each of the infections listed on the right, please answer the following questions using these answer choices:  Influenza  Clostridium difficile associated diarrhea  • “Y” = Yes • “N” = No • “?” = I don’t know 1) Do you have a written policy to prevent transmissions?  Antibioticresistant infections (eg infections caused by MRSA or VRE)  2) Do you have educational material readily available for visitors and family? 3) Do you have a tracking system (manual or electronic) to provide the ICP with quick, easy and accurate data about these types of infections for your facility?  Quick & easy Accurate  4) Does the ICP collect data on these infections? 5) Do you have a simple, written definition that staff can use to determine whether a resident has this infection?  6) If you answered yes to Question 5, what is the source of your definitions? Our facility or group of facilities  7)  Our health region  Public Health Agency of Canada (Health Canada)  Another Canadian source  An international source  I don’t know  Please select all of the methods your ICP uses to collect data on infections and outbreaks in your facility: Chart reviews  Kardex or vital signs chart reviews  Staff reporting of infections to ICP  Participation in facility rounds  Laboratory culture reviews  Monitoring of antibiotic use  I don’t know  For each of the following questions, please select the best answer to the right. 8) How often does your facility use data collected on infections to review trends?  Rarely or never  9) How often does your facility use data collected on infections to prepare a report?  Rarely or never  Routinely every few months Every 6 to 12 months  At least monthly At least quarterly  I don’t know I don’t know  (If they are never prepared please skip to Section 4) 10) Please circle all stakeholders who are given the infection control report:  Facility administration  Infection control committee members  Facility physicians  Facility staff  Family council  11) How often is the infection control report presented to and reviewed by the infection control committee or by your facility administration?  Rarely or never  As needed only  Regularly  I don’t know  12) How often does your facility use the infection control report to implement improvements or corrective actions in the facility?  Never  Sometimes  Often  I don’t know  154  E  F  G  H  Annual resident influenza immunization  Pneumococcal immunization for residents  New staff Tuberculosis screening  I New resident Tuberculosis screening  D  Staff work restrictions during influenza outbreaks  C  Annual staff influenza immunization  “Y” = Yes “N” = No “?” = I don’t know  B  Cleaning & sterilization  • • •  A Hand hygiene  For each of the infection control issues listed on the right, please answer the following questions using these answer choices:  Additional precautions for residents with infections  Section 4: Infection Prevention & Control Activities  1) Do you have a written policy for this? (If ‘No’, skip to next column) 2) Is the policy readily available to staff in their work area? 3) Is the language used in the policy clear and concise? 4) Are there policy reminders, memory aids or checklists in the work area? 5) Did the staff participate in the policy’s development? 6) Did the ICP participate in policy development? 7) Is there a system for monitoring the compliance of staff with the policy? 8) Do you think that this policy is followed correctly at least 80% of the time by staff?  9)  If you have a written hand hygiene policy, in what year was it established? ___________  10) How frequently are your facility policies and procedures reviewed?  Rarely or never  Only when problems arise  Regularly (eg. Yearly)  I don’t know  11) How much paid educational time does your ICP receive?  None  < 1 work day per year  1 to 5 work days per year  > 5 work days per year  I don’t know  12) On average, how much paid educational time does your front-line staff receive for infection prevention and control (Includes on and off facility education)?  None  < 1 work day per year  1 to 5 work days per year  > 5 work days per year  I don’t know  13) Does your front-line staff have easy access to educational material on infection control in their work area? 14) On average how frequently do you have family meetings in which you discuss resident infection prevention and control issues?  Yes Rarely or never  Once per year  No  I don’t know  More than once per year  I don’t know  155  APPENDIX D: I N V ITAT IO N  L E T T E R F O R P IL O T S T U D Y  (P H A S E 3) DEPARTMENT OF HEALTH CARE AND EPIDEMIOLOGY Faculty of Medicine Mather Building, 5804 Fairview Avenue Vancouver, B.C. V6T 1Z3 Tel: (604) 822-2772 Fax: (604) 822-4994 Website: www.healthcare.ubc.ca  We are writing to enlist the help of the senior manager most knowledgeable of infection prevention and control from within your facility in validating a new survey based on the 2004 Public Health Agency of Canada infection control recommendations for long term care, as well as the five principles of patient safety from the Institute of Medicine. This survey is being developed and validated for the purpose of evaluating structures and processes that may enable long term care (LTC) facilities to prevent and control infections. Numerous studies have shown the utility and effectiveness of infection control programs in acute care hospitals, but almost none have done so for LTC facilities. Because of this shortage of research, no surveys have been developed and validated specifically for LTC. Existing surveys have largely been based on extrapolations from structures and processes shown to be effective in acute care. This has occurred even though all expert groups agree that infection prevention and control programs and processes in LTC and acute care need to and should be fundamentally different. If you are interested in participating in this study the 7-page survey and a cover letter will be sent to you by electronic mail. The survey can be completed and returned electronically, or printed and sent by Canada Post. The research team consists of Drs. Samuel Sheps and David Matheson from UBC, Dr. Bonnie Henry from the BC Centre for Disease Control, and Valerie Schall, Master’s student in Health Care and Epidemiology at UBC, and the study is being done in collaboration with BC’s Provincial Infection Control Network. The analysis of this survey will be part of a graduate thesis and is not being financed by an external grant. If you have any concerns about your rights during the study, please phone the Research Subject Information Line in the UBC Office of Research Services at 604-822-8598. Please let us know by e-mail reply to ___________ if you are interested in participating in this important project. Please also feel free to e-mail any member of the research team if you have any questions about this project. On behalf of the research team, I would like to thank you for your consideration, Valerie Schall Sam Sheps, MA, MD, FRCPC Director, WRTC Prof. & Director MSc/PhD program, Dep. Health Care and Epidemiology, UBC  David Matheson, M.Math, MD, FRCPC DMMD Consultants Inc. Associate Professor, UBC Vancouver, BC, Canada  Bonnie Henry, MDMPH, FRCPC, Physician Epidemiologist BCCDC Epi. Services Associate Professor, UBC Vancouver BC, Canada  Valerie Schall, RN, BScN Masters student Health Care and Epi., UBC ICP, Fraser Health, B.C, Canada  156  APPENDIX E: C O V E R  L E T T E R F O R P IL O T S T U D Y  (P H A S E 3)  DEPARTMENT OF HEALTH CARE AND EPIDEMIOLOGY Faculty of Medicine Mather Building, 5804 Fairview Avenue Vancouver, B.C. V6T 1Z3 Tel: (604) 822-2772 Fax: (604) 822-4994 Website: www.healthcare.ubc.ca  November 19th, 2007  STUDY TITLE: Infection Prevention & Control Effectiveness and Safety: Validation of a Survey for Long Term Care Facilities  We are inviting you to participate in a study that will help to establish the validity and reliability of a new survey based on the 2004 Public Health Agency of Canada infection control recommendations for long term care (LTC), as well as the five principles of patient safety from the Institute of Medicine. This survey is being developed and validated for the purpose of evaluating structures and processes that may enable long term care (LTC) facilities to prevent and control infections. Please complete the survey that has just sent to you electronically in a fillable PDF format. You will find that other questions have been posed at the end of the survey so that you can provide us with your feedback. Completion of the survey should take approximately 30 minutes, and the completed self-administered survey can be automatically sent back to the research team by pressing the “Submit by Email” button on the first page of the survey (the return e-mail is preprogrammed). Please note that you cannot save your responses electronically unless you have Acrobat Professional installed on your computer. If you wish to save a copy of your responses or if you prefer to enter your responses manually, please press the “Print Form” button on Page 1 of the survey. In order to ensure confidentiality, completed questionnaires will be kept secured and shared only with our research team members. The analysis of this survey will be part of a graduate thesis and is not being financed by an external grant. Please note that, when questionnaires are completed and returned, we will assume that consent has been given for participation in the study. Follow-up calls may be made to answer any questions and to ensure that surveys are returned within two weeks. Study findings will be made available to all participants as soon as the study is completed. All investigators are listed below with their contact information. If you have any questions regarding this study, please contact Valerie Schall or any other member of the research team using the contact information provided below. If you have any concerns about your rights during the study, please phone the Research Subject Information Line in the UBC Office of Research Services at 604-822-8598. The entire research team wishes to thank you for your interest and participation.  Sam Sheps, MA, MD, FRCPC Director, WRTC Prof. & Director MSc/PhD program, Dep. Health Care and Epidemiology, UBC  David Matheson, M.Math, MD, FRCPC DMMD Consultants Inc. Associate Professor, UBC Vancouver, BC, Canada  Bonnie Henry, MDMPH, FRCPC, Physician Epidemiologist BCCDC Epi. Services Associate Professor, UBC Vancouver BC, Canada  Valerie Schall, RN, BScN Masters student Health Care and Epi., UBC ICP, Fraser Health, B.C, Canada  157  APPENDIX F: Q U E S T IO N 1 – R E F O R M AT T E D W O R K S H E E T 1 Question 1: Looking at the 2nd and 3rd rows of each matrix, are all key infection control structural and process components and concepts listed for each dimension? _____________________________________________________________________  Leadership index: Row 1  Index Name  Row 2  Key concepts  Row 3  Concept components  Row 4  Operationalized components  Suggested change #1: Comments regarding suggestion  Leadership Index Physician  IC Committee  Role  Time  Education  Leader vs consultant  Hours per month  One of (micro, epi, ID, IC)  Degree of leadership provided  Presence (Y/N)  Local vs regional  Does "Degree of leadership provided" incorporate reporting structure? Or should it be listed separately? •  When developing the matrix, it was recognized that reporting is an important piece for many of the components within multiple indices (i.e. the chain of communication is important for all structural and process components). Therefore the plan was to directly measure this component in the surveillance index. This index was chosen because reporting is one of the 3 key activities described in the CDC definition for surveillance.  •  Reporting structure is a good indicator of the stakeholders involved how decisions are made regarding infection control processes.  Options for change  1. Add “Reporting Structure” as an additional concept of “IC committee” (in addition to “Degree of leadership provided”). 2. Do not add “Reporting Structure” as it is adequately measured in the Surveillance Index under “Report access by stakeholders” and “Use of report for Continuous Quality Improvement”.  Your vote (and any comments)  Suggested change #2: Comments regarding suggestion  Would administrative support be considered here? •  In my experience our admin leaders have influenced our direction.  •  The assumption when developing the matrix was that a local infection control committee would involve the facility administrative staff; perhaps this is a false assumption and their involvement needs to be clearly articulated.  •  Admin support is, of course, essential, and I think these comments are trying to get at the same issue as I was with suggestions about including reporting structure. There should be some way to measure this.  158  Options for change  1. Add “Administrative Support” as an additional concept of “IC committee” (in addition to “Degree of leadership provided”). 2. Do not add “Administrative Support” because this support is implied if the facility has a local IC committee.  Your vote (and any comments)  Suggested change #3: Comments regarding suggestion  What about Pharmacy leadership? •  We have had a lot of information re antibiotic resistance etc. from them.  •  This may be better positioned in Partnership and Resources?  •  Admin support is, of course, essential, and I think these comments are trying to get at the same issue as I was with suggestions about including reporting structure. There should be some way to measure this.  Options for change  1. Add “Pharmacy” as a component of the Leadership index 2. Leave “Pharmacy” out as it is a component of the Partnership & Resources index. If “Reporting Structure” is added to the Leadership index, assess reporting to Pharmacy.  Your vote (and any comments)  Suggested change #4: Comments regarding suggestion  What about a dedicated epidemiologist? Or is the presence of one assumed? •  The “physician” component was supposed to provide information about an epidemiologist (see row 4): “Education” assesses the specialty of the physician or any infection control courses they have taken. Also, “Role” differentiates between a dedicated physician and a consultant.  •  I believe a lot of the comments re epidemiologist would fall under "Partnerships and resources" - although not many LTCF would have access to an epidemiologist.  Options for change  1. Add “Epidemiologist” as a separate component from “Physician” 2. Do not add “Epidemiologist” as a separate component because it is already measured under “Physician”  Your vote (and any comments)  Suggested change #5: Comments regarding suggestion  Change physician education. •  I believe most of the answers will be "none" - in LTC, most physicians are Family Physicians, not specialists and have not had "infection control training".  •  This is exactly the type of data we are trying to gather regarding physician leaders or  159  consultants. The fact that most do not have infection control related infection is important considering it is an important recommendation from Health Canada. In addition, it has been shown to be one of the most important factors in successfully preventing and controlling infections in acute care. Options for change  1. Modify the “Education” component for Physician (If you select this choice please describe the change you feel is needed) 2. Do not modify the “Education” component for Physician  Your vote (and any comments)  160  ICP index: Row 1  Index Name  ICP Index  Row 2  Key concepts  ICP  Row 3  Concept components  Row 4  Operationalized components  Suggested change #1: Comments regarding suggestion  Time  Education  Experience  Hours per 100 LTRC beds per week  Courses in IC, epi (0, 1, >1)  In years (large > 2 yrs)  Should components include certification and professional association? •  I think certification is important; professional association may be interesting and help to identify organizations support of the infection control program and processes.  •  Certification and professional association were left out when developing this index because there is no data to suggest that they have an affect on infection rates. Moreover, there is no Canadian source of certification.  Options for change  1. Add “Certification and Professional Association” as a component of ICP 2. Do not add “Certification and Professional Association” as a component of ICP  Your vote (and any comments)  Suggested change #2: Comments regarding suggestion  Add a code to the abbreviations •  Options for change  It is time consuming to look back and forth to a code, but it does save space on the survey 1. Add a code 2. Use limited acronyms that are defined at the beginning of the survey (e.g. ICP or IP&CP), and spell out all other words.  Your vote (and any comments)  Suggested change #3: Comments regarding suggestion  Change ICP to IP&CP •  CHICA now, as have many of the health Authorities have changed our titles to Infection Prevention and Control, with more focus put on the prevention aspect of the job.  •  We are asking that senior managers of the facilities answer the survey. These professionals are unlikely to be aware of the new terminology and the new acronym may make the survey more confusing for them (and therefore may decrease the reliability of the survey responses).  161  Options for change  1. Change ICP to IP&CP 2. Do not change ICP to IP&CP  Your vote (and any comments)  162  Policy & Strategy index: Row 1  Index Name  Policy & Strategy Index  Row 2  Key concepts  Policies & Procedures  Row 3  Concept components Development  Row 4  Operationalized components  Written policy (Y/N)  Suggested change #1:  Quality (ICP or consultant involvement) (Y/N)  Product selection quality  Implementation  Staff education (Y/N)  Estimate of compliance (proportion of time staff are compliant)  Monitoring system for compliance (Y/N)  ICP participation (Y/N)  No Changes  Comments regarding suggestion Options for change  1. Accept index as is 2. Review index again with the group  Your vote (and any comments)  163  Partnership & resources index:  Suggested change #1:  BCCDC outbreak reports  Books  Journals Access to any IC journal (Y/N)  Hand hygiene (Y/N)  Routine & additional precautions (Y/N)  One of GI or influenza regular reports  National guidelines Access to any book on IC or epi  Access (days per week)  Bacteriology  Virology  Access (Y/N)  Educational material  Access (days per week)  Secretarial support  Access (Y/N)  Internet Access (Y/N)  Lab services  Appropriate software for surveillance  Computer  IT for ICP  Access (Y/N)  Liaison with PH  Row 4  Operationalized components  Row 3  Partnership support  Frequency  Key conce pts  IC program in other facility  Row 2  Partnership & Resources Index  Frequency  Index Name  Concept component s Support from  Row 1  Under "Partnership support" - what about Occupational Health?  Comments regarding suggestion Options for change  1. Add “Liaison with Occupational Health” as an additional component of “Partnership support” 2. Do not add “Partnership with Occupational Health Staff” as an additional component of “Partnership support”  Your vote (and any comments)  Suggested change #2: Comments regarding suggestion Options for change  Not sure whether Pharmacy/Epidemiologist relationships would be here versus leadership? •  I think these should be included in this section. Epidemiologists are not common outside the lower mainland. Some areas do not have Medical Microbiologists either. 1. Add “Partnership with Pharmacy” as an additional component of “Partnership Support” 2. Remove the “Physician” component of the Leadership index and add “Epidemiologist consultant” as an additional component of “Partnership Support” 3. Both options 1 and 2 4. Neither options 1 nor 2  Your vote (and any comments)  164  Surveillance index: Row 1  Index Name  Row 2  Key concepts  Row 3  Concept components  Row 4  Surveillance Index Data collection Simple written standardized definitions for infections  Operationalized components  Use (Y/N)  Suggested change #1: Comments regarding suggestion  Surveillance for high M&M infections Case finding using one of: Charts, Kardex, vitals record, lab, rounds, staff report  Types of infections: pneumonia, UTI, MRO, RI, viral gastro, CDAD, flu, skin infections  Data analysis  Reporting  Monitoring of Antibiotic use  Early detection of outbreaks  Report prepared  Report access by stakeholders  Use of report for CQI  Done (Y/N)  Number of unusual infections or outbreaks detected after 1 case or 2 cases  At least one per year (Y/N)  Access of: Staff, physicians, administration, IC committee  Report review and improveme nts made (Y/N)  Another source of case finding is through "antibiotic reviews" which are often a computer printout from pharmacy. •  This is important because in LTC, diagnostic tests are often not done, there are often too many kardexs and charts to review on a regular basis and often charting is incomplete.  •  This is already being measured in the component “Monitoring of Abx Use”  Options for change  1. Change the component “Monitoring of ABX Use” to specify antibiotic reviews. If this option is selected, please specify the recommended change. 2. Do not change the component “Monitoring of ABX Use”.  Your vote (and any comments)  Suggested change #2: Comments regarding suggestion  Options for change  Under "Data Analysis", a key concept would be identifying trends in routine surveillance data, implementation of corrective actions etc. •  This is more of a long term view rather than simply "Early detection of outbreaks". Although I can understand why "# of outbreaks" appears both here and in the "Control Process Index", I think the identification of more subtle trends is critical.  •  Surveillance using long term trends is being measured indirectly under “Reporting”. If reports are regularly prepared, these are then distributed to stakeholders such as physicians, staff, administration and the IC committee, and the report is regularly reviewed and improvements are made based on those reports, then we are seeing that subtle long term trends are used for continuous quality improvement (CQI). 1. Add “Trend review and corrective action” as an additional component of “Data Analysis” and keep “Reporting” component as is. 2. Move “Use of Report (or trends) for CQI” from “Reporting” to “Data Analysis” 3. Do not make the above changes  Your vote (and any comments)  165  Control index:  Suggested change #1: Comments regarding suggestion  Options for change  Yearly staff flu immunization program  Influenza prevention Work restrictions during outbreaks (Y/N)  New staff screening TB skin test (Y/N)  Immunization  Occupational transmission prevention  Immunization status (Y/N)  Admission screening  One time pneumococcal (Y/N)  Staff  Yearly flu (Y/N)  ICP  Visitors & family  TB (Y/N)  Rapid control  Preventative care for residents  Education  Availability of material  Row 4  Concept components  Operationalized components  Row 3  Outbreak prevention & management  Access (hours per year)  Key concepts  Paid education time (0, 1 day, >1 day)  Row 2  Control Index  Access to IC courses or conf. (Y/N)  Index Name  Number of outbreaks detected after 2 cases  Row 1  Should we measure adequacy of communication system? Both internal to the facility and back to the IC team. •  I agree with the communications comment - i.e. does IC report rates to front line staff? This is often reported as one of the more effective means of lowering infection rates. Communicating with the IC committee gets back to the Leadership Index.  •  “Report Access by Stakeholders” in the “Reporting” component of the Surveillance index should already measure this. It could also be moved or added to the Leadership index but I don’t want to do too much overlapping of indices. 1. Add a communication component to the Control index and do not change the Surveillance index (i.e. Do not remove the “Report Access by Stakeholders” component). If you select this option, please describe the recommended modification. 2. Move the “Report Access by Stakeholders” component from the Surveillance index to the Control index. 3. Do not make the above changes  Your vote (and any comments)  166  APPENDIX G: Q UESTION 2 – S AFETY P RINCIPLE 1: P ROVIDE  LEADERSHIP  Black = Named by at least 50%, Grey = Named by at least 33%, Light orange = Named by <20% Leadership Index: 1. Physician –Role –Time –Education 2. IC reporting structure -Reporting to administration -Financial support 3. IC committee –Active participation of senior management –Degree of leadership provided  Partnership & Resources Index: 1. Partnership support –Support from IC program in other facility –Partnership with pharmacy –Epidemiologist consultant –Liaison with OHS –Liaison with PH 2. IT for ICP –Computer –Internet –Appropriate software for surveillance –Secretarial support 3. Lab services –Virology –Bacteriology  Surveillance Index: 1. Data collection –Simple standardized definitions for infections –Surveillance for high M&M infections –Monitoring antimicrobial use 2. Data analysis–Trend review & corrective action –Early detection of outbreaks 3. Reporting –Report preparation –IC report provided to stakeholders –Use of report for CQI  4. Educational material –Journals –Books –BCCDC outbreak reports –National guidelines  ICP Index: 1. ICP -Time -Education -Additional duties -Experience  Policy & Strategy Index: 1. Policies & procedures –Development –Implementation –Product selection quality  Control Index: 1.  Outbreak prevention & management –Rapid staff reporting of suspected outbreak –Rapid control 2. Education –ICP –Staff –Family & visitors 3.  Preventative care for residents –Admission screening –Immunization 4.  Occupational transmission prevention –New staff screening –Influenza prevention  167  APPENDIX H: Q UESTION 2 – S AFETY P RINCIPLE 2: R ESPECT  HUMAN LIMITS IN  PROCESS DESIGN  Black = Named by at least 50%, Grey = Named by at least 33%, Light orange = Named by <20% Leadership Index: 1. Physician –Role –Time –Education 2. IC reporting structure -Reporting to administration -Financial support  3. IC committee –Active participation of senior management –Degree of leadership provided  Partnership & Resources Index: 1. Partnership support –Support from IC program in other facility –Partnership with pharmacy –Epidemiologist consultant –Liaison with OHS –Liaison with PH 2. IT for ICP –Computer –Internet –Appropriate software –Secretarial support 3. Lab services –Virology –Bacteriology  Surveillance Index: 1. Data collection –Simple standardized definitions for infections –Surveillance for high M&M infections –Monitoring antimicrobial use 2. Data analysis–Trend review & corrective action –Early detection of outbreaks 3. Reporting –Report preparation –IC report provided to stakeholders –Use of report for CQI  4. Educational material –Journals –Books –BCCDC outbreak reports –National guidelines  ICP Index: 1. ICP •Time •Education •Experience •Additional duties  Policy & Strategy Index: 1. Policies & procedures –Development –Implementation –Product selection quality  Control Index: 1.  Outbreak prevention & management –Rapid staff reporting of suspected outbreak –Rapid control 2. Education –ICP –Staff –Family & visitors 3. Preventative care for residents –Admission screening –Immunization 4.  Occupational transmission prevention –New staff screening –Influenza prevention  168  APPENDIX I: Q UESTION 2 – S AFETY P RINCIPLE 3: P ROMOTE  EFFECTIVE TEAM  FUNCTIONING  Black = Named by at least 50%, Grey = Named by at least 33%, Light orange = Named by <20% Leadership Index: 1. Physician –Role –Time –Education 2. IC reporting structure -Reporting to administration -Financial support  3. IC committee –Active participation of senior management –Degree of leadership provided  Partnership & Resources Index: 1. Partnership support –Support from IC program in other facility –Partnership with pharmacy –Epidemiologist consultant –Liaison with OHS –Liaison with PH 2. IT for ICP –Computer –Internet –Appropriate software –Secretarial support 3. Lab services –Virology –Bacteriology  Surveillance Index: 1. Data collection –Simple standardized definitions for infections –Surveillance for high M&M infections –Monitoring antimicrobial use 2. Data analysis–Trend review & corrective action –Early detection of outbreaks 3. Reporting –Report preparation –IC report provided to stakeholders –Use of report for CQI  4. Educational material –Journals –Books –BCCDC outbreak reports –National guidelines  ICP Index: 1. ICP •Time •Education •Experience •Additional duties  Policy & Strategy Index: 1. Policies & procedures –Development –Implementation –Product selection quality  Control Index: 1.  Outbreak prevention & management –Rapid staff reporting of suspected outbreak –Rapid control 2. Education –ICP –Staff –Family & visitors 3. Preventative care for residents –Admission screening –Immunization 4.  Occupational transmission prevention –New staff screening –Influenza prevention  169  APPENDIX J: Q UESTION 2 – S AFETY P RINCIPLE 4: A NTICIPATE  THE UNEXPECTED  Black = Named by at least 50%, Grey = Named by at least 33%, Light orange = Named by <20% Leadership Index: 1. Physician –Role –Time –Education 2. IC reporting structure -Reporting to administration -Financial support  3. IC committee –Active participation of senior management –Degree of leadership provided  Partnership & Resources Index: 1. Partnership support –Support from IC program in other facility –Partnership with pharmacy –Epidemiologist consultant –Liaison with OHS –Liaison with PH 2. IT for ICP –Computer –Internet –Appropriate software –Secretarial support 3. Lab services –Virology –Bacteriology  Surveillance Index: 1. Data collection –Simple standardized definitions for infections –Surveillance for high M&M infections –Monitoring antimicrobial use 2. Data analysis–Trend review & corrective action –Early detection of outbreaks 3. Reporting –Report preparation –IC report provided to stakeholders –Use of report for CQI  4. Educational material –Journals –Books –BCCDC outbreak reports –National guidelines  ICP Index: 1. ICP •Time •Education •Experience •Additional duties  Policy & Strategy Index: 1. Policies & procedures –Development –Implementation –Product selection quality  Control Index: 1.  Outbreak prevention & management –Rapid staff reporting of suspected outbreak –Rapid control 2. Education –ICP –Staff –Family & visitors 3. Preventative care for residents –Admission screening –Immunization 4.  Occupational transmission prevention –Screening of new staff –Influenza prevention  170  APPENDIX K: Q UESTION 2 – S AFETY P RINCIPLE 5: C REATE  A  LEARNING ENVIRONMENT  Black = Named by at least 50%, Grey = Named by at least 33%, Light orange = Named by <20% Leadership Index: 1. Physician –Role –Time –Education 2. IC reporting structure -Reporting to administration -Financial support 3. IC committee -Active participation of senior management –Degree of leadership provided  Partnership & Resources Index: 1. Partnership support –Support from IC program in other facility –Partnership with pharmacy –Epidemiologist consultant –Liaison with OHS –Liaison with PH 2. IT for ICP –Computer –Internet –Appropriate software –Secretarial support 3. Lab services –Virology –Bacteriology  Surveillance Index: 1. Data collection –Simple standardized definitions for infections –Surveillance for high M&M infections –Monitoring antimicrobial use 2. Data analysis–Trend review & corrective action –Early detection of outbreaks 3. Reporting –Report preparation –IC report provided to stakeholders –Use of report for CQI  4. Educational material –Journals –Books –BCCDC outbreak reports –National guidelines  ICP Index: 1. ICP •Time •Education •Experience •Additional duties  Policy & Strategy Index: 1. Policies & procedures –Development –Implementation –Product selection quality  Control Index: 1.  Outbreak prevention & management –Rapid staff reporting of suspected outbreak –Rapid control 2. Education –ICP –Staff –Family & visitors 3. Preventative care for residents –Admission screening –Immunization 4.  Occupational transmission prevention –Screening of new staff –Influenza prevention  171  APPENDIX L: I NDICES  AFTER  Q UESTION 4 Leadership Index  Physician  Role  IC leadership structure  Time  Leader vs consultant  Education Makes reference to regional, national or < 30 min/month, 30 min international to <1hr/month, 1 to One of resident safety <5hrs/month, >5 micro, epi, initiatives (Rarely, hrs/month ID, IC) occasionally, often)”  Reporting to administration  Clear assignment of IC oversight  Yes/No  Yes/No  IC Committee  Financial support  Active participation of senior management  Percentage of Program financed by facility admin & HA  Yes/No  Knowledge & understanding of regional, national or Degree of international resident safety leadership initiatives provided  Local vs regional  Yes/No  ICP Index ICP Time  Education  Resident / family partnership  Within facility partnership  Additional duties  # of FTE (OHS, resident ICPs per Courses in safety, admin, number of Certification IC, epi (0, 1, educator, charge (CIC) >1) beds nurse)  Experience In years (large > 2 yrs)  Housekeeping  Liaison with OHS  Resident / family / staff meetings  Nursing  Type of communication related to Type of communication related to IC (A)None, B)only outbreak IC (A)None, B)only outbreak control, C)IP&C initiatives and control, C)IP&C initiatives and outbreak control outbreak control  Type of communication related to IC (A)None, B)only outbreak control, C)IP&C initiatives and outbreak control Average frequency  Policy & Strategy Index  Written IC G&O (Y/N)  Pandemic / Emergency plan  Plan in place for facility (Y/N)  Plan for surge capacity  Clear IC G&O  G&O reviewed & updated regularly (Y/N)  Blameless reporting of errors Initiative supported by Admin (Y/N)  Initiative supported by Admin (Feedback given to staff on changes made based on IC reports)  Open communication flow within facility  Admin support demonstrated regarding IC concerns from staff (Rarely sometimes, often)  Admin support for IC initiatives demonstrated to staff (eg. In newsletter or other correspondence) (Rarely, sometimes, often)  Financial support of IC initiatives (Y/N)  IC promotion by Admin  ICP participation (Y/N)  Quality (ICP or consultant involvement) (Y/N)  Regularly scheduled reviews (Y/N)  Staff participation (Y/N)  Checklists (Y/N)  Clear, concise language (Y/N)  Written policies for a subset of key IC issues (Y/N)  Readily available in work area (Y/N)  Product Implementati Selection on quality  Development  Emergency Preparedness Strategy  Safety Culture  Plan in place for facility (Y/N)  Policies & Procedures  172  Influenza prevention Estimate of compliance Monitoring system for compliance  173  Yes/No  Hand hygiene (Y/N)  Routine & additional precautions (Y/N)  Educational material  Frequency  Occupational transmission prevention  Improvements made based on report (Y/N)  IC report provided to stakeholders  Done (Y/N)  Surveillance Index Data analysis  Books BCCDC outbreak reports  Monitoring system in place (Y/N)  Journals  Access to any book on IC or epi (Y/N) One of GI or influenza regular reports (Y/N)  Access to any IC journal (Y/N)  Bacterio.  Frequency of report presented to and reviewed by IC committee or admin (Never, as needed only, yearly and as needed, quarterly and as needed, more than quarterly)  Report prepared  Given to: Staff, physicians, administration, IC committee  Lab services  Estimate (from management) that staff follow policy >80% of time  New staff screening  Access (days per week)  Access (days per week)  Appropriate software for Secretarial surveillance support Virology  Yearly staff flu vac. Program (Y/N)  Admission screening Immunization  At least quarterly, Every 6-12 months, Rarely or never, I don't know  Early detection of outbreaks  Written policy for work restrictions during outbreaks (Y/N)  Preventative care for residents  Access (Y/N)  Access (Y/N)  IT for ICP  TB skin test (Y/N)  Corrective action  Immun. status (Y/N)  Visitors & family  Done routinely (Y/N) For influenza outbreaks in last 24 months, time from onset of 2nd case to report to ICP or PH (Percentage reported)  Monitoring Trend of Abx use review  One time pneumo (Y/N)  Education  Access (Y/N)  Partnership with PH Computer Internet  Yearly flu (Y/N)  Data collection  Done routinely (Y/N)  Access (Y/N)  Supportive (Rarely, sometimes, often)  External Partnership support  TB (Y/N)  Staff Done (Y/N)  Yes/No  Quick & easy access to accurate data  Avail. of material (on influenza, gastro, ARO, C-diff, UTI, pneumo, TB )  Access (paid hours per year)  ICP  Accessible (Rarely, sometimes, often)  Accessible (Rarely, sometimes, often) Supportive (Rarely, sometimes, often)  Supportive (Rarely, sometimes, often)  Accessible (Rarely, sometimes, often)  Supportive (Rarely, sometimes, often) Partnership with pharmacy Epi. Consultant  Materials used are approved by IC (Y/N)  Types of infections: influenza, CDAD, ARO  Surveillance for high M&M infections  Access to IC materials (on ward?)  Outbreak control  ICP involvement in new staff IC orientation (frequency)  Rapid control  Paid educ. time (0, 1 day, >1 day)  Simple written standardized definitions for infections  Case finding using one of: Charts, Kardex, vitals record, lab, rounds, staff report  Source  Use (Y/N)  Accessible (Rarely, sometimes, often)  Support from IC program in other facility  Access to IC courses or conf. (Y/N)  Average attack rate during influenza outbreaks in last 12 & 24 months  Average length of influenza outbreaks (in days) in last 12 & 24 months  Partnership & Resources Index Budget  National Dedicated guidelines to IC  Reporting Use of report for CQI  Control Index  Error analysis Regular IC audits  APPENDIX M: Matrix of scatterplots – Correlation between number of beds and staffing The following scatterplots and linear regression analyses describe the weak association found between the number of beds of sample facilities (the x-axis if looking down the left column) and the following variables: numbers of Registered Nurses (RN), Licensed Practical Nurses (LPN), Care Aides, RNs plus LPNs, and all front-line staff.  In a scatterplot, a correlation is identified when there is a distinctive pattern, such as a line, formed by the dots. The R-squared value of 0.33 shows that this regression model (and therefore the staffing variables) explains little of the variability in the number of beds reported by facility Senior Managers. The large p-values also show that there is no significant association between any of the variable and the number of beds. Models using each of the variables described in the five scatterplots individually also corroborated these findings: no significant association was found.  Linear regression results from a model with numbers of Registered Nurses (RNs), Licenced Practical Nurses (LPNs), and Care Aides as explanatory variables for the number of beds in sample facilities Coefficients Value Standard Error t-test statistic P-value 59.96 21.89 2.74 0.03 Intercept 2.74 Number of RNs 2.38 1.15 0.28 -1.97 Number of LPNs 2.37 -0.83 0.43 0.56 Number of Care Aides 0.43 1.28 0.24 F-statistic 1.30 0.34 R-squared 0.33  174  175  APPENDIX N: Draft 3 of survey ( After pilot study ) Instructions: The following survey asks important questions regarding the infection surveillance and control resources available to, and activities of, your facility. We request that the senior manager most knowledgeable with your facility infection prevention and control activities complete this survey.    Name of the primary person completing the survey: _____________________________ Role: ____________________________     Name of any person providing assistance with the survey (if applicable): ______________________ Role: ____________________________  Section 1: General Facility Information 1)  How many open residential care beds does your facility have? ________beds  2)  How many Full-Time Equivalents (FTE) does your facility have of the following categories of directcare staff: Total # FTE for facility:  Registered Nurses _________  Licensed Practical Nurses __________  Care Aides ________  Section 2: Infection Control Leadership & Resources 1)  Is there a nurse or other person primarily responsible for the infection prevention and control program in your facility?  Yes, someone working in our  Yes, we have an ICP from the  No, facility has no primary person  facility has this mandate  health authority  responsible for this  I don’t know  If yes, this person will be called ICP from now on. If no, please skip to question 11  For each question below, please circle the best answer to the right and fill any blanks: 2) If yes to question 1, when was this ICP position firs created? 3) How many college or university courses has your ICP taken specifically in infection control or epidemiology? 4) How many years of experience does your ICP have specifically in the infection prevention and control role?  In the year:_________ 0 courses  >1 course  I don’t know  2 yrs or less  > 2 years  I don’t know  Yes  No  I don’t know  5) Is your ICP certified in infection prevention and control? (i.e. CIC) 6) How many Full Time Equivalents (FTE) does your ICP work? (Your ICP may cover numerous facilities, but please enter their total FTEs working as an ICP anywhere) 7) To which member of your administration or management does your ICP report issues?  1 course  # of FTE:_______  I don’t know  Name:_____________________________________ Role:______________________________________  176  8)  If you share an ICP with other facilities (e.g. you have an ICP from the health authority), please state the number of facilities they cover: a. Number of acute care hospitals: ________ b. Number of long term care facilities: _________  9)  If your ICP works from within your facility, please circle all other duties your ICP has within or outside your facility?  Occupational Health & Safety Charge Nurse  10)  Resident Safety ICP for other long term care facilities (# of facilities:____)  Facility Senior management Other duties: ___________________  Does your ICP have access to the following resources either at your facility or in their external office? (For each resource, please select one of the answers to the right with an ‘x’) Resources  a. b. c. d. e. f. g. h. i. j. k.  11)  Staff Educator (not related to infection prevention & control) I don’t know  Yes  No  I don’t know  Computer Internet Appropriate software for surveillance (e.g. Excel or EpiInfo) Secretarial support Journals on infection control Influenza and / or enteric outbreak reports for BC Resource books on infection control or epidemiology Health Canada guidelines for routine practices (AKA Standard Precautions) and additional precautions Health Canada guidelines for hand-hygiene, cleaning, disinfection and sterilization A budget dedicated to the infection control program Subsidized educational experiences in infection control (e.g. Workshops or conferences) How many days per week can you obtain the following lab tests? f. Bacterial culture: ________days per week g. Viral test for influenza or gastroenteritis: __________days per week h. Please describe any concerns with specimen turn-around time (i.e. the time between specimen collection and the availability of the results from the laboratory): ______________________________________________________________________________  12)  Please circle one type of infection control physician support your facility has (this question refers to one particular person with this mandate).  Infection Control Physician Leader  Infection Control Physician Consultant (does not lead but is  No Infection Control Physician support  I don’t know  consulted when needed) If yes, from now on this person will be called your infection control physician. If no, please skip to question 15  177  13)  Please circle every type of specialty training that your infection control physician has had: Infection control  14)  30 min to <1 hr/month  Infectious Diseases  I don’t know  1 to 4hrs/month  >4hrs/month  I don’t know  On average, your infection control physician makes reference to regional, national or international resident safety initiatives: (circle the best answer) Rarely or never  16)  Microbiology  On average, circle the number of hours per month that your infection control physician spends on infection prevention and control activities for your facility? < 30 min/month  15)  Epidemiology  Sometimes  Often or always  I don’t know  For each of the three following activities, please select the answer that best describes the frequency of participation of your ICP or infection control physician by placing an ‘x’ in one of the boxes to the right: Activities  Never  Sometimes  Usually  Always  I don’t know  (a) Development of infection control educational material for staff (b) New staff orientation (c) Providing expertise related to infection control product selection (e.g. Handhygiene products, germicidal agents, wound care products)  17)  Please select the best description of the type of communication the ICP or infection control physician usually has with the following departments in relation to infection control: None or rare Only outbreak Outbreak control & infection I don’t Departments control  prevention and control initiatives  know  (a) Housekeeping: (b) Occupational Health & Safety: (c) Nursing:  18)  Please select the best estimates of the accessibility and amount of support received from each of the following types of external infection control resources when these have been needed in the past: Never or rarely Sometimes Often I don’t Never External resources know  (a) Pharmacy  needed  Accessible Supportive  (b) Epidemiologist consultant  Accessible Supportive  (c) Public Health Unit or Medical Health Officer (d) Infection control program from another facility or acute care hospital  Accessible Supportive Accessible Supportive  178  19)  I estimate that our infection control program (i.e. ICP, infection control physician, infection control strategies, etc…) is financed _________% by our facility administration, and _________% by the health authority.  Please answer each of the following questions by selecting with an ‘x’ one of the boxes to the right: Yes  No  I don’t know  20) Does your facility have a policy of admitting only residents who have a Do Not Resuscitate order? 21) Does your facility have clearly written infection control goals and objectives? 22) Are your infection control goals and objectives regularly reviewed and updated? 23) Does your facility keep records of staff immunization status? 24) Does your facility have a written plan that can readily be implemented in case of emergency (e.g. Influenza pandemic)? 25) Does this plan contain a method of increasing the number of direct-care staff and/or the number of resident beds if needed during an emergency such as an Influenza pandemic? 26) Are infection control audits routinely done in your facility to identify potential issues? 27) If yes to Question 24, how frequently are the audits done?  28)  Approximately every _______months  I don’t know  Is there an infection control committee that acts as an infection control resource for your facility?  Yes, a local committee for Yes, a regional Yes, both a local and regional our facility committee committee If there is no committee, please skip to Question 29  No committee  I don’t know  Please answer the following questions by selecting one of the boxes to the right with an ‘x’: Rarely or never  Sometimes  Often or always  I don’t know  29) On average, your infection control committee members make reference to regional, national or international resident safety initiatives: 30) Is the facility senior management actively involved in this committee? 31) Does your facility administration provide financial support for short-term infection prevention and control initiatives? 32) Does your facility administration demonstrate their support of infection prevention and control initiatives in other ways (e.g. Written support in newsletters or other correspondence)? 33) In your opinion, do staff members feel free to speak up to senior management if they see something that may negatively impact residents? 34) Are staff given feedback about changes put into place based on infection control or other safety related reports? 35) In your opinion, does senior management seriously consider staff suggestions for improving resident care?  179  Section 3: Infection Surveillance Activities For each of the infections listed on the right, please answer the following questions using these answer choices:  Influenza  Clostridium difficile associated diarrhea  • “Y” = Yes • “N” = No • “?” = I don’t know 1) Do you have a written policy to prevent transmissions?  Antibioticresistant infections (eg infections caused by MRSA or VRE)  2) Do you have educational material readily available to hand out to visitors and family? (e.g. pamphlets) 3) Do you have a tracking system (manual or electronic) to provide the ICP with quick, easy and accurate data about these types of infections for your facility?  Quick & easy Accurate  4) Does the ICP collect data on these infections? 5) Do you have a simple, written definition that staff can use to determine whether a resident has this infection?  6) If you answered yes to Question 5, what is the source of your definitions? Our facility or group of facilities  8)  Our health region  Public Health Agency of Canada (Health Canada)  Another Canadian source  An international source  I don’t know  Please select all of the methods your ICP uses to collect data on infections and outbreaks in your facility: Chart reviews  Kardex or vital signs chart reviews  Staff reporting of infections to ICP  Participation in facility rounds  Laboratory culture reviews  Monitoring of antibiotic use  I don’t know  For each of the following questions, please select the best answer to the right. 8) How often does your facility use data collected on infections to review trends?  Rarely or never  9) How often does your facility use data collected on infections to prepare a report?  Rarely or never  Routinely every few months Every 6 to 12 months  At least monthly At least quarterly  I don’t know I don’t know  (If they are never prepared please skip to Section 4) 10) Please circle all stakeholders who are given the infection control report:  Facility administration  Infection control committee members  Facility physicians  Directcare staff  Family council  11) How often is the infection control report presented to and reviewed by the infection control committee or by your facility administration?  Rarely or never  As needed only  Regularly  I don’t know  12) How often does your facility use the infection control report to implement improvements or corrective actions in the facility?  Never  Sometimes  Often  I don’t know  180  E  F  G  H  Annual resident influenza immunization  Pneumococcal immunization for residents  New staff Tuberculosis screening  I New resident Tuberculosis screening  D  Staff work restrictions during influenza outbreaks  C  Annual staff influenza immunization  “Y” = Yes “N” = No “?” = I don’t know  B  Cleaning & sterilization  • • •  A Hand hygiene  For each of the infection control issues listed on the right, please answer the following questions using these answer choices:  Additional precautions for residents with infections  Section 4: Infection Prevention & Control Activities  1) Do you have a written policy for this? (If ‘No’, skip to next column) 2) Is the policy readily available to staff in their work area? 3) Is the language used in the policy clear and concise? 4) Are there policy reminders, memory aids or checklists in the work area? 5) Did direct-care staff participate in the policy’s development? 6) Did the ICP participate in policy development? 7) Is there a system for monitoring the compliance of staff with the policy? 8) Do you think that this policy is followed correctly at least 80% of the time by staff?  9) If you have a written hand hygiene policy, in what year was it first established? (This is a marker for the beginning of your Infection Prevention and Control program) ___________ 10) How frequently are your facility policies and procedures reviewed?  Rarely or never  Only when problems arise  Regularly (eg. Yearly)  I don’t know  11) How much paid educational time does your ICP receive?  None  < 1 work day per year  1 to 5 work days per year  > 5 work days per year  I don’t know  12) On average, how much paid educational time does your direct-care staff receive for infection prevention and control (Please estimate including all inservices)?  None  < 1 work day per year  1 to 5 work days per year  > 5 work days per year  I don’t know  13) Does your front-line staff have easy access to educational material on infection control in their work area?  Yes  No  I don’t know  181  

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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0066441/manifest

Comment

Related Items