Behind the Mask

BEHIND THE MASK: Determinants of Nurses’ Adherence to Recommended Use of Facial Protective Equipment to Prevent Occupational Transmission of Communicable Respiratory Illness in Acute Care Hospitals

Kathryn Anne Nichol

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Institute of Medical Science University of Toronto

BEHIND THE MASK: Determinants of Nurses’ Adherence to Recommended Use of Facial Protective Equipment to Prevent Occupational Transmission of Communicable Respiratory Illness in Acute Care Hospitals

Kathryn Anne Nichol

Doctor of Philosophy

Institute of Medical Science University of Toronto

2010 Abstract (max 350w)

Background - Communicable respiratory illness is a serious occupational threat to healthcare workers. A key reason for occupational transmission is failure to implement appropriate barrier precautions. Facial protective equipment, including surgical masks, respirators and eye/face protection, is the least adhered to type of personal protective equipment used by healthcare workers, yet it is an important barrier precaution against communicable respiratory illness.

Objectives - To describe nurses‘ adherence to recommended use of facial protective equipment and to identify the factors that influence adherence.

Methods - A two-phased study was conducted. Phase 1 was a cross-sectional survey of nurses in selected units of six acute care hospitals in Toronto, Canada. Phase 2 was a direct observational study of critical care nurses.

Results – Of the 1074 nurses who completed surveys (82% response rate), 44% reported adherence to recommended use of facial protective equipment. Multivariable analysis revealed four organizational predictors of adherence: ready availability of equipment, regular training and fit testing, organizational support for health and safety, and good communication. Following the survey, 112 observations in 14 intensive care units were conducted that showed a 44% competence rate with proper use of N95 respirators. Common gaps included failure to verify the seal and touching the face piece. Multivariable analysis revealed knowledge of recommended use of facial protective equipment as a significant predictor of competence.

Discussion – Despite the SARS experience and the resulting investment in our public health system, nurses‘ adherence to recommended use of facial protective equipment and competence in effective use of N95 respirators remains suboptimal. To improve adherence, organizational leaders should focus on equipment availability, training and fit testing, organizational support for health and safety, and positive communication. To improve competence in effective use of N95 respirators, strategies to increase knowledge should be implemented. These efforts should assist to reduce occupational transmission of communicable respiratory illness and foster a healthier and safer working environment for nurses. (314 w)

iii

Acknowledgments

I would like to acknowledge my supervisor – Dr. D. Linn Holness – for supporting and guiding me through this degree; my thesis committee – Dr. Allison McGeer, Dr. Phil Bigelow, Dr. Linda O‘Brien-Pallas and Dr. James Scott – for their time and expertise; my former employer - Ms. Joseline Sikorski at OSACH - for encouraging me to further my education and providing a flexible work schedule; my research assistant - Ms. Sarah Hayday - for her hard work and dedication to the project; the Centre for Research Expertise in Occupational Disease at St. Michael‘s Hospital and the University of Toronto and the Research Advisory Council of the Workplace Safety and Insurance Board for generous funding; and, of course, my family – Ian, Austin and Kristianne – for keeping life fun and well balanced along the way.

―People underestimate the importance of diligence as a virtue. No doubt this has something to do with how supremely mundane it seems. It is defined as ‗the constant and earnest effort to accomplish what is undertaken‘. There is a flavor of simplistic relentlessness to it, yet it is the only true prerequisite of great accomplishment.‖ Dr. Atul Gawande, from his book Better: A Surgeon’s Notes on Performance

―All diseases have two causes – one pathological, and one political.‖ Dr. Rudolph Virchow (1821 – 1902)

―Between the extremes of panic and complacency lies the solid ground of vigilance.‖ Dr. Margaret Chan, Director-General of the World Health Organization, July 2, 2009

Table of Contents

Acknowledgments ...... iv

Table of Contents ...... v

List of Abbreviations ...... ix

List of Tables ...... xi

List of Figures ...... xiii

List of Appendices ...... xiv

1 CHAPTER 1: Introduction ...... 1

1.1 Background to the Problem ...... 1

1.2 Problem Statement ...... 2

1.3 Purpose ...... 2

1.4 Hypothesis ...... 2

1.5 Study Objectives ...... 3

1.5.1 Phase 1 – Cross-Sectional Survey ...... 3

1.5.2 Phase 2 – Direct Observational Study of Critical Care Nurses ...... 4

2 CHAPTER 2: Literature Review ...... 5

2.1 History of Standards for the Protection of Healthcare Workers from Occupational Transmission of Communicable Respiratory Illness ...... 6

2.2 Healthcare Workers‘ Adherence to Infection Control Practices ...... 13

2.2.1 Critical Appraisal Methods ...... 13

2.2.2 Hand Hygiene ...... 14

2.2.3 Medical Gloves and Universal Precautions ...... 14

2.2.4 Vaccination/Immunization ...... 15

2.2.5 Facial Protective Equipment ...... 16

2.3 Theoretical Models and Self-Protective Behaviours ...... 19

2.3.1 Value-Expectancy Models ...... 19

2.3.2 Behaviour Change Models ...... 22

2.3.3 Environmental/Contextual Model: The PRECEDE Model ...... 24

2.4 Safety Climate and Adherence ...... 29

2.5 The Relationship between Worker Safety and Patient Safety ...... 30

2.5.1 History of Patient Safety in Canada ...... 31

2.5.2 Link between Patient and Worker Safety ...... 32

2.5.3 Measuring Patient Safety Climate ...... 34

2.6 Synthesis of the Literature ...... 35

2.7 Critical Review of Methods Chosen ...... 36

2.7.1 Studies Comparing Observed and Self-Reported Adherence ...... 36

2.7.2 Intervention Studies ...... 37

2.7.3 Observational Studies of Adherence ...... 38

2.7.4 Measuring Competence ...... 38

2.8 Pilot Study ...... 40

3 CHAPTER 3: Methods ...... 41

3.1 Study Design ...... 41

3.2 Enrolment ...... 42

3.2.1 Hospital Enrolment ...... 42

3.2.2 Subject Enrolment ...... 43

3.3 Data Collection ...... 43

3.3.1 Cross-Sectional Survey ...... 43

3.3.2 Direct Observational Study ...... 44

3.4 Measures ...... 45

3.4.1 Survey Tool ...... 45 vi

3.4.2 Unit Observation Record ...... 49

3.4.3 Participant Observation Record and Guide ...... 50

3.4.4 Explanatory Variables ...... 51

3.4.5 Outcome Measures ...... 52

3.4.6 Data Management ...... 53

3.5 Statistical Analysis ...... 53

3.5.1 Descriptive Analysis ...... 53

3.5.2 Exploratory Factor Analysis ...... 53

3.5.3 Kappa Statistic ...... 54

3.5.4 Reliability Testing ...... 54

3.5.5 Bivariate Analysis ...... 55

3.5.6 Multivariable Analysis ...... 55

3.6 Sample Size ...... 55

3.6.1 Phase 1 – Cross Sectional Survey ...... 55

3.6.2 Phase 2 – Direct Observational Study ...... 57

4 CHAPTER 4: Results ...... 58

4.1 Phase 1 – Adherence to Recommended Use of Facial Protection ...... 58

4.1.1 Descriptive Analysis ...... 58

4.1.2 Exploratory Factor Analysis ...... 62

4.1.3 Reliability Testing ...... 65

4.1.4 Bivariate Analysis ...... 65

4.1.5 Multivariable Analysis ...... 69

4.2 Phase 1 - Patient Safety Climate ...... 72

4.3 Phase 1 – Unit Observations ...... 74

4.4 Phase 2 – Participant Observations ...... 75

4.4.1 Descriptive Analysis ...... 75 vii

4.4.2 Bivariate Analysis ...... 80

4.4.3 Multivariable Analysis ...... 81

4.5 Relationship between Self-Reported Adherence and Observed Competence ...... 81

5 CHAPTER 5: Discussion ...... 83

5.1 Demographics ...... 83

5.2 Adherence to Recommended Use of Facial Protection ...... 84

5.3 Training and Fit Testing ...... 88

5.4 Health Effects ...... 89

5.5 Determinants of Adherence ...... 90

5.5.1 Demographic Determinants ...... 91

5.5.2 Individual Determinants ...... 92

5.5.3 Environmental Determinants ...... 93

5.5.4 Organizational Determinants ...... 95

5.5.5 Safety Climate ...... 98

5.6 Patient Safety ...... 100

5.7 Strengths and Limitations ...... 102

5.7.1 Strengths ...... 102

5.7.2 Limitations ...... 103

6 CHAPTER 6: Conclusions ...... 106

7 CHAPTER 7: Future Directions ...... 108

Bibliography ...... 111

Appendices ...... 133

viii

List of Abbreviations

AIDS Acquired Immunodeficiency Syndrome

BScN Bachelor of Science in Nursing degree

BN Bachelor of Nursing degree

CDC Centers for Disease Control and Prevention

CIDRAP Center for Infectious Disease Research and Policy

CNO Chief Nursing Officer

CPSI Canadian Patient Safety Institute

CSA Canadian Standards Association

FPE Facial Protective Equipment

GTA Greater Toronto Area

HEPA High Efficiency Particulate Air filter

HIV Human Immunodeficiency Virus

ICU Intensive Care Unit

IOM Institute of Medicine

MOHLTC Ontario Ministry of Health and Long-Term Care

MOL Ontario Ministry of Labour

MRSA Methicillin-resistant Staphylococcus aureus

NHIS National Health Interview Survey

NIOSH National Institute for Occupational Safety and Health

OHSAH Occupational Health and Safety Agency for Healthcare in British Columbia

PHAC Public Health Agency of Canada

PIDAC Provincial Infectious Diseases Advisory Committee

PPE Personal Protective Equipment

PRECEDE Predisposing, Reinforcing, and Enabling Constructs in Educational Diagnosis and Evaluation

RN Registered Nurse

RPN Registered Practical Nurse

SARS Severe Acute Respiratory Syndrome

SAS Statistical Analysis Software

TB Tuberculosis

TPB Theory of Planned Behaviour

TTM Transtheoretical Model

TRA Theory of Reasoned Action

WHO World Health Organization

List of Tables

Table 1: Response Choice Definitions for Unit Observation Record ...... 50

Table 2: Cross Sectional Survey Sample Size Calculations ...... 56

Table 3: Sample Size Calculations for Variable 'Tenure as a Nurse' ...... 56

Table 4: Sample Size Calculations for Observational Study ...... 57

Table 5: Demographic Characteristics of Survey Population ...... 58

Table 6: Summary of Responses to Items in the Adherence Scale ...... 59

Table 7: Summary of Responses to Knowledge of Use of Facial Protection Scale ...... 60

Table 8: Summary of Responses to Knowledge of Transmission of Influenza Scale ...... 61

Table 9: Summary of Responses to Exposure Scale ...... 62

Table 10: Factor Loadings of Items in Organizational Support Scale ...... 64

Table 11: Reliability of Scales Used to Measure Constructs within the Survey ...... 65

Table 12: Bivariate Analysis of Association between Demographic Factors and Adherence ..... 66

Table 13: Bivariate Analysis of Association between Individual Factors and Adherence ...... 67

Table 14: Bivariate Analysis of Association between Environmental Factors and Adherence ... 68

Table 15: Bivariate Analysis of Association between Organizational Factors and Adherence ... 69

Table 16: Adjusted Odds Ratios for Adherence to Recommended Use of Facial Protection ...... 71

Table 17: Summary of Responses to Patient Safety Climate Scale ...... 72

Table 18: Bivariate Analysis of Association between Worker Safety Climate Measures and Patient Safety Climate ...... 73

Table 19: Bivariate Analysis of Association between Patient Safety Climate Scale and Adherence ...... 73

Table 20: Descriptive Findings for Unit Observations ...... 74

Table 21: Unit Observations Presented by Unit Type ...... 74

Table 22: Comparison of Self-Report Data and Unit Observations ...... 75

Table 23: Reasons for Observation Not Completed ...... 76

Table 24: Demographic Characteristics of the Observational Study Population and Comparison of Critical Care Nurses who were Observed and Not Observed ...... 77

Table 25: Descriptive Findings for Observational Study ...... 79

Table 26: Bivariate Analysis of Association between Explanatory Variables and Competence . 80

Table 27: Adjusted Odds Ratios for Competence with Recommended Use of an N95 Respirator ...... 81

Table 28: Relationship between Self-Reported Adherence and Observed Competence ...... 82

xii

List of Figures

Figure 1: PRECEDE Model ...... 25

Figure 2: Model of Compliance with Universal Precautions ...... 26

Figure 3: Theoretical Model for Factors Associated with Self-Protective Behavior at Work ...... 28

Figure 4: Factor Analysis Scree Plot of Eigenvalues ...... 63

Figure 5: Breakdown of Observations by Type and Level of Precaution ...... 78

Figure 6: Preliminary Organizational Model of Adherence ...... 99

Figure 7: Final Organizational Model of Adherence ...... 101

xiii

List of Appendices

Appendix 1 - Annotated Bibliography of Adherence Literature ...... 134

Appendix 2 - Published Manuscript of the Pilot Study ...... 142

Appendix 3 - Two Page Proposal Summary ...... 151

Appendix 4 - Letter of Information ...... 154

Appendix 5 - Letter of Information and Consent ...... 156

Appendix 6 - Facial Protection Questionnaire ...... 160

Appendix 7 - Unit Observation Record ...... 169

Appendix 8 - Participant Observation Record ...... 170

Appendix 9 - Participant Observation Guide ...... 172

Appendix 10 - Case Definitions for Explanatory Variables ...... 173

xiv 1

1 CHAPTER 1: Introduction 1.1 Background to the Problem

Communicable respiratory illness is a serious occupational threat to the health and safety of healthcare workers (Lowe & Wilder-Smith, 2005). Concern over occupational transmission of communicable respiratory illness in the health sector is not new. The spread of tuberculosis, varicella zoster (chicken pox) and influenza among healthcare workers have been well documented in the literature (Lowe & Wilder-Smith, 2005). In recent years, concern for the well- being of healthcare workers has escalated. In 2003, occupational transmission of severe acute respiratory syndrome (SARS) highlighted the vulnerability of healthcare workers. In his third Commission Report to investigate SARS, Justice Archie Campbell highlighted the heavy burden of disease that fell on nurses, doctors and other health workers (The SARS Commission, 2006). Of the 375 people who contracted SARS in Ontario, 72% were infected in a healthcare setting. Of this group, 45% were health workers and two nurses and one physician died of SARS-related causes (The SARS Commission, 2006). Very recently, the international public health community has been in high alert monitoring an outbreak of Influenza A (H1N1). This novel influenza caused mild-to-moderate disease initially in Mexico and quickly spread throughout the world. In addition to emerging infections, the global threat of bioterrorism is present and being taken very seriously by Canadian provincial and federal governments and health agencies worldwide.

Communicable respiratory illness can be spread by the airborne, droplet and/or contact routes. The use of barrier precautions is an important strategy to prevent occupational transmission of respiratory illness among healthcare workers. Experts in the prevention and control of occupational hazards agree that the use of personal protective equipment should always be a last resort. However, for healthcare workers, personal protective equipment may be the first line of defense, in addition to other preventive measures which may be adopted concurrently or later when the mode of transmission of the infectious disease is determined (Chia, 2005). In a review of the scientific literature on the efficacy of personal protective equipment to prevent the transmission of SARS, researchers determined that failure to use appropriate barrier precautions was a primary contributor to occupational transmission (Yassi, et al., 2004). Adherence to safe work practices to protect against transmission of blood borne disease has been well studied and found to be substandard. Safety climate dimensions and other

2 organizational factors have been shown to be the main influence on adherence. Adherence to safe work practices to prevent the spread of communicable respiratory illness, such as immunization, has also been found to be poor. Of all standard preventive practices, the use of facial protective equipment (FPE) (including eye, face and respiratory protection) had the least adherence by healthcare workers. Understanding why healthcare workers fail to appropriately use facial protective equipment has not been well researched.

1.2 Problem Statement

The use of facial protective equipment is an important strategy to prevent occupational transmission of communicable respiratory illness among healthcare workers, yet adherence to recommended use is suboptimal. As the predominant occupation in the health sector and as the health worker with the most patient interaction, nurses are at high risk for occupational transmission of communicable respiratory illness. Improving nurses‘ adherence to recommended use of facial protective equipment should assist to reduce occupational transmission of illness. This study used self report and observational data to identify important determinants of adherence that can be used to implement strategies and interventions to improve adherence, reduce illness and enhance the work and health of nurses.

1.3 Purpose

The overall purpose of this study was to describe nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness and to determine the individual, environmental and organizational factors that influence adherence.

1.4 Hypothesis

Organizational factors and other safety climate dimensions will be predictive of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in the acute healthcare setting.

1.5 Study Objectives

1.5.1 Phase 1 – Cross-Sectional Survey

Phase 1 of the study was a cross sectional survey of nurses who regularly used facial protective equipment to prevent occupational transmission of communicable respiratory illness.

Primary Objectives 1. To describe nurses‘ reported adherence to the recommended use of facial protective equipment in acute care hospitals. 2. To describe the reported health outcomes of nurses as a result of occupational exposure to communicable respiratory illness. 3. To describe the organizational, environmental and individual factors that influence nurses‘ adherence to facial protective equipment. 4. To determine the relative importance of organizational, environmental and individual factors in predicting nurses‘ adherence to recommended use of facial protective equipment.

Secondary Objectives 1. To describe nurses‘ perceptions of worker safety climate and patient safety climate and determine if there is a relationship between the two measures. 2. To observe cleanliness and orderliness of the work environment and availability of facial protective equipment at the unit level. 3. To compare self-report and observational measures of cleanliness and orderliness of the work environment and availability of facial protective equipment. 4. To evaluate reliability of the scales within the survey tool.

1.5.2 Phase 2 – Direct Observational Study of Critical Care Nurses

Phase 2 was a direct observational study of nurses in the critical care setting donning, using, doffing and disposing of facial protective equipment used to prevent occupational transmission of communicable respiratory illness.

Primary Objectives 1. To observe nurses‘ adherence to recommended use of facial protective equipment in the critical care setting. 2. To describe the organizational, environmental and individual factors that influence observed adherence with facial protective equipment. 3. To determine the relative importance of organizational, environmental and individual factors in predicting observed adherence with the use of facial protective equipment.

Secondary Objectives 1. To compare self-report and observational measures of adherence to recommended use of facial protective equipment. 2. To evaluate the feasibility of conducting a direct observational study of the use of facial protective equipment in the critical care environment.

2 CHAPTER 2: Literature Review

With the 2003 outbreak of SARS and the current belief that the world is overdue for a serious influenza pandemic, there is heightened concern about communicable respiratory illnesses and their impact on health services and workers. Healthcare workers are at high risk for occupational transmission of communicable respiratory illness as they provide care to patients with acute illness before the agent of disease has been identified. They are required to carry out invasive procedures during which exposure opportunities are high and work in congested and fast-paced environments. It is also important to note that the healthcare industry is unlike other industries as their product is human health and well being (Institute of Medicine, 2008). Patient interaction confounds the usual barriers to safety as healthcare workers have been found to sacrifice their own safety for what they feel is better patient care and communication (Institute of Medicine, 2008). Unfortunately, efforts to appropriately protect healthcare workers are further hindered by a lack of data on transmission of respiratory pathogens such as influenza, the personal protective equipment that effectively protects workers and how to ensure adherence to safe use of the equipment.

Personal protective equipment (PPE) is one component of a continuum of safety efforts to prevent occupational transmission of disease. Occupational health and safety experts have traditionally followed a hierarchy of control technologies when planning and implementing safety programs. This hierarchy describes three general categories of controls: 1) engineering controls, followed by 2) administrative and work practice controls, followed by 3) personal protective equipment (Thorne, Khozin, & McDiarmid, 2004). Engineering controls target the source of the hazard and include strategies such as negative pressure isolation, enclosure and high efficiency particulate aerosol (HEPA) filtration. Engineering controls are seen as the most protective as they do not rely on human interaction for success. If further protection is necessary, administrative or work practice controls targeting the path between the hazard and the worker can be implemented. These include signage, training, quarantine, policies and procedures and medical surveillance. Personal protective equipment is the last tier of protection and targets hazards right at the worker. PPE includes barrier protection to prevent skin, mucous membrane and respiratory exposures such as gloves, gowns, goggles, masks and respirators.

Prior to the 1980s, the use of PPE in the healthcare sector was largely confined to surgical settings and was primarily intended to protect patients from healthcare workers. The emergence of HIV/AIDS introduced widespread use of medical gloves in healthcare facilities and the resurgence of tuberculosis resulted in increased use of respiratory protection in respirology clinics and units. It was only in 2003, with the SARS outbreak, that the need for widespread use of respirators, masks and eye protection to protect healthcare workers from transmission of disease from patients or the work environment was taken seriously. Unfortunately, adherence to recommended use of personal protective equipment in the health sector has been shown to be consistently substandard, highlighting the complexity of the problem and the need for a novel approach to improvement.

This review of the literature will chronicle the history of infection control standards regarding communicable respiratory illness and the emergence of widespread use of facial protection as an important barrier to occupational transmission of this type of disease. A review of the literature on adherence to infection control practices, including facial protection, will be presented supported by several theoretical models of importance. The PRECEDE model, a model to understand behaviour change that involves the context within which the person operates, will be introduced as the model used to frame this study. As important constructs in this model, a review of organizational factors and other safety climate dimensions of adherence will be presented including evidence of their relevance to worker and patient safety. The literature review will conclude with a critical appraisal of methods chosen and a summary of the results of a pilot study undertaken to inform the current study.

2.1 History of Standards for the Protection of Healthcare Workers from Occupational Transmission of Communicable Respiratory Illness

The need for infection control standards to protect healthcare workers originated in the mid 1800s from ―fever hospitals‖ that were utilized for the care of patients with specific communicable pathogens of major public health concern, such as smallpox, diphtheria, and tuberculosis (Public Health Agency of Canada, 1999). As these diseases were eradicated or became less common, care was transferred to special isolation wards in general hospitals and eventually to isolation rooms on patient care units. The first protection standards were

7 developed based primarily on a communicable agent‘s mode of transmission. Over time, healthcare workers identified problems with this strategy for categorizing diseases as it often led to either excessive or insufficient use of precautions. As a result, disease-specific precautions that included patient behavioural characteristics were developed and care providers could choose between these category- or disease-specific systems (Public Health Agency of Canada, 1999). This approach was more comprehensive but required more advanced decision-making skills on the part of the healthcare provider.

A major development that changed protective standards for healthcare workers forever was the realization that patients could present with a communicable disease such as human immunodeficiency virus or HIV, without demonstrating any signs or symptoms (Public Health Agency of Canada, 1999). It became clear that healthcare workers could no longer rely on category- or disease-specific systems alone to identify adequate isolation precautions. This led to the introduction of a system of universal precautions in 1987 in the United States where protection from exposure to the blood and body fluids of all patients was implemented. Over time, these precautions were extended to include secretions, moist body substances and non- intact skin and supplemented other precautions including those to prevent transmission of respiratory illness.

In the mid 1990s, the Centers for Disease Control and Prevention (CDC) in the United States revised their isolation guidelines and adopted an approach that combined the best recommendations from each of the three previous systems; category- and disease-specific systems and universal precautions, into a system now called standard precautions. Canada quickly followed, releasing the current guideline Routine Practices and Additional Precautions for Preventing the Transmission of Infection in Health Care in 1999. This guideline utilizes a two-tiered system of routine practices for all patients and three categories of additional precautions for known or suspected routes of transmission (airborne, droplet and contact). Additional precautions also take specific patient characteristics and behaviours into consideration. Within these guidelines airborne transmission is defined as occurring from small droplets or droplets that have dried out (droplet nuclei) that are less than 5 microns in diameter and can remain suspended in the air for long periods of time. Droplet transmission occurs from large droplets that are equal to or over 5 microns in diameter and are propelled a short distance (approximately 1 metre/3 feet or less) through the air. Contact transmission occurs from direct

8 or indirect contact with a person or object. Some organisms may be transmitted via more than one route (Public Health Agency of Canada, 1999). The guideline includes hand hygiene, aseptic practice and cleaning and disinfection as important practices to prevent the spread of disease.

Current recommendations for facial protective equipment exist within these guidelines. Communicable respiratory illnesses spread via the airborne route require the use of a high- efficiency dust/mist mask that filters particles one micron in size, has a 95% filter efficiency and provides a tight facial seal (less than 10% leak), such as an N95 respirator certified by the National Institute for Occupational Safety and Health (NIOSH) in the United States. Eye protection is indicated as per routine patient care practices; if the procedure or patient care activity is likely to generate splashes or sprays of blood, body fluids, secretions or excretions then the worker should use eye protection. The facial protective equipment recommended to prevent transmission of respiratory illness spread via the droplet route includes the use of a surgical/procedure mask if within 1 metre of the patient. Eye protection is indicated as per routine patient care practices except for care of children with symptoms of acute respiratory infection. In these cases, eye protection must be worn if within 1 metre of a coughing pediatric patient or performing procedures that may result in coughing. Contact precautions do not require the use of facial protection equipment except when indicated by routine practices (Public Health Agency of Canada, 1999).

A second major development that evolved in the 1990s was the increased complexity of the clinical presentation of tuberculosis (TB). Tuberculosis is a well-known communicable bacterial infection spread via the airborne route and Canada has had standards for the prevention and control of TB transmission since 1972. In the 1990s, the spread of the human immunodeficiency virus (HIV) and the emergence of drug resistant strains of tuberculosis made patient care complicated and some cases incurable (The Lung Association, 2009). These issues continue to be of concern today. In Canada in 2007, the TB rate was 4.7 per 100,000 with a total of 1,547 new cases. In recent years, there have been several outbreaks of TB in homeless shelters and among other marginalized populations in Canada. Ontario continues to have the most TB cases of any province and approximately 1/3 of Canadian TB cases live in the Greater Toronto Area (GTA). Ontario, and in particular the GTA, also has the highest rate of drug

9 resistant TB in Canada. One in every six TB patients in the GTA has a TB strain resistant to at least one first-line drug (Public Health Agency of Canada, 2007).

Although the increased complexity of tuberculosis did renew interest in occupational transmission of communicable respiratory disease, it was the SARS outbreaks of 2003 that really brought it to the forefront. Interestingly, the first hospital SARS case in Toronto was initially thought to be tuberculosis (National Advisory Committee on SARS and Public Health, 2003). SARS emerged and spread rapidly resulting in 8096 cases and 774 deaths worldwide (World Health Organization, 2003). In Canada, Toronto was the epicenter of the outbreak and, of the 375 people who contracted SARS in Ontario, 270 people (72%) were infected in a healthcare setting. Of this group, 45 were health workers and two nurses and one physician died of SARS- related causes (The SARS Commission, 2006). SARS demonstrated highly efficient transmission in healthcare facilities. This highlighted the vulnerability of our modern healthcare system to communicable respiratory illness and the inadequacy of our current infection control guidelines and practices (Gopalakrishna, et al., 2004).

During SARS, the Scientific Advisory Committee to the Ontario Ministry of Health and Long-Term Care (MoHLTC) was charged with developing quarantine guidelines and hospital directives covering topics such as restricted access, isolation precautions, employee screening, and patient transfers (National Advisory Committee on SARS and Public Health, 2003). Among the many new guidelines and directives was a guidance document on preventing communicable respiratory illness in the acute healthcare setting - Preventing Respiratory Illnesses, Protecting Patients and Staff: Infection Control and Surveillance Standards for Febrile Respiratory Illness (FRI) in Non-Outbreak Conditions in Acute Care Hospitals (Ontario Ministry of Health and Long-Term Care, 2003). This document identified febrile respiratory illnesses such as SARS as primarily spread via large droplets and contact and recommended precautions based on these routes of transmission. It was later updated by the Provincial Infectious Diseases Advisory Committee (PIDAC) of the MoHLTC - Preventing Febrile Respiratory Illnesses: Protecting Patients and Staff - Best Practices in Surveillance and Infection Prevention and Control for Febrile Respiratory Illness (FRI), excluding Tuberculosis, for All Ontario Health Care Settings (Provincial Infectious Diseases Advisory Committee, 2006) maintaining requirements for routine and droplet precautions.

Reason for concern over occupational transmission of communicable respiratory illness went beyond SARS. In 2004, the year after SARS, the first human case of avian influenza A (H7) in British Columbia occurred in a person who was involved in culling of infected birds (World Health Organization, 2004). Although high and low pathogenic strains of avian influenza had been present in wild and domestic birds since the late 1800s, the majority of human cases linked to avian influenza outbreaks had only been identified since 1997; and mostly in Asia, Europe and Africa (Public Health Agency of Canada, 2006). The case in British Columbia and subsequent Canadian cases raised concern regarding the opportunity for ongoing genetic mutation or viral re-assortment. Health officials were concerned simultaneous infection with human influenza and avian influenza viruses may result in an exchange of genes and the development of a new influenza subtype with efficient human-to-human transmission and pandemic potential (Public Health Agency of Canada, 2006). As a result, additional guidelines and documents were developed specifically for avian influenza including the Operational Plan During an Outbreak of Avian Influenza in the Domestic Poultry Population(s) (Ontario Ministry of Health and Long-Term Care, 2006) and the Human Health Issues related to Avian Influenza in Canada document (Public Health Agency of Canada, 2006). Due to the pandemic potential of a mutated strain, these guidelines recommend following airborne precautions for the protection of healthcare workers including the use of an N95 respirator.

The response from the medical community regarding the array of guidelines and variety of modes of transmission for the different types of influenza was vibrant and contradictory. In 2006, Dr. Raymond Tellier, a renowned microbiologist from the Hospital for Sick Children in Toronto, reviewed the published literature and concluded the evidence showed influenza A virus could be transmitted via the airborne route (Tellier, 2006). Almost immediately, in 2007, Dr. Michael Gardam, an infectious disease physician from the University Health Network in Toronto, published an article with his colleagues concluding transmission of influenza A occurs at close range rather than over long distances, suggesting that airborne transmission is unlikely to be of significance (Brankston, Gitterman, Hirji, Lemieux, & Gardam, 2007). This disparate interpretation of the published literature by well known and highly regarded health experts resulted in a renewed research interest in transmission of influenza in healthcare facilities.

In 2007, in the context of updating the Canadian Pandemic Influenza Plan for the health sector, PHAC asked the Council of Canadian Academies to appoint an independent expert panel

11 to assess the current science of transmission of influenza and the contribution of N95 respirators and surgical masks in preventing transmission. The panel found there was evidence for the persistent survival of influenza virus in ambient air under common environmental conditions suggesting that long range inhalation transmission of influenza was possible. However, direct evidence of its contribution to influenza transmission was sparse. Regarding protective measures, the panel found evidence that N95 respirators protect against the inhalation of nasopharyngeal, tracheobronchial and alveolar sized particles. The efficiency of surgical masks to block these particles was found to be either highly variable or unknown (Council of Canadian Academies, 2008). This report served to further highlight the divergent opinions that existed within the infection prevention and control community and spawned researchers to conduct new studies on influenza transmission and protection (Fabian, et al., 2008) (Blachere, et al., 2009).

The year 2007 also brought the release of the final report of the SARS Commission. Justice Archie Campbell of the Superior Court of Justice was commissioned by the Ontario Government to conduct an independent investigation into the introduction and spread of SARS. The final report, Spring of Fear, contained more than 80 recommendations impacting both the Ontario Occupational Health and Safety Act and the Health Protection and Promotion Act and their regulations and how the two provincial ministries impacting worker safety, the Ministry of Labour (MOL) and the MoHLTC, need to operate during emergencies (The SARS Commission, 2006). A key recommendation regarding worker safety from the Campbell Report was the adoption of the precautionary principle as a guiding principle throughout Ontario‘s health system. The precautionary principle states that reasonable steps to reduce risk need not await scientific certainty. In other words, in the absence of evidence on the mode of transmission of a communicable illness, the more protective guidelines should always be adopted.

Although the government has not formally adopted the precautionary principle into legislation, its utilization was obvious when increased influenza A (H1N1) activity was identified in Mexico and subsequently spread throughout the world. In Canada, guidelines were developed recommending healthcare workers use respiratory protection when within 2 metres of a patient with suspected influenza-like illness (Public Health Agency of Canada, 2009). Respiratory protection was defined as a surgical mask and eye or face protection, or an N95 respirator and eye or face protection. Healthcare workers were directed to wear a surgical mask

12 only if the patient was willing and able to wear a surgical mask themselves or if they had a weak or no cough, otherwise an N95 respirator was recommended.

In the United States, the CDC had slightly different requirements and recommended all healthcare personnel who enter the rooms of patients in isolation with confirmed, suspected, or probable novel H1N1 influenza should wear a fit-tested disposable N95 respirator or better (Centers for Disease Control and Prevention, 2009). In response to the disparate recommendations between the two countries, the CDC asked the Institute of Medicine (IOM) to convene a committee to provide recommendations regarding the necessary respiratory protection for healthcare workers against influenza A (H1N1). The committee reviewed the available literature and concluded that healthcare workers caring for H1N1 influenza patients should wear fit-tested N95 respirators, not surgical masks, to protect them from the virus (Institute of Medicine of the National Academies, 2009). At the same time, the IOM called for additional research on influenza transmission and the effectiveness of various respiratory protection tools in clinical settings, along with efforts to develop new respiratory protection technologies to enhance worker safety and comfort.

The history of standards for the protection of healthcare workers from occupational transmission of communicable respiratory illness is colourful and has resulted in the development of numerous guidelines; some conflicting and controversial. There are guidelines based on mode of transmission (airborne, droplet, contact), specific disease (tuberculosis, avian influenza, influenza A (H1N1)), for specific conditions and procedures (institutional outbreaks, pandemics and aerosol-generating procedures) and then there is the precautionary principle for emerging infections and mutated/reorganized agents. Although these guidelines were designed to protect healthcare workers, critical information is still missing. Making decisions regarding adequate protective practices is not an easy task. To navigate these resources and choose the appropriate one(s) for their situation, healthcare leaders and front line managers must have advanced knowledge of the issues involved in transmission of communicable respiratory illness and possess excellent assessment skills to evaluate the needs of their organization and their workers. Even when decisions are made and policies and procedures are established; those with accountability for practice must ensure they are adhered to.

2.2 Healthcare Workers’ Adherence to Infection Control Practices

Currently, many guidance documents on the type of protective equipment to use under certain circumstances are available. Unfortunately, the existence of recommendations and guidelines for widespread use of protective equipment does not always translate into actual work practices. Research has shown healthcare workers‘ adherence to safe work behaviours to prevent the spread of communicable disease, such as proper use of PPE, is historically poor. This section of the literature review summarizes studies that have examined adherence to safe work behaviours to prevent the spread of blood borne disease, influenza immunization in the healthcare sector and the proper use of facial protective equipment to prevent transmission of communicable respiratory illness.

2.2.1 Critical Appraisal Methods

A search of the literature on healthcare workers‘ adherence to infection control practices was conducted in November, 2005 using the common health search engines MEDLINE® and the Cumulative Index to Nursing and Allied Health Literature®. Search criteria specified sources dated within the last 15 years. Initially, key search terms ―compliance‖ and ―adherence‖ were matched with work behaviour terms ―hand washing‖ ―hand hygiene‖ ―immunization‖ ―personal protective equipment‖ facial protective equipment‖ ―masks‖ ―gloves‖ ―respiratory protection‖ ―N95 respirators‖ ―eye protection‖ ―universal precautions‖ and ― routine practices‖ to produce 890 hits. Reference lists of relevant articles were also examined and further sources of importance identified. PubMed Central was used to locate these sources and identify similar studies of interest. Studies were considered for inclusion for review if the following criteria were met: English language literature, date of publication within the last 15 years, peer-reviewed literature, the main outcome variable was adherence to infection control practices and the context was the health sector.

Thirty-two key references that focused on measuring adherence to infection control practices in the health sector and/or examining determinants of adherence were identified as highly relevant to this area of study (Appendix 1 - Annotated Bibliography of Adherence Literature). Between November, 2005 (initial literature search) and June, 2009 (initial preparation of dissertation), the University of Minnesota Center for Infectious Disease Research

& Policy (CIDRAP) website and mailing list was used to keep informed of new guidelines, research studies and political events involving infectious disease. Regular scans of the tables of contents of journals of interest (American Journal of Infection Control, Infection Control and Hospital Epidemiology, Canadian Medical Association Journal, Emerging Infectious Diseases, Journal of Occupational and Environmental Medicine) were also conducted to identify new literature of importance.

2.2.2 Hand Hygiene

Hand hygiene is the leading measure to prevent healthcare-associated infection and includes hand washing with soap and water and the use of alcohol-based hand sanitizers. Studies over the last 30 years have consistently shown adherence to proper hand hygiene practices in healthcare facilities to be substandard. One author compared 12 studies on adherence to hand hygiene between 1981 and 1999 and found that average adherence to hand hygiene recommendations varied between hospital units, among professional categories of healthcare workers, and according to working conditions, as well as according to the definitions used in different studies. Adherence rates from all studies were substandard at <50% (Pittet, 2001). A more recent study assessed hand hygiene adherence rates at US healthcare facilities by measuring product usage and providing feedback about adherence. Results showed that although hand hygiene adherence can increase when monitoring is combined with feedback, overall rates still occur at or below 50% for both ICUs and non-ICUs (McGuckin, Waterman, & Govednik, 2009).

2.2.3 Medical Gloves and Universal Precautions

Glove use is widely recommended in the healthcare setting for two main reasons: 1) to prevent microorganisms which may be present on healthcare workers‘ hands from being transmitted to patients, and; 2) to reduce the risk of healthcare workers acquiring infections from patients. Many studies have found adherence to proper glove use to be poor. One author cited six studies that found adherence to recommended glove use to be between 16 and 63% (DeJoy, Searcy, Murphy, & Gershon, 2000). Another study found that 38% of nurses reported adhering to universal precautions for 8 of their last 10 patients during venipuncture and only 16% indicated that they had applied universal precautions for all of them (Godin, Naccache, Morel, & Ebacher, 2000). In a study of emergency department personnel, a third report found adherence to

15 universal precautions to be 45% or less (Evanoff, et al., 1999). A final study examined healthcare workers use of gowns and gloves, as well as hand hygiene, to prevent the spread of methicillin- resistant Staphylococcus aureus (MRSA) and observed a 28% adherence rate (Waqqas, Huor, Brassard, & Loo, 2002). In the examination of determinants of adherence to proper glove use, early studies focused on the importance of individual-level factors such as age, occupation, tenure, personality traits and education (Kelen, et al., 1990) (Hammond, Eckes, Gomez, & Cunningham, 1990). More recent studies have focused on organizational and safety climate determinants of compliance (Gershon, et al., 2000) (Yassi, et al., 2007).

2.2.4 Vaccination/Immunization

Annual vaccination against influenza is recommended for all healthcare workers as it has been shown to reduce transmission of influenza in healthcare settings, reduce staff illness and absenteeism, and reduce influenza-related morbidity and mortality among patients at increased risk for severe illness (Centers for Disease Control and Prevention, 2006). In the United States, data from the National Health Interview Survey (NHIS) reported vaccination coverage of only 40% among healthcare workers in a 2003 survey (Centers for Disease Control and Prevention, 2005). Steiner et al. reported that despite recommendations from the Advisory Committee on Immunization Practices, annual influenza immunization rates of healthcare workers ranged from 2 – 60% (Steiner, Vermeulen, Mullahy, & Hayney, 2002). In a Canadian study of house staff across residency programs, a vaccination rate of 50% was reported (Lester, McGeer, Tomlinson, & Detsky, 2003). Another study examined influenza vaccination adherence among healthcare workers in a German University Hospital and found a self-reported adherence rate of 26.9% (Wicker, Rabenau, Doerr, & Allwinn, 2008).

Numerous studies have reported the barriers and facilitators to influenza vaccine uptake (Centers for Disease Control and Prevention, 2006). Barriers reported include fear of adverse side effects, insufficient time or inconvenience, perceived ineffectiveness of the vaccine, medical contraindication, perceived low risk of contracting influenza, and fear of needles. Factors facilitating vaccine acceptance included a desire for self-protection, previous receipt of influenza vaccine, a desire to protect patients, and belief in the effectiveness of the vaccine (Centers for Disease Control and Prevention, 2006). A recent review of the literature on attitudes and predictors of influenza vaccination of healthcare workers in hospitals found previous receipt of

16 influenza vaccine, belief in the vaccine‘s effectiveness and older age to be predictive of vaccination uptake (Hollmeyer, Hayden, Poland, & Buchholz, 2009).

2.2.5 Facial Protective Equipment

Adherence to recommended use of personal protective equipment to prevent the spread of communicable respiratory illness has not received the same amount of research interest as hand hygiene, glove use, universal precautions and influenza immunization. This was likely due to the fact that prior to 2003 and the SARS outbreak, facial protective equipment (FPE), including surgical masks, eye protection and respirators, was not taken seriously by healthcare administrators, unit managers or front line workers for protection against occupational transmission of communicable respiratory illness from patients or the work environment.

Several studies have examined adherence to recommended respirator use for caring for patients with suspected or confirmed tuberculosis. One study looked at three US hospitals over three years and found that healthcare workers wore appropriate respiratory protection with tuberculosis patients 44 to 97% of the time (Tokars, et al., 2001). A second study showed adherence to respirator use to be 57% when the diagnosis of tuberculosis was unconfirmed and 84% when it was confirmed (Kellerman, Saiman, San Gabriel, Besser, & Jarvis, 2001). In an investigation of 3 hospitals in California, a third report found only one hospital adhered to the minimum requirements for respiratory protection (Sutton, Nicas, Reinisch, & Harrison, 1998).

After SARS, there was a marked increase in research interest in safe work practices to prevent occupational transmission of communicable respiratory illness. In the United States, a group of investigators from the Center for Disease Control and Prevention retrospectively reviewed healthcare workers that worked during the SARS outbreaks and their adherence to respirator use. Sixty-six healthcare workers reported exposure to a patient who was coughing and later found to be SARS positive, yet 40% of these workers did not use a respirator (Park, et al., 2004). In Toronto, a study of SARS in critical care nurses showed 16 of 32 nurses who entered a SARS patient‘s room at least once consistently wore an N95 respirator (Loeb, et al., 2004). A second study of the Toronto SARS outbreak showed the self-reported consistent adherence rate to recommended barrier precautions varied depending on the diagnosis. Adherence was higher (84.6%) when a confirmed diagnosis of SARS had been made and lower

17 when patients were not thought to have SARS but precautions had been ordered, including the use of a mask, gown and gloves (45.3%) (Shigayeva, et al., 2007).

After the World Health Association confirmed human-to-human transmission of avian influenza in February of 2004 (Brown, 2004), a number of studies were published examining adherence to protective practices to prevent transmission of this communicable respiratory illness. One study, conducted in response to the 2004 outbreak of avian influenza (H7N3) among poultry in British Columbia, examined adherence to recommended protective measures and associated human infections. Results showed among 65 people who entered barns with infected birds, 55 (85%), 54 (83%) and 36 (55%) reported always wearing gloves, mask or goggles, respectively (Skowronski, et al., 2007). In another report on avian influenza, researchers from the United Kingdom conducted a retrospective cohort study of poultry workers who had been potentially exposed to infected material and measured completeness of use of personal protective equipment. Protective equipment ―always used‖ included disposable gloves (67%), face-fitted mask (51%), other mask (24%), and protective goggles (19%) (Morgan, et al., 2009).

Reports on adherence to respiratory protection to prevent transmission of the novel influenza A (H1N1) are also available. One report from the Centers for Disease Control and Prevention in the United States provided the results of an analysis of H1N1 cases in healthcare workers in the first weeks of the epidemic. Among 12 health workers believed to have been infected by patients, only 3 reported always using a surgical mask or an N95 respirator, none reported consistent use of eye protection and none reported always using gloves, gown, and a mask or respirator (Centers for Disease Control and Prevention, 2009). A study of the initial cases of influenza A (H1N1) in Mexico showed 22 of 190 healthcare workers caring for the first three confirmed cases developed respiratory symptoms. After infection control measures were strictly enforced, including the use of N95 respirators, goggles, gowns and gloves, no further health workers became ill (Perez-Padilla, et al., 2009).

Of all the personal protective equipment used as protection against occupational transmission of communicable disease, facial protection is highlighted as the most problematic (Yassi, Moore, Fitzgerald, Bigelow, Hon, & Bryce, 2005). In a study of emergency personnel, failure to wear a mask, followed by eye protection, was found to be the most common major

break in adhering to universal precautions (Evanoff, et al., 1999). A second study of practices against blood borne pathogens found behaviours least adhered to included wearing disposable face masks and wearing protective eye shields (McGovern, Vesley, Kochevar, Gershon, Rhame, & Anderson, 2000). A third study at one Toronto hospital reported that masks were identified as the most bothersome infection control precaution during the SARS outbreak (Nickell, et al., 2004). A recent study of healthcare workers adherence to hand hygiene, glove use, eye protection and respiratory protection for coughing patients with influenza-like illness showed proper use of respiratory protection (33%) and proper use of eye protection (22%) by nurses to be the most problematic (Turnberg, et al., 2008). Many barriers to using facial protection have been identified by front line healthcare workers including physical discomfort, difficulty breathing, adverse skin reactions, anxiety and communication breakdown (Osborne, 2003).

This research study aimed to examine facial protective equipment and workers adherence to its‘ recommended use. This has been identified as a critical area requiring research by several authorities on occupational health and infection control including the Occupational Health and Safety Agency for Healthcare (OHSAH) in British Columbia, Canada and the National Institute for Occupational Safety and Health (NIOSH) in the United States. In their review of the literature to identify knowledge gaps and research priorities for effective protection against occupationally-acquired respiratory infectious diseases, OHSAH found that failure to implement appropriate barrier precautions and hand hygiene was responsible for most nosocomial transmission of SARS (Yassi, et al., 2004). Understanding why there was a failure to implement appropriate precautions and how to best promote adherence in the future was identified as a research priority. In 2006, NIOSH asked the Institute of Medicine (IOM) to conduct a study on the personal protective equipment needed by healthcare workers in the event of an influenza pandemic. One of three critical areas identified for expeditious research and policy action was adherence to the proper use of personal protective equipment to prevent communicable respiratory illness. In particular, the IOM recommended that research focus on the identification and dissemination of best practices for improving compliance and the human factors and behavioral issues related to ease and effectiveness of equipment use for extended periods and in patient care-interactive work environments (Institute of Medicine, 2007).

2.3 Theoretical Models and Self-Protective Behaviours

A variety of theoretical models have been developed to explain why people adopt or reject self-protective behaviours. In a review of the theoretical models that have been used to analyze and predict health-related behaviour, three categories of models were identified; value- expectancy, behavior change and environmental/contextual models (DeJoy, 1996). Following are brief descriptions of the value-expectancy and behaviour change models and a more in depth description of the PRECEDE model that was chosen to frame this research study.

2.3.1 Value-Expectancy Models

The first category of models includes those that are based on a person‘s value expectancy. These are decision making or cost-benefit models that have occupied the health literature for the last 30 years (DeJoy, 1996). Value-expectancy models are based on the notion that individuals estimate the seriousness of risks, evaluate the costs and benefits of various actions and then choose a specific course of action that will maximize a desired outcome (DeJoy, 1996). Three prominent examples are the Health Belief Model (Becker, 1974), the Theory of Reasoned Action/Theory of Planned Behavior (Ajzen & Fishbein, 1980) and Protection Motivation Theory (Rogers, 1983). Although these three models are slightly different, all three emphasize a person‘s threat-related beliefs or perceptions.

2.3.1.1 Health Belief Model

The Health Belief Model was designed specifically to explain health behaviour. The model predicts that an individual‘s perceived threat of disease will determine their likelihood of adopting healthy behaviours. The individual‘s perceived threat of disease is determined by: 1) individual perceptions of susceptibility to disease and disease severity 2) modifying factors such as demographic variables and sociopsychological characteristics and 3) appraisal of the benefits and costs of/barriers to the health behaviour. Certain cues to action may trigger or initiate a change in behaviour including internal cues such as adverse health symptoms or external cues such as social pressure. Individuals likely to be adherent perceive themselves as more vulnerable to the disease, perceive the disease as more serious and have evaluated the benefits of the health behaviour to outweigh the costs (Janz, Champion, & Strecher, 2002).

There is a wide range of studies supporting the Health Belief Model in explaining healthy lifestyle behaviours including dental care, dieting, AIDS risk-reduction, breast self- examination and sunscreen use (Tang & Wong, 2004). Several articles examine the application of the Health Belief Model to improving adherence to generic infection control practices (Seto, 1995) (Kretzer & Larson, 1998). In a study carried out in Hong Kong post-SARS, the Health Belief Model was used to identify determinants of individual‘s practice of wearing a face mask to prevent SARS. Perceived susceptibility to SARS, cues to action (government guidance to wear face masks) and perceived benefits of wearing a face mask were predictive of the target SARS preventive behaviour (Tang & Wong, 2004).

2.3.1.2 Theory of Reasoned Action /Theory of Planned Behavior

The Theory of Reasoned Action (TRA)/Theory of Planned Behavior (TPB) asserts the most important determinant of behaviour is an individual‘s behavioural intention. The determinants of intention include attitudes towards performing the behaviour and the subjective norms associated with the behaviour. Attitude is determined by the person‘s beliefs about outcomes or attributes of performing the behaviour. Subjective norms are determined by the person‘s normative beliefs and whether important referent individuals approve or disapprove of the behaviour (Montano, Kasprzyk, & Taplin, 2002). The TPB is an extension of the TRA and adds an additional construct concerned with perceived control over performance of the behaviour. The construct of perceived behavioural control accounts for factors outside of the individual‘s control that may affect their intention and behaviour (Montano, Kasprzyk, & Taplin, 2002). Similar to the Health Belief Model, the TRA/TPB states individuals evaluate benefits and drawbacks of health behaviours depending on the probability they will occur. In contrast to the Health Belief Model, this model gives an important role to social influences on individuals, such as the perception that subjects have of other reference individuals, subjective norms and the motivation of individuals to comply with these norms and references (DeJoy, 1996).

The TRA/TPB has been applied to a number of healthy lifestyle behaviours including exercise, weight loss, use of child car seats and smoking (DeJoy, 1996). Two studies modeled hand washing behavior among nurses on the basis of the TRA/TPB (O‘Boyle, Henly, & Larson, 2001) (Whitby, McLaws, & Ross, 2006) and one demonstrated adherence to proper hand

21 washing was predicted by nurses‘ beliefs in the benefits of the activity, peer pressure of senior physicians and administrators, and role modeling (Whitby, McLaws, & Ross, 2006).

2.3.1.3 Protection Motivation Theory

Rogers (Rogers, 1983) developed Protection Motivation Theory which is an expansion of the Health Belief Model that includes two additional factors: self-efficacy and response efficacy. The model categorizes components under two types of appraisal; 1) coping appraisal – including self efficacy, response efficacy and response costs and 2) threat appraisal – including severity, vulnerability and intrinsic rewards. Self-efficacy is the belief that one can successfully enact the recommended behavior. Response efficacy is the effectiveness of the recommended behavior in removing or preventing possible harm. Severity refers to the degree of harm from the unhealthy behavior and vulnerability is the probability that one will experience harm (Prentice-Dunn & Rogers, 1986). The construct of fear within the threat appraisal can be generated through persuasive messages designed to scare or frighten people into complying with a particular message (DeJoy, 1996). A recent example of the use of fear within the threat appraisal to evoke behaviour change was the 2008 Workplace Safety and Insurance Board‘s graphic public awareness campaign. This campaign utilized TV commercials, cinema spots, print ads, transit shelter ads, Internet advertising and the prevent-it.ca website to showcase horrific scenes from workplace accidents in an effort to reduce occupational injuries and illnesses (Workplace Safety and Insurance Board, 2008).

The Protection Motivation Theory has not been as well tested as other models. Initial studies applying this model mostly involved fear-arousing communications and attitude change (DeJoy, 1996). In a study of nurses‘ knowledge, attitudes and health beliefs on safe use of cytotoxic drugs in Israel, one author found significant correlations between several components of Protection Motivation Theory (perceived susceptibility, barriers, benefits and self-efficacy) and use of protective measures (Ben-Ami, Shaham, Rabin, Melzer, & Ribak, 2001). A very recent qualitative study employed Protection Motivation Theory to understand the origins of SARS risk perceptions and their impact on precautionary actions of Chinese residents in the United Kingdom and the Netherlands. The authors found that information from Asia heightened risk perceptions and protective behavior among the study population and, when combined with

22 low efficacy regarding precautionary measures, avoidance-based precautionary actions and unnecessary alarm appeared to dominate (Jiang, et al., 2009).

The Health Belief Model, the Theory of Reasoned Action/Theory of Planned Behaviour and Protection Motivation Theory are models based on a person‘s value expectancy and emphasize how individuals‘ attitudes, beliefs and expectations influence their reactions to various health threats. Although the individual level factors that influence behaviour are important, the main limitation of using these models to explain workplace self protective behaviour is they fail to address the situational or environmental factors in the workplace that combine with individual determinants to influence behaviour (DeJoy, 1996) (Kretzer & Larson, 1998).

2.3.2 Behaviour Change Models

A second category of models used to explain why people adopt or reject self-protective behaviours includes those that focus on the process of behaviour change. Examples include the Transtheoretical Model (Prochaska & DiClemente, 1982) and the Precaution-Adoption Process (Weinstein, 1988). These models describe the behaviour change process as a series of stages. People at different stages in the process require different interventions to move them towards healthy behaviours.

2.3.2.1 Transtheoretical Model

The Transtheoretical Model (TTM) stems from an analysis of different theories of psychotherapy and consists of five core constructs: stages of change, processes of change, decisional balance, self-efficacy and temptation. In the TTM, change is a process involving progression through six stages: precontemplation, contemplation, preparation, action, maintenance and termination (DeJoy, 1996). In addition, Prochaska included relapse as a feedback loop to reflect return from action or maintenance to an earlier stage. Specific processes of change are identified for each stage. For example, for movement from precontemplation to contemplation, the processes of consciousness-raising, dramatic relief and environmental reevaluation are identified. Similarly, the model indicates that interventions to change behaviour must also be stage-matched to be successful. Similar to value-driven models, the construct of decisional balance reflects the individuals relative weighing of the pros and cons of the change

23 and the construct of self-efficacy is the situation-specific confidence to cope. Temptation refers to the intensity of urges to engage in a specific habit in the midst of a difficult situation (Prochaska, Redding, & Evers, 2002).

The Transtheoretical Model has been used to examine healthy lifestyle behaviours such as smoking cessation (Aveyard, Lawrence, Cheng, Griffin, Croghan, & Johnson, 2006) (Aveyard, Massey, Parsons, Manaseki, & Griffin, 2009) and diabetic dietary interventions (Salmela, Poskiparta, Kasila, Vahasarja, & Vanhala, 2009). The TTM has also been used as a framework for defining stages of workers‘ acceptance of hearing protection devices (Raymond & Lusk, 2006). The authors tested a new hearing protection device staging algorithm. Results showed the identification of distinct stages of hearing protection device use reflects the behavioral diversity among workers and supports the need for stage-specific training. Stages of change theory has also been applied to hand hygiene adherence (Cole, 2006). The author suggests that the TTM explains the antagonism nurses feel towards educators and infection control practitioners when they try to force change in practice when staff are not ready. To enhance and sustain motivation to conduct hand hygiene, the author suggests motivational interviewing to assist the nurse to work through their ambivalence about behaviour change that exists at the pre-contemplation and contemplation stages (Cole, 2006).

2.3.2.2 Precaution-Adoption Process Model

The Precaution-Adoption Process Model is another model that focuses on the process of behaviour change (Weinstein, 1988). Weinstein felt that most traditional behavioral theories apply only to people who are engaged by the threat and do not take into account different stages in which people might be with regard to the adoption of the behavior. Similar to the Transtheoretical Model, Weinstein described the process of adopting or quitting a behavior as a logical sequence of qualitatively different cognitive stages ranging from complete ignorance to regular performance of behaviour. A refined version of these stages includes: (1) being unaware of the issue, (2) being aware of the issue but not personally engaged, (3) being engaged and deciding what to do, (4) having decided not to act, (5) having decided to act, (6) acting, and (7) maintenance (Weinstein & Sandman, 1992). A main feature of this model is the focus on personal susceptibility demonstrating the idea that, for many different hazards, people consider their own risk to be considerably less than that of other people (DeJoy, 1996). Unlike the TTM,

24 the Precaution-Adoption Process Model asserts that people usually pass through the stages in order (Weinstein, Lyon, Sandman, & Cuite, 1998).

The Precaution-Adoption Process Model has been used to study a variety of personal health behaviors such as colorectal screening (Costanza, et al., 2005) and smoking cessation (Borrelli, et al., 2002) and has been evaluated in several studies of home radon testing (Weinstein & Sandman, 1992). No evidence of the model being used to examine self protective behaviours in the workplace was found.

The Transtheoretical Model and the Precaution-Adoption Process Model describe behaviour change as a process and suggest different kinds of information and interventions will have different levels of relevance depending on where a person is in the process. There are several limitations in applying these models to safe work behaviours. One health behaviour expert argued that as ‗‗educated professionals‘‘ healthcare providers should always be at the contemplation stage of behaviour change if their shared goal is to improve practice (Cole, 2006). This notion is supported by findings that multi-modal campaigns are highly successful at reinforcing safe work behaviour as the majority of staff already have an intention to comply (Randle, Clarke, & Storr, 2006) (Cole, 2006). A second limitation, similar to those of value- expectancy models, is the lack of an interactionist perspective. In the workplace, providing workers with information on hazards will likely raise awareness and establish personal susceptibility but organizational and environmental barriers may arrest the actual changing and maintenance of safe work practices.

2.3.3 Environmental/Contextual Model: The PRECEDE Model

A third category of models takes social and environmental factors and the context in which the individual functions into consideration. These models go beyond person-focused variables, take a more ecological approach to analyzing the determinants of health behaviour (DeJoy, 1996) and are better suited to examine work-related health behaviours. The main example of this type of model is the PRECEDE model (Green, Kreuter, Deeds, & Partridge, 1980) (Green & Kreuter, 1991) (Figure 1). PRECEDE is an acronym for predisposing, reinforcing and enabling causes in educational diagnosis and evaluation. This model was originally developed as a planning framework for health education programs and emphasizes the identification of the behavioural causes of health problems and the analysis of

25 factors related to these causes. The model suggests three sets of diagnostic factors that drive the development of intervention strategies: predisposing factors, enabling factors and reinforcing factors.

Predisposing Factors Behaviour Reinforcing Health Quality of Factors Life

Enabling Environment Factors

Figure 1: PRECEDE Model (Green, 2007)

Predisposing factors are the characteristics of the individual that facilitate or hinder self- protective behaviour such as their beliefs, attitudes and values. Enabling factors are the objective aspects of the environment or system that block or promote self-protective action such as skill, knowledge, and availability and accessibility of equipment and resources. Reinforcing factors are the rewards or punishments that follow or are anticipated as a consequence of behaviour such as performance feedback, social approval or disapproval received from coworkers, supervisors and managers, and other safety climate dimensions (Green, Kreuter, Deeds, & Partridge, 1980) (Green & Kreuter, 1991). The PRECEDE model has been applied to healthy lifestyle behaviours including eating disorders in athletes (Benson & Taub, 1993) and the self-management behaviours of parents with asthmatic children (Chiang, Huang, & Lu, 2003).

The PRECEDE model has also been adapted for application to self protective behaviour at work (DeJoy, 1986) and used to examine the factors related to adoption or rejection of safe work behaviours in the healthcare sector. In 1992, a NIOSH study of three large acute care hospitals in the United States was carried out using the PRECEDE model to examine healthcare workers‘ compliance with universal precautions to prevent transmission of AIDS/HIV and other blood borne pathogens. Factors examined included individual (predisposing) factors including perceived knowledge of universal precautions and perceived value of preventive actions; job/task (enabling) factors such as job hindrances or barriers and workload; and environmental-

26 organizational (reinforcing) factors such as availability of equipment and safety climate (DeJoy, Murphy, & Gershon, 1995) (Figure 2). Numerous analyses of the data set produced multiple reports.

Predisposing Factors

Enabling Factors Compliance with UP

Reinforcing Factors

Figure 2: Model of Compliance with Universal Precautions (DeJoy, Searcy, Murphy, & Gershon, 2000)

One of two early reports on the NIOSH study examined the three levels of factors related to compliance with universal precautions among nurses at one of the three hospitals (DeJoy, Murphy, & Gershon, 1995). Results showed that job/task and environmental-organizational factors were the best predictors of nurses‘ adherence to universal precautions to prevent occupational transmission of blood-borne disease. A second early report described the predictors of adherence among all healthcare workers at all three study hospitals and found perceived organizational commitment to safety (reinforcing factor) to be the most significant predictor of adherence (Gershon, et al., 1995).

The first of three subsequent reports on the NIOSH study examined predictors of nurses‘ adherence to general precautions and personal protective equipment at all three hospitals and found predisposing factors were relatively unimportant for compliance with personal protective equipment (DeJoy, Searcy, Murphy, & Gershon, 2000). Following this, an analysis of healthcare workers‘ compliance at one of the three centres showed perception of a strong organizational safety climate to be significantly associated with compliance (McGovern, Vesley, Kochevar, Gershon, Rhame, & Anderson, 2000). The physician group was also examined in a third report and enhanced compliance was found to be associated with knowledge, training, the perception that protective measures were effective and the perception of organizational commitment to

27 safety (Michalsen, et al., 1997). Subsequent studies have used DeJoy‘s adapted model to examine the link between the effect organizational factors have on compliance and worker and patient outcomes such as workplace exposure incidents (Gershon, et al., 2000) (Lundstrom, Pugliese, Bartley, Cox, & Guither, 2002).

The PRECEDE model was also used to frame the research study carried out by investigators at the Occupational Health and Safety Agency for Healthcare (OHSAH) in British Columbia, Canada to identify knowledge gaps and research priorities for effective prevention against SARS and other occupationally-acquired infectious respiratory diseases (Yassi, et al., 2004). The study was funded by the Ontario Hospital Association through The Change Foundation and was comprised of an extensive review of the literature and 15 focus groups of healthcare workers in Toronto, Ontario and Vancouver, British Columbia. The theoretical framework divided factors associated with self-protective behaviour at work into three categories: organizational, environmental and individual factors (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005) (Figure 3). Results from both the literature review and the focus groups showed healthcare workers felt organizational factors in the workplace such as communication, training and policies and procedures had a greater influence on their ability to protect themselves from respiratory pathogens while at work than environmental and individual factors (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005). Priorities for further research identified by this study included examining the measures of safety climate in a healthcare institution and how to best facilitate an organizational culture that promotes safety (Yassi, et al., 2004).

Environmental Factors – Availability of resources, equipment and supplies; availability of engineering Organizational Factors - controls; environmental Individual Factors – Management’s decontamination expectations; specific Knowledge; perception of policies & procedures; risk; beliefs/attitudes; past safety climate; availability experience; peer of training and education influence; socio- programs demographic factors

Behavioural Intentions: Intention to comply with infection control/ occupational health guidelines

Figure 3: Theoretical Model for Factors Associated with Self-Protective Behavior at Work (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005)

A follow-up study carried out by OHSAH again employed this theoretical framework and examined the relationship between organizational, environmental, and individual factors and self-reported compliance with regard to general infection control practices (Yassi, et al., 2007). Findings showed compliance of workers with general infection control practices were significantly affected by the organization and environment they work in but not by their personal beliefs or attitudes.

Each of the three categories of models, value-expectancy, behavior change and environmental/contextual, contribute to understanding self-protective behaviours. A review of studies examining personal health behaviours and workplace self-protective behaviours using each of the three types of models informs us of the important constructs to focus upon when examining a very specific behaviour – adherence to recommended use of facial protective equipment. Value-expectancy models offer the person-focused variables that may play a role in adherence. Stage models emphasize that adherence is dynamic and comprised of different stages. Although informative, both of these models lack the interactionist perspective that is critical to worker self-protective behaviours: the organizational and environmental characteristics of the setting within which the work is carried out. Therefore, environmental/contextual models, specifically the PRECEDE model and Moore‘s theoretical framework, were used to frame this

29 study of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals.

2.4 Safety Climate and Adherence

A review of the literature outlined previously demonstrates the importance of a strong organizational climate of safety to worker adherence to safe work behaviours. Safety climate is based on the organizational and environmental characteristics of a company or institution. These characteristics, including health and safety policies and procedures, training programs and dedicated resources, equipment and supplies, communicate commitment to employee health, safety and wellness. Where these programs are in place, safety perceptions are more favourable and workers are less likely to engage in unsafe acts (Clarke, 2006). Safety climate refers to the shared perceptions of employees about the safety of their work environment (Gershon, et al., 2000).

Safety climate was first studied in the manufacturing and industrial sectors (Cohen, 1977) (Zohar, 1980). This research identified several key organizational and environmental characteristics of a workplace with a strong safety climate including management‘s involvement in safety programs, high status and rank for safety officers, strong safety training and communication programs, good housekeeping and recognition for safe performance (Gershon, et al., 2000). Subsequent studies have examined safety climate in construction (Dedobbeleer & Beland, 1991) (Gillen, Baltz, Gassel, Kirsch, & Vaccaro, 2002), wood processing plants (Varon & Mattila, 2000) and offshore environments (Mearns, Whitaker, & Flin, 2003) and found a strong safety climate was linked to better adherence to safe work behaviours and fewer injuries.

Safety climate within the healthcare setting was initially examined in the mid 1990s due to enhanced concern over transmission of HIV/AIDS related to poor compliance with universal precautions (DeJoy, Murphy, & Gershon, 1995). Researchers developed a new thirteen-item scale to measure safety climate that assessed respondents‘ perceptions of the extent of their hospital‘s commitment to safety in general and to universal precautions in particular. Specific questions were developed to address the existence of workplace safety committees, safety manuals and written procedures, safety specialists on staff, policies related to supervision, accountability regarding safety, policies on reporting violations, safety training and the accessibility and availability of protective equipment and engineering controls. Findings showed

30 compliance with universal precautions was strongly correlated to organizational safety climate (Gershon, et al., 1995). This tool was further revised and used to examine predictors of nurses‘ acceptance of an intravenous catheter safety device (Rivers, Aday, Frankowski, Felknor, White, & Nichols, 2003) and adherence to safe needle precautions among hospital workers (Vaughn, McCoy, Beekmann, Woolson, Torner, & Doebbeling, 2004).

In a subsequent study of safe work practices to prevent occupational transmission of blood borne pathogens, the same investigative team revised the safety climate questionnaire based on qualitative data and a review of the literature. A new 20-item hospital safety climate scale was developed and included items categorized into six organizational dimensions: equipment availability, management support, absence of job hindrances, feedback/training, cleanliness/orderliness and minimal conflict/good communication. The questionnaire was tested on hospital-based healthcare workers and senior management support was found to be associated with higher compliance and fewer workplace exposure incidents (Gershon, et al., 2000). Using the same data, a condensed version of this safety climate scale (6-items) was validated and showed acceptable internal consistency, reliability and validity (Hahn & Murphy, 2008). The 20-item scale has been theoretically applied to adherence to hazardous drug guidelines (McDiarmid & Condon, 2005) and modified to address respiratory exposures (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008) (Turnberg & Daniell, 2008). A slightly revised version of the scale used to address respiratory exposures by Nichol et al. was employed in this study of adherence to recommended use of facial protection equipment.

2.5 The Relationship between Worker Safety and Patient Safety

2.5.1 History of Patient Safety in Canada

Patient safety has been on the international health agenda since the release of the Institute of Medicine report, To Err is Human, in 1999. Based on findings from two research studies, this report estimated at least 44,000 people, and perhaps as many as 98,000 people, die in American hospitals each year as a result of medical errors that could have been prevented. One of the report‘s main conclusions was that the majority of medical errors do not result from individual recklessness or the actions of a particular group. More commonly, errors are associated with faulty systems, processes, and conditions that lead people to make mistakes or fail to prevent them (Institute of Medicine, 1999).

Following the release of the IOM report, the National Steering Committee on Patient Safety was formed in Canada. In September 2002, this committee issued a comprehensive report, Building a Safer System, which proposed a national integrated strategy for improving patient safety in the Canadian healthcare system. One of its key recommendations was the establishment of a Canadian Patient Safety Institute (CPSI), intended to promote innovative solutions and to facilitate collaboration among governments and stakeholders to enhance patient safety. In December 2003, Health Canada officially created and announced funding for the CPSI (Canadian Patient Safety Institute, 2008).

Since 2003, patient safety has been a national priority in Canada and is defined as ―the reduction and mitigation of unsafe acts within the healthcare system, as well as through the use of best practices shown to lead to optimal patient outcomes‖ (Davidson Dick, Weisbrod, Gregory, Dyck, & Neudorf, 2006). Those involved in and leading Canada‘s patient safety agenda positioned patient safety as a distinct and separate entity from worker safety. This viewpoint is slowly changing. Risk-critical industries, such as aviation, have, for many years, recognized the potential for human error, and have adopted holistic health and safety systems to prevent, mitigate, and manage potential risk of injury to staff, customers, and the public (Helmreich, 2000). The connection between a safe, engaged workforce and a profitable business model with few or no catastrophic incidents has been recognized and acted upon. Healthcare is also a high-risk industry but has only recently discovered the relationship between patient and worker safety, and that quality care and the reduction of adverse patient outcomes is dependent upon a healthy and safe workforce and environment.

In a study of critical incidents and adverse events in Canadian hospitals in 2000, 7.5% of adult clients experienced an adverse event during hospitalization, and 36.9% were said to be preventable (Baker, et al., 2004). These figures suggest that, of the almost 2.5 million annual hospital admissions in Canada similar to the type studied, about 185 000 are associated with an adverse event and close to 70 000 of these are potentially preventable (Baker, et al., 2004). Like errors in the aviation industry, medical errors are becoming seen as inherent in the work of all healthcare professionals resulting from an interplay of complex factors and rarely due to the carelessness or misconduct of an individual (Kuhn & Youngberg, 2002). These errors can result in real or potential harm to any healthcare participant including employees, patients and public visitors.

2.5.2 Link between Patient and Worker Safety

As a result of the realization that patient safety and worker safety are intricately linked, there has been a surge of research examining the relationship between these constructs. The areas that have shown a positive relationship include shift work and fatigue, staffing levels and workload, hazard-based safety programs and organizational safety climate. The literature also revealed several studies demonstrating a negative relationship between patient and worker safety.

The first area demonstrating a positive relationship between patient and worker safety was shift work and fatigue. One study examined the log books of nurses and found the risks of making an error were significantly increased when work shifts were longer than twelve hours, when nurses worked overtime, or when they worked more than forty hours per week (Rogers, Hwang, Scott, Aiken, & Dinges, 2004). A study of errors committed by medical residents found those who worked fewer hours and got more sleep had less than half the rate of attentional failures while working during on-call nights (Lockley, et al., 2004). A randomized control trial in the intensive care unit setting demonstrated that modification of intern work schedules reduced rates of serious medical errors by 26% (Landrigan, et al., 2004). A study of physician health and wellness found that the detrimental effects of long working hours and lack of sleep were linked to poor quality and substandard client care (Taub, Morin, Goldrich, Ray, & Benjamin, 2006).

A second area of focus has been nurse staffing levels and workload. The National Survey of the Work and Health of Nurses revealed that 27% of nurses surveyed felt that quality of care had deteriorated over the previous year due to fewer staff (67%) and too many clients (38%)

(Canadian Institute for Health Information, 2006). A systematic review of the literature on the effects of nurse staffing on patient, nurse and hospital outcomes published between 1980 and 2003 found richer nurse staffing was associated with lower failure-to-rescue rates, lower inpatient mortality rates and shorter hospital stays (Lang, Hodge, Olson, Romano, & Kravitz, 2004). A discussion article examining the link between nursing variables and patient outcomes stated Canadian healthcare leaders must act upon the working conditions of nurses, specifically staffing ratios and skill mix, as they have a direct relationship to client outcomes such as satisfaction, morbidity, and mortality (Nicklin & Graves, 2005).

Specific hazard-based programs to protect health workers from injuries and illness have also been examined for their effect on patient outcomes. One example is in the area of communicable disease prevention (Yassi & Hancock, 2005). Vaccinating healthcare workers has been used for many years to control influenza outbreaks in patient populations (Carman, et al., 2000). During the 2003 outbreaks of SARS in Toronto, the healthcare setting was the main site of disease transmission. As a result, protecting healthcare workers was the main defense against spread of SARS to patients and to the public at large (Nicholson, 1998). Unfortunately, in Toronto vaccinating healthcare workers alone was not enough to stop the outbreak. Another example is the prevention of musculoskeletal injury. The use of ceiling lifts to reduce musculoskeletal injuries in staff resulted in improved reports of comfort and perception of safety among nursing home patients (Yassi & Hancock, 2005). The introduction of a toileting program using lift equipment also resulted in reduced resident agitation and a lower risk of aggressive behaviours (Engst, Chhokar, Robinson, Earthy, & Yassi, 2004).

Employee perceptions of the safety climate of their work environment have also been examined in relation to patient outcomes. One study examined the safety climate in three general hospitals in Israel and found less treatment errors on units where employees perceived worker safety to be a high corporate priority (Katz-Navon, Naveh, & Stern, 2005). A secondary analysis of data from an international study examined Canadian nurses working in 292 acute care hospitals in three provinces. Findings showed strong nursing leadership created conditions for a supportive work climate, which in turn directly affected patient safety outcomes including reductions in falls, nosocomial infections, medication errors and patient complaints (Spence Laschinger & Leiter, 2006). Other safety climate variables identified to cause patient errors in healthcare included lack of communication or miscommunication, lack of safety procedures,

34 inadequate supervision, breaks in continuity of care and excessive workload (Yassi & Hancock, 2005).

Reports demonstrating a negative or inverse relationship between efforts to protect workers and patient safety also exist. Two studies examined MRSA rates before, during and after the SARS outbreak in Toronto (Poutanen, Vearncombe, McGeer, Gardam, Large, & Simor, 2005) and Hong Kong (Yap, et al., 2004) and showed either no change (Toronto) or increased MRSA rates (Hong Kong) during the outbreak. Authors concluded this could have been due to the reuse of personal protective equipment as a result of equipment shortages or because healthcare workers were primarily concerned with protecting themselves and may have either not clearly understood or neglected measures intended to prevent infection transmission between patients. Another report suggested that diminished speech intelligibility associated with respirators worn by healthcare workers may result in adverse patient outcomes (Radonovich, Yanke, Cheng, & Bender, 2010).

2.5.3 Measuring Patient Safety Climate

A positive relationship between employee safety and patient safety seems to lie within the organizational characteristics or safety climate of a healthcare workplace. Scales to measure the worker safety climate of an organization have been developed and tested (Gershon, et al., 1995) (Gershon, et al., 2000) (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008). Similar scales to measure the patient safety climate of an organization exist. A literature review of tools used to measure patient safety climate in healthcare settings revealed nine surveys; five for general use, two for use at the unit level and two for specific settings (pharmacy and transfusion) (Colla, Bracken, Kinney, & Weeks, 2005). Tools ranged in size from very large (194 items) to concise (21-items). The shortest scale, the 21-item Safety Climate Survey, and its condensed version, the Safety Climate Scale (10-items), were adapted from tools used in the aviation industry (Colla, Bracken, Kinney, & Weeks, 2005). The Safety Climate Survey was used to measure patient safety climate in 12 hospitals in 4 European countries and the United States (Sexton, Thomas, & Helmreich, 2000). The Safety Climate Scale was used to measure patient safety climate in a large urban medical centre in the United States (Pronovost, et al., 2003). Both instruments were administered in a study of four university-affiliated intensive care units (N=313) in Hamilton, Ontario, Canada and tested for internal consistency and test-retest

35 reliability (Kho, Carbone, Lucas, & Cook, 2005). Results showed both scales had construct validity and sufficient reliability. Using Cronbach‘s alpha, the internal consistency of the Safety Climate Survey and Safety Climate Scale was 0.86 and 0.80, respectively. Test re-test reliability of the scales was 0.92. As a result, the shorter Safety Climate Scale was chosen for this study of adherence to facial protective equipment.

The literature has shown that organizational factors and other safety climate dimensions are important determinants in healthcare worker adherence to safe work behaviours (DeJoy, Searcy, Murphy, & Gershon, 2000) (Gershon, et al., 2000). As healthcare worker safety is closely tied to patient safety, organizational factors should also be examined in association with patient outcomes (Lundstrom, Pugliese, Bartley, Cox, & Guither, 2002). Surveys have been designed to measure patient safety climate (Colla, Bracken, Kinney, & Weeks, 2005) and can be incorporated into occupational health and safety research initiatives examining hospital safety climate and worker outcomes. This study of adherence to recommended use of facial protection employed the Safety Climate Scale (Pronovost, et al., 2003) to investigate the relationship between nurses‘ perceptions of worker safety climate and patient safety climate.

2.6 Synthesis of the Literature

The history of infection control standards regarding communicable respiratory illness has provided important background information on the development of protective standards and the emergence of facial protection as an important barrier to occupational transmission of disease. Important gaps in the literature still exist indicating that further research is necessary to ensure standards for the protection of workers are evidence-based. The literature on adherence to infection control practices, including facial protection, demonstrated that substandard adherence has been a historical problem where solutions have either had minimal impact or proven to be unsustainable. The area of blood borne disease has been fairly well studied using various models and approaches; examining both general and specific work practices. The majority of studies were survey-based yielding descriptive and correlational results. A few studies examined objective outcomes such as needle stick injuries, workplace exposures and MRSA transmission rates.

Only recently has literature emerged examining communicable respiratory illness. Several studies have been published on adherence to general work practices but no literature was

36 found examining facial protective equipment specifically, demonstrating this to be a very new area of investigation. Again, most studies were survey-based yielding descriptive and correlational results. Observational studies were uncommon due to the challenges of observing work practices under isolation conditions that often require closed doors and curtains. Those that were carried out utilized substandard definitions of adherence in an effort to compensate for logistical barriers. Recent survey-based studies used contextual models that examined the context within which the person operates. A contextual model that has been used within the health sector, the PRECEDE model, was used to frame this study. As important constructs in this model, a review of organizational factors and other safety climate dimensions of adherence were presented including evidence of their relevance to worker and patient safety. The link between worker and patient safety was identified as an important area of research as strategies employed to enhance worker safety may not only result in a healthier workforce but a safer and healthier patient population as well.

2.7 Critical Review of Methods Chosen

In an effort to devise a rigorous study design, a review of the methodological literature on measuring adherence was carried out. Several studies examining adherence to safe work behaviours showed reported behaviour did not always predict actual, observed behaviour. Studies comparing the two methodologies and testing interventions were carried out in the areas of hand hygiene and universal precautions. Observational studies of safe work practices to prevent the spread of communicable respiratory illness were conducted but partial definitions of adherence were used due to barriers to observing work practices in an isolation setting. While observations of work practices were shown to reveal more accurate measures of adherence, where barriers to conducting observations existed, unplanned or on-the-spot demonstrations were suggested as alternate measures of adherence. As a result, the observational phase of this study was designed to provide opportunities for true observations and on-the-spot demonstrations to be carried out.

2.7.1 Studies Comparing Observed and Self-Reported Adherence

Observational studies of hand hygiene activities, including hand washing and the use of alcohol-based hand sanitizer, were identified as fairly simple to carry out due to the high number and central location of sinks and pumps in a health institution. One study of hand hygiene

37 adherence in 120 critical care nurses reported a substandard correlation between observed and self-reported adherence (O‘Boyle, Henly, & Larson, 2001). A second study of 71 healthcare professionals at one British hospital demonstrated self-reported hand hygiene behaviour had no relationship with actual observations (Jenner, Fletcher, Watson, Jones, Miller, & Scott, 2006). A third discussion paper reported that the majority of studies demonstrated estimates in hand hygiene self-reporting exceeded observed performance (Rickard, 2004).

Observational studies of adherence to universal precautions, including hand hygiene and the use of gowns, gloves, masks and goggles, were identified as more difficult to conduct as the donning of specific protective equipment often occurred inside the patient‘s room and/or behind closed curtains. An observer standing in the hallway would not be able to see the worker donning and using the equipment. To mitigate this barrier, one group of American investigators compared observed and self-reported compliance with universal precautions in an open-concept emergency department setting. Findings showed self-reported rates of compliance with universal precautions were significantly higher than observed rates (Henry, Campbell, & Maki, 1992). The collection of self-report and observational data in this study provided the opportunity to compare these two measures of adherence.

2.7.2 Intervention Studies

Several studies have tested interventions to improve observed adherence to safe work behaviours. One US study conducted in the critical care setting observed the effects of performance monitoring and feedback on hand hygiene adherence. Findings showed multilevel feedback on infection control process measures and rates of healthcare associated infections significantly improved hand hygiene adherence (Assanasen, Edmond, & Bearman, 2008). A second study observed compliance with contact precautions to control transmission of antibiotic- resistant organisms. Findings showed implementing a method of feedback and accountability enhanced compliance with hand hygiene and the use of gowns, gloves and masks and reduced transmission rates of MRSA (Cromer, et al., 2004). A Canadian study of observed compliance with proper use of personal protective equipment found online infection control courses transferred knowledge and improved hand hygiene and proper equipment selection and use (Hon, et al., 2008).

2.7.3 Observational Studies of Adherence

Observational studies of adherence with safe work practices to prevent transmission of communicable respiratory illness were not abundant due to the numerous organizational and environmental barriers present in the healthcare setting. These barriers included the logistical difficulty of conducting observations of patient care activities from outside an isolation room that has curtains or blinds and the lack of natural opportunity to observe a worker using facial protective equipment. To manage these barriers, several studies within the health sector used only partial measures to define adherence to proper use of an N95 respirator. One US study of three hospitals with high incidence rates of tuberculosis defined compliance as ―observed donning an N95 respirator correctly when entering the patient room‖ (Sutton, Nicas, & Harrison, 2000). A second study of two US hospitals also examined compliance with TB control practices and defined compliance as ―worker chose to use an N95 respirator over a surgical mask‖ (Tokars, et al., 2001). A third observational study in the pediatric health setting defined compliance as ―observed to be wearing an N95 respirator correctly when entering and exiting the patient‘s room‖ (Kellerman, Saiman, San Gabriel, Besser, & Jarvis, 2001). In a Brazilian study, evaluation of N95 respirator use was conducted and workers were randomly observed for the use of a respirator when entering and exiting high-risk areas or performing high-risk procedures. Investigators also quantitatively measured the adequacy of facial fit and facial-seal leakage when the worker left the patient room (Biscotto, Pedroso, Starling, & Roth, 2005).

2.7.4 Measuring Competence

In anticipation of the presence of major barriers to conducting true observations of adherence to recommended use of facial protective equipment in the critical care setting, this study provided the opportunity to measure competence through requested or on-the-spot demonstrations. Where a natural opportunity to covertly observe use of facial protection did not present itself after two attempts, the observer would ask the participant to demonstrate the proper donning, doffing and disposal of personal protective equipment required for a patient on airborne precautions. Through this demonstration, the observer could measure the competence of the participant with recommended use of an N95 respirator.

Competence can be defined as the demonstrated ability to apply knowledge and skills (International Organization for Standardization, 2000). Competence is different from adherence.

Adherence refers to the degree to which an individual follows a set of guidelines in a real-life setting (Kak, Burkhalter, & Cooper, 2001). The relationship between competence (can do) and adherence (does do) is complex: the first does not always predict the second (Southgate & Dauphinee, 1998) (While, 1994). Competence is someone‘s capacity to perform, whereas adherence is their actual performance within a real-life setting. While competence does not always lead to effective performance or adherence (While, 1994), it could be seen as one of its many determinants.

In the nursing literature, much attention has been given to the measurement of competence versus performance. Manual skills were identified as the foundation of nursing performance and continue to be the focus of evaluation in the academic environment. Many disagree with this competence-based model of training and evaluation and identify additional areas of importance for consideration (While, 1994). These include assessment skills, cognitive skills, decision making, and the ability to deal with unexpected, complicated or challenging situations requiring creativity and sensitivity.

Although competence is different from adherence, some parallels can be drawn. One author claimed that competence assessments can have predictive value for performance in actual practice, if efficiency and consultation times are taken into consideration (Rethans, Sturmans, Drop, Van der Vleutens, & Hobus, 1991). Another author identified writing short clinical reports under time pressure may require similar skills to writing exam answers (McFarlane, Goldney, & Kalney, 1989). Regardless, there is a large body of literature suggesting that competence on its own is not sufficient for adequate professional practice (While, 1994).

This critical review of methods chosen emphasizes the inadequacy of the self-report approach when examining adherence to safe work behaviours and suggests that adherence should be measured by actual observation. To address logistical barriers, observational studies of practices to prevent transmission of communicable respiratory illness have traditionally utilized incomplete definitions of adherence. The Ontario Ministry of Health and Long Term Care and the Public Health Agency of Canada outlines the steps to follow when donning, using, doffing and disposing of an N95 respirator (Ontario Ministry of Health and Long-Term Care, 2003). Guidance on donning specific brands of N95 respirators are also provided by each manufacturer (3M, 2004). This study attempted to use a more complete definition of adherence based on

40 government and manufacturers guidelines. Measures of observed adherence incorporated those steps that were identified as critical to the protection of a workers‘ health. Where barriers precluded observations of adherence, demonstrations were requested and measures of competence with recommended use of an N95 respirator were recorded.

2.8 Pilot Study

A pilot study was carried out in 2006 to test a new survey tool and provide the rationale to pursue this larger, more comprehensive study of the factors that influence nurses use of facial protective equipment. The pilot study used a cross-sectional survey design and targeted nurses that regularly used facial protection at two acute care hospitals in downtown Toronto, Canada. Following ethics approval at both sites, two focus groups were held to test the survey tool.

Study participants included 177 nurses who regularly used respirators, surgical masks and eye protection. Results showed that environmental and organizational factors were significant predictors of nurses‘ compliance with the use of facial protective equipment. These self-reported factors included cleanliness and orderliness of the workplace, ready availability of equipment, media coverage of communicable respiratory diseases, organizational support for health and safety and an absence of job hindrances. Internal consistency testing demonstrated construct validity for 10 of 13 measures within the survey. Results from this pilot study provided the justification to pursue this area of research further and were used to inform this study. See Appendix 2 - Published Manuscript of the Pilot Study - for a copy of the paper outlining the results of this study.

3 CHAPTER 3: Methods 3.1 Study Design

This study used a two-phased approach. Phase 1 of the study was a cross-sectional survey of nurses (registered nurses and registered practical nurses) who regularly used facial protective equipment to prevent occupational transmission of communicable respiratory illness. Phase 2 was a direct observational study of nurses donning, using, doffing and discarding facial protective equipment in the intensive care unit setting.

The nursing occupation was chosen for several reasons. Research on the Canadian workforce in 2002 showed that the absenteeism rate related to illness and injury for full-time registered nurses was 83% higher than the rate for the full-time workforce (O'Brien-Pallas, et al., 2004). During SARS, healthcare workers experienced the heaviest burden of illness, and attack rates differed among the professions. In a Japanese study, attack rates were highest among nurses (35%), followed by physicians (16%) and administrative staff (2%) (Kho D. , 2005). In Toronto, two nurses died of SARS-related causes (The SARS Commission, 2006). Nursing is also the largest single profession in the healthcare sector, making up 60% of all regulated health professionals (Registered Nurses Association of Ontario, 2004). Nurses spend more face-to-face time with patients on the front-line than any other type of healthcare worker and dominate mid- level management with a variety of organizational- and system-level responsibilities (Stirling, 2004). The 2005 National Survey of the Work and Health of Nurses revealed that nurses reported an increased prevalence of risk factors that contributed to high rates of injury and illness including restructuring, job instability, increased workloads, more overtime, increased verbal aggression, frequent shift changes and an aging workforce (Canadian Institute for Health Information, 2006). Being a non-practicing registered nurse herself, the graduate student researcher also has a personal interest in studying nurses.

A cross-sectional study design was chosen to provide a current "snapshot" of adherence to facial protective equipment and to determine if there was a relationship between adherence and a set of predetermined individual, environmental and organizational variables. This study design is cost-effective and can be carried out in a busy, congested and complex work environment with minimal interruption of workflow. Convenience or opportunity sampling was used and attempts to reach as many nurses as possible were made to ensure the results were

42 representative of the population. Random sampling and a mail-out survey were not used as the research team was concerned this strategy would yield a low response rate within the target population.

The ICU setting was chosen for the observational study for several reasons. The frequency of use of facial protective equipment may be higher in this setting as critically ill patients may be more infectious at the time of their admission and may require closer staff- patient interaction and more aerosol-generating procedures than patients on regular care units (Muller & McGeer, 2006). A Canadian study of SARS among critical care nurses found nurses who assisted with suctioning and/or manipulated patients‘ oxygen masks were at higher risk of contracting the illness (Loeb, et al., 2004). The physical make-up of this setting also provided the best opportunities to conduct true observations of nurses‘ work practices. Reasons for this include: close proximity of the nurses‘ work/documentation area to the patient care area made it easy for the researcher to observe the nurse from the nursing station; the use of large glass windows to see into patient care areas allowed the researcher to observe the nurse while wearing facial protection and caring for the patient; the use of large patient assignment boards allowed the researcher to identify patient assignments for regular care and coverage during breaks; and, rapid turnover of patients provided a higher probability that the observer would be present when participating nurses were assigned to patients requiring respiratory precautions.

3.2 Enrolment

3.2.1 Hospital Enrolment

Eight acute care hospitals in the Greater Toronto Area that did not participate in the pilot were contacted to participate in the study and six accepted. Initially, the Vice President and/or Chief Nursing Officer (VP/CNO) of the hospital was contacted by the researcher to schedule an appointment to discuss recruitment of the hospital as a site for study. A two-page summary of the study (Appendix 3) and the survey tool were provided for review and a more detailed research proposal was made available upon request. The VP/CNO was chosen as the primary contact to ensure senior nursing commitment to the study. This senior nursing commitment facilitated ethics approvals and assisted with securing support for the study from unit managers, nursing union officials, infection control practitioners and front line nurses. Ethics approval was obtained from the administering hospital, the University of Toronto and all participating

43 hospitals. Each hospital required a customized version of the consent form on hospital letterhead. All ethics approvals were renewed annually.

3.2.2 Subject Enrolment

Once participation was confirmed and ethics approval received, the VP/CNO provided the researcher with a list of units where facial protection was regularly worn by nursing staff. Units where nurses regularly used facial protection were targeted for this study in an attempt to collect accurate and reliable information. These units included intensive care units (ICUs), emergency departments, step-down units, respirology units or clinics and other in-patient medical units that regularly received patients with respiratory symptoms.

Forty-six units within the six acute care hospitals were identified as employing nursing staff that used facial protective equipment at least monthly to prevent occupational transmission of communicable respiratory illness. Each unit manager was contacted by the researcher and a face-to-face meeting was scheduled to discuss participation in the study. The two-page summary of the study and the survey tool were provided for review and a more detailed research proposal was made available upon request. All unit managers agreed to participate in the study. In one hospital, two units were identified as sharing the same staff. These units were combined for study purposes, resulting in a total of 45 participating units.

Appointments to recruit subjects and collect data on the units were scheduled through each unit manager. The researcher conducted four visits to each unit for two to four hours each visit. In the majority of cases, survey distribution was done on an individual-basis with the researcher walking through the unit distributing the information sheet and questionnaire. In some cases, at the request of the unit manager, the information sheet and questionnaire were handed out by the researcher during an education session or staff meeting.

3.3 Data Collection

3.3.1 Cross-Sectional Survey

Phase 1 of the study, the cross-sectional survey, commenced in October, 2007. Potential participants were provided with an information sheet outlining the study (Appendix 4) and the questionnaire. Time was provided for nurses to read the information sheet and ask any questions.

They were informed that participation was voluntary and individual responses would not be identified in any presentation or publication arising from the study results. Consent was implied when participants agreed to complete the questionnaire. Coffee vouchers were offered as tokens of appreciation ($5 value).

Nurses filled out the cross-sectional survey during work hours. If participants were able to complete the questionnaire during the visit, the researcher collected the survey back the same day. If they were not able to complete the questionnaire during the visit, a drop box was made available on the unit. At the same time the surveys were being distributed, information on the cleanliness and orderliness of the unit and the ready availability of facial protection was collected by the researcher on a unit-by-unit basis. Agreement between objective unit observations and self-report data from the survey was planned in an attempt to validate these measures. The total number of nurses working on the unit and the number of nurses present on study days were obtained to calculate response rates and participation rates.

In all units except the Intensive Care Units (ICU) the survey was anonymous. In the ICU, the surveys had number identifiers to facilitate matching observed data from Phase 2 of the study with self-report survey data from Phase 1. During Phase 1 of the study, nurses working in the Intensive Care Unit were recruited to participate in Phase 2. Nurses were asked to read the information sheet, complete the survey and provide written consent for a second researcher to return to the unit one to three months later to observe the nurse donning, using, doffing and disposing of facial protective equipment. ICU nurses could choose to only complete the survey, complete the survey and consent to be observed at a later date or not to participate at all. See Appendix 5 for a generic version of the information sheet and consent form used for nurses working in the ICU.

3.3.2 Direct Observational Study

Phase 2 of the study, the observational study conducted in the intensive care setting, was commenced in January, 2008 and was carried out by a trained research assistant. Training provided included review of the Ministry of Health and Long Term Care Routine Practices and Additional Precautions document, review of two educational videos on use of personal protective equipment to prevent occupational transmission of disease in the healthcare sector, a tour of a large ICU and a demonstration of the proper donning, use, doffing and disposal of facial

45 protective equipment by the ICU Clinical Educator. After training was completed, the research assistant was able to demonstrate the correct use of facial protective equipment successfully.

To schedule dates for observations, the unit administrator provided the research assistant with the work schedules of consenting nurses. At the start of each visit, the clerical assistant on the unit verified that the nurse was present and, if necessary, what his/her patient assignment was. Attempts were made to conduct covert observations – where the nurse was unaware that they were being observed or unaware that they were being observed as part of a specific research study. Covert observations control for the Hawthorne Effect. This effect occurs when people change their behaviour because they know they are being observed, not because of objective changes in their circumstances (Gale, 2004). Other strategies employed to control for this effect included using a trained research assistant who had not visited the units before, conducting the observational study 1 – 3 months after survey data was collected and attempting to observe the nurse from the nursing station.

Two attempts were made to observe nurses donning, using, doffing and disposing of facial protective equipment without their knowledge. If a satisfactory natural opportunity for covert observation did not occur, the observer identified themselves and requested a demonstration of the steps involved in donning, use, doffing and disposal of the facial protective equipment required when caring for a patient on airborne respiratory precautions. Only one observation was recorded per individual to ensure data was independent. Again, agreement between self-reported adherence and observed adherence was planned in an attempt to validate the outcome variable.

3.4 Measures

3.4.1 Survey Tool

This study used an 8-page, 84-item questionnaire that consisted of 13 different sections including:

A. Background Information B. Knowledge of Droplet and Airborne Spread Disease I. Using Facial Protection

II. Transmission of Influenza C. Effectiveness of Preventive Actions D. Exposure E. Perception of Risk F. Personal Barriers to Use G. Use of Facial Protective Equipment (Adherence) H. Cleanliness/Orderliness of the Workplace I. Availability of Facial Protection J. Training K. Media Coverage of Communicable Diseases L. Organizational Support M. Patient Safety Climate

The questionnaire was a revised version of the survey tool used in the pilot study. Selection of scales and questionnaire items for the original tool were, for the most part, taken from established instruments and input on content was solicited from the Ontario Nurses Association, nursing management from each pilot study hospital and occupational health and infection control experts on facial protection and airborne and droplet spread illness. Two focus groups were held to test the pilot study tool. See Appendix 2 - Published Manuscript of the Pilot Study - for details on the development of the original tool. Revisions to the pilot study tool were based on the results of reliability testing, factor analysis, the identification of outliers and expert consultation.

Initially, revisions to the pilot study tool were based on the results of reliability testing. Three of the thirteen scales demonstrated a substandard reliability score – knowledge, effectiveness of preventive actions and perception of risk. A value of ≥ 0.7 is reported to be an acceptable reliability coefficient (Nunnaly, 1978). The 8-item scale to measure knowledge of droplet and airborne spread disease demonstrated a reliability coefficient of 0.32. This low score was likely due to multiple content domains within the scale including proper use of facial protection, transmission of infection, cleaning and disinfection and government guidelines. Only scales measuring a single trait or content domain reveal a high reliability score. In an attempt to improve the reliability rating, two new knowledge scales were employed in the current study. The first scale examined participants‘ knowledge of the proper use of facial protective equipment to prevent transmission of droplet and airborne spread communicable respiratory illness. This

47 scale consisted of 8 items and was based on provincial government guidelines and educational tools developed for the acute healthcare sector (Ontario Ministry of Health and Long-Term Care, 2006) (Ontario Ministry of Health and Long-Term Care, 2007). The second scale examined participants‘ knowledge of transmission of influenza. This scale consisted of 6 items and was based on information published by the US Department of Labor (Occupational Safety and Health Administration, 2007).

The 3-item scale used in the pilot study questionnaire to measure perception of risk demonstrated a reliability coefficient of 0.64. When the item measuring risk was separated from the two items measuring impact the raw score increased to 0.79, suggesting that these domains need to be measured separately. In the revised questionnaire items measuring impact were reworded to measure level of concern due to risk. For example, ―If I contracted an infectious respiratory disease through work the impact on my own health would be minimal‖ was revised to read ―I am concerned I may catch a communicable respiratory illness at work‖.

The 4-item scale used in the pilot study questionnaire to measure perception of effectiveness of preventive actions demonstrated a reliability coefficient of 0.50. This could have been due to the discrepancy between nurses‘ perception of effectiveness of respirators and hand hygiene and their perception of effectiveness of surgical masks and eye protection. Respirators and hand hygiene were generally perceived to be more effective in preventing transmission of disease than surgical masks and eye protection. In the revised questionnaire the item examining perception of effectiveness of hand hygiene was dropped as hand hygiene does not fall under the umbrella of facial protective equipment. The rest of the items were left the same with the knowledge that individual pieces of facial protective equipment may need to be examined individually.

Further revisions to the pilot study tool were based on factor analysis. Factor analysis on the four constructs measured at the organizational level (organizational support, absence of job hindrances, feedback and conflict/communication) showed that two constructs (organizational support and conflict/communication) measured similar domains. In the revised questionnaire, all four constructs were combined and included 17 items. The grouping was given the heading Organizational Support with the intention to conduct further exploratory factor analysis to determine an appropriate factor solution.

Finally, several revisions to the pilot study tool were based on the identification of outliers and consultation with experts. Within the Background Information section, an item inquiring whether the participant was a registered nurse (RN) or registered practical nurse (RPN) was added as several unit managers indicated that both professional groups used facial protective equipment during patient care activities. As a result of including the RPN group, a fourth option to select for level of education, Certificate, was added. Frequency of Use of Facial Protection was added to this section as after discussion with the thesis committee it was identified to be a demographic characteristic. Consistent measurements of age, tenure on the unit and tenure as a nurse in completed years were used. Within the Training section, an item asking for the last year that training was received was added as length of time between training was identified as a potential variable of interest. Within the Organizational Support section, three potential variables of interest were added including: a functioning joint health and safety committee; readily available written policies for eye and respiratory protection; and, supervisory enforcement of legislation and workplace policy regarding proper use of facial protection. One item regarding professional accountability was removed as it was identified as more of a personal level variable than an organizational level variable.

For the most part, a 5-point Likert response scale of strongly agree/agree/neutral/ disagree/strongly disagree was used to allow participants to be expansive with their response. For knowledge scales, an additional response item of ‗don‘t know‘ was included. A Likert scale is a type of psychometric response scale often used in questionnaires (Likert, 1932), and is the most widely used scale in survey research. When responding to a Likert questionnaire item, respondents specify their level of agreement to a statement. Likert scale responses are used to aggregate answers from several questions by taking the mean of the answers or the sum of the answers. With shorter scales such as the 5-point scale used in this study, equal differences between answers cannot be assumed; therefore data should be analyzed as ordinal in nature (Behan & Behan, 1954) (Henkel, 1975). Where applicable, data collected via the survey tool were treated as ordinal data and reduced to a dichotomous variable for analysis. A response of strongly agree/agree indicated presence of the variable and a response of neutral/disagree/strongly disagree indicated absence of the variable. Case definitions of explanatory variables are discussed in Section 3.4.4.

Additional measures were also added to the tool including the knowledge scale on transmission of influenza (Occupational Safety and Health Administration, 2007), a new measure of worker health outcome (based on the pilot tool exposure scale) and a well-tested scale for measuring patient safety climate (Colla, Bracken, Kinney, & Weeks, 2005) (Kho, Carbone, Lucas, & Cook, 2005). Participants were invited to share any additional comments they had at the end of the questionnaire. The questionnaire was completed within 20 minutes by most participants. See Appendix 6 for a copy of the questionnaire used in this study.

3.4.2 Unit Observation Record

Due to the lack of an existing tool, a new tool to record observations of cleanliness, orderliness and ready availability of FPE at the unit level was developed (Appendix 7). Measures of the availability of facial protective equipment (N95 respirators, surgical masks and eye protection) were made at point-of-use meaning on a cart or shelf right outside the patients room or within a short walking distance (up to 3 metres away). This measure of availability was used as studies have shown ready availability of protective equipment predicted increased adherence to universal precautions (DeJoy, Searcy, Murphy, & Gershon, 2000) (Green-McKenzie, Gershon, & Karkashian, 2001). In the latter study, workers were found to be almost 3 times more likely to wear a respirator or mask if it was readily available and 4.5 times more likely to wear a gown. A well-known tool for assessing the availability of alcohol-based hand rub and clean gloves also used a point-of-use or ―near patient‖ measurement (Institute for Healthcare Improvement, 2006). A 5-point Likert response scale of strongly agree/agree/inconsistent/ disagree/strongly disagree was used to allow the researcher to be expansive with her response. Response choice definitions are outlined in Table 1. Opportunities to provide additional comments were incorporated. To ensure consistency, only the research assistant evaluated the unit and completed this tool. One record was completed for each unit.

Response Definition Choice

All rooms of patients requiring respiratory precautions had the facial protective Strongly Agree equipment available right outside the patient room or within 3 metres.

Most rooms of patients requiring respiratory precautions had the facial protective Agree equipment available right outside the patient room or within 3 metres.

Half of the rooms of patients requiring respiratory precautions had the facial Inconsistent protective equipment available right outside the patient room or within 3 metres.

Most rooms of patients requiring respiratory precautions had the facial protective Disagree equipment available further than 3 metres away or not available at all.

Strongly All rooms of patients requiring respiratory precautions had the facial protective Disagree equipment available further than 3 metres away or not available at all.

Table 1: Response Choice Definitions for Unit Observation Record

3.4.3 Participant Observation Record and Guide

A new 2-page observation tool was developed for this study (Appendix 8 - Participant Observation Record). Although several observation tools to measure adherence to proper hand hygiene (World Health Organization, 2006) (Institute for Healthcare Improvement, 2006) (Ontario Ministry of Health and Long-Term Care, 2006) (Pashman, Bradley, Wang, Higa, Fu, & Dembry, 2007) were available, no tools to measure adherence to recommended use of facial protective equipment were found. As discussed in Section 2.7.3, due to the presence of barriers, available reports on observational studies in this area have traditionally utilized incomplete definitions of adherence.

The Public Health Agency of Canada outlines the decision making process that healthcare workers must follow when determining the adequate facial protective equipment to use in specific circumstances (Public Health Agency of Canada, 1999). They Ontario Ministry of Health and Long-Term Care outlines the steps to follow when donning, using, doffing and disposing of facial protective equipment (Ontario Ministry of Health and Long-Term Care, 2003). Guidance on donning specific brands of facial protective equipment, such as N95 respirators, is also provided by each manufacturer (3M, 2004). This study attempted to use more complete definitions of adherence based on these government and manufacturers guidelines.

The observation record used in this study provided opportunities to observe nurses using facial protective equipment while caring for patients on different levels of respiratory precautions and while caring for their own patients or covering for another nurse during lunch and breaks. The three levels of precautions included: contact (level 1), contact/droplet (level 2) or airborne (level 3) respiratory precautions (Public Health Agency of Canada, 1999). Specific items of interest were identified under each level of precaution in accordance with government and manufacturers‘ guidelines. An opportunity to provide further information was provided at the end of the form. Each of the six participating organizations was given an opportunity to review the observation tool in the event their corporate policies differed from government and/or manufacturer wording or recommendations. All organizations agreed that the tool was applicable in its original form. A short observation guide was provided to help the research assistant to carry out the observations efficiently and consistently (Appendix 9 - Participant Observation Guide).

3.4.4 Explanatory Variables

Using Moore‘s model (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005), explanatory variables were categorized as demographic, individual, environmental or organizational. The majority of items were analyzed as dichotomous variables due to the ordinal nature of the 5-point scales and the variable interval difference between scale ranks. Appendix 10 presents the explanatory variables categorized using Moore‘s framework and provides the case definition for each as it was used in the analysis. The survey also included a measure of patient safety climate. This 10-item scale was taken from the literature (Pronovost, et al., 2003) (Kho, Carbone, Lucas, & Cook, 2005) and used to evaluate nurses‘ perceptions of the organizations commitment to patient safety. It was analyzed as a categorical variable in concordance with analysis used in the current study and in the literature (Pronovost, et al., 2003). A supportive patient safety climate was defined as answering strongly agree/agree with all 10 patient safety items. Patient safety climate was compared with three measures of worker safety climate (organizational support, absence of job hindrances and communication) to determine the presence of a relationship between the two constructs.

3.4.5 Outcome Measures

In Phase 1 of the study, the key outcome measure was self-reported adherence to recommended use of facial protection. Adherence was measured with an 8-item scale requiring response using a 5-point Likert scale of always/mostly/sometimes/rarely/never. This measure was developed based on Canadian federal and provincial government guidelines on preventing occupational transmission of communicable respiratory disease (Public Health Agency of Canada, 1999) (Ontario Ministry of Health and Long-Term Care, 2006) and the College of Nurses of Ontario Practice Standard on Infection Prevention and Control (College of Nurses of Ontario, 2005). This self-report measure of adherence had a Cronbach‘s Alpha test for reliability score of 0.78 in the pilot study. Consistent with explanatory measures, this outcome measure was analyzed as a dichotomous variable (adherent/not adherent) due to the ordinal nature of scale ranks and due to the absence of a threshold for adherence; even moderate adherence can result in occupational transmission of disease. A response of always or mostly to at least seven of the eight items (score ≥ 28/40 or 70%) was the definition of adherent. This definition was used as a comparator for the pilot study, was consistent with methodology found in the literature (Gershon, et al., 1995) (Yassi, et al., 2007) and ensured inclusion of at least one item on adherence to eye protection. All other responses indicated non-adherence.

In Phase 2 of the study, the key outcome measures were either adherence to recommended use of facial protective equipment (if a true observation was carried out) or competence with recommended use of an N95 respirator (if a demonstration was requested). Both outcome measures were analyzed as dichotomous variables. Participants were deemed to be adherent/competent if they carried out at least five of six critical steps when using facial protective equipment for a patient on airborne respiratory precautions or carried out at least four of five critical steps when using facial protective equipment for a patient on droplet respiratory precautions. All other cases were deemed to be non-adherent/not competent. These definitions were the result of discussions with the thesis committee in the absence of guidance from the literature. The six critical steps required for airborne respiratory precautions included: chose N95 respirator, positioned straps correctly, adjusted nose piece, conducted seal check, refrained from touching face piece during use/doffing and discarded directly into waste receptacle. The five critical steps required for droplet respiratory precautions included: chose surgical or procedure mask, positioned straps correctly, adjusted nose piece,

53 refrained from touching face piece during use/doffing and discarded directly into waste receptacle. As current Canadian guidelines direct healthcare workers to use eye protection when there is a risk of generating splash or spray to the face/eyes (Public Health Agency of Canada, 1999) (Public Health Agency of Canada, 2002), use of eye protection was not included in the case definitions of observed adherence or competence.

3.4.6 Data Management

Surveys completed from all units except the ICU were anonymous. Names were not recorded so participants could not be linked to any answers provided. In the ICU, number identifiers were recorded so self report data could be matched with observational data. Results were compiled using aggregate data only. Electronic data was transported via memory key with password protection. All electronic data is stored on a St. Michael‘s Hospital computer located in a locked and dedicated research unit at the hospital. Hard copy data is stored in a locked cabinet in the same secure research unit and will be kept for five years after the completion of the study. After five years, data will be shredded or deleted.

3.5 Statistical Analysis

3.5.1 Descriptive Analysis

Data entry was carried out through Microsoft Access and Microsoft Excel and subsequently analyzed using SAS (Statistical Analysis Software) Version 9.1.3 (SAS Institute, Inc., 2005). Initially, the data was cleaned and edited. Summary descriptive statistics were computed on all variables including frequency tables for categorical variables and means, medians, ranges and standard deviations for continuous variables. Where unit type was examined as an explanatory variable, the unit with shared employees was dropped (N=44 units) as one unit was general medicine and the other was the emergency department.

3.5.2 Exploratory Factor Analysis

In the pilot study, factor analysis on the four constructs measured at the organizational level (organizational support, absence of job hindrances, feedback and conflict/communication) indicated that two constructs (organizational support and conflict/communication) measured similar domains. In the current study, all four constructs were combined to include 17 items.

Further factor analysis was done on this grouping, entitled Organizational Support, to determine an appropriate factor solution.

Exploratory factor analysis was performed in accordance with methods suggested in the literature for organizational studies (Conway & Huffcutt, 2003) (Ford, MacCallum, & Tait, 1986) using the principal components analysis with an oblique rotation (promax) which allowed factors to be correlated to each other. For factor analysis to yield stable results, an accepted guide regarding sample size stating 50 cases is very poor, 100 is poor, 200 is fair, 300 is good, 500 is very good, and 1000 or more is excellent was utilized (Comrey & Lee, 1992). To determine the number of factors to extract, we used Kaiser‘s criterion, suggested by Guttman and adapted by Kaiser, that considers factors with an eigenvalue greater than one as common factors (Nunnaly, 1978). To interpret factor loadings, we followed generally accepted guidelines indicating loadings above 0.6 to be high and those below 0.4 to be low (Hair, Anderson, Tathan, & Black, 1998). Items with secondary factor loadings of more than half the primary factor loading were considered for exclusion (Ford, MacCallum, & Tait, 1986).

3.5.3 Kappa Statistic

Cohen's Kappa Coefficient was used to measure agreement of categorical data. The Kappa Statistic compares the agreement against that which might be expected by chance and can be thought of as the chance-corrected proportional agreement. Possible values range from +1 (perfect agreement) via 0 (no agreement above that expected by chance) to -1 (complete disagreement). Generally accepted guidelines to evaluate strength of agreement stating 0.00 is poor agreement, 0.01 - 0.20 is slight agreement, 0.21 – 0.40 is fair agreement, 0.41 – 0.60 is moderate agreement, 0.61 – 0.80 is substantial agreement and 0.81 – 1.00 is almost perfect were utilized (Landis & Koch, 1977).

3.5.4 Reliability Testing

Cronbach‘s Alpha (α) Test for Reliability was employed to describe the reliability of scales used to measure constructs within the questionnaire including adherence and each set of diagnostic factors. To determine acceptable reliability coefficients, we followed generally accepted guidelines indicating values ≥ 0.7 to be acceptable (Nunnaly, 1978).

3.5.5 Bivariate Analysis

Bivariate analysis was done using the Chi-square Significance Test for categorical data and the Wilcoxon Rank Sum Test (or Mann-Whitney Test) for non-normally distributed continuous variables (age, tenure as a nurse, tenure on the unit). If 25% or greater of cells had expected cell counts less than five, the Fisher‘s Exact (two-sided) Test was used instead of the Chi-square Test and typically yielded a slightly higher level of significance. A p-value of ≤ 0.05 was used to determine statistical significance. Although multiple bivariate comparisons were made, a Bonferroni adjustment to avoid Type 1 errors was not carried out as the comparisons were interpreted only to inform multivariable analysis.

3.5.6 Multivariable Analysis

Multivariable logistic regression analysis was used to examine the direct effects of variables of importance on adherence taking the effects of other variables into consideration. Variables were chosen for inclusion based on their level of association with adherence as determined by the bivariate analysis, as well as variables of importance from the pilot study and the literature. In situations where there were large numbers of risk factors, Hosmer and Lemeshow‘s guidance to keep those risk factors whose inclusion reached a reasonably liberal significance level, such as 0.2, was used (Hosmer & Lemeshow, 2000). The generally accepted principal of maintaining only 10 events per predictor to get reasonably stable estimates of the regression coefficients (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996) was also used to determine the number of variables to include in the stepwise logistic regression model. The c- statistic was used to measure goodness-of-fit of the logistic model. To determine acceptable discriminative power, we followed generally accepted guidelines indicating values between 0.7 and 0.8 were acceptable, and values between 0.8 and 0.9 were excellent (Hosmer & Lemeshow, 2000).

3.6 Sample Size

3.6.1 Phase 1 – Cross Sectional Survey

Sample size calculations for Phase 1 of the study, the cross sectional survey, were done using SAS Version 8 (SAS Institute, 1999) and were based on results from the pilot study (Table

2). Calculations using variables that showed a significant association with compliance at the bivariate and/or multivariable levels in the pilot study were taken into consideration. All calculations were done assuming a statistical power of 0.8 and an alpha level of 0.05 (2-sided).

Sample size needed to Sample size needed to Variable of Interest detect an increase in the detect an increase in the odds ratio of 50% in odds ratio of 100% in adherence adherence Organizational Support*† 845 282

Media Coverage*† 927 306

Cleanliness/Orderliness* 1082 390 Availability of Facial 806 272 Protection* Absence of Job Hindrances* 816 275 Frequency of Use of Facial † 1526 505 Protection Work Status† 1636 542

* significant p≤0.05 at the bivariate level † significant p≤0.05 at the multivariable level

Table 2: Cross Sectional Survey Sample Size Calculations

Sample size calculations were also done using the variable ‗tenure as a nurse‘ which was significantly associated with compliance at the bivariate and multivariable levels in the pilot study (Table 3). Calculations were done using the power procedure in SAS v.8 assuming a statistical power of 0.8, equal weighting between groups and an alpha level of 0.05 (2-sided). The standard deviation (SD) from the mean used for the calculation was 9.0 (averaged from the pilot study results of a SD of 8.84 from the mean of the non-compliant group and a SD of 9.98 from the mean of the compliant group).

Mean difference that can be Sample Size detected between groups (y) 1.0 2545 3.0 285 5.0 104

Table 3: Sample Size Calculations for Variable 'Tenure as a Nurse'

Based on these calculations, the sample size for the cross-sectional survey was established at 300 nurses. This value had adequate power to see odds ratio increases in variables

57 of interest that were similar to those of the pilot study and to detect a mean difference in tenure as a nurse of 3 years between adherent and non-adherent groups. This sample size was logistically possible to achieve in a healthcare setting where the data collection time period was limited and where front line nurses were exceedingly busy.

3.6.2 Phase 2 – Direct Observational Study

Sample size calculations for Phase 2 of the study, the direct observational study of ICU nurses, were also done using SAS Version 8 (SAS Institute, 1999) and were based on a compliance rate of 42% reported in the pilot study (Table 4). It is recognized that if a proportion is between 40 and 60%, then 50% should be used for sample size calculation, as it would lead to a larger sample size (Macfarlane, 1997). Calculations for both 42% and 50% are provided. All calculations assumed a confidence level of 95%.

Proportion Compliant Half-Width Sample Size

42% 5.0 374

42% 10.0 94

50% 5.0 385

50% 10.0 97

Table 4: Sample Size Calculations for Observational Study

Based on these calculations, the sample size for the observational study was established at 100 independent observations. Calculations showed that this value had adequate power to detect an adherence rate similar to the rate found in the pilot study +/- 10%. This sample size was logistically possible to achieve in the intensive care setting as data collection was scheduled to be carried out over the winter and early spring when rates of patients with respiratory illness increases.

4 CHAPTER 4: Results 4.1 Phase 1 – Adherence to Recommended Use of Facial Protection

4.1.1 Descriptive Analysis

A total of 1313 surveys were distributed and 1074 completed surveys were returned for a response rate of 82%. This exceeded the projected enrolment figure of 300 nurses and had sufficient power to see odds ratio increases in variables of interest that were similar to those of the pilot study and to detect a mean difference in tenure as a nurse of at least 3 years between adherent and non-adherent groups. This sample size served to be logistically possible to achieve in a healthcare setting where the data collection time period was limited to four visits per unit and where front line nurses were exceedingly busy. At the time of the study, 2127 full and part time nurses were employed on the 46 units resulting in a study participation rate of 51%. Results showed the majority of respondents were female registered nurses working full-time (Table 5).

Demographic Characteristics of Survey Population (N=1074)

Variable Level n (%) Gender Female 976 (91) RN 1018 (95) Nurse Type RPN 54 (5) Certificate 42 (4) Diploma 496 (47) Education Degree 487 (46) Masters 38 (4) Work Status Full-time 816 (77) Supervisor Status Yes 502 (47) Age, y 38.8 (20–67)* Tenure as a nurse, y 14.4 (0-49)* Tenure on the unit, y 6.5 (0-36)* Community 581 (54) Hospital Type Teaching 493 (46) Critical Care 393 (37) Unit Type Emergency 308 (29) Medicine 355 (34) * sample mean (range) Table 5: Demographic Characteristics of Survey Population

Demographic differences of significance between unit type and nurse type were also examined. Nurses working in emergency departments were younger and had less tenure as a nurse and less tenure on the unit than nurses working in critical care or on medical units. There were more male nurses working in critical care and emergency than on medical units. Registered nurses and nurses with Master‘s degrees were more likely to work in critical care areas. Part-time nurses were more likely to work in emergency departments. More nurses working on medical units held supervisory or charge roles than emergency and critical care units. More Registered Practical Nurses worked in community hospitals and on medical units.

Table 6 provides a summary of responses to items within the adherence scale. Results showed that 44% of survey respondents met the case definition of adherence and answered always or mostly to at least 7 of the 8 items within the adherence scale. Twenty-five percent answered always or mostly to all 8 items. The items with the lowest reported adherence included ―I wear eye protection (face shield, goggles, fitted eye protection) when within 1 metre of a coughing patient that I suspect may have a droplet spread disease‖ (51%) and ―I conduct a seal check each time I put on an N95 respirator‖ (53%).

Summary of Responses to Items in the Adherence Scale (N=1074) Item Always/Mostly n (%) I wear an N95 respirator when caring for a patient that I suspect 873 (82) may have an airborne spread disease. I wear an N95 respirator when caring for a patient who is in airborne precautions even if I don't believe that they have an 896 (84) airborne spread disease. I wear an N95 respirator when caring for a patient that has been 1022 (96) diagnosed with an airborne spread disease. I conduct a seal check each time I put on an N95 respirator 567 (53) I wear a surgical mask or an N95 respirator when within 1 metre of 742 (70) a patient that I suspect may have a droplet spread disease. I wear a surgical mask when within 1 metre of a patient that has a 909 (86) diagnosed droplet spread disease. I wear eye protection (face shield, goggles, fitted eye protection) when within 1 metre of a coughing patient that I suspect may have 544 (51) a droplet spread disease. I wear eye protection when within 1 metre of a coughing patient 662 (62) that has a diagnosed droplet spread disease.

Table 6: Summary of Responses to Items in the Adherence Scale

Table 7 presents a summary of responses to items within the knowledge of recommended use of facial protection scale. A choice of Don’t Know or an item left blank was coded as incorrect. Results showed that 46% of the survey respondents met the case definition of knowledgeable and answered at least 7 of the 8 items within the scale correctly. Fifteen percent of respondents answered all the items correctly.

Summary of Responses to Knowledge of Recommended Use of Facial Protection Scale (N=1074) Correct Item n (%) Fit testing of N95 respirators is done to ensure the make, model and size 1042 (97) of the respirator seals to my face. T Wearing a disposable N95 respirator will help protect me from catching 982 (91) an airborne communicable disease. T An N95 respirator filters out a minimum of 95% of particles or droplets 815 (76) that may be communicable. T There is no recommended way to remove an N95 respirator. F 740 (69) Wearing a surgical mask will help protect me from catching a droplet 517 (48) spread communicable disease. T It is good practice to reuse a surgical mask several times as protection 857 (80) from a communicable disease. F Acceptable eye protection includes regular reading glasses. F 798 (74) Reusable eye shields or goggles used as protection from a 827 (77) communicable disease should be disinfected after each use. T T – correct answer is True, F – correct answer is False

Table 7: Summary of Responses to Knowledge of Use of Facial Protection Scale

Table 8 presents a summary of responses to items within the knowledge of transmission of influenza scale. Again, a choice of Don’t Know or an item left blank was coded as incorrect. Results showed that 56% of the survey respondents met the case definition of knowledgeable and answered at least 5 of the 6 items within the scale correctly. Twenty-seven percent answered all the items correctly.

Summary of Responses to Knowledge of Transmission of Influenza Scale (N=1074) Correct Item n (%) Influenza is thought to be primarily spread through large droplets (droplet 891 (83) transmission) that directly contact the nose, mouth or eyes. T These droplets are produced when infected people cough, sneeze or talk, sending the relatively large infectious droplets and very small sprays 1020 (95) (aerosols) into the nearby air and into contact with other people. T Large droplets can only travel a limited range; therefore, people should 707 (66) limit close contact (within 6 feet) with others when possible. T To a lesser degree, human influenza is spread by touching objects contaminated with influenza viruses and then transferring the infected 903 (84) material from the hands to the nose, mouth or eyes. T Influenza may also be spread by very small infectious particles (aerosols) 822 (77) traveling in the air. T The contribution of each route of exposure to influenza transmission is uncertain at this time and may vary based upon the characteristics of the 529 (49) influenza strain. T T – correct answer is True, F – correct answer is False

Table 8: Summary of Responses to Knowledge of Transmission of Influenza Scale

Information on exposure and experience with occupational transmission of illness and adverse health effects are presented in Table 9. Just under a third of respondents indicated they had contracted a communicable respiratory illness through work. The survey allowed for respondents to provide further information about the nature of these illnesses. Forty-one respondents provided comments and identified influenza, general upper respiratory tract infection, SARS, Norwalk virus, common cold and seroconversion to TB positive as communicable respiratory illnesses acquired through their work. Eight respondents indicated severe adverse mental health effects from working during the SARS outbreak.

Summary of Responses to Exposure Scale (N=1074) Strongly Agree/Agree Item Survey Identifier n (%) I have been exposed to a communicable 27 875 (82) respiratory illness at work I have contracted a communicable respiratory 28 313 (30) illness through work. I have suffered adverse physical health effects as a result of exposure to a communicable 29 175 (16) respiratory illness at work. I have suffered adverse mental health effects as a result of exposure to a communicable 30 94 (9) respiratory illness at work. Contracted an illness, suffered adverse physical health effects OR suffered adverse mental 28, 29 or 30 369 (35) health effects

Table 9: Summary of Responses to Exposure Scale

4.1.2 Exploratory Factor Analysis

Exploratory factor analysis was done on the 17 items grouped under the heading of Organizational Support to determine an appropriate factor solution. According to accepted guidelines regarding factor analysis and sample size (Comrey & Lee, 1992), 1074 completed surveys was adequate to yield stable results. Initially, a four-factor model, based on findings from the literature (Gershon, et al., 2000) and similar to methods followed in the pilot study, was applied. Based on evidence of cross-loading within the four-factor solution, a subsequent three- factor model was applied. This model was supported by the Kaiser-Guttman rule as only three factors had eigenvalues greater than 1.0 (Figure 4). The first eigenvalue was 8.06 and accounted

63 for 47.42% of the variance, the second was 1.38 and accounted for 8.14% of the variance and the third was 1.24 and accounted for 7.31% of the variance.

Figure 4: Factor Analysis Scree Plot of Eigenvalues

All items, except for one, demonstrated primary factor loadings of ≥ 0.4. One item (On my unit, employees are encouraged to identify unsafe work practices amongst themselves) demonstrated a primary loading of 0.39 and a cross-loading of more than half the primary loading. The item was not excluded as the primary loading was extremely close to 0.4, the factor had more than three strong loadings and the item was part of a well-tested safety climate scale (Gershon, et al., 2000). Eigenvalues represent the variance explained by each factor. Total eigenvalue represents cumulative variance explained and when divided by the number of items in the scale should be greater than 60% to represent a good model. In this analysis, the three- factor model explained 63% of the variance. Final communality estimates are provided for each

item. The communality measures the percent of variance in a given variable explained by all the factors jointly and may be interpreted as the reliability of the indicator. Communality estimates ranged from 0.46 to 0.78 indicating the factor model worked fairly well for all indicators. Table 10 describes the results of the final three-factor model.

Factor Loadings of Items in Organizational Support Scale (N=1074) Original Final Construct from Item New Construct Factor 1 Factor 2 Factor 3 Communality Nichol et. al. Estimate (2008) The protection of workers from occupational exposure to communicable Organizational Organizational -0.11 0.89 -0.06 0.65 respiratory disease is a high priority where support support I work. My workplace has a functioning joint Organizational New item -0.1 0.83 -0.01 0.59 health and safety committee. support On my unit, all reasonable steps are taken Organizational Organizational to minimize hazardous job tasks and 0.14 0.69 0.09 0.69 support support procedures. On my unit, employees are encouraged to Organizational Organizational become involved in health and safety 0.26 0.65 -0.04 0.65 support support matters. Managers on my unit do their part to ensure employees protection from Organizational Organizational 0.37 0.59 -0.08 0.68 occupational exposure to communicable support support respiratory disease. In my current work area, written policies Organizational for eye and respiratory protection are New item 0.07 0.52 0.22 0.49 support readily accessible. My job duties do not interfere with my Absence of job Absence of job -0.09 0.09 0.8 0.66 being able to use facial protection. hindrances hindrances I have enough time in my work to use Absence of job Absence of job 0.03 0.05 0.8 0.72 facial protection properly. hindrances hindrances I can provide good quality care to my Absence of job Absence of job 0.07 -0.14 0.84 0.67 patients while wearing facial protection. hindrances hindrances On my unit, employees are encouraged to identify unsafe work practices amongst Feedback Communication 0.39 0.25 0.23 0.52 themselves. On my unit, unsafe work practices that aren't resolved are corrected by Feedback Communication 0.58 0.12 0.11 0.53 supervisors. In my current work area, my coworkers support me in following safe work Feedback Communication 0.46 0.09 0.25 0.46 practices. In my current work area, supervisors Feedback Communication 0.6 0.21 0.02 0.58 regularly discuss safety at staff meetings. There are good employee relations on my Conflict/ Communication 0.89 -0.11 -0.02 0.67 unit. communication Management and staff on my unit support Conflict/ Communication 0.96 -0.09 -0.06 0.78 one another. communication On my unit, there is open communication Conflict/ Communication 0.87 -0.02 -0.02 0.71 between supervisors and staff. communication My supervisor enforces compliance with legislation and workplace policy regarding New item Communication 0.66 0.21 0.02 0.66 proper use of facial protections. Eigenvalue 8.06 1.38 1.24

Table 10: Factor Loadings of Items in Organizational Support Scale

Additional exploratory factor analysis was done on the survey scales used to measure individual and environmental factors. Results showed primary factor loadings > 0.4 under the original groupings and no evidence of cross loading indicating scales were independent.

4.1.3 Reliability Testing

Table 11 describes the reliability of scales used to measure constructs within the questionnaire using Cronbach‘s coefficient alpha (α) raw scores. Twelve of fifteen scales demonstrated acceptable reliability coefficients. Similar to pilot study results, both knowledge scales and the effectiveness of preventive actions scale demonstrated substandard reliability scores (< 0.70).

Reliability of Scales used to Measure Constructs within the Survey Variable Survey Scale α (raw) Category Identifier Knowledge - Using facial protection 10-17 0.42 Knowledge - Transmission of influenza 18-23 0.53 Effectiveness of preventive actions 24-26 0.35 Individual Health effects – personal experience (self) 28-30 0.79 Perception of risk 32-34 0.85 Personal barriers to use 35a-37c 0.92 Outcome Adherence 38-45 0.75 Cleanliness/orderliness of workplace 46-48 0.89 Environmental Availability of facial protection 49-51 0.7 Media Coverage 56-57 0.89 Training 52a-52f 0.92 Organizational support 58-63 0.87 Organizational Absence of job hindrances 64-66 0.78 Communication 67-74 0.9 Patient Safety Patient safety climate 75-84 0.87

Table 11: Reliability of Scales Used to Measure Constructs within the Survey

4.1.4 Bivariate Analysis

Table 12 describes the bivariate analysis of association between demographic variables and adherence. Due to low numbers in the ‗certificate‘ and ‗Master‘s degree‘ groups, the education variable was reduced to those respondents with a university degree (Bachelor‘s and Master‘s) and those without a degree (certificate and diploma). Four demographic variables

showed a significant relationship (p ≤ 0.05) with adherence: age, tenure as a nurse, hospital type and unit type. Nurses working in emergency reported the lowest rates of adherence to recommended use of facial protection (19%).

Bivariate Analysis of Association between Demographic Factors and Adherence Chi-square Significance Test used unless otherwise indicated. Variable Non-adherent Adherent Total Variable Level P-value Category n (%) n (%) n (%) Gender Female 530 (91) 413 (92) 943 (91) 0.65 Age, y* Mean (SD) 37.9 (9.9) 39.9 (10.3) 38.8 (10.1) 0.0037 Nurse type RN 555 (95) 427 (95) 982 (95) 0.92 Degree 301 (59) 209 (41) 510 (50) Education 0.09 No Degree 277 (54) 239 (46) 516 (50) Tenure as a nurse, y* Mean (SD) 13.6 (10.4) 15.5 (10.6) 14.4 (10.4) 0.003 Tenure on the unit, y* Mean (SD) 6.3 (6.5) 6.8 (6.8) 6.5 (6.6) 0.21 Work status Full-time 440 (76) 345 (77) 785 (77) 0.73 Supervisory status Yes 259 (25) 221 (22) 480 (47) 0.14 A 74 (13) 37 (8) 111 (11) Demographic B 110 (19) 79 (17) 189 (18) C 108 (18) 77 (17) 185 (18) Hospital 0.11 D 81 (14) 72 (16) 153 (15) E 43 (7) 31 (7) 74 (7) F 168 (29) 156 (35) 324 (31) Community 335 (57) 224 (50) 559 (54) Type of hospital 0.012 Teaching 249 (43) 228 (50) 477 (46) Critical Care 212 (37) 170 (38) 382 (38) Type of unit Emergency 210 (37) 87 (19) 297 (29) <0.0001 Medicine 148 (26) 191 (43) 339 (33) * Wilcoxon Rank Sum Test for non-normally distributed continuous variables used.

Table 12: Bivariate Analysis of Association between Demographic Factors and Adherence

Table 13 describes the bivariate analysis of association between individual factors and adherence. Six individual variables showed a significant relationship (p ≤ 0.05) with adherence. Four showed a positive relationship: frequent use of facial protection, knowledge of proper use of facial protection, knowledge of transmission of influenza, and the perception that preventive actions (use of N95 respirators, surgical masks and eye protection) were effective; and two showed a negative relationship: a history of contracting an occupational illness or suffering adverse health effects and reporting personal barriers to using facial protection. N95 respirators were reported to have the most barriers to use with physical discomfort (59%), difficulty breathing (54%) and communication breakdown (43%) as the most common barriers reported. Forty-two percent of respondents indicated difficulty seeing as the most common barrier to using eye protection. Physical discomfort (41%) and difficulty breathing (40%) were identified as the most common barriers to using surgical masks.

Bivariate Analysis of Association between Individual Factors and Adherence Chi-square Significance Test used.

Variable Non-adherent Adherent Total Variable Level P-value Category n (%) n (%) n (%)

Frequency of use of Daily/weekly/monthly 489 (85) 402 (92) 891 (88) 0.002 facial protection Rarely/never 83 (15) 36 (8) 119 (12) Knowledge - Using Knowledgeable 247 (42) 226 (50) 473 (46) 0.014 facial protection Knowledge - Transmission of Knowledgeable 306 (52) 273 (60) 579 (56) 0.01 influenza Effectiveness of Perceived preventive 153 (26) 145 (33) 298 (29) 0.03 preventive actions actions to be effective Reported contracting Health effects - occupational illness or personal experience 218 (38) 138 (31) 356 (35) 0.02 Individual suffering adverse (self) health effects Reported exposure Health effects - and adverse health personal experience outcome to family, 280 (48) 200 (44) 480 (46) 0.24 (others) close friend or colleague Perceived Perception of risk 407 (70) 312 (69) 719 (70) 0.74 occupational risk Reported personal Personal barriers to barriers to using facial 412 (82) 273 (71) 685 (77) 0.0002 use protection

Table 13: Bivariate Analysis of Association between Individual Factors and Adherence

Table 14 describes the bivariate analysis of association between environmental factors and adherence. Two environmental variables showed a positive significant relationship (p ≤ 0.05) with adherence: reporting ready availability of facial protective equipment; and, reporting that the media influenced their risk perception and work practices.

Bivariate Analysis of Association between Environmental Factors and Adherence Chi-square Significance Test used.

Variable Non-adherent Adherent Total Variable Level P-value Category n (%) n (%) n (%)

Cleanliness/ Reported unit to be 166 (29) 152 (34) 318 (31) 0.08 orderliness clean/orderly Reported facial Availability of protection to be 271 (47) 292 (65) 563 (55) <0.0001 facial protection Environmental readily available Reported media influenced risk Media coverage 241 (42) 247 (55) 488 (48) <0.0001 perception and work practices

Table 14: Bivariate Analysis of Association between Environmental Factors and Adherence

Table 15 describes the bivariate analysis of association between organizational factors and adherence. Knowledge of N95 was categorized as an organizational variable as several hospitals had this information on employee identification badges. Regarding training and fit testing, 51% of respondents reported being trained and fit tested within the previous two years. All five organizational variables showed a significant relationship (p ≤ 0.05) with adherence.

Bivariate Analysis of Association between Organizational Factors and Adherence Chi-square Significance Test used. Non- Variable Adherent Total Variable Level adherent P-value Category n (%) n (%) n (%) Trained and fit tested in Trained and fit the last 2 years and 211 (42) 235 (62) 446 (51) <0.0001 tested trained on 5/6 items Knowledge of Reported knowing the N95 261 (45) 232 (51) 493 (48) 0.03 N95 they were fit tested for Reported organizational Organizational support for health and 142 (25) 205 (47) 347 (34) <0.0001 support Organizational safety Reported absence of job Absence of job hindrances to working 282 (49) 291 (65) 573 (56) <0.0001 hindrances safely Reported good Communication communication in 100 (18) 159 (36) 259 (26) <0.0001 workplace

Table 15: Bivariate Analysis of Association between Organizational Factors and Adherence

4.1.5 Multivariable Analysis

Eighteen variables with a bivariate p-value <0.10 (tenure as a nurse, education, hospital, unit type, frequency of use of facial protective equipment, knowledge – using facial protection, knowledge – transmission of influenza, effectiveness of preventive actions, personal experience with adverse health effects (self), personal barriers to use, cleanliness/orderliness of the unit, availability of facial protection, media coverage, training and fit testing, knowledge of N95, organizational support, absence of job hindrances and communication) were selected for inclusion in a stepwise logistic regression model. In accordance with the generally accepted principal of 10 events per predictor (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996) and based on securing 308 events of adherence, the inclusion of 18 variables in the model was acceptable. Due to the positive relationship between age and tenure as a nurse (p<0.0001), only tenure as a nurse was selected for inclusion in modeling. As the bivariate p-values for these two variables were almost the same, tenure was selected over age to remain consistent with the pilot

70 study. Due to possible inherent contributory bias, modeling including and not including the variable ‗frequency of use of facial protective equipment‘ was carried out and results were almost identical. The variable ‗hospital‘ was used instead of ‗hospital type‘ due to the concern that the number of community hospitals (four) and teaching hospitals (two) were too low to be representative.

Table 16 describes the results of multivariable modeling. Six predictors of adherence were revealed: unit type; more frequent use of facial protective equipment; ready availability of facial protective equipment; training and fit testing within the last two years and training on five of six key items; organizational support for health and safety; and, good communication in the workplace. To ensure clinical significance of predictor variables, backwards stepwise selection to 10 variables was conducted and revealed the same six predictors of adherence. Key interactions between organizational level variables were examined and showed the effect of training and fit testing depended on an absence of job hindrances (p=0.025). This model yielded a c-statistic of 0.77. This value is considered to be an acceptable measure of goodness-of-fit (Hosmer & Lemeshow, 2000) and indicates the model assigns a higher probability to a correct case 77% of the time.

Adjusted Odds Ratios for Adherence to Recommended Use of Facial Protection Variable Odds Ratio Variable Level Pr > ChiSq Category (95% CI) Tenure as a nurse Mean 0.31 1.01 (0.99, 1.03) Education Degree vs. no degree 0.15 1.34 (0.9, 1.98 B vs. A 0.82 1.09 (0.51, 2.31) C vs. A 0.14 1.77 (0.83, 3.8) Hospital D vs. A 0.31 1.45 (0.69, 3.25) Demographic E vs. A 0.13 2.06 (0.8, 5.25 F vs. A 0.13 1.74 (0.86, 3.51) Emergency vs. Critical <0.0001 0.41 (0.26, 0.64) Unit type Care Medical vs. Critical Care 0.32 1.25 (0.8, 1.96) Daily/weekly/monthly vs. Frequency of use 0.0029 2.45 (1.36, 4.42) rarely/never Knowledge - Using facial Knowledgeable vs. not 0.16 1.3 (0.9, 1.87) protection Knowledge - Knowledgeable vs. not 0.18 1.28 (0.89, 1.83) Transmission of Influenza Perceived preventive Effectiveness of actions to be effective vs. 0.08 1.42 (0.96, 2.09) preventive actions Individual not Contracted occupational illness or suffered Health effects (self) 0.0504 0.68 (0.46, 1.001) adverse health effects vs. not Reported personal Personal barriers to using barriers to using facial 0.27 0.78 (0.5, 1.21) facial protection protection vs. no personal barriers Cleanliness/Orderliness Perceived unit 0.88 0.97 (0.64, 1.46) of unit clean/orderly vs. no Perceived facial Availability of facial Environmental protection to be available 0.03 1.53 (1.05, 2.22) protection vs. no Reported media Media coverage 0.88 1.03 (0.71, 1.5) influenced work vs. not Trained and fit tested vs. Training and fit testing 0.009 1.66 (1.32, 2.43) not Knew N95 fit tested for Knowledge of N95 0.14 1.31 (0.91, 1.88) vs. didn’t know Organizational support Perceived organizational 0.003 1.98 (1.27, 3.09) Organizational for health and safety support vs. not Absence of job Perceived absence of job 0.64 0.91 (0.61, 1.34) hindrances hindrances vs. not Perceived positive Communication communication in 0.002 2.13 (1.31, 3.44) workplace vs. not

Table 16: Adjusted Odds Ratios for Adherence to Recommended Use of Facial Protection

4.2 Phase 1 - Patient Safety Climate

One of the secondary objectives of this study was to examine the relationship between nurses‘ perceptions of patient safety climate and worker safety climate within their organizations. Survey instructions directed nurses to focus specifically on patient safety when answering items in this scale. Table 17 provides a summary of responses to items within the patient safety climate scale. The scale was analyzed as a categorical variable in concordance with analysis used in this study and in the literature (Pronovost, et al., 2003). Results showed that 17% of survey respondents identified a positive patient safety climate in their organization by answering strongly agree or agree to all 10 items within the scale.

Summary of Responses to Patient Safety Climate Scale (N=1074) Strongly Agree/ Item Agree, n (%) The senior leaders in my hospital listen to me and care about my 442 (42) concerns. The physician and clinical leaders in my area listen to me and care 602 (56) about my concerns. Leadership is driving us to be a safety centered institution. 576 (54) My suggestions about safety would be acted upon if I expressed them 545 (51) to management. Management/leadership does not knowingly compromise safety 590 (56) concerns for productivity. I am encouraged by my colleagues to report any safety concerns I 811 (76) may have. I know the proper channels to direct questions regarding patient 830 (78) safety. I am satisfied with the availability of clinical leadership. 666 (62) I believe that most adverse events occur as a result of multiple system 786 (74) failures and are not attributable to one individual's actions. I am aware that patient safety has become a major area for 836 (79) improvement in this institution.

Table 17: Summary of Responses to Patient Safety Climate Scale

Nurses‘ perceptions of patient safety climate were compared with three measures of worker safety climate (organizational support, absence of job hindrances and communication) to determine the presence of a relationship between the two constructs. Results are described in Table 18 and provide evidence of a positive relationship.

Bivariate analysis of Association between Worker Safety Climate Measures and Patient Safety Climate Chi-square Significance Test used. Lower rating Strong on patient patient Variable Total Variable Level safety safety P-value Category n (%) climate climate scale, n (%) n (%) Reported Organizational organizational 213 ( 26) 133 ( 76) 346 (34) <0.0001 support support for health and safety Reported absence Organizational Absence of job of job hindrances to 425 (50) 148 (82) 573 (56) <0.0001 hindrances working safely Reported good Communication communication in 135 (16) 126 (71) 261 (26) <0.0001 workplace

Table 18: Bivariate Analysis of Association between Worker Safety Climate Measures and Patient Safety Climate

A significant relationship between perception of a strong patient safety climate and self- reported adherence to recommended use of facial protective equipment was also found (Table 19) and further supports an important relationship between worker safety climate and patient safety climate.

Bivariate Analysis of Association between Patient Safety Climate Scale and Adherence Chi-square Significance Test used. Non-adherent Adherent Total Variable Variable Level P-value Category n (%) n (%) n (%)

Reported a positive Patient Safety Patient Safety patient safety 74 (13) 100 (23) 174 (17) <0.0001 Climate climate

Table 19: Bivariate Analysis of Association between Patient Safety Climate Scale and Adherence

4.3 Phase 1 – Unit Observations

Observations regarding the cleanliness and orderliness of the unit and the ready availability of facial protection were recorded by the researcher on a unit-by-unit basis during Phase 1 of the study (Table 20). The majority of units were observed to be clean (87% strongly agreed/agreed) and just over half were observed to be orderly (53% strongly agreed/agreed). Eye protection was identified as the least readily available piece of facial protection equipment (38%).

Descriptive Findings for Unit Observations (N=45) n (%) Variable Strongly Strongly Agree Inconsistent Disagree Agree Disagree Clean 25(56) 14(31) 3(7) 2(4) 1(2) Orderly 15(33) 9(20) 6(13) 11(24) 4(9) N95 Respirators 20(44) 7(16) 11(24) 3(7) 4(9) Available Surgical Masks 28(62) 2(4) 7(16) 5(11) 3(7) Available Eye Protection 14(31) 3(7) 5(11) 5(11) 18(40) Available

Table 20: Descriptive Findings for Unit Observations

Unit observations were also examined based on unit type (Table 21) and eye protection was more often readily available in critical care areas than emergency departments or medical units (p=0.0006). Presence of the variable was indicated if the observer recorded strongly agree/agree.

Unit Observations Presented by Unit Type (N=44) Unit Type n (%) Variable Critical Care Emergency Medicine Total Clean 16 (89) 9 (100) 13 (76) 38 (86) Orderly 10 (56) 7 (78) 6 (35) 23 (52) N95 Respirators Available 14 (78) 5 (56) 7 (41) 26 (59) Surgical Masks Available 14 (78) 4 (44) 11 (65) 29 (66) Eye Protection Available 13 (72) 2 (22) 2 (12) 17 (39)* *statistically significant difference – Chi-square significance test used.

Table 21: Unit Observations Presented by Unit Type

As the literature review revealed self-reported behaviour does not always predict actual behaviour, a comparison of survey and observational data regarding cleanliness and orderliness of the work environment and ready availability of facial protective equipment was carried out. To include reports from all responding nurses, the definition used to determine presence of the variable was ≥ 75% agreement between responding nurses. Agreement <75% indicated absence of the variable. This definition was the result of thesis committee discussions. Results of the comparison are presented in Table 22 and generally demonstrate slight agreement between measures (Landis & Koch, 1977). Agreement between measures of availability of eye protection was fair. Both survey and observational data showed eye protection to be the least readily available piece of facial protective equipment, particularly in emergency and on medical units.

Comparison of Self-Report Data and Unit Observations (N=45) Agreement rating Kappa # pairs in according to Variable Coefficient of agreement (%) Landis & Koch, Agreement 1977 Clean 21 (47) 0.1 Slight Orderly 21 (47) -0.005 Poor N95 respirators 24 (53) 0.02 Slight available Surgical masks 30 (67) 0.04 Slight available Eye protection 30 (67) 0.3 Fair available

Table 22: Comparison of Self-Report Data and Unit Observations

4.4 Phase 2 – Participant Observations

4.4.1 Descriptive Analysis

A total of 444 information sheets and consent forms were provided to critical care nurses and 140 consented to participate in Phase 2 of the study (32%). Of those critical care nurses who completed a survey (336), forty-two percent consented to participate in the observational study as well. Of the 140 consenting nurses, 112 observations were completed (80%). Table 23 describes the reasons for attrition. The main reason for an observation not being carried out was consenting nurses were absent from work during all observer visits. These absences were despite visits being based on each nurses work schedule. Illness and last minute schedule switches were the main reasons for absence from a scheduled shift. A second reason was the 1-3 month interval

76 between Phase 1 (survey) and Phase 2 (observations). During this time period ten nurses went on extended leave (three on maternity leave, three on extended sick leave and four on extended leave of absence) and five were no longer working on the unit (three transferred to another unit and two left their nursing job at the hospital).

Reasons for Observation Not Completed (N=28)

Reason n (%) Absent during all observer visits 11 (39) On extended leave of absence 10 (36) No longer works on unit 5 (18) Opted out of observational study 2 (7)

Table 23: Reasons for Observation Not Completed

One hundred and twelve observations exceeded the projected enrolment figure of 100 and had sufficient power to detect an adherence rate similar to the rate found in the pilot study +/- 10%. At the time of the study, 859 full and part time nurses were employed on the 14 critical care units resulting in a phase 2 study participation rate of 13%. See Table 24 for the demographic characteristics of the observational study population. There were no statistically significant differences in the demographic characteristics of the critical care nurses that were observed (N=112) and the critical care nurses that were not observed (including those that consented to be observed but weren‘t) (N=281). In addition, there were no statistically significant differences in the demographic characteristics between the survey population (N=1074) and the observational study population (N=112) except those that would be expected (hospital type, unit type and nurse type).

Demographic Characteristics of the Observational Study Population and Comparison of Critical Care Nurses who were Observed and Not Observed Chi-square Significance Test used unless otherwise indicated. Observed Not Observed Variable Level N = 112 N = 281 P-value n (%) n (%) Gender Female 99 (88) 254 (90) 0.55 Nurse Type RN 112 (100) 281 (100) n/a Certificate 4 (3.5) 4 (1) Diploma 54 (48) 130 (47) Education 0.45 Degree 51 (46) 132 (47) Masters 3 (2.5) 13 (5) Work Status Full-time 87 (78) 215 (77) 0.82 Supervisor Status Yes 24 (22) 124 (44) 0.12 Age, y* 39.7 (21-61) 40.7 (20-63) 0.44

Tenure as a nurse, y* 15.3 (0-39) 16 (0-40) 0.78

Tenure on the unit, y* 7.4 (0-30) 7.3 (0-30) 0.96

Community 49 (44) 150 (53) Hospital Type 0.08 Teaching 63 (56) 131 (47) Critical Unit Type 112 (100) 281 (100) n/a Care * sample mean (range) * Wilcoxon Rank Sum Test for non-normally distributed continuous variables used.

Table 24: Demographic Characteristics of the Observational Study Population and Comparison of Critical Care Nurses who were Observed and Not Observed

Of the 112 observations carried out in the intensive care setting, ten were true observations and 102 were demonstrations. The low number of true observations was due to participating nurses not being assigned to care for patients on respiratory precautions after two attempts by the observer. As the majority of observations were demonstrations, we were not able to measure adherence but were able to measure competence. As discussed in Section 2.7.4, the two constructs are different. Competence can be defined as the demonstrated ability to apply knowledge and skills whereas adherence refers to the degree to which an individual follows a set of guidelines in a real-life setting. In three cases of the ten true observations, nurses donned facial protective equipment within the observer‘s field of view but then went into a patient room with closed curtains/blinds so the observer was not able to see the nurse using, doffing or disposing of facial protective equipment. All observations and demonstrations were of nurses using facial protection for patients requiring either droplet (Level 2) precautions or airborne (Level 3) precautions. Figure 5 describes the breakdown of observations by type and level of precaution.

Airborne (98) Demonstration (102) Droplet (4) Direct Observational Study (N=112) Airborne (2) True Observation (10) Droplet (8)

Figure 5: Breakdown of Observations by Type and Level of Precaution

Table 25 provides the descriptive findings for the direct observational study. Participants were deemed to be competent if they carried out at least four of five critical steps when using facial protective equipment for a patient on droplet respiratory precautions or carried out at least five of six critical steps when using facial protective equipment for a patient on airborne respiratory precautions. All nurses demonstrated competence when using a surgical mask for patients on droplet precautions and 44% (43/98) demonstrated competence when using a disposable N95 respirator for patients on airborne precautions. Due to the low number of complete Level 2 (droplet precaution) observations (11), analysis and discussion of results was focused on competence with proper use of N95 respirators. The most common gaps in competence with recommended use of an N95 respirator included failure to verify the seal and touching the potentially contaminated face piece. Interestingly, even though the use of eye protection is based on the risk of splash/spray to the eyes/face, 74% of nurses demonstrating use of facial protection for patients on airborne precautions donned eye protection. The majority (80%) of nurses indicated the respirator they used during the demonstration was the one they were fit tested for. For the 19 nurses that used a respirator they weren‘t fit tested for, the most common reason stated was their respirator wasn‘t available on the unit.

Descriptive Findings for Observational Study (N=112) Variable Level n (%) Observational Study (N=112) True Observation 10 (9) Type of Observation Demonstration 102 (91) Droplet - Level 2 12 (11) Level of Precaution Airborne - Level 3 100 (89) Droplet - Level 2 Observations (N=12) True Observation 8 (67) Type of Observation Demonstration 4 (33) 1. Chose surgical mask 12(100) 2. Straps positioned correctly 12 (100) Five critical steps 3. Adjust nose piece 11 (92) 4. Refrain from touching face piece 10 (91)* 5. Direct disposal 11 (100)* Carried out 5/5 steps 9 (82)* Level of Competence Carried out at least 4/5 steps 11 (100)* Eye Protection Wore eye protection 4 (33) Airborne - Level 3 Observations (N=100) True Observation 2 (2) Type of Observation Demonstration 98 (98) 1. Chose N95 respirator 100 (100) 2. Straps positioned correctly 76 (76) 3. Adjust nose piece 92 (92) Six Critical Steps 4. Seal check 24 (24) 5. Refrain from touching face piece 39 (40)** 6. Direct disposal 98 (100)** Carried out 6/6 steps 11 (11)** Carried out at least 5/6 steps 43 (44)** Level of Competence Carried out at least 4/6 steps 79 (81)** Carried out at least 3/6 steps 93 (95)** Carried out at least 2/6 steps 98 (100)** Eye Protection Wore eye protection 74 (74) Fit Tested for this respirator Yes 79 (80) *n=11 because one nurse went behind curtains during use and doffing/disposal **n=98 because two nurses went behind curtains during use and doffing/disposal

Table 25: Descriptive Findings for Observational Study

4.4.2 Bivariate Analysis

Table 26 describes the bivariate analysis of association between explanatory variables of interest and competence. Variables of interest were identified from the competence literature as well as through thesis committee discussions.

Bivariate Analysis of Association between Explanatory Variables and Competence with Recommended Use of a Disposable N95 Respirator (N=98) Chi-square Significance test used unless otherwise indicated. Not Variable Competent Total p- Variable Level competent Category n (%) n (%) value n (%) Gender Female 49 (89) 37 (86) 88 (88) 0.65 Age, y* Mean (SD) 38.6 (10.4) 40.5 (8) 39.4 (9.4) 0.27 Education Degree vs. no degree 32(58) 19 (44) 51 (52) 0.17 Tenure as a Mean (SD) 14 (9.6) 16.4 (9) 15 (9.4) 0.22 Demographic Nurse, y* Tenure on the Mean (SD) 7 (7.2) 8.3 (7) 7.6 (7) 0.26 Unit, y* Work Status Full time 43 (80) 31 (72) 74 (76) 0.39 Supervisor Supervised in last 12 28 (52) 22 (51) 50 (52) 0.95 Status months Frequency of Daily/weekly/monthly 44 (85) 40 (93) 84 (88) 0.2 Use of FP vs. rarely/never Individual Knowledge - Using Facial Knowledgeable 24 (45) 27 (63) 51 (52) 0.06 Protection Cleanliness/ Orderliness of Clean and orderly 27 (49) 14 (33) 41 (42) 0.1 Environmental Unit Availability of Facial protection 36 (65) 23 (55) 59 (61) 0.29 Facial Protection available Training and Fit Trained and fit tested 17 (36) 12 (32) 29 (34) 0.66 Testing Fit Testing Fit tested in last 2 yrs 43 (78) 34 (81) 77 (79) 0.74 Organizational Perceived 16 (30) 9 (22) 25 (26) 0.4 Organizational Support for H&S organizational support Absence of Job Perceived absence of 36 (65) 20 (48) 56 (58) 0.08 Hindrances job hindrances Perceived good Communication 9 (17) 6 (15) 15 (16) 0.79 communication * Wilcoxon Rank Sum Test for non-normally distributed continuous variables used.

Table 26: Bivariate Analysis of Association between Explanatory Variables and Competence

4.4.3 Multivariable Analysis

Five variables with a bivariate p-value ≤0.20 (education, frequency of use of facial protective equipment, knowledge – using facial protection, cleanliness/orderliness of the unit, and absence of job hindrances) and one variable of theoretical importance (fit tested within the last two years) were selected for inclusion in initial stepwise logistic regression modeling. The fit testing variable was included as training on how to competently use an N95 respirator is often part of the fit testing procedure. Based on the study output of 41 competent events, the generally accepted principal of maintaining only 10 events per predictor (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996) suggested the inclusion of only 4 variables. Therefore, the two variables demonstrating the highest p-values (education and cleanliness/orderliness) in initial modeling were dropped. Final multivariable logistic regression analysis revealed one predictor of competence: knowledge of recommended use of facial protection (Table 27). A wide confidence interval indicates further testing involving a larger sample size would be beneficial. This model yielded a c-statistic of 0.7 which is considered to be an acceptable measure of goodness-of-fit (Hosmer & Lemeshow, 2000).

Adjusted Odds Ratios for Competence with Recommended Use of an N95 Respirator

Variable Odds Ratio Variable Level Pr>ChiSq Category (95% CI) Daily/weekly/monthly Frequency of Use 0.12 0.25 (0.05, 1.4) vs. rarely/never Individual Knowledge - Using Facial Knowledgeable 0.02 2.9 (1.2, 7.2) Protection Environmental Fit Tested in last 2 years Fit Tested 0.33 0.58 (0.19, 1.76) Perceived absence Organizational Absence of Job Hindrances 0.08 2.3 (0.92, 5.6) of job hindrances

Table 27: Adjusted Odds Ratios for Competence with Recommended Use of an N95 Respirator

4.5 Relationship between Self-Reported Adherence and Observed Competence

As a secondary objective of the observational study, a comparison of self-report and observational measures of adherence was planned. As the majority of observations were demonstrations, only competence was evaluated. Therefore a comparison of self-reported

82 adherence to recommended use of facial protective equipment and observed competence with recommended use of an N95 respirator was conducted (Table 28). Results revealed no significant relationship (p=0.23) and poor agreement between measures (K=0.12). This result was expected as competence assessments did not take the work setting or context in which the worker functions into consideration.

Relationship between Self-Reported Adherence and Observed Competence (N=98)

Self-Reported Adherence Outcome Variable Total Not adherent Adherent

Not competent 34 (35%) 20 (21%) 54 (56%) Observed Competence Competent 22 (23%) 21 (22%) 43 (44%)

Total 56 (58%) 41 (42%) 97 (100%)

Table 28: Relationship between Self-Reported Adherence and Observed Competence

5 CHAPTER 5: Discussion

The overall purpose of this study was to describe nurses‘ adherence to recommended use of facial protective equipment and to determine the factors that are associated with adherence. Moore‘s theoretical framework that divided factors associated with self-protective behaviour at work into three categories; organizational, environmental and individual factors, was employed. Based on research examining healthcare workers‘ adherence to practices to prevent the transmission of blood borne disease, it was hypothesized that organizational level factors and other safety climate dimensions would be most predictive of nurses‘ adherence to recommended use of facial protective equipment. Study findings provided support for this hypothesis. Secondary objectives aimed to examine the relationship between perceptions of worker safety climate and patient safety climate and to test the feasibility of study methodology in the acute healthcare setting. A new organizational model of adherence is introduced and findings are discussed as they relate to the literature and how they may influence future policy, practice and research. Strengths and limitations of the study are presented.

5.1 Demographics

According to the 2007 Canadian Nurses Association RN Workforce Profiles by Area of Responsibility, this cross-sectional survey sample of 1074 nurses was representative of the population of nurses working in medicine/surgery, critical care and emergency in Canadian acute care hospitals (Canadian Nurses Association, 2009) regarding age and gender. Differences were noted regarding work status and education. Approximately 57% of the RN Workforce Profile population worked full-time and 77% of the study sample reported full-time work status. Differences in work status may have been due to the introduction of the Ontario Ministry of Health and Long-Term Care‘s health human resource strategy – HealthForceOntario – in May, 2006. This strategy was implemented to attract and retain healthcare professionals to Ontario with a focus on recruiting full-time nurses. Regarding education, 33% of the nursing workforce in the 2007 profile was educated at or beyond the baccalaureate level, and half of the study sample was educated at this level. Again, differences in education between the two groups were expected as diploma nursing programs have been phased out in Ontario. In 2001, the provincial government announced that effective 2005 all new RNs must have a degree. Since then, students have been taking either a collaborative college-university nursing program or a four-year

84 university nursing program - both leading to a Bachelor of Science in Nursing degree (BScN) or a Bachelor of Nursing degree (BN).

Comparisons were also made between the observational study of 112 nurses and the survey population. No statistically significant differences in demographic characteristics were observed except those which would be expected; hospital type (teaching), unit type (critical care) and nurse type (RN). In addition, there were no statistically significant differences between the critical care nurses who were observed (112) and critical care nurses who were not observed (281), including those who consented to be observed but weren‘t.

5.2 Adherence to Recommended Use of Facial Protection

Nurses‘ adherence to recommended use of facial protective equipment was found to be substandard with only 44% of survey respondents reporting adherence. This is consistent with reports in the literature regarding hand hygiene (Pittet, 2001) (McGuckin, Waterman, & Govednik, 2009), medical gloves and universal precautions (DeJoy, Searcy, Murphy, & Gershon, 2000) (Godin, Naccache, Morel, & Ebacher, 2000) (Evanoff, et al., 1999) (Waqqas, Huor, Brassard, & Loo, 2002) and immunization (Centers for Disease Control and Prevention, 2005) (Steiner, Vermeulen, Mullahy, & Hayney, 2002) (Lester, McGeer, Tomlinson, & Detsky, 2003) (Wicker, Rabenau, Doerr, & Allwinn, 2008). Studies examining adherence to recommended respirator use for suspected or confirmed tuberculosis also reported inferior adherence (Tokars, et al., 2001) (Kellerman, Saiman, San Gabriel, Besser, & Jarvis, 2001) (Sutton, Nicas, Reinisch, & Harrison, 1998). More recently, studies demonstrating similar substandard adherence rates to recommended use of facial protective equipment for patients with other types of communicable respiratory illness have been published involving healthcare workers caring for patients with SARS (Park, et al., 2004) (Loeb, et al., 2004) (Shigayeva, et al., 2007), avian influenza (Skowronski, et al., 2007) (Morgan, et al., 2009) and influenza A (H1N1) (Centers for Disease Control and Prevention, 2009) (Perez-Padilla, et al., 2009). This historical problem of substandard adherence to safe work practices in the healthcare sector, including proper use of facial protective equipment, highlights the complexity of the problem and the need for a novel approach to improvement.

In this study, adherence was defined as answering always or mostly to seven of eight items identified by federal and provincial government sources as critical to the protection of

85 workers. Initially, this definition may seem fairly stringent but non-adherence with even one item could result in transmission of disease. Therefore, the self-reported adherence rate of 44% may actually be an overestimate, signaling an even bigger concern than reported. This is particularly important as this study was conducted during regular, non-outbreak conditions when non- adherence would have had a lower chance of resulting in occupational transmission of disease. Non-adherence during outbreak conditions could have a more severe impact and result in extensive disease transmission to workers, patients and the public. This highlights the importance of implementing strategies to enhance adherence now in order to mitigate the potential of catastrophic disease transmission during future outbreaks.

Of the eight items measured to assess adherence, two demonstrated particularly low rates. The first was wearing eye protection when within 1 metre of a coughing patient suspected to have a communicable respiratory illness (51% adherence). Historically, studies examining ocular transmission of communicable disease have been focused on the use of eye protection in the context of dental infection control practice, during operative procedures or for protection against traumatic or chemical injuries (Yassi, et al., 2004) and have not addressed ocular exposure to bio-aerosols. As a result, evidence supporting the use of eye protection has been limited. Recent animal studies (Olofsson, Kumlin, Dimock, & Arnberg, 2005) (Belser, Wadford, Xu, Katz, & Tumpey, 2009) have shown the eye to be an important portal of entry for avian influenza viruses. This new evidence provides support for the use of eye protection against communicable respiratory pathogens and the inclusion of eye protection in the definition of adherence to recommended facial protection. Further research in this area is needed.

The second item demonstrating a particularly low self-reported adherence rate was conducting a seal check when donning an N95 respirator (53%). The observational study also identified the user seal check as an important gap in nurses‘ competence with recommended use of an N95 respirator. Data from the Centers for Disease Control and Prevention indicated that for any given N95 respirator, the percentage of subjects who achieve an adequate fit ranges from 0 to 88% (Centers for Disease Control and Prevention, 1998). As a result, authorities recommend that fit testing be carried out on each individual to find a brand of respirator that achieves an adequate fit and that the user performs a seal check after donning the respirator to test for gross leaks. Recent studies attempting to demonstrate the value of conducting a seal check to validate the fit of a respirator have raised questions about this practice (Derrick, Chana, Gomersalla, &

Luib, 2005) (Delaney, McKay, & Freeman, 2003). Regardless, current guidelines stand and a user seal check was included in the definition of adherence to recommended use of facial protection used in this study. Poor adherence in the areas of eye protection and user seal check indicated that adherence may be lower for items for which there is the least evidence. This suggests that better evidence is important for protection to ensure that efforts to improve adherence are focused on practices that are effective.

To carry out a more robust evaluation of adherence, this study was designed to include self report and observational measures. Due to only a small number of participating nurses being assigned to care for patients on respiratory precautions, the majority of observations were demonstrations and observed adherence could not be evaluated. Instead, competence was assessed, specifically competence with recommended use of an N95 respirator. Findings showed 44% of observed nurses‘ demonstrated competence when using an N95 respirator despite the Toronto SARS experience and the resulting investment in and attention to worker protection against communicable respiratory illness. In the literature, observational studies of adherence with safe work practices were limited due to a myriad of organizational and environmental barriers and no studies were found that examined competence in place of adherence. To mitigate barriers, most studies used only partial measures of adherence (Sutton, Nicas, Reinisch, & Harrison, 1998) (Tokars, et al., 2001) (Kellerman, Saiman, San Gabriel, Besser, & Jarvis, 2001) (Biscotto, Pedroso, Starling, & Roth, 2005) and findings were likely overestimates of true observed adherence. One study used an open concept emergency department to control for environmental barriers to conducting observations (Henry, Campbell, & Maki, 1992) but still had logistical problems. Even though barriers to evaluating observed adherence were anticipated in this study, comprehensive definitions of outcome variables were used to provide more accurate findings. To validate the definition of competence, a relationship between competence and conducting a user seal check was demonstrated: of those who did a seal check (24/100), 22 were competent (92%) and 2 were not competent (8%) (p<0.001).

The critical steps to follow when donning, using, doffing and disposing of an N95 respirator when caring for a patient with a suspect or diagnosed airborne disease were identified according to government guidelines and discussions by the thesis committee. The inclusion of Step 1 – choice of an N95 respirator over a surgical mask, was deemed to be very important as it signified the nurses‘ understanding of the different modes of disease transmission and the

87 associated equipment that would provide adequate protection. The study showed 100% of nurses chose the correct piece of equipment. This could be a direct reflection of the attention paid to the debate over N95 versus surgical mask that has been playing out at the system level since 2003. Step 2 - straps positioned correctly, and Step 3 – adjust nose piece, were also identified as critical due to the importance of a tight seal with use of an N95 respirator. The study showed 76% of nurses positioned the straps correctly and 93% adjusted the nose piece according to the manufacturer‘s guidelines for each type of N95 respirator. These steps would be less critical with a surgical mask as there is less need for a tight seal. Step 4 – seal check, was discussed earlier in this section. Step 5 – refrain from touching the potentially contaminated face piece, was identified as the most problematic step after conducting a seal check. Only 40% of nurses refrained from touching the face piece during use and disposal of the N95 respirator. The likelihood of this error was significantly higher during use of an N95 respirator than a surgical mask [p<0.001], although only 11 observations of surgical mask use and disposal were made. This finding is supported in the recent literature demonstrating high levels of discomfort reported by healthcare workers (Baig, Knapp, Eagan, & Radonovich, 2010) when using N95 respirators and demonstrates the need for future research into a new generation of respirators for use by healthcare workers as protection against communicable respiratory illness.

Interventions to improve uptake of actions involving recommended use of eye protection, conducting a user seal check and refraining from touching the face piece were identified as priorities in both phases of this study and require particular attention at the policy and practice level. At the policy level, organizational leaders could implement specific protocols regarding these actions and specific steps to follow when using facial protective equipment for patients on respiratory precautions could be documented, posted and communicated to staff. Regular training and education could highlight these practices as requiring specific attention. At the practice level, unit leaders could evaluate use of eye protection, adherence to seal checks and avoidance of touching the face piece through supervisor, peer and equipment audits. The value of auditing practice was illustrated by the finding that, during the observational study, 74% of nurses demonstrating recommended use of facial protective equipment for patients with an airborne disease chose to don eye protection even though the observer didn‘t mention whether there was a risk present of splash/spray to the eyes. During audits, incidents of non-adherence

88 should be immediately identified, corrected and followed up. Non-adherence/adherence could also be part of performance evaluations and merit awards.

5.3 Training and Fit Testing

Study results showed 49% of respondents reported they had not received adequate training and fit testing within the last two years; again, despite the local SARS experience and widespread system investment in healthcare worker safety. Educational elements that were identified as basic to an adequate training program included: the different communicable respiratory diseases that workers may be exposed to, modes of transmission, how workers should protect themselves, and how to properly use a surgical mask, eye protection and an N95 respirator (including how to conduct a seal check). Fit testing and training were combined in this variable as healthcare workers often receive education on all or some of these elements during the fit testing process. The two year time frame was utilized due to the Canadian Standards Association (CAN/CSA-Z94.4-02 (R2008) Selection, Use and Care of Respirators) guidance that fit testing be carried out every two years. Similar substandard rates of training and fit testing were reported in the literature (Lau, et al., 2004) (Kim, Jeffe, Evanoff, Mutha, Freeman, & Fraser, 2001).

These results highlight the need for attention to be given to regular cycles of education, refresher training and verification of knowledge transfer through post testing and auditing. Various types and methods of education have been suggested in the literature including in- service programs, group discussions, lectures, workshops, on-line education or through the distribution of printed materials (Yassi, et al., 2004). This study found 42% of nurses who had received training and fit testing in the last 2 years reported non-adherence to recommended use of facial protective equipment. This finding highlights the need for future research on what formative training and continuing education strategies are most effective in transferring and sustaining knowledge regarding good infection control practices.

Related to substandard reports of quality and frequency of training and education, study findings showed fewer than half of nurses demonstrated adequate knowledge of recommended use of facial protective equipment (46%) and only 56% demonstrated adequate knowledge of transmission of influenza. Regarding knowledge of use of FPE, the item that was answered incorrectly the most often was ―Wearing a surgical mask will help protect me from catching a

89 droplet spread communicable disease‖. Almost all respondents were aware of the reason for conducting fit testing indicating a good understanding of airborne transmission and the importance of a tight face seal. Regarding knowledge of transmission of influenza, the item that was answered incorrectly the most often was ―The contribution of each route of exposure to influenza transmission is uncertain at this time and may vary based upon the characteristics of the influenza strain‖. As the literature search indicated, even experts in the field cannot agree on the mode of transmission of influenza. Regardless, it is important for nurses to be aware of the current state of the evidence that impacts decisions regarding their protection.

These findings highlight the importance of ensuring knowledge is being transferred during educational interventions and sustained between educational cycles. Particular attention should be paid to educating nurses on the protective properties of a surgical mask and the current evidence on mode of transmission of influenza. At the practice level, these findings suggest there may be an important role for periodic knowledge assessments where nurses use on-line quizzes or short tests to identify gaps in knowledge. Based on assessment results, these strategies should trigger the appropriate educational intervention such as a review of written material, a refresher course or, where knowledge gaps are significant, repeating an entire educational session.

5.4 Health Effects

Through the cross-sectional survey, nurses reported occupational exposure to communicable respiratory illness, subsequent disease transmission and adverse health effects experienced. Eighty-two percent of nurses reported being exposed to a communicable respiratory illness at work. Reflecting on the type of units participating in this study - critical care, emergency and medical units designated for patients with respiratory illnesses – this number should be at or near 100%. Study findings may be due to some nurses‘ perceptions that they are well protected at work but are more likely due to nurses underestimation of their risk. Of those that reported being exposed, 30% indicated they had contracted a communicable respiratory illness through work and identified influenza, general upper respiratory tract infection, SARS, the common cold and seroconversion to TB positive as the communicable respiratory illnesses acquired. Although these results were not verified through hospital occupational injury/illness reports, the literature on occupational transmission of SARS indicated 45% of individuals infected in a healthcare setting were health workers and two nurses and one physician died of

SARS-related causes (The SARS Commission, 2006). During the more recent influenza A (H1N1) outbreak, occupational transmission was reported fairly early in the outbreak timeline (Perez-Padilla, et al., 2009) (Wicker, et al., 2009) and to-date one nurse has died of complications resulting from influenza A (H1N1) infection (California Nurses Association, 2009). Of those who reported they had contracted SARS at work, eight nurses indicated they had suffered or were still suffering severe adverse mental health effects as a result. These findings highlight the vulnerability of healthcare workers to occupational transmission of communicable respiratory illness and provide support for ongoing attention to and focus on worker protection.

When the relationship between nurses who had contracted an illness, suffered adverse physical health effects or suffered adverse mental health effects and adherence to recommended use of facial protection was examined, a negative association was discovered. Nurses who reported they had contracted an occupational illness or suffered adverse health effects were significantly more likely to report non-adherence. Even though this finding was not anticipated, it could be the result of nurses who have not suffered an ill effect thinking they are adherent when they actually aren‘t. It could also be due to those nurses who were adherent and got sick anyway felt protection was pointless.

5.5 Determinants of Adherence

Using Moore‘s theoretical framework, demographic, individual, environmental and organizational determinants of adherence to recommended use of facial protective equipment were examined. Multivariable analysis revealed six predictors of adherence: unit type (demographic), at least monthly use of facial protective equipment (individual), ready availability of equipment (environmental), adequate training and fit testing within the last two years (organizational), organizational support for health and safety (organizational) and good communication in the area of health and safety (organizational). The effect of training and fit testing was shown to be dependent on an absence of job hindrances. The majority of predictors fell within the categories of organizational and environmental level variables supporting the original hypothesis and the importance of a climate of safety in a workplace. Moore‘s model was also used to identify determinants of competence with recommended use of facial protective equipment. Multivariable analysis revealed one significant predictor of competence: knowledge of recommended use of facial protective equipment (individual).

5.5.1 Demographic Determinants

In the literature, demographic factors such as gender, education level and occupation have not been found to be consistently associated with adherence to infection control procedures (Gershon, et al., 1995). This study was consistent with this finding and did not show a relationship between gender, nurse type, education, tenure on the unit, work status, or supervisory status and adherence. A relationship between two demographic variables, age and tenure as a nurse, and adherence was demonstrated at the bivariate level but was lost on adjusted, multivariable analysis.

Unit type was revealed as a significant predictor of adherence. Emergency room nurses were 40% less likely to report adherence than critical care nurses. Reports in the literature were conflicting. Several studies found stratification by unit type revealed adherence to safe work behaviours was better among intensive care units than in other types of units (Clock, Cohen, Behta Pharm, Ross, & Larson, 2009) (Duggan, Hensley, Khuder, Papadimos, & Jacobs, 2008). Another report found rates of non-adherence to be highest in emergency departments (Creedon, et al., 2008). Differences in safe work behaviours of nurses working in critical care units, emergency departments and in-patient medical units could lie in the work environment. Emergency room nurses may report higher non-adherence due to the fast-paced, busy work environment where clinical practice is focused on making rapid diagnoses, managing crises and dealing with unanticipated events.

Unit level differences were also reported regarding ready availability of facial protective equipment and cleanliness and orderliness of the work environment. Both survey and observational data showed critical care units were significantly more likely to have eye protection readily available than emergency or medical units. This could reflect the higher risk of splash/spray hazards in critical care due to more invasive and aerosol-generating procedures being carried out. Regardless of type, all units should have protective equipment readily available to workers. Critical care nurses were also more likely to report a clean and orderly unit than emergency or medical units but similar results were not found in the observational study.

5.5.2 Individual Determinants

Similar to demographic factors, individual factors such as knowledge, perception of risk, beliefs and attitudes and past experience, have not been found to be consistently associated with adherence to infection control practices (DeJoy, Murphy, & Gershon, 1995) (Gershon, et al., 1995) (DeJoy, Searcy, Murphy, & Gershon, 2000) (McGovern, Vesley, Kochevar, Gershon, Rhame, & Anderson, 2000). This study found similar results and showed only one individual factor to be significantly associated with adherence upon multivariable analysis. Nurses who reported daily, weekly or monthly use of facial protective equipment were 2.5 times more likely to report adherence than nurses who reported using FPE rarely or never. Although this relationship may have some inherent contributory bias, it is important to note the implications it has on policy and practice. Organizations could designate a relatively limited number of staff to care for patients with communicable respiratory illness. This would reduce the risk of inadvertent exposure to other healthcare workers. Other benefits of cohorting include: concentrating resources in one area, making it possible to designate certain equipment, such as portable x-ray units, for use with patients and thereby reduce the risk of possible fomite transmission; allowance for concentrated and intensive infection control activities, such as careful monitoring of compliance with precautions and entry screening; and, the physical separation of patients from others in the hospital (Srinivasan, et al., 2004). Limiting the number of trained and experienced workers could also be a problem, particularly during an outbreak when the regular cohort of workers may be depleted through absences due to illness or redeployment or of insufficient number to meet the demand.

In the observational study of nurses‘ competence with recommended use of an N95 respirator, multivariable analysis revealed one individual level variable as the sole predictor to competence. Nurses who were knowledgeable in the recommended use of facial protective equipment were 2.9 times more likely to demonstrate competent use of an N95 respirator. As competence was measured outside of the work environment, it makes sense that it would be influenced by an individual level factor such as knowledge. Measuring competence through worker demonstration eliminates the interaction between the worker and the work environment that is so critical when measuring adherence.

In the nursing literature, much attention has been given to the measurement of competence versus performance. For comparison to this study, performance is consistent with adherence as it reflects worker actions or behaviours that are influenced by the work environment. Manual skills were identified as the foundation of nursing performance and continue to be the focus of evaluation in the academic environment. Many disagree with this competence-based model of training and evaluation for the same reasons that we chose to use a framework for evaluating adherence that explored the work environment and the safety climate of the organization. Those who disagree suggest evaluating assessment skills, cognitive skills, decision making, and the ability to deal with unexpected, complicated or challenging situations requiring creativity and sensitivity (While, 1994) in addition to competence to provide a more comprehensive model for evaluating acceptable clinical practice.

This study of competence with recommended use of facial protective equipment has generated several important areas of future research. The first is examining the relationship between competence and adherence. Several reports in the literature have reflected on this issue. One author claimed that competence assessments can have predictive value for performance in actual practice, if efficiency and consultation times are taken into consideration (Rethans, Sturmans, Drop, Van der Vleutens, & Hobus, 1991). Another author identified writing short clinical reports under time pressure may require similar skills to writing exam answers (McFarlane, Goldney, & Kalney, 1989). Quantifying this relationship could show that while competence does not always lead to effective performance or adherence, it could be seen as one of its many determinants. This area of research is particularly important as studies of observed adherence to recommended use of facial protective equipment are logistically difficult in the acute healthcare sector. This leads to the second area of future research that has stemmed from this work: the examination of different study methodologies that could mitigate these barriers.

5.5.3 Environmental Determinants

Environmental factors including cleanliness and orderliness of the work environment and ready availability of equipment have been shown to be predictors of adherence to infection control practices (Gershon, et al., 2000) (Lau, et al., 2004). This study was consistent with the literature supporting a relationship between availability of equipment and adherence and found nurses who reported ready availability of surgical masks, respirators and eye protection were 1.5

94 times more likely to report adherence to recommended use of FPE. The measure of ―ready‖ availability used in this study was supported by the literature examining compliance with universal precautions (DeJoy, Searcy, Murphy, & Gershon, 2000) (Green-McKenzie, Gershon, & Karkashian, 2001) and was defined as equipment being available at point-of-use, meaning on a cart or shelf right outside the patients room, or within a short walking distance (up to 3 metres away) (Institute for Healthcare Improvement, 2006).

Challenges to accurately assessing ready availability of equipment were identified during the study and demonstrated when a comparison between survey reports and observational data revealed slight to fair agreement between measures. Some units stored respiratory precautions carts off the unit until the patient arrived. Locations of carts stored on the units varied including the clean utility room, in the hall or in the patients‘ room. Other units had different carts for airborne and droplet precautions. Finally, each nurse was fit-tested for a specific make, model and size of N95 respirator and the availability of each nurses‘ respirator should factor into a measure of availability. This is further supported by the finding during the observational study that 20% of nurses did not use the N95 they were fit tested for during their demonstration as it wasn‘t available. From a policy and practice perspective, it is important for organizational leaders to develop a system to ensure ready availability of individual-specific equipment while accounting for unit level differences in cart type and storage requirements. Organizations should have dedicated and readily accessible areas to store facial protective equipment and the resources to audit these areas to ensure they remain well-stocked.

Media coverage is an external environmental factor and has also been shown to influence risk perception and uptake of safe work practices (Lymer, Richt, & Isaksson, 2004) (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008). Findings from this study did not demonstrate a significant relationship between media coverage and adherence to recommended use of facial protective equipment. This could be due to the timing of the study. During 2007 and 2008 when data was being collected for this study, communicable respiratory illness was not a media headline like it was post-SARS (2003) or during the influenza A (H1N1) outbreak (2009). This study also represented smaller community hospitals that weren‘t affected by the SARS outbreak as dramatically as the teaching hospitals involved in the pilot study (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008), although all six participating hospitals reported they had cared for patients with SARS. This investigation also involved a much larger

95 study population that allowed for greater power than those reported in the literature (Lymer, Richt, & Isaksson, 2004) (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008). From a social marketing perspective, these findings raise questions about the use of media outlets such as newspaper, television and radio to communicate with the public during health crises and suggest that timing could play an important role in effectiveness. During the influenza A (H1N1) outbreak, media was used to communicate the government‘s vaccine deployment strategy in reaction to mass line-ups and public confusion regarding priority groups. Later in the outbreak, media was also used to encourage uptake of vaccination. The success or failure of this ―reactive‖ strategy is currently under debate and highlights the need for future research in this area.

Although supported in the literature, this study did not demonstrate a significant relationship between self-reported cleanliness and orderliness of the work environment and adherence to recommended use of facial protective equipment. This could be due to the subjective nature of this variable. Study participants may have had different definitions of the terms ―clean‖ and ―orderly‖. When assessing cleanliness and orderliness, nurses may have assessed the patient room they were working in, their pod or section of the unit or their unit as a whole. In addition, survey reports and observations of cleanliness and orderliness revealed only slight agreement. Other factors may have affected agreement between these two measures. Survey data was collected one to three months before unit observations were made allowing for opportunity for the cleanliness or orderliness of the unit to change. Also, several units were undergoing construction and renovation which would negatively impact on cleanliness and orderliness and exaggerate differences due to time lapsed.

5.5.4 Organizational Determinants

Organizational factors were identified in the literature as the most important determinant of adherence to safe work behaviours to prevent the transmission of blood borne disease (DeJoy, Murphy, & Gershon, 1995) (Gershon, et al., 1995) (DeJoy, Searcy, Murphy, & Gershon, 2000) (McGovern, Vesley, Kochevar, Gershon, Rhame, & Anderson, 2000) (Michalsen, et al., 1997), and adherence to general infection control procedures to prevent communicable respiratory illnesses (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005) (Yassi, et al., 2007). This examination of nurses‘ adherence to recommended use of facial protective equipment to prevent

96 the spread of communicable respiratory illness demonstrated findings consistent with these reports. Three organizational level variables showed a significant relationship with adherence: training and fit testing, organizational support and good communication. These findings highlight the importance of implementing measures that support an organizational climate of safety in healthcare.

While only half of the survey population reported being trained and fit tested within the last two years, those that were trained and fit tested were 1.66 times more likely to report adherence with recommended use of facial protective equipment. Similar results were reported in the literature. In a study of 1,716 healthcare workers from three acute care hospitals, the infection control training factor was associated with safe practices (Gershon, et al., 1995), and in a study involving 322 hospital-based physicians, hours of training was significantly associated with universal precautions practice compliance (Michalsen, et al., 1997). Another study examined the influence of safety climate factors on bloodborne pathogen exposures among 789 employees of a large urban research medical centre and found feedback and training to be associated with a reduction in blood and body fluid exposure incidents (Gershon, et al., 2000). During the SARS outbreak, one study demonstrated an association of developing SARS with clinicians who had received less than two hours of respiratory infection control training (Lau, et al., 2004). Previous sections in this report discuss the need for regular cycles of education, refresher training, verification of knowledge transfer and further research into identifying the formative training and continuing education strategies that are most effective in transferring knowledge regarding recommended use of facial protective equipment (Section 5.3).

In an examination of key interactions between organizational level variables, this study showed the effect of training and fit testing on adherence was dependent on an absence of job hindrances such as a heavy workload or a lack of time. This finding supports the classification of fit testing and training as an organizational level variable. For nurses to attend training and fit testing, the organization must have processes in place to provide coverage on the unit for the duration of the session.

The second organizational level variable that was identified as a significant predictor of adherence was organizational support for health and safety. Nurses who perceived their organization supported the health and safety of employees were almost 2 times more likely to

97 report adherence to recommended use of facial protective equipment. This finding is also consistent with the literature related to universal precautions (McGovern, Vesley, Kochevar, Gershon, Rhame, & Anderson, 2000) (Rivers, Aday, Frankowski, Felknor, White, & Nichols, 2003) (Gershon, et al., 2000) and adherence with general infection control practices to prevent the spread of communicable respiratory illness (Moore, Gilbert, Saunders, Bryce, & Yassi, 2005). This relationship highlights how important it is that nurses think their health and safety is valued by their employer as it suggests those who perceive they are valued will engage in healthier work behaviours. At the policy level, interventions targeted at developing and communicating policies and procedures related to eye and respiratory protection and establishing steps to follow in proper use of facial protective equipment should be carried out. At the unit level, front line managers should involve workers in issues related to FPE and enforce adherence with legislation and workplace policy.

Good communication regarding health and safety was the third predictor of adherence. Nurses who perceived good communication regarding health and safety within their organization were 2.13 times more likely to report adherence to recommended use of facial protective equipment. Again, this finding was consistent with the literature. One study conducted in the emergency department setting found healthcare workers‘ compliance with barrier precautions improved when staff was notified of the arrival of patients by ambulance staff. This pre- notification resulted in 92% of staff using appropriate protection compared with 63% when patients arrived unannounced (Digiacomo, et al., 1997). In a review of the effectiveness of various interventions aimed at changing clinical practice, authors reported evidence that educational outreach visits, posted reminders, interactive educational meetings and other multi- faceted interventions were effective in improving the transfer of information into practice. Passive or one-dimensional interventions, such as mailing out new recommendations, were generally not found to be effective but could be part of a broader strategy (Bero, Grilli, Grimshaw, Harvey, Oxman, & Thomson, 1998).

The communication scale used in the survey also included elements of feedback on performance including ―On my unit, employees are encouraged to identify unsafe work practices amongst themselves‖ and ―On my unit, unsafe work practices that aren‘t resolved are corrected by supervisors‖. In the literature feedback on performance has been shown to predict adherence. In a study of nurses‘ adherence to universal precautions, feedback was identified as a predictor of

98 compliance but the type of feedback or communication used was not identified (DeJoy, Murphy, & Gershon, 1995). In another study of blood borne pathogen exposures, the frequency of exposure incidents was significantly lower when employees received high levels of safety- related feedback (Gershon, et al., 2000). A third study examined a group of Thai healthcare workers and found much higher compliance with glove use and hand washing during a peer feedback intervention, but noted that the effect lessened with the passing of time (Moongtui, Gauthier, & Turner, 2000).

The association between good communication and adherence has implications at the policy and practice level. At the system and organizational levels, leaders should develop communication strategies regarding infection prevention and control for regular and outbreak situations. Strategies should encompass multiple approaches to information transfer. At the unit level, leaders should facilitate and reinforce information sharing, evaluate knowledge transfer and ensure adherence to health and safety practices are communicated to employees during regular performance evaluations. An important area of future research identified earlier includes the use of media outlets as communication tools (Section 5.5.3). Several of the hospitals participating in this study had installed televisions in common areas and were broadcasting their own hospital channel. An evaluation of knowledge transfer to workers, patients and the public through this communication strategy would be a useful area of future research.

5.5.5 Safety Climate

Each of the three organizational predictors of adherence (training and fit testing, organizational support and good communication), along with the environmental level predictor ―ready availability of facial protective equipment‖, are important components of an organization‘s safety climate. The literature review demonstrated the importance of a strong climate of safety to worker adherence to safe work behaviours in the healthcare setting (Gershon, et al., 1995) (Rivers, Aday, Frankowski, Felknor, White, & Nichols, 2003) (Vaughn, McCoy, Beekmann, Woolson, Torner, & Doebbeling, 2004) (Gershon, et al., 2000) (McDiarmid & Condon, 2005) (Nichol, Bigelow, O'Brien-Pallas, McGeer, Manno, & Holness, 2008). Safety climate is based on the organizational and environmental characteristics of a company or institution as these characteristics signify commitment to employee health, safety and wellness.

Training is an organizational variable of particular importance to safety climate. Organizations that provide regular training and fit testing are demonstrating to their workers that they value their health and safety. The importance of training to enhancing adherence may be less related to building knowledge or skills and more related to demonstrating commitment to keeping workers informed (Yassi, et al., 2007). Training and education can also be used as communication conduits that will allow healthcare workers to feel confident in their organizations abilities to manage existing and emerging infections. To this extent, investment in training should not only be motivated by the desire to change beliefs but also be focused on ensuring that healthcare workers are aware of organizational policies, procedures and expectations.

The results of this study demonstrate the important relationship between specific organizational level variables and adherence to recommended use of facial protective equipment. Based on these findings, a preliminary organizational model of adherence is provided (Figure 6). Organizational leaders can use this model to focus efforts to enhance adherence on implementing strategies to ensure facial protective equipment is readily available, provide regular training and fit testing, demonstrate organizational support for health and safety through policies, procedures, resource allocation and accountability and by establishing a multi-faceted communication strategy regarding health and safety.

Absence of Job Hindrances

Training and Fit Testing

Ready Availability of Equipment Adherence to Reduction in Recommended Transmission of Healthy Use of Facial Communicable Worker Organizational Protective Respiratory Support for Equipment Illness Health and Safety

Good Health and Safety Communication

Figure 6: Preliminary Organizational Model of Adherence

100

5.6 Patient Safety

The concept of safety climate in the healthcare industry has important differences from other industries such as steel factories, offshore environments and nuclear power plants. In healthcare, the results of a safe environment directly affect not only an organization‘s employees but also its customers – patients and the public. As a secondary objective this study examined the relationship between worker safety and patient safety and found a significant relationship between three organizational factors that supported a strong worker safety climate - organizational support, absence of job hindrances and communication - and nurses‘ perceptions of strong patient safety climate. A significant relationship between a strong patient safety climate and reported adherence to recommended use of facial protective equipment was also found and further supports an important link between worker safety climate and patient safety climate.

Only recently has research emerged examining the link between these two constructs. Studies demonstrating a positive or direct relationship between patient safety and worker safety have been reported in the areas of shift work and fatigue (Rogers, Hwang, Scott, Aiken, & Dinges, 2004) (Lockley, et al., 2004) (Landrigan, et al., 2004) (Taub, Morin, Goldrich, Ray, & Benjamin, 2006), nurse staffing levels and workload (Canadian Institute for Health Information, 2006) (Lang, Hodge, Olson, Romano, & Kravitz, 2004) (Nicklin & Graves, 2005), and the existence of specific hazard based programs (Yassi & Hancock, 2005) (Nicholson, 1998) (Carman, et al., 2000) (Engst, Chhokar, Robinson, Earthy, & Yassi, 2004) and are discussed earlier in this report (Section 2.5). Safety climate has also been examined in relation to patient outcomes and shown to have a positive impact on several patient safety indicators including treatment errors, falls, nosocomial infections and complaints (Katz-Navon, Naveh, & Stern, 2005) (Spence Laschinger & Leiter, 2006) (Yassi & Hancock, 2005). Reports demonstrating a negative or inverse relationship between efforts to protect workers and patient safety also exist (Poutanen, Vearncombe, McGeer, Gardam, Large, & Simor, 2005) (Yap, et al., 2004). Further research into this connection could tease out the positive and negative effects of efforts to enhance worker safety on patient safety indicators and outcomes. It would be interesting to identify the common factors and independent drivers of worker safety and patient safety. Moore‘s framework of categorizing drivers as individual, environmental and organizational could be employed. Work to-date suggests that the environmental and organizational drivers to patient and worker safety could be similar, whereas the individual drivers may differ.

101

Based on findings from this study demonstrating a significant relationship between three organizational factors that supported a strong worker safety climate and nurses‘ perceptions of strong patient safety climate, an amended version of Figure 6 is presented (Figure 7). This final organizational model of adherence suggests that strategies employed at the organizational and environmental level to enhance worker safety may result in a safer and healthier patient population as well.

Absence of Job Hindrances

Training and Fit Testing

Good Health and Healthy Safety Patient Communication

Figure 7: Final Organizational Model of Adherence

These findings are of significance to those committed to advancing the occupational health and safety agenda within the healthcare sector. Patient safety has been a national priority in Canada since 2003 yet only recently have policy leaders had access to evidence to show that quality patient care and the reduction of adverse patient outcomes is partly dependent upon a healthy and safe workforce and environment. Unifying the safety of all parties in the healthcare sector – employees, patients and the public – at the national strategy level will serve to greatly advance healthcare worker health, safety and wellness.

102

5.7 Strengths and Limitations

5.7.1 Strengths

There are several important strengths of this study. Phase 1, the cross sectional survey, had a large sample population and a high response rate. Results of the study were consistent with the literature and supported the study hypothesis. A more complete definition of adherence was used than what was reported in the literature. Finally, factor analysis and reliability testing demonstrated psychometric stability of the majority of scales within the survey instrument.

The study included a large survey sample population and had a high response rate (82%). The response rate was higher than what is reported in the literature (Turnberg, et al., 2008) (Gershon, et al., 1995). A study of major influence to this work did not report a response rate (Yassi, et al., 2007). The high response rate in this study may have been due to the Toronto SARS experience yielding a nursing workforce interested in and supportive of research in this area. Almost all nurses who were offered an information sheet and survey accepted it. Data collection methods allowing the researcher to spend time on the units distributing and collecting surveys and talking to nurses likely contributed to the high participation rate. Coffee vouchers were a welcome token of appreciation for participation in the study.

Findings from this study were consistent with the literature regarding adherence to infection control practices. Moore‘s theoretical framework was highly applicable to this examination of adherence. Results supported the study hypothesis that organizational factors and other safety climate dimensions will be predictive of nurses‘ adherence to recommended use of facial protective equipment.

In the literature, observational studies of adherence with safe work practices to prevent transmission of communicable respiratory illness were not abundant due to the numerous organizational and environmental barriers present in the healthcare setting. These barriers included the logistical difficulty of conducting observations of patient care activities from outside an isolation room that has curtains or blinds and the lack of natural opportunity to observe a worker using facial protective equipment. To manage these barriers, several studies within the health sector used only partial measures to define adherence to proper use of an N95 respirator. Even though barriers were anticipated, this study used a more complete definition of

103 adherence based on government and manufacturers guidelines. Measures of observed adherence incorporated those steps that were identified as critical to the protection of a workers‘ health. Where barriers precluded observations of adherence, demonstrations were requested and measures of competence with recommended use of an N95 respirator were recorded.

Factor analysis and factor reliability were conducted to test the psychometric properties of the survey tool. Exploratory factor analysis yielded a three-factor model for organizational level items within the survey. In the pilot study, exploratory factor analysis on the four constructs measured at the organizational level (organizational support, absence of job hindrances, feedback and conflict/communication) demonstrated overlapping constructs. The newly revealed three-factor model demonstrated minimal cross-loading and consisted of a six- item scale to measure organizational support for health and safety, a three-item scale to measure absence of job hindrances to using facial protective equipment and an eight-item scale to measure communication regarding health and safety. Additional exploratory factor analysis was done on the survey scales used to measure individual and environmental factors and results supported original groupings. Internal consistency of each factor was examined using Cronbach‘s reliability co-efficient. Twelve of fifteen scales demonstrated adequate factor reliability. These results demonstrate psychometric stability of the majority of scales within the survey. Further testing of the survey tool could examine if the dimensions hold across different healthcare populations, settings and safety applications over time.

5.7.2 Limitations

Limitations of the study differed according to phase. For both phase 1 and phase 2 of the study, results may not be generalizable to all professions, geographic locations or type of healthcare facility as this study examined nurses working in acute care facilities in the Greater Toronto Area. Inclusion of only two teaching and four community hospitals in the study precluded the examination of hospital type as an explanatory variable.

In phase 1 the utilization of a cross-sectional study design precluded the determination of causality. This design was chosen as it provided a "snapshot" of adherence to facial protective equipment and was appropriate for the objective of describing the relationship between adherence and a set of predetermined individual, environmental and organizational variables. This study design was also chosen as it was cost-effective and could be carried out in a busy,

104 congested and complex work environment with minimal interruption of workflow. Self- selection bias may have been a limitation as participation in the study was voluntary and a decision to participate may have been correlated with traits that affect the study. Random sampling and a mail-out survey were not used as the research team was concerned this strategy would yield a low response rate within the target population. Self-selection bias was minimized by the high participation rate as 51% of nurses employed on the units involved in this study took part in the cross-sectional survey.

Subject recall and social desirability bias may have been a problem given the reliance on self-report data for phase 1 of this study. Survey anonymity would have assisted to reduce social desirability bias. Self-reported adherence has been found to be higher than actual adherence (Henry, Campbell, & Maki, 1992) (O‘Boyle, Henly, & Larson, 2001) (Rickard, 2004). The observational study was designed to mitigate this limitation but we were unable to measure adherence as planned. The collection of objective measures to validate self-report data, particularly organizational and environmental factors, such as evidence of policies and procedures, training and fit testing records and communication protocols would have assisted to validate the data and should be considered in future studies.

Reliability testing for 3 of the explanatory measures fell below the acceptable range (< 0.70). Similar results were reported in the pilot study. The scale to measure knowledge of recommended use of facial protection demonstrated a reliability coefficient of 0.42. Although attempts were made to develop a knowledge scale that measured a single content domain, the umbrella term ‗facial protection‘ represents three distinct types of personal protective equipment (respirators, surgical masks, and eye protection) and is required for two categories of pathogen transmission (airborne and droplet). For these reasons, a high reliability coefficient cannot be expected. The same might be said for the scale to measure knowledge of transmission of influenza as current evidence indicates all three categories of transmission (airborne, droplet and contact) play a role in the spread of influenza. The development of knowledge scales with a single content domain may not be possible within the realm of facial protective equipment. The scale to measure perception of effectiveness of preventive measures demonstrated a reliability coefficient of 0.35. This is likely due to the discrepancy between nurses‘ perceptions of effectiveness of the three types of facial protective equipment: 96% of nurses felt they could reduce their risk of catching a communicable respiratory disease by using an N95 respirator;

105

40% felt they could reduce their risk by using a surgical mask; and, 60% felt they could reduce their risk by using eye protection. A similar difference in perception of effectiveness of the three types of facial protective equipment was found in the pilot study and suggests equipment may need to be evaluated on an individual basis in future studies.

Multi-level modeling of adherence to recommended use of facial protective equipment was not carried out as part of this study due to the focus on organizational level factors. In addition, ethics approval was not granted to study unit level differences. Future analysis of this data set could include multilevel modeling as nurses were sampled from different units and hospitals and it is reasonable to assume that nurses within the same unit would be more homogeneous on certain measures than nurses from different units. Likewise, nurses from the same hospital could be more similar than nurses from different hospitals.

When conducting observations of the cleanliness and orderliness of the unit and ready availability of equipment, only single observations were conducted during different phases and at different times of the study. Emergency and medical unit observations were conducted during phase 1 and critical care unit observations were conducted during phase 2. This made it difficult to make useful comparisons of self-report data and observations as unit conditions may have changed drastically over the three month time difference.

Finally, during the observational study, only single observations of nurses using facial protection were conducted. Due to a limited number of nurses caring for patients on respiratory precautions, the majority of observations were demonstrations and we were unable to observe adherence. There was a 20% study attrition rate in phase 2, mainly due to leaves of absence, nurses no longer working on the unit or last minute shift changes. Only two participants changed their mind about participating. Multivariable analysis of the predictors of competence found knowledge to be a significant predictor but a wide confidence interval suggested a larger sample size was needed.

106

6 CHAPTER 6: Conclusions

This study highlights the historical problem of substandard adherence to safe work practices in the health sector and the need for a novel approach to this complex issue. Findings demonstrated that despite the SARS experience and the resulting investment in our public health system, nurses‘ adherence to recommended use of facial protective equipment remains suboptimal and nurses continue to be vulnerable to contracting occupational respiratory illness. Using Moore‘s theoretical framework that categorizes determinants of healthy work practices into individual, environmental and organizational factors, results showed that organizational factors and other safety climate dimensions were the primary determinants of adherence to recommended use of facial protective equipment. To improve adherence, organizational leaders should focus on organizational and environmental supports including ready availability of protective equipment, regular cycles of training and fit testing, organizational policies, procedures and protocols that are documented, implemented and enforced and a multi-modal communication strategy regarding infection prevention and control for regular and outbreak situations. These interventions should assist to strengthen the safety climate of the organization and foster a safer and healthier work environment for nurses.

Through the observational study, competence with recommended use of an N95 respirator to prevent airborne transmission of communicable respiratory illness was also assessed to be suboptimal. Common gaps included failure to verify the seal and touching the face piece. As competence was assessed external to the situational or ecological context of the work environment, the primary predictor was knowledge of recommended use of facial protective equipment; an individual factor influencing safe work behaviour. To enhance competence, organizational leaders and front line managers should focus on strategies to increase knowledge. Although study findings did not demonstrate a relationship between self-reported adherence and observed competence, further research could examine the role of competence as a predictor of adherence.

The relationship between nurses‘ perceptions of a strong patient safety climate and a strong worker safety climate was examined as a secondary objective of the study. Recently, research has emerged linking the two constructs and this study served as an important opportunity for further investigation. Results supported an important relationship and suggest

107 strategies employed at the organizational and environmental level to enhance worker safety may result in a safer and healthier patient population as well.

Finally, psychometric properties of the survey tool and the feasibility of conducting an observational study in the critical care environment were examined. Results of exploratory factor analysis and reliability testing demonstrate psychometric stability of the majority of scales within the survey instrument and support further testing of this tool. The main challenge to conducting covert observations of recommended use of facial protective equipment in the critical care environment was a lack of patients with communicable respiratory illness. This was despite conducting the study over the winter season when influenza rates are typically higher and on units where patients with acute respiratory disease are cared for. Regardless, the study allowed for examination of competence in place of adherence and results contributed to the body of research on preventing occupational transmission of communicable respiratory illness in the acute healthcare sector.

108

7 CHAPTER 7: Future Directions

This study has generated several important areas of future research in the areas of improving adherence to infection control practices, understanding the relationship between competence and adherence, examining the link between worker and patient safety climate and evaluating observational study methodologies in the acute healthcare setting.

Adherence to recommended use of facial protective equipment was found to be substandard, particularly in the areas of using eye protection and conducting a user seal check for an N95 respirator. Current literature on the benefit of these practices to reducing occupational transmission of communicable respiratory illness is minimal and, in the case of the user seal check, questions the value of the practice. Future research examining ocular transmission of respiratory disease and the protective value of the user seal check are needed to ensure efforts to improve adherence are focused on practices that are effective. Another area of particular concern was nurses touching the potentially contaminated face piece during use and doffing. The likelihood of this error was significantly higher for respirators than surgical masks and the literature provides ample reports of high levels of discomfort associated with N95 respirators. These findings demonstrate the need for future research in to a new generation of respirators for use by healthcare workers as protection against communicable respiratory illness.

The most important determinants of adherence to recommended use of facial protective equipment were found to be the organizational characteristics and other safety climate dimensions of the workplace. These included the ready availability of facial protective equipment, regular training and fit testing, organizational support for health and safety through the development, implementation and enforcement of policies, procedures and protocols and multi-modal communication strategies regarding infection prevention and control. Ready availability of facial protective equipment proved to be difficult to assess as protective equipment is personal (for example the specific N95 respirator that a nurse is fit tested for) and hospitals use different cart and storage systems and strategies. Further research examining different methodologies to evaluate ready availability of equipment would be valuable. Regarding training and fit testing, study findings showed only half of the survey population reported being trained and fit tested within the last two years and, for those who had received training, 42% reported non-adherence to recommended use of facial protective equipment. These

109 findings highlight the need for future research on what formative training and continuing education strategies are most effective in transferring and sustaining knowledge regarding good infection control practices. As a follow-up to this study, intervention studies testing different strategies targeting these safety climate dimensions are needed.

Another interesting area of future research is the use of media outlets such as newspaper, television and radio to communicate with the public during health crises. Although media was not shown to be a predictor of adherence in this study, good communication practices were. The evaluation of novel approaches to communication, such as hospitals having televisions in common areas and broadcasting their own channels would be an interesting area of study. Knowledge transfer to all three health system parties - workers, patients and the public - could be examined.

Competence with recommended use of an N95 respirator for a patient on airborne precautions was examined in Phase 2 of this study. Similar to adherence, competence was found to be substandard among critical care nurses. The primary determinant of competence was found to be knowledge of recommended use of facial protective equipment. No relationship between self-reported adherence and observed competence was found in this study but one could hypothesize that while competence does not always lead to effective performance or adherence, it could be seen as one of its many determinants. Further research examining the relationship between competence and adherence would be valuable. This area of research is particularly important as studies of observed adherence to recommended use of facial protective equipment are logistically difficult in the acute healthcare sector.

This study demonstrated a positive relationship between nurses‘ perceptions of a strong patient safety climate and worker safety climate. Studies from the literature demonstrated examples of both positive and negative relationships between the two constructs. Further research into this connection could tease out the positive and negative effects of efforts to enhance worker safety on patient safety indicators and outcomes. It would be interesting to identify the common factors and independent drivers of worker safety and patient safety. Moore‘s framework of categorizing drivers as individual, environmental and organizational could be employed. Work to-date suggests that the environmental and organizational drivers to patient and worker safety could be similar, whereas the individual drivers may differ.

110

Finally, the examination of different methodologies for investigating adherence to facial protective equipment could be part of future research studies. Using the current methodology, we were unable to measure observed adherence but were able to measure competence. Future work in this area is needed. Regarding statistical analysis, multi-level modeling of adherence to recommended use of facial protective equipment would allow for additional variance components that take into account the different levels of heterogeneity across units and hospitals. Further testing of the survey tool would assist in determining if the dimensions hold across different healthcare populations, settings and safety applications over time.

The future areas of research generated by this study provide exciting opportunities for researchers to advance the current body of knowledge on adherence to important infection control practices. It also provides opportunities to merge fields of study that have traditionally not been linked. This important work will serve to better protect healthcare workers and strengthen the workforce that is so critical to the success of our Canadian public health system.

111

Bibliography

3M. (2004, March 26). 3M Fitting Instructions for 3M Filtering Facepiece Respirators. Retrieved July 30, 2009, from 3M: http://multimedia.3m.com/mws/mediawebserver?66666UuZjcFSLXTtMXMc5x46EVuQEcuZg Vs6EVs6E666666--

Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. New Jersey: Prentice-Hall.

Assanasen, S., Edmond, M., & Bearman, G. (2008). Impact of 2 different levels of performance feedback on compliance with infection control process measures in 2 intensive care units. American Journal of Infection Control , 36 (6), 407-13.

Aveyard, P., Lawrence, T., Cheng, K., Griffin, C., Croghan, E., & Johnson, C. (2006). A randomized controlled trial of smoking cessation for pregnant women to test the effect of a transtheoretical model-based intervention on movement in stage and interaction with baseline stage. British Journal of Health Psychology , 11 (Pt 2), 263-78.

Aveyard, P., Massey, L., Parsons, A., Manaseki, S., & Griffin, C. (2009). The effect of transtheoretical model based interventions on smoking cessation. Social Science and Medicine , 68 (3), 397-403.

Baig, A., Knapp, C., Eagan, A., & Radonovich, L. (2010). Health care workers' views about respirator use and features that should be included in the next generation of respirators. American Journal of Infection Control , in press.

Baker, R., Norton, P., Flintoft, V., Blais, R., Brown, A., Cox, J., et al. (2004). The Canadian adverse events study: The incidence of adverse events among hospital patients in Canada. Canadian Medical Association Journal , 170 (11), 1678-86.

Becker, M. (1974). The health belief model and personal health behavior. Health Education Monographs , 2 (4), 409-19.

112

Behan, F., & Behan, R. (1954). Football Numbers (Continued). American Psychologist , 9 (6), 262-263.

Belser, J., Wadford, D., Xu, J., Katz, J., & Tumpey, T. (2009). Ocular infection of mice with influenza A (H7) viruses: A site of primary replication and spread to the respiratory tract. Journal of Virology , 83 (14), 7075-84.

Ben-Ami, S., Shaham, J., Rabin, S., Melzer, A., & Ribak, J. (2001). The influence of nurses‘ knowledge, attitudes, and health beliefs on their safe behavior with cytotoxic drugs in Israel. Cancer Nursing , 24 (3), 192-200.

Benson, R., & Taub, D. (1993). Using the PRECEDE model for causal analysis of bulimic tendencies among elite women swimmers: Predisposing, reinforcing, and enabling causes in educational diagnosis. Journal of Health Education , 24 (6), 360-8.

Bero, L., Grilli, R., Grimshaw, J., Harvey, E., Oxman, A., & Thomson, M. (1998). Closing the gap between research and practice: An overview of systematic reviews of interventions to promote implementation of research findings by health care professionals. British Medical Journal , 317 (7156), 465-8.

Biscotto, C., Pedroso, E., Starling, C., & Roth, V. (2005). Evaluation of N95 respirator use as a tuberculosis control measure in a resource-limited setting. International Journal of Tuberculosis and Lung Disease , 9 (5), 545-9.

Blachere, F., Lindsley, W., Pearce, T., Anderson, S., Fisher, M., Khakoo, R., et al. (2009). Measurement of airborne influenza virus in a hospital emergency department. Clinical Infectious Diseases , 48, 438-40.

Borrelli, B., McQuaid, E., Becker, B., Hammond, K., Papandonatos, G., Fritz, G., et al. (2002). Motivating parents of kids with asthma to quit smoking: The PAQS project. Health Education Research , 17 (5), 659-669.

Brankston, G., Gitterman, L., Hirji, Z., Lemieux, M., & Gardam, M. (2007). Transmission of influenza A in human beings. The Lancet Infectious Diseases , 7 (4), 257-265.

113

Brown, H. (2004). WHO confirms human-to-human avian flu transmission. The Lancet , 363, 462.

California Nurses Association. (2009, July 31). Swine flu fatal to local cancer nurse. Retrieved December 17, 2009, from California Nurses Association: http://www.calnurses.org/media- center/in-the-news/2009/july/swine-flu-fatal-to-local-cancer-nurse.html

Canadian Institute for Health Information. (2006, December). Findings from the 2005 national survey of the work and health of nurses. Retrieved July 2, 2009, from Canadian Institute for Health Information: http://secure.cihi.ca/cihiweb/dispPage.jsp?cw_page=AR_1588_E

Canadian Nurses Association. (2009, July). RN Workforce Profiles by Area of Responsibility. Retrieved December 2009, 14, from Canadian Nurses Association: http://www.cna- aiic.ca/CNA/documents/pdf/publications/2007_RN_Profiles_e.pdf

Canadian Patient Safety Institute. (2008). Overview. Retrieved July 2, 2009, from Canadian Patient Safety Institute: http://www.patientsafetyinstitute.ca/English/About/overview/Pages/default.aspx

Carman, W., Elder, A., Wallace, L., McAulay, K., Walker, A., Murray, G., et al. (2000). Effects of influenza vaccination of health-care workers on mortality of elderly people in long-term care: a randomised controlled trial. The Lancet , 355 (9198), 93-7.

Centers for Disease Control and Prevention. (2005). Estimated Influenza Vaccination Coverage Among Adults and Children: United States, September 1, 2004 - January 31, 2005. Morbidity and Mortality Weekly Report , 54 (12), 304-307.

Centers for Disease Control and Prevention. (2006). Influenza vaccination of health-care personnel. Morbidity and Mortality Weekly Report , 55 (RR02), 1-16.

Centers for Disease Control and Prevention. (2009, May 13). Interim Guidance for Infection Control for Care of Patients with Confirmed or Suspected Novel Influenza A (H1N1) Virus Infection in a Healthcare Setting. Retrieved September 8, 2009, from Centers for Disease Control and Prevention: http://www.cdc.gov/h1n1flu/guidelines_infection_control.htm

114

Centers for Disease Control and Prevention. (1998, December 11). Morbidity and Mortality Weekly Report. Retrieved December 16, 2009, from Laboratory Performance Evaluation of N95 Filtering Facepiece Respirators, 1996: http://www.cdc.gov/mmwr/preview/mmwrhtml/00055954.htm

Centers for Disease Control and Prevention. (2009). Novel Influenza A (H1N1) virus infections among health-care personnel: United States, April--May 2009. Morbidity and Mortality Weekly Report , 58 (23), 641-5.

Chia, S. (2005). Use of respiratory protection among health care workers and emerging infectious diseases. The Global Occupational Health Network Newsletter , Winter (8), 6-7.

Chiang, L., Huang, J., & Lu, C. (2003). Educational diagnosis of self-management behaviors of parents with asthmatic children by triangulation based on PRECEDE-PROCEED model in Taiwan. Patient Education and Counseling , 49 (1), 19-25.

Clarke, S. (2006). The relationship between safety climate and safety performance: A meta- analytic review. Journal of Occupational Health Psychology , 11 (4), 315-27.

Clock, S., Cohen, B., Behta Pharm, M., Ross, B., & Larson, E. (2009). Contact precautions for multidrug-resistant organisms: Current recommendations and actual practice . American Journal of Infection Control , in press.

Cohen, A. (1977). Factors in successful occupational safety programs. Journal of Safety Research , 9, 168-78.

Cole, M. (2006). Using a motivational paradigm to improve handwashing compliance. Nurse Education in Practice , 6 (3), 156-162.

Colla, J., Bracken, A., Kinney, L., & Weeks, W. (2005). Measuring patient safety climate: A review of surveys. Quality and Safety in Health Care , 14 (5), 364-6.

College of Nurses of Ontario. (2005). Infection Prevention and Control Practice Standard. Toronto: College of Nurses of Ontario.

115

Comrey, A., & Lee, H. (1992). A first course in factor analysis: Second edition. Hillsdale, New Jersey: Lawrence Earlbaum Associates, Inc.

Conway, J., & Huffcutt, A. (2003). A review and evaluation of exploratory factor analysis practices in organizational research. Organizational Research Methods , 6 (2), 147-168.

Costanza, M., Luckmann, R., Stoddard, A., Avrunin, J., White, M., Stark, J., et al. (2005). Applying a stage model of behavior change to colon cancer screening. Preventive Medicine , 41 (3-4), 707-719.

Council of Canadian Academies. (2008, January). Influenza transmission and the role of personal protective respiratory equipment: An assessment of the evidence. Retrieved June 11, 2009, from Science Advice: http://www.scienceadvice.ca/documents/(2007-12- 19)_Report_in_Focus_-_Influenza.pdf

Creedon, S., Slevin, B., De Souza, V., Mannix, M., Quinn, G., Boyle, L., et al. (2008). Hand hygiene compliance: exploring variations in practice between hospitals. Nursing Times , 104 (49), 32-35.

Cromer, A., Hutsell, S., Latham, S., Bryant, K., Wacker, B., Smith, S., et al. (2004). Impact of implementing a method of feedback and accountability related to contact precautions compliance. American Journal of Infection Control , 32 (8), 451-5.

Davidson Dick, D., Weisbrod, L., Gregory, D., Dyck, N., & Neudorf, K. (2006). Case study: On the leading edge of new curricula concepts: Systems and safety in nursing education. Canadian Journal of Nursing Leadership , 19 (3), 34-42.

Dedobbeleer, N., & Beland, F. (1991). A safety climate measure for construction sites. Journal of Safety Research , 22 (2), 97-103.

DeJoy, D. (1986). A Behavioral diagnostic model for self-protective behavior in the workplace. Professional Safety , 31 (12), 26-30.

DeJoy, D. (1996). Theoretical models of health behavior and workplace self-protective behavior. Journal of Safety Research , 27 (2), 61-72.

116

DeJoy, D., Murphy, L., & Gershon, R. (1995). The influence of employee, job/task, and organizational factors on adherence to universal precautions among nurses. International Journal of Industrial Ergonomics , 16, 43-55.

DeJoy, D., Searcy, C., Murphy, L., & Gershon, R. (2000). Behavioral-diagnostic analysis of compliance with universal precautions among nurses. Journal of Occupational Health Psychology , 5 (1), 127-141.

Delaney, L., McKay, R., & Freeman, A. (2003). Determination of known exhalation valve damage using a negative pressure user seal check method on full facepiece respirators. Journal of Applied Occupational and Environmental Hygiene , 18, 237-43.

Derrick, J., Chana, Y., Gomersalla, C., & Luib, S. (2005). Predictive value of the user seal check in determining half-face respirator fit. Journal of Hospital Infection , 59, 152-5.

Digiacomo, J., Hoff, W., Rotondo, M., Martin, K., Kauder, D., Anderson, H., et al. (1997). Barrier precautions in trauma resuscitation: Real-time analysis utilizing videotape review. The American Journal of Emergency Medicine , 15 (1), 34-9.

Duggan, J., Hensley, S., Khuder, S., Papadimos, T., & Jacobs, L. (2008). Inverse correlation between level of professional education and rate of handwashing compliance in a teaching hospital. Infection Control and Hospital Epidemiology , 29 (6), 534–538.

Engst, C., Chhokar, R., Robinson, D., Earthy, A., & Yassi, A. (2004). Implementation of a scheduled toileting program in a long-term care facility. Journal of the American Association of Occupational Health Nurses , 52 (10), 427-35.

Evanoff, B., Kim, L., Mutha, S., Jeffe, D., Haase, C., Anderekc, D., et al. (1999). Compliance with universal precautions among emergency department personnel caring for trauma patients. Annals of Emergency medicine , 33 (2), 160-165.

Fabian, P., McDevitt, J., DeHaan, W., Fung, R., Cowling, B., Chan, K., et al. (2008). Influenza virus in human exhaled breath: An observational study. PLoS ONE , 3 (7), 1-6.

Ford, J., MacCallum, R., & Tait, M. (1986). The application of exploratory factor analysis in applied psychology: A critical review and analysis. Personnel Psychology , 39, 291-314.

117

Ford, K., MacCallum, R., & Tait, M. (1986). The application of exploratory factor analysis in applied psychology: A critical review and analysis. Personnel Psychology , 39 (2), 291 - 314.

Gale, E. (2004). The Hawthorne studies—a fable for our times? Quarterly Journal of Medicine , 97, 439-49.

Gershon, R., Karkashian, C., Grosch, J., Murphy, L., Escamilla-Cejudo, A., Flanagan, P., et al. (2000). Hospital safety climate and its relationship with safe work practices and workplace exposure incidents. American Journal of Infection Control , 28 (3), 211-21.

Gershon, R., Vlahov, D., Felknor, S., Vesley, D., Johnson, P., Delclos, G., et al. (1995). Compliance with universal precautions among health care workers at three regional hospitals. American Journal of Infection Control , 23 (4), 225-36.

Gillen, M., Baltz, D., Gassel, M., Kirsch, L., & Vaccaro, D. (2002). Perceived safety climate, job demands, and coworker support among union and nonunion injured construction workers. Journal of Safety Research , 33 (1), 33-51.

Godin, G., Naccache, H., Morel, S., & Ebacher, M. (2000). Determinants of nurses adherence to universal precautions for venipunctures. American Journal of Infection Control , 28 (5), 359-364.

Gopalakrishna, G., Choo, P., Y., L., Tay, B., Lim, Y., Khan, A., et al. (2004). SARS transmission and hospital containment. Emerging Infectious Diseases , 10 (3), 395-400.

Green, L. (2007, May 5). A resource for instructors, students, health practitioners, and researchers using: The PRECEDE-PROCEED model for health program planning and evaluation. Retrieved June 19, 2009, from The Precede-Proceed Model of Health Program Planning & Evaluation : http://www.lgreen.net/precede.htm

Green, L., & Kreuter, M. (1991). Health promotion planning: An educational and environmental approach, 2nd Edition. Mountain View, California: Mayfield.

Green, L., Kreuter, M., Deeds, S., & Partridge, K. (1980). Health education planning: A diagnostic approach. Palo Alto, California: Mayfield.

118

Green-McKenzie, J., Gershon, R., & Karkashian, C. (2001). Infection control practices among correctional healthcare workers: Effect of management attitudes and availability of protective equipment and engineering controls. Infection Control and Hospital Epidemiology , 22 (9), 555- 559.

Hahn, S., & Murphy, L. (2008). A short scale for measuring safety climate. Safety Science , 46, 1047-1066.

Hair, J., Anderson, R., Tathan, R., & Black, W. (1998). Multivariate data analysis with readings: Fifth edition. Englewood Cliffs, NJ: Prentice-Hall.

Hammond, J., Eckes, J., Gomez, g., & Cunningham, D. (1990). HIV, trauma and infection control: Universal precautions are universally ignored. Journal of Trauma-Injury Infection and Critical Care , 30 (5), 555-561.

Helmreich, R. (2000). On error management: Lessons from aviation. British Medical Journal , 320, 781-5.

Henkel, R. (1975). Part-whole correlations and the treatment of ordinal and quasi-interval data as interval data. Pacific Sociological Review , 18 (1), 3-26.

Henry, K., Campbell, S., & Maki, M. (1992). A comparison of observed and self-reported compliance with universal precautions among emergency department personnel at a Minnesota public teaching hospital: Implications for assessing infection control programs. Annals of Emergency Medicine , 21 (8), 940-6.

Hollmeyer, H., Hayden, F., Poland, G., & Buchholz, U. (2009). Influenza vaccination of health care workers in hospitals: A review of studies on attitudes and predictors. Vaccine , 27 (30), 3935-3944.

Hon, C., Gamage, B., Bryce, E., LoChang, J., Yassi, A., Maultsaid, D., et al. (2008). Personal protective equipment in health care: Can online infection control courses transfer knowledge and improve proper selection and use? American Journal of Infection Control , 36 (10), e33-7.

Hosmer, D., & Lemeshow, S. (2000). Applied logistic regression: Second Edition. New York: John Wiley & Sons, Inc.

119

Institute for Healthcare Improvement. (2006). How-to Guide: Improving Hand Hygiene - A Guide for Improving Practices among Health Care Workers. Retrieved September 11, 2009, from The Society for Healthcare Epidemiology of America: http://www.shea- online.org/Assets/files/IHI_Hand_Hygiene.pdf

Institute of Medicine of the National Academies. (2009). Respiratory Protection for Healthcare Workers in the Workplace against Novel H1N1 Influenza A. Washington, D.C.: The National Academiew Press.

Institute of Medicine. (2008). Preparing for an influenza pandemic: Personal protective equipment for healthcare workers. Washington, DC: The National Academies Press.

Institute of Medicine. (2007). Report Brief - Preparing for an influenza pandemic: Personal protective equipment for healthcare workers. Retrieved June 12, 2009, from Institute of Medicine: http://iom.edu/Object.File/Master/46/098/Influenza%20Report%20Brief%20web.pdf

Institute of Medicine. (1999, November). To Err is Human: Building a Safer Health System. Retrieved July 2, 2009, from Institute of Medicine: http://www.iom.edu/Object.File/Master/4/117/ToErr-8pager.pdf

International Organization for Standardization. (2000). ISO 9000 Quality Management. Geneva, Switzerland: ISO.

Janz, N., Champion, V., & Strecher, V. (2002). The health belief model. In K. Glanz, B. Rimer, & F. Lewis, Health Behavior and Health Education: Theory, Research and Practice (Third Edition) (pp. 45-66). San Francisco: Jossey-Bass.

Jenner, E., Fletcher, B., Watson, P., Jones, F., Miller, L., & Scott, G. (2006). Discrepancy between self-reported and observed hand hygiene behaviour in healthcare professionals. Journal of Hospital Infection , 63 (4), 418-22.

Jiang, X., Elam, G., Yuen, C., Voeten, H., de Zwart, O., Veldhuijzen, I., et al. (2009). The perceived threat of SARS and its impact on precautionary actions and adverse consequences: A qualitative study among Chinese communities in the United Kingdom and the Netherlands. International Journal of Behavioral Medicine , 16 (1), 58-67.

120

Kak, N., Burkhalter, B., & Cooper, M. (2001, July). Measuring the Competence of Healthcare Providers. Retrieved September 8, 2009, from USAID Health Care Improvement Project: http://www.hciproject.org/methods&tools/PDFs/comptence.pdf

Katz-Navon, T., Naveh, E., & Stern, Z. (2005). Safety climate in health care organizations: A multidimensional approach. Academy of Management Journal , 48 (6), 1075-1089.

Kelen, G., DiGiovanna, T., Celentano, D., Kalainov, D., Bisson, L., Junkins, E., et al. (1990). Adherence to universal (barrier) precuations during interventions on critically ill and injured emergency department patients. Journal of Acquired Immune Deficiency Syndromes , 3 (10), 987-94.

Kellerman, S., Saiman, L., San Gabriel, P., Besser, R., & Jarvis, W. (2001). Observational study of the use of infection control interventions for mycobacterium tuberculosis in pediatric facilities. Pediatric Infectious Disease Journal , 20 (6), 566-573.

Kho, D. (2005). Emerging infections among health care workers: The Severe Acute Respiratory Syndrome (SARS) experience. The Global Occupational Health Network Newsletter , Winter (8), 3-4.

Kho, M., Carbone, M., Lucas, J., & Cook, D. (2005). Safety climate survey: Reliability of results from a multicenter ICU survey. Quality and Safety in Health Care , 14 (4), 273-8.

Kim, L., Jeffe, D., Evanoff, B., Mutha, S., Freeman, B., & Fraser, V. (2001). Improved compliance With universal precautions in the operating room following an educational intervention. Infection Control and Hospital Epidemiology , 22 (8), 522-4.

Kretzer, E., & Larson, E. (1998). Behavioral interventions to improve infection control practices. American Journal of Infection Control , 26 (3), 245-53.

Kuhn, A., & Youngberg, B. (2002). The need for risk management to evolve to assure a culture of safety. Quality and Safety in Health Care , 11, 158-62.

Landis, J., & Koch, G. (1977). The measurement of observer agreement for categorical data. Biometrics , 33, 159-174.

121

Landrigan, C., Rothschild, J., Cronin, J., Kaushal, R., Burdick, E., Katz, J., et al. (2004). Effect of reducing interns' work hours on serious medical errors in intensive care units. New England Journal of Medicine , 351 (18), 1838-48.

Lang, T., Hodge, M., Olson, V., Romano, P., & Kravitz, R. (2004). Nurse-patient ratios: A systematic review on the effects of nurse staffing on patient, nurse employee, and hospital outcomes. Journal of Nursing Administration , 34 (7-8), 326-37.

Lau, J., Fung, K., Wong, T., Kim, J., Wong, E., Chung, S., et al. (2004). SARS transmission among hospital workers in Hong Kong. Emerging Infectious Diseases , 10 (2).

Lester, R., McGeer, A., Tomlinson, G., & Detsky, A. (2003). Use of, effectiveness of, and attitudes regarding influenza vaccine among house staff. Infection Control and Hospital Epidemiology , 24, 839-44.

Likert, R. (1932). A technique for the measurement of attitudes. Archives of Psychology , 140, 1- 55.

Lockley, S., Cronin, J., Evans, E., Cade, B., Lee, C., Landrigan, C., et al. (2004). Effect of reducing interns' weekly work hours on sleep and attentional failures. New England Journal of Medicine , 351 (18), 1829-1837.

Loeb, M., McGeer, A., Henry, B., Ofner, M., Rose, D., Hlywka, T., et al. (2004). SARS among critical care nurses, Toronto. Emerging Infectious Diseases , 10 (2), 251-5.

Lowe, J., & Wilder-Smith, A. (2005). Infectious respiratory illnesses and their impact on healthcare workers: A review. Annals Academy of Medicine , 34 (1), 105-110.

Lundstrom, T., Pugliese, G., Bartley, J., Cox, J., & Guither, C. (2002). Organizational and environmental factors that affect worker health and safeety and patient outcomes. American Journal of Infection Control , 30 (2), 93-106.

Lymer, U., Richt, B., & Isaksson, B. (2004). Blood exposure: factors promoting health care worker's compliance with guidelines in connection with risk. Journal of Clinical Nursing , 13, 547-54.

122

Macfarlane, S. (1997). Conducting a descriptive survey: 2. Choosing a sampling strategy. Tropical Doctor , 27 (1), 14-21.

McDiarmid, M., & Condon, M. (2005). Organizational safety culture/climate and worker compliance with hazardous drug guidelines: Lessons from the blood-borne pathogen experience. Journal of Occupational and Environmental Medicine , 47 (7), 740-9.

McFarlane, A., Goldney, R., & Kalney, R. (1989). A factor analytic study of clinical competence in undergraduate psychiatry. Medical Education , 23, 422-28.

McGovern, P., Vesley, D., Kochevar, L., Gershon, R., Rhame, F., & Anderson, E. (2000). Factors affecting universal precautions compliance. Journal of Business and Psychology , 15 (1), 149-161.

McGuckin, M., Waterman, R., & Govednik, J. (2009). Hand hygiene compliance rates in the United States: A one-year multicenter collaboration using product/volume usage measurement and feedback. American Journal of Medical Quality , 24 (3), 205-13.

Mearns, K., Whitaker, S., & Flin, R. (2003). Safety climate, safety management practice and safety performance in offshore environments. Safety Science , 41 (8), 641-80.

Michalsen, A., Delclos, G., Felknor, S., Davidson, A., Johnson, P., Vesley, D., et al. (1997). Compliance with universal precautions among physicians. Journal of Occupational and Environmental Medicine , 39 (2), 130-37.

Montano, D., Kasprzyk, D., & Taplin, S. (2002). The theory of reasoned action and the theory of planned behavior. In K. Glanz, B. Rimer, & F. Lewis, Health Behavior and Health Education: Theory, Research and Practice (Third Edition) (pp. 67-98). San Francisco: Jossey-Bass.

Moongtui, W., Gauthier, D., & Turner, J. (2000). Using peer feedback to improve handwashing and glove usage among Thai healthcare workers. American Journal of Infection Control , 28, 365-9.

Moore, D., Gilbert, M., Saunders, S., Bryce, E., & Yassi, A. (2005). Occupational health and infection control practices related to severe acute respiratory syndrome: Health care worker

123 perceptions. Journal of the American Association of Occupational Health Nurses , 53 (6), 257- 66.

Morgan, O., Kuhne, M., Nair, P., Verlander, N., Preece, R., McDougal, M., et al. (2009). Personal protective equipment and risk for avian influenza (A7N3). Emerging Infectious Diseases , 15 (1), 59-62.

Muller, M., & McGeer, A. (2006). Febrile respiratory illness in the intensive care unit setting: An infection control perspective. Current Opinion in Critical Care , 12 (1), 37-42.

National Advisory Committee on SARS and Public Health. (2003). Learning from SARS - Renewal of Public Health in Canada. Retrieved June 11, 2009, from Public Health Agency of Canada: http://www.phac-aspc.gc.ca/publicat/sars-sras/naylor/

Nichol, K., Bigelow, P., O'Brien-Pallas, L., McGeer, A., Manno, M., & Holness, L. (2008). The individual, environmental and organizational factors that influence nurses' use of facial protection to prevent occupational transmission of communicable respiratory illness in acute care hospitals. American Journal of Infection Control , 36 (7), 481-7.

Nicholson, K. (1998). Human influenza. In K. Nicholson, R. Webster, & A. Hay, Textbook of Influenza (pp. 219-64). London: Blackwell Science.

Nickell, L., Crighton, E., Tracy, C., Al-Enazy, H., bolaji, H., Hanjrah, S., et al. (2004). Psychosocial effects of SARS on hospital staff: Survey of a large tertiary care institution. Journal of the Canadian Medical Association , 170 (5), 793-798.

Nicklin, W., & Graves, E. (2005). Nursing and patient outcomes: It's time for healthcare leadership to respond. Healthcare Management Forum , 18 (1), 9-13, 40-5.

Nunnaly, J. (1978). Psychometric Theory. New York, NY: McGraw-Hill.

O‘Boyle, C., Henly, S., & Larson, E. (2001). Understanding adherence to hand hygiene recommendations: The theory of planned behaviour. American Journal of Infection Control , 29, 352-360.

124

O'Brien-Pallas, L., Shamian, J., Thomson, D., Alksnis, C., Koehoorn, M., Kerr, M., et al. (2004). Work-related disability in Canadian nurses. Journal of Nursing Scholarship , 36 (4), 352-7.

Occupational Safety and Health Administration. (2007). Guidance on preparing workplaces for an influenza pandemic. Retrieved August 5, 2009, from Occupational Safety and Health Administration: http://www.osha.gov/Publications/influenza_pandemic.html#spread_between_people

Olofsson, S., Kumlin, U., Dimock, K., & Arnberg, N. (2005). Avian influenza and sialic acid receptors: More than meets the eye? Lancet Infectious Diseases , 5, 184-8.

Ontario Ministry of Health and Long-Term Care. (2003). Emergency Infection Control Kit: Donning and Removal of Personal Protective Equipment. Retrieved July 30, 2009, from Ontario Ministry of Health and Long-Term Care: http://www.health.gov.on.ca/english/providers/program/emu/emerg_kit/kit_donning.html

Ontario Ministry of Health and Long-Term Care. (2006, February). Hand Hygiene Observation Tool. Retrieved September 11, 2009, from Just Clean Your Hands: http://www.justcleanyourhands.ca/pdf/9_5_Observation_Tool_19Feb08.pdf

Ontario Ministry of Health and Long-Term Care. (2007, April). Infection Prevention and Control Core Competency Education for Acute Care: Routine practices module. Retrieved August 5, 2009, from Ontario Ministry of Health and Long-Term Care: http://www.health.gov.on.ca/english/providers/program/infectious/routine_practices/scorm/Routi nePractices.html

Ontario Ministry of Health and Long-Term Care. (2006). Operational Plan During an Outbreak of Avian Influenza in the Domestic Poultry Operation(s). Retrieved June 11, 2009, from Ontario Ministry of Health and Long-Term Care: http://www.health.gov.on.ca/english/public/program/emu/avian/avian_op_plan.pdf

Ontario Ministry of Health and Long-Term Care. (2006). Preventing febrile respiratory illnesses: Protecting patients and staff (Revised Edition). Retrieved August 5, 2009, from Ontario Ministry of Health and Long-Term Care:

125 http://www.health.gov.on.ca/english/providers/program/infectious/diseases/best_prac/bp_fri_080 406.pdf

Ontario Ministry of Health and Long-Term Care. (2003). Preventing Respiratory Illnesses, Protecting Patients and Staff - Recommended Infection Control and Surveillance Standards. Retrieved June 11, 2009, from Ministry of Health and Long-Term Care: http://www.health.gov.on.ca/english/providers/program/pubhealth/sars/docs/docs2/dir_122303_a cute_care_nonoutbreak.pdf

Osborne, S. (2003). Influences on compliance with standard precautions among operating room nurses. American Journal of Infection Control , 31 (7), 415-23.

Park, B., Peck, A., Kuehnert, M., Newbern, C., Smelser, C., Comer, J., et al. (2004). Lack of SARS transmission among healthcare workers, United States. Emerging Infectious Diseases , 10 (2), 244-248.

Pashman, J., Bradley, E., Wang, H., Higa, B., Fu, M., & Dembry, L. (2007). Promotion of hand hygiene techniques through use of a surveillance tool. Journal of Hospital Infection , 66 (3), 249- 254.

Peduzzi, P., Concato, J., Kemper, E., Holford, T., & Feinstein, A. (1996). A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology , 49, 1373-9.

Perez-Padilla, R., de la Rosa-Zamboni, D., Ponce de Leon, S., Hernandez, M., Quiñones-Falconi, F., Bautista, E., et al. (2009). Pneumonia and respiratory failure from swine-origin Influenza A (H1N1) in Mexico. New England Journal of Medicine , 361, 1-10.

Pittet, D. (2001). Improving adherence to hand hygiene practice: A multidisciplinary approach. Emerging Infectious Diseases , 7 (2), 234-240.

Poutanen, S., Vearncombe, M., McGeer, A., Gardam, M., Large, G., & Simor, A. (2005). Nosocomial acquisition of methicillin-resistant staphlococcus aureau during an outbreak of severe acute respiratory syndrome. Infection Control and Hospital Epidemiology , 26 (2), 134-7.

126

Prentice-Dunn, S., & Rogers, R. (1986). Protection motivation theory and preventive health: Beyond the Health Belief Model. Health Education Research , 1 (3), 153-161.

Prochaska, J., & DiClemente, C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice , 20, 161-173.

Prochaska, J., Redding, C., & Evers, K. (2002). The transtheoreticalmodel and stages of change. In K. Glanz, B. Rimer, & F. Lewis, Health Behavior and Health Education: Theory, Research and Practice (Third Edition) (pp. 99-120). San Francisco: Jossey-Bass.

Pronovost, P., Weast, B., Holzmueller, C., Rosenstein, B., Kidwell, R., Haller, K., et al. (2003). Evaluation of the culture of safety: Survey of clinicians and managers in an academic medical center. Quality and Safety in Health Care , 12 (6), 405-10.

Provincial Infectious Diseases Advisory Committee. (2006). Preventing Febrile Respiratory Illnesses: Protecting Patients and Staff - Best Practices in Surveillance and Infection Prevention and Control for Febrile Respiratory Illness (FRI), excluding Tuberculosis, for All Ontario Health Care Settings. Retrieved June 11, 2009, from Ontario Ministry of Health and Long-Term Care: http://www.health.gov.on.ca/english/providers/program/infectious/diseases/best_prac/bp_fri_080 406.pdf

Public Health Agency of Canada. (2007). Canadian Tuberculosis Standards - 6th Edition. (R. Long, & R. Ellis, Eds.) Retrieved June 11, 2009, from Public Health Agency of Canada: http://www.phac-aspc.gc.ca/tbpc-latb/pubs/pdf/tbstand07_e.pdf

Public Health Agency of Canada. (2006). Human Health Issues related to Avian Influenza in Canada . Retrieved June 11, 2009, from http://www.phac-aspc.gc.ca/publicat/daio-enia/1- eng.php#jmp-lan1

Public Health Agency of Canada. (2006, October 25). Human Health Issues related to Avian Influenza in Canada. Retrieved June 11, 2009, from Public Health Agency of Canada: http://www.phac-aspc.gc.ca/publicat/daio-enia/pdf/nat-ai-guide-2006_e.pdf

127

Public Health Agency of Canada. (2009, May 11). Interim Guidance: Infection prevention and control measures for Health Care Workers in Acute Care Facilities - Human Cases of H1N1 Flu Virus (Human Swine Flu). Retrieved June 11, 2009, from Public Health Agency of Canada: http://www.phac-aspc.gc.ca/alert-alerte/swine-porcine/pdf/interim_guidance_infection_control- eng.pdf

Public Health Agency of Canada. (2002, March). Prevention and Control of Occupational Infections in Healthcare. Retrieved December 2, 2009, from Public Health Agency of Canada: http://www.phac-aspc.gc.ca/publicat/ccdr-rmtc/02vol28/28s1/index.html

Public Health Agency of Canada. (1999). Routine Practices and Additional Precautions for Preventing the Transmission of Infection in Health Care - Revision of Isolation and Precaution Techniques. Retrieved June 11, 2009, from Public Health Agency of Canada: http://www.phac- aspc.gc.ca/publicat/ccdr-rmtc/99vol25/25s4/

Radonovich, L., Yanke, R., Cheng, J., & Bender, B. (2010). Diminished speech intelligibility associated with certain types of respirators worn by healthcare workers. Journal of Occupational and Environmental Hygiene , 7, 63-70.

Randle, J., Clarke, M., & Storr, J. (2006). Hand hygiene compliance in healthcare workers. Journal of Hospital Infection , 64 (3), 205-9.

Raymond, D., & Lusk, L. (2006). Staging workers' use of hearing protection devices: Application of the transtheoretical model. Journal of the American Association of Occupational Health Nurses , 54 (4), 165-172.

Registered Nurses Association of Ontario. (2004, June). SARS Unmasked: Final Report on the Nursing Experience with SARS in Ontario. Retrieved July 30, 2009, from Registered Nurses Association of Ontario: http://www.rnao.org/Storage/24/1891_SARS_Report_June_04.pdf

Rethans, J., Sturmans, F., Drop, R., Van der Vleutens, C., & Hobus, P. (1991). Does competence of general practitioners predict their performance? Comparison between examination setting and actual practice. British Medical Journal , 303, 1377-80.

128

Rickard, N. (2004). Hand hygiene: Promoting compliance among nurses and health workers. British Journal of Nursing , 13 (7), 404-10.

Rivers, D., Aday, L., Frankowski, R., Felknor, S., White, D., & Nichols, B. (2003). Predictors of nurses' acceptance of an intravenous catheter safety device. Nursing Research , 52 (4), 249-255.

Rogers, A., Hwang, W., Scott, L., Aiken, L., & Dinges, D. (2004). The working hours of hospital staff nurses and patient safety. Health Affairs , 23 (4), 202-212.

Rogers, R. (1983). Cognitive and psychological processes in fear appeals and attitude change: A revised theory of protection motivation. In J. Cacioppo, & R. Petty, Social Psychophysiology (pp. 153-176). New York: Guilford Press.

Salmela, S., Poskiparta, M., Kasila, K., Vahasarja, K., & Vanhala, M. (2009). Transtheoretical model-based dietary interventions in primary care: A review of the evidence in diabetes. Health Education Research , 24 (2), 237-52.

SAS Institute, I. (1999). SAS Software Version 8.0. Cary, North Carolina: SAS Institute, Inc.

SAS Institute, Inc. (2005). SAS Software Version 9.1. Cary, NC: SAS Institute, Inc.

Seto, W. (1995). Staff compliance with infection control practices: Application of behavioural sciences. Journal of Hospital Infection , 30 (Supp), 107-15.

Sexton, B., Thomas, E., & Helmreich, R. (2000). Error, stress and teamwork in medicine and aviation: Cross sectional surveys. British Medical Journal , 320 (7237), 745-9.

Shigayeva, A., Green, K., Raboud, J., Henry, B., Simor, A., Vearncombe, M., et al. (2007). Factors associated with critical-care healthcare workers' adherence to recommended barrier precautins during the Toronto severe acute respiratory syndrome outbreak. Infection Control and Hospital Epidemiology , 28 (11), 1275-1283.

Skowronski, D., Li, Y., Tweed, S., Tam, T., Petric, M., David, S., et al. (2007). Protective measures and human antibody response during an avian influenza H7N3 outbreak in poultry in British Columbia, Canada. Canadian Medical Association Journal , 176 (1), 47-53.

129

Southgate, L., & Dauphinee, D. (1998). Maintaining standards in British and Canadian medicine: The developing role of the regulatory body. British Medical Journal , 316, 697-700.

Spence Laschinger, H., & Leiter, M. (2006). The impact of nursing work environments on patient safety outcomes. The Journal of Nursing Administrtion , 36 (5), 259-67.

Srinivasan, A., McDonald, L., Jernigan, D., Helfand, R., Ginsheimer, K., Jernigan, J., et al. (2004). Foundations of the Severe Acute Respiratory Syndrome preparedness and response plan for healthcare facilities. Infection Control and Hospital Epidemiology , 25 (12), 1020-5.

Steiner, M., Vermeulen, L., Mullahy, J., & Hayney, M. (2002). Factors influencing decisions regarding influenza vaccination and treatment: A survey of healthcare workers. Infection Control and Hospital Epidemiology , 23 (10), 625-27.

Stirling, B. (2004). Nurses and the control of infectious disease: Understanding epidimiology and disease transmission is vital to nursing care. The Canadian Nurse , 100 (9), 16-21.

Sutton, P., Nicas, M., & Harrison, R. (2000). Tuberculosis isolation: Comparison of written procedures and actual practices in three California hospitals. Infection Control and Hospital Epidemiology , 21 (1), 28-32.

Sutton, P., Nicas, M., Reinisch, F., & Harrison, R. (1998). Evaluating the control of tuberculosis among healthcare workers: Adherence to CDC guidelines of three urban hospitals in California. Infection Control and Hospital Epidemiology , 19 (7), 487-93.

Tang, C., & Wong, C. (2004). Factors influencing the wearing of facemasks to prevent the severe acute respiratory syndrome among adult Chinese in Hong Kong. Preventive Medicine , 39, 1187-1193.

Taub, S., Morin, K., Goldrich, M., Ray, P., & Benjamin, R. (2006). Physician health and wellness. Occupational Medicine , 56 (2), 77-82.

Tellier, R. (2006). Review of aerosol transmission of influenza A virus. Emerging Infectious Diseases , 12, 1657-62.

130

The Lung Association. (2009). Tuberculosis: Information for Health Care Providers - 4th Edition. Retrieved June 11, 2009, from The Lung Association: https://lung.healthdiary.ca/Guest/Product.aspx?IDS=yaQAZ%2F8w5Dph%2FsSJ8eILtw%3D%3 D

The SARS Commission. (2006). Spring of Fear, Volumes 1-3. Toronto, Canada: Ontario Ministry of Health and Long-Term Care.

Thorne, C., Khozin, S., & McDiarmid, M. (2004). Using the hierarchy of control technologies to improve healthcare facility infection control: Lessons from Severe Acute Respiratory Syndrome. Journal of Occupational and Environmental Medicine , 46 (7), 613-22.

Tokars, J., McKinley, G., Otten, J., Woodley, C., Sordillo, E., Caldwell, J., et al. (2001). Use and efficacy of tuberculosis infection control practices at hospitals with previous outbreaks of multidrug-resistant tuberculosis. Infection Control and Hospital Epidemiology , 22 (7), 449-455.

Turnberg, W., & Daniell, W. (2008). Evaluation of a healthcare safety climate measurement tool. Journal of Safety Research , 39, 563-568.

Turnberg, W., Daniell, W., Seizas, N., Simpson, T., Van Buren, J., Lipkin, E., et al. (2008). Appraisal of recommended respiratory infection control practices in primary care and emergency department settings. American Journal of Infection control , 36, 268-75.

Varon, U., & Mattila, M. (2000). The safety climate and its relationship to safety practices, safety of the work environment and occupational accidents in eight wood processing plants. Accident Analysis and Prevention , 32, 761-9.

Vaughn, T., McCoy, K., Beekmann, S., Woolson, R., Torner, J., & Doebbeling, B. (2004). Factors promoting consistent adherence to safe needle precautions among hospital workers. Infection Control and Hospital Epidemiology , 25 (7), 548-55.

Waqqas, A., Huor, P., Brassard, P., & Loo, V. (2002). Compliance with methicillin-resistant Staphylococcus aureus precautions in a teaching hospital. American Journal of Infection Control , 30 (7), 430-433.

Weinstein, N. (1988). The precaution adoption process. Health Psychology , 7, 355-386.

131

Weinstein, N., & Sandman, P. (1992). A model of the precaution adoption process: Evidence from home radon testing. Health Psychology , 11 (3), 170-80.

Weinstein, N., Lyon, J., Sandman, P., & Cuite, C. (1998). Experimental evidence for stages of health behavior change: The precaution adoption process model applied to home radon testing. Health Psychology , 17 (5), 445-453.

While, A. (1994). Competence versus performance: Which is more important. Journal of Advanced Nursing , 20, 525-531.

Whitby, M., McLaws, M., & Ross, M. (2006). Why healthcare workers don‘t wash their hands: A behavioral explanation. Infection Control and Hospital Epidemiology , 27 (5), 484-492.

Wicker, S., Rabenau, H., Bickel, M., Wolf, T., Brodt, R., Brandt, C., et al. (2009). Novel influenza H1N1/2009: virus transmission among health care worker. Deutsche Medizinische Wochenschrift , 134 (48), 2443-6.

Wicker, S., Rabenau, H., Doerr, H., & Allwinn, R. (2008). Influenza vaccination compliance among health care workers in a german university hospital. Infection , epub ahead of print.

Workplace Safety and Insurance Board. (2008, October 20). News Release - You Can Never Be Too Safe. Retrieved June 17, 2009, from Workplace Safety and Insurance Board: http://www.wsib.on.ca/wsib/wsibobj.nsf/LookupFiles/NewsReleaseWorkplaceSafetyAwareness/ $File/CoreCampaign2008.pdf

World Health Organization. (2004). Avian influenza A(H7) human infections in Canada. Retrieved June 11, 2009, from http://www.who.int/csr/don/2004_04_05/en/index.html

World Health Organization. (2003). Summary of probable SARS cases with onset of illness from 1 November 2002 to 31 July 2003 . Retrieved June 11, 2009, from http://www.who.int/csr/sars/country/table2004_04_21/en/index.html

World Health Organization. (2006). Tools for Evaluation and Feedback. Retrieved September 11, 2009, from World Health Organization: http://www.who.int/gpsc/5may/tools/evaluation_feedback/en/index.html

132

Yap, F., Gomersall, C., Fung, K., Ho, P., Ho, O., Lam, P., et al. (2004). Increase in methicillin- resistant Staphylococcus aureus acquisition rate and change in pathogen pattern associated with an outbreak of severe acute respiratory syndrome. Clinical Infectious Diseases , 39 (4), 511-6.

Yassi, A., & Hancock, T. (2005). Patient safety - worker safety: Building a culture of safety to improve healthcare worker and patient well-being. Healthcare Quarterly , 8 (Special Issue), 32- 38.

Yassi, A., Bryce, E., Moore, D., Janssen, R., Copes, R., Bartlett, K., et al. (2004). Protecting the faces of health care workers: knowledge gaps and research priorities. A Report to the Change Foundation. . Vancouver, British Columbia, Canada: Occupational Health and Safety Agency for Healthcare.

Yassi, A., Lockhart, K., Copes, R., Kerr, M., Corbiere, M., Bryce, E., et al. (2007). Determinants of healthcare workers' compliance with infection control procedures. Healthcare Quarterley , 10 (1), 44-52.

Yassi, A., Moore, D., Fitzgerald, J., Bigelow, P., Hon, C., & Bryce, E. (2005). Research Gaps in Protecting Healthcare Workers From SARS and Other Respiratory Pathogens: An Interdisciplinary, Multi-Stakeholder, Evidence-Based Approach. Journal of Occupational and Environmental Medicine , 47 (1), 41-50.

Zohar, D. (1980). Safety climate in industrial organizations: Theoretical and applied implications. Journal of Applied Psychology , 65 (1), 96-102.

133

Appendices

Appendix 1 - Annotated Bibliography of Adherence Literature

Appendix 2 - Published Manuscript of the Pilot Study

Appendix 3 - Two Page Proposal Summary

Appendix 4 - Letter of Information

Appendix 5 - Letter of Information and Consent

Appendix 6 - Facial Protection Questionnaire

Appendix 7 - Unit Observation Record

Appendix 8 - Participant Observation Record

Appendix 9 - Participant Observation Guide

Appendix 10: Case Definitions for Explanatory Variables

134

Appendix 1 - Annotated Bibliography of Adherence Literature

Type of Tool Author Study Year Title Model Sample Variables Intervention Measures Findings Available Demographics, compliance with double gloving Compliance Yes - and eye rate, Standard Influences protection, HBM influences on Compliance rate 72.1%. Precautions on 227/500 scales compliance Some demographics and Descriptive, compliance Australian (perception of with UP and influenced compliance. Occupational correlational with Health operating risk, severity, occupational Perception of risk, benefits Exposure Osborne, design - mail standard Belief room benefits, exposure and barriers influenced Reporting Sonya out survey 2003 precautions Model nurses barriers) reporting compliance. Survey Demographics, employee Influences personal on characteristics, Job/task and organizational Descriptive, compliance 451/1100 job/task factors, Compliance factors were best predictors correlational with nurses from organizational rate, of compliance. Proposed Yes - Health design - mail standard PRECED one large factors, predictors of new model of adherence and Care Worker DeJoy, D. out survey* 1995 precautions E Model US hospital adherence to UP compliance intervention strategy. Questionnaire* Compliance rate 31 - 38%. Predictors of compliance include knowledge, UP Demographics, training, perception of employee effectiveness, organizational Influences personal commitment. Predictors of on characteristics, non-compliance include older Descriptive, compliance 322 job/task factors, Compliance age, high work stress and correlational with physicians/ organizational rate, perceived conflict of interest Yes - Health design - mail standard PRECED 1716 US factors, predictors of between patient and worker Care Worker Michalsen, A. out survey* 1995 precautions E Model HCWs adherence to UP compliance safety. Questionnaire* Work Systems Minimization of risk requires a Analysis of work systems approach that Compliance considers individual, job/task with UP and among environmental/organizational DeJoy, D. Commentary 1996 HCWs factors. n/a

135

Predisposing, enabling and reinforcing factors influenced Demographics, Compliance general compliance. employee rate, Enabling and reinforcing Influences personal predictors of factors influenced compliance on characteristics, compliance with PPE. Indirect effects of Descriptive, compliance job/task factors, with general +ve safety climate on correlational with 902 organizational UP and with enabling work environment Yes - Health design - mail standard PRECED nurses/171 factors, PPE predicted compliance the Care Worker DeJoy, D. out survey* 2000 precautions E Model 6 US HCWs adherence to UP specifically best. Questionnaire*

Safety Review of climate - two studies assessming (HCW and manageme retail nt and workers) organizatio looking at nal HCW (see safety influences PRECED above) and DeJoy, D. climate 2004 on safety E Model retail n/a Compliance varies from high (glove, sharps disposal) to low (protective clothing, eye Demographics, protection). Compliance employee strongly correlated with org Influences personal commitment, perceived on characteristics, conflict of interest, risk taking Descriptive, compliance 1716 /3000 job/task factors, Compliance tendencies, perception of correlational with US HCWs organizational rate and risk, knowledge and training. Yes - Health design - mail standard PRECED from 3 large factors, correlates of Also r/t being female, nurse Care Worker Gershon, R. out survey* 1995 precautions E Model hospitals adherence to UP compliance vs. physician. Questionnaire* Demographics, Factors associated with employee compliance include longer personal tenure in job, increased characteristics, knowledge of HIV, Secondary factors job/task factors, Compliance conservative attitute toward analysis of affecting 606 HCWs organizational rate and risky behaviour, perception of Yes - Health McGovern, Gershon, UP PRECED at one large factors, correlates of strong safety climate and Care Worker P. 1995 2000 compliance E Model US hospital adherence to UP compliance training in PPE. Questionnaire* hospital safety climate and 20-item hospital safety its climate scale. Sr. mgmt relationship support, absence of with safe workplace barriers and work demographics, correlates of cleanliness of workplace cross- practices compliance, compliance, significantly related to sectional and 789/1240 exposure correlates of compliance. Sr. mgmt yes - 20-item descriptive workplace PRECED HCWs one incident history, exposure support and training/feedback hospital safety Gershon, R. study 2000 exposures E Model US hospital safety climate incidents related to exposures. climate scale

136

Descriptive and Knowledge, observationa observed l design - compliance Knowledge rate high but questionnair rate, doesn't correlate with e (assess correlates of compliance. Most common knowledge) To assess 30 Demographics, compliance, reasons for non-compliance and knowledge residents knowledge, common were time, inconvenience and observation of and and 31 US compliance, reasons for presumption pt not infected. (adherence compliance med reasons for non- non- Correlates with compliance Helfgott, A. with UP) 1998 with UP students compliance compliance include less experience. No TPB scale (intention, attitude, Theory of subjective norm, Planned perceived Behaviou control, 38% compliance for 8/10 Determinan r, Theory perceived belief, venipunctures, 16% for ts of of role belief) 10/10. Perception of control Descriptive, adherence Interpers 105/238 completed at key factor in compliance. corrrelational to UP for onal rural baseline, compliance, Non-compliance r/t training design - mail venipunctur Behaviou Quebec compliance at 3 correlates of and sub-optimal working Godin, G. out survey 2000 e r nurses months later compliance conditions No Using force Using PPE is a clinical field decision with ethical analysis to dimensions. Behaviour is a promote function of ind'l and use of PPE environmental factors. FFA as part of is an analysis of the driving IC and restraining factors that White, C.M. Commentary 1992 procedures influence behaviour. n/a 4-stage framework for workplace self-protective behaviour is proposed: hazard appraisal, decision making, initiation and Theoretical Value- adherence. 5 factors are of Models of expectan primary or secondary Health cy, importance during each Behaviour environm stage: threat-related beliefs, and ent/conte response efficacy, self- Workplace xtual and efficacy, facilitating conditions self- behaviou and safety climate. Each Literature protective r change stage should have specific DeJoy, D. Review 1996 behaviour models interventions. n/a

137

PMT scales (perceived susceptibility, influences severity, Knowledge predicted Case control of benefits, barriers compliance. Perceived study - knowledge, Health 61 hospital and self susceptibility and perceived questionnair attitudes Belief (exposed) efficacy), benefit predicted compliant. e, random and health Model, and compliance, Older nurses more observation beliefs on Protectio community barriers, knowledge, compliance. Control group to validate safe use of n (control) knowledge, life compliance, more strictly compliant. self report of cytotoxic Motivatio nurses in habits, correlates of Perceived barriers predicted Ben-Ami, S. compliance 2001 drugs n Theory Isreal demographics compliance non-compliance. No Review of respiratory Infectious pathogens that are known to respiratory cause nosocomial outbreaks illnesses - influenza, pertussis, and their tuberculosis, SARS. Include impact of incidence, transmission and Low, J. Review 2005 HCWs prevention.

Improving Barriers to HH, risk factors for Literature adherence noncompliance, strategies Pittet, D. Review 2001 to HH proposed to promote HH n/a HCWs caring for 30 children During 30% of visits door with remained open for 10 min. confirmed During 20% of visits no resp or protection was worn. HCWs IC suspected more likely to wear RP when intervention TB were TB confirmed. 2/3 of non- s to prevent observed Clinical HCW failed to wear proper TB in for 242 characteristic RP. 1/2 of patients failed to Observation pediatric hours (656 Adherence with s of patients, wear masks when leaving Kellerman, S. al study 2001 facilities visits) IC practices compliance room n/a

541 observation s of HCWs HCW caring for compliance 52 patients with TB in Observation control respiratory Adherence with 64/541 violations - half by LoBue, P. al study 1999 policies isolation IC practices Compliance physicians in training n/a

138

Literature reivew of current efforts being made to establish links between health care workers and patient outcomes. Focus is on worker H&S concerns that are affected by the organization of work and physical environment. Overall, studies provide Org and evidence of direct effects on env factors work performance and that affect suggest indirect effects on worker H&S quality of patient care. Lundstrom, Literature and patient Magnet Strongest links found with T. Review 2002 outcomes Hospitals infections. n/a High risk perception Perception of significantly related to low risk, perception of counter perception of measures at the institutional counter level. Nurses and docs had 7282/9978 measures, different knowledge of HCWs from Perception of knowledge, preventive measures, SARS risk 7 tertiary risk and concept of concept of institutional Descriptive perceptions hospitals in countermeasure insititutional measures and perception of Imai, T. survey 2005 in HCWs Japan s for SARS measures risk. no All hospital inpatients who had active TB or were placed in TB isolation and HCWs 8.6% patients admitted to Use and assigned to patient info, resp isolation had TB. 19% efficacy of wards characteristics of implementati with TB weren't isolated on TB IC where TB isolation rooms, on of TB first day. Most isolation practices at patients HCW/patient TB control rooms were neg pressure. hospitals were skin test results, measures, 44 - 97% of ppl entering Prospective, with treated - 2 HCW use of pt/HCW isolation rooms wore observationa previous US respiratory seroconversi appropriate respirator. No Tokars, J. l study 2001 outbreaks hospitals protection on evidence of transmission. n/a

139

demographics, medical history, influenza 1718 exposure Factors vaccinated potential and influencing HCWs and reasons for vaccination 62% vaccination rate, decisions 482 non- receiving/non rate, factors avoiding lost work r/t regarding vaccinated receiving influencing vaccination, low risk r/t non- influenza HCWs from vaccine, decision, vaccination. 1/3 would refuse Descriptive vaccination 1 hospital in willingness to willingness to vaccination if had to pay Steiner, M. survey 2002 and tx US pay pay $10+. no profession, time of day, time of week, gender, location in hospital, nurse/patient 488 ratio, proportion observation of patients on compliance 28% compliance rate. Compliance s of HCWs precautions on rate, Compliance lowest among Observation with MRSA in hospital ward, correlates of physicians, highest among Afif, W. al study 2002 precautions in Montreal precautions compliance Ots/PTs. n/a

DeJoy's compliance 34% compliance, lowest for behaviou rate, eyewear, face mask and ral 216/350 FT correlates of protective clothing. diagnosti health care compliance, Demographic factors not c self- employees demographics, determine predictive of compliance cross protectiv from psychosocial relationship (except lower age). Safety sectional Compliance e model Marylands' factors, work- between climate and job satisfaction study - with UP in (PRECE 29 related factors, compliance predictive of compliance.14% questionnair correctional DE correctional compliance with and had exposures. Compliance Gershon, R. e 1999 HC facilities model) facilities Ups exposures inversely r/t exposures. Yes perception of demographics, protective perception of measures by controversy over what was risk, perception occupation, protective or not. Job title, PPE use by of protective predictors of reported contact with SARS cross HCWs 10236/1455 measures, belief in patients, area of work and sectional during 4 HCWs in impact of SARS protective IES scores predicted for study - SARS 9 HC outbreak on measures agrrment that paper and/or questionnair outbreak in settings in work and home (facial surgical masks were Chia, S. e 2004 Singapore Singapore life protection) protection against SARS. Yes

140

304 observation 34% major breaks of invasive s of procedures - most common invasive was failure to wear mask then Compliance procedures, eye protection. 55% minor with UP in 752 most breaks in patient encounters. Observation Emergency observation minor and major common Surgeons least likely to use al study - Dept. s of patient breaks in barrier minor and precautions - residents most Evanoff, B. videotapes 1999 personnel encounters precautions major breaks likely. n/a occupation and SARS outbreak had work history, significant psychosocial concerns about effects on hospital staff. SARS, use and Effects differed wrt effects of SARS psychosocial occupation and risk precautionary effects of perception. Effect on families 2001/4283 measures, SARS and lifestyle substantial. employees sociodemographi outbreak, Masks reported most cross Psychosoci at c char's and correlates of bothersome IC precaution. sectional al effects of Sunnybrook GHQ-12 effects, most Emotional distress tied to study - SARS on and (measure of problematic reduced ability to do one's job questionnair tertiary level Women's 3 recent emotional precautionar due to precautionary Nickell, L. e 2004 HCWs sites distress) y measures measures. no Predictors of nurses'acce cross ptance of acceptance - sectional an IV 649/742 Organizational satisfied, training and safety climate study - catheter Safety nurses at (safety climate, recommend predictive of acceptance. questionnair safety Climate one US training) and to others, Shorter tenure at hospital Rivers, D. e 2003 device Model hospital Individual compliance predictive of acceptance. no

charge nurse, informal tape- leaders, students, IC nurses, recorded type of work, availability of factors interviews equipment, blood-exposure promoting Grounde with 9 incidents and media- HCW d Theory nurses, 6 coverage of IDs are qualitative compliance Approac nursing described as potentially Lymer, U. interviews 2004 with UP h assistants important for compliance

141

110 HCWs with exposure within droplet demographics, range to 6 PPE use, SARS exposure patients - characteristics, 45 no patient events, lack of mask, 72 presence during SARS no eye medical transmissio protection, procedures, Descriptive n among 40 skin-to- clinical S&S, transmission Park, B. cohort study 2004 US HCWs skin serum samples of SARS no transmission of SARS no

Knowledge gaps and research priorities for effective protection against occ OHSAH/ Literature acquired nosocomial spread of SARS Change Review and resp PRECED mostly due to failure to use Foundation focus groups 2004 disease E Model barrier precautions n/a 1. characterize SARS-related behaviour intentions of HCWs regarding compliance with protective measure and ability and willingness to treat SARS patients and to accept quarantine 2. explore environmental, organizational and individual determinants barriers and of HCW intentions by cross facilitators conducting focus groups of sectional to front line HCWs and ind'ls study - implementin responsible for implementing questionnair g protective control measures 3. survey e, focus measures 1875 HCWs nurses 4. audit workplaces groups, unpu against in BC, 750 organizational to assess organizational and hospital blish SARS for PRECED in Ontario factors - safety environmental ability to Yassi, A. audits ed HCWs E Model (Total 2625) climate, SARS, prevent future outbreaks. yes

142

Appendix 2 - Published Manuscript of the Pilot Study The individual, environmental, and organizational factors that influence nurses‘ use of facial protection to prevent occupational transmission of communicable respiratory illness in acute care hospitals

Kathryn Nichol, MHSc,a Philip Bigelow, PhD,b Linda O‘Brien-Pallas, PhD,c Allison McGeer, MD,d Mike Manno, MSc,a and D. Linn Holness, MDa Toronto, Canada

From the Centre for Research Expertise in Occupational Disease,a University of Toronto and St. Michael‘s Hospital; Institute for Work and Healthb; Faculty of Nursing,c University of Toronto; Mount Sinai Hospital,d Toronto, Canada. Address correspondence to Kathryn Nichol, MHSc, Centre for Research Expertise in Occupational Disease, c/o Department of Occupational and Environmental Health, St. Michael‘s Hospital, 30 Bond St, Toronto, ON M5B 1W8, Canada. E-mail: [email protected]. Disclosures: none. 0196-6553/$34.00 Copyright ª 2008 by the Association for Professionals in Infection Control & Epidemiology, Inc. doi:10.1016/j.ajic.2007.12.004

Abstract

Background: Communicable respiratory illness is an important cause of morbidity among nurses. One of the key reasons for occupational transmission of this illness is the failure to implement appropriate barrier precautions, particularly facial protection. The objectives of this study were to describe the factors that influence nurses‘ decisions to use facial protection and to determine their relative importance in predicting compliance.

Methods: This cross-sectional survey was conducted in 9 units of 2 urban hospitals in which nursing staff regularly use facial protection.

Results: A total of 400 self-administered questionnaires were provided to nurses, and 177 were returned (44% response rate). Less than half of respondents reported compliance with the recommended use of facial protection (eye/face protection, respirators, and surgical masks) to prevent occupational transmission of communicable respiratory disease. Multivariate analysis showed 5 factors to be key predictors of nurses‘ compliance with the recommended use of facial protection. These factors include full-time work status, greater than 5 years tenure as a nurse, at least monthly use of facial protection, a belief that media coverage of infectious diseases impacts risk perception and work practices, and organizational support for health and safety.

Conclusion: Strategies and interventions based on these findings should result in enhanced compliance with facial protection and, ultimately, a reduction in occupational transmission of communicable respiratory illness. (Am J Infect Control 2008;36:481-7.)

Communicable respiratory illness is an important cause of morbidity among health care workers each year. In Ontario, this problem gained international recognition with the outbreak of severe acute respiratory syndrome (SARS) in Toronto in 2003. In Toronto, SARS resulted in the deaths of 3 health care workers and ongoing morbidity in many others. In a review of the scientific literature on the efficacy of personal protective equipment to prevent the transmission of SARS, it was determined that failure to implement appropriate barrier precautions was one of the key reasons for occupational transmission of communicable respiratory disease.1 Facial protection (respirators, surgical masks, and eye/face protection) was identified as the personal protective equipment that was least complied with by health care workers, yet it is an important barrier precaution against respiratory illness.2-4 Understanding why health care workers fail to appropriately use facial protection has not been well researched.

The Predisposing, Reinforcing, and Enabling Factors in Educational Diagnosis and Evaluation (PRECEDE) model provides a framework to examine the factors that influence health behavior.5 This model has been adapted for application to self-protective behavior at work.6 Predisposing factors are individual characteristics such as beliefs, attitudes, and values. Enabling factors are environmental factors such as knowledge and availability of equipment and resources. Reinforcing factors are organizational factors such as communication, policies and procedures, and management commitment to health and safety.1

143

The PRECEDE model was used to frame recent literature reviews to identify knowledge gaps and research priorities for effective prevention against occupationally acquired communicable respiratory diseases.1 These reviews categorized factors as organizational, environmental, and individual. Results showed that an abundance of studies have examined factors related to compliance with universal precautions among health care workers,7-10 and the majority have found that environmental and organizational factors played a more important role in compliance than individual factors. Results also showed that there has been very little work done regarding factors that influence compliance of health care workers with precautions taken to protect against communicable respiratory illness. The authors suggest that determinants of adherence to universal precautions are likely applicable to many types of self protective behavior, including those taken to prevent transmission of communicable respiratory illness.1

The objectives of this study were to describe the individual, environmental, and organizational factors that affect nurses‘ compliance with the use of facial protection and to determine the relative importance of these factors. Once the most influential factors are identified, strategies and interventions to enhance compliance can be formulated, tested, and implemented to reduce the prevalence of worker morbidity and mortality from communicable respiratory illness.

METHODS

Study design and study participants

This research study was designed as a preliminary investigation to inform a larger and more comprehensive project. A cross-sectional survey design was used. The study took place in 2 acute care hospitals in Toronto, Canada. The vice president and/or chief nursing officer of each hospital were approached to discuss recruitment of the hospital as a site for study. Ethics approval was obtained from the research ethics boards in each hospital.

Units in which nurses were more likely to require regular use of facial protection were identified by the vice president and/or chief nursing officer of each hospital. Five units from one hospital and 4 units from another agreed to participate. Approximately 500 full- and part-time nurses were employed on the 9 units at the time of the study. Taking summer vacation activity, limited time available to collect data on the units, and the demanding work environment into consideration, the researcher aimed to collect completed surveys from 150 nurses or 30% of the total population. Estimating a 35% to 40% response rate, 400 questionnaires were distributed over a 6-week period during the summer of 2006. The researcher (K.N.) spent 2 to 4 days over different shifts on each unit handing out and collecting surveys. A survey drop box was also placed on each unit.

Survey tool

A new, 5-page, 61-item questionnaire was developed for the study. The questionnaire was divided into 5 parts: demographics, individual factors, compliance, environmental factors, and organizational factors. Part 1 included 7 items that measured basic demographic data. Respondents answered by checking boxes or filling in the blanks. Part 2 examined individual factors that may influence compliance including knowledge of droplet and airborne spread respiratory disease (8 items), perception of effectiveness of preventive actions (4 items), exposure history (2 items), perception of occupational risk (3 items), and personal barriers to the use of facial protection (13 items). The measure to assess knowledge was developed based on government guidelines on preventing the occupational transmission of airborne and droplet spread communicable disease in health care facilities.11 A response scale of true/false/don‘t know was used. The remaining individual factors were measured using elements and scales from established tools.8,12-14 Most questions used a 5-point Likert response scale of strongly agree/agree/neutral/ disagree/strongly disagree for answers.

Part 3 of the survey examined compliance with recommended use of facial protection. An 8-item scale was developed to measure compliance based on government guidelines.11 Participants were asked about their compliance with N95 respirators, surgical masks, and eye protection when they suspected a patient had a communicable respiratory illness and when a diagnosis had been made. Responses were measured with a 5- point Likert scale of always/mostly/sometimes/rarely/ never. ‗‗Compliant‘‘ was defined as answering always or mostly to at least 7 of the 8 items.

144

Part 4 of the survey examined environmental factors that could influence use of facial protection including training (8 items), cleanliness/orderliness of the workplace (3 items), availability of facial protection (3 items), and media coverage of infectious diseases (2 items). Training measures included frequency and content of the training program, as well as fit testing of respirators. The measures for the training program content were determined using government guidelines,11 and a response scale of yes/no/don‘t know was used. Cleanliness and orderliness of the workplace and availability of facial protective equipment were measured using elements from established scales.15,16 Measures related to media coverage were based on a qualitative study that examined the factors that influenced compliance with bloodborne precautions.10 A 5-point Likert response scale of strongly agree/agree/neutral/disagree/ strongly disagree was used.

Organizational factors were measured in part 5 of the survey and included organizational support for health and safety (5 items), absence of job hindrances (3 items), peer and supervisor feedback (4 items), and conflict/communication (3 items). These factors were measured using elements from established scales.15,16 Where necessary, wording was changed to make the statements relevant to preventing the transmission of respiratory illness as opposed to bloodborne illness. A 5-point Likert response scale of strongly agree/agree/neutral/disagree/strongly disagree was used. At the end of the survey, nurses were encouraged to share any further relevant information on the back of the last page.

Pretesting of the survey tool

The questionnaire was reviewed by occupational health and infection control experts on facial protection and airborne and droplet spread illness, senior nursing officials at both hospitals, and one local of the Ontario Nurses Association. Two focus groups were held to test the survey tool. Nurses participating in the focus groups completed the survey and provided feedback on the accuracy and format of the tool and the time it took for completion. As a result of the focus groups, minor changes were made to the tool. The final version of the questionnaire contained 61 items and was completed by most respondents within 15 minutes.

Statistical analysis

All statistical analysis was performed using Statistical Analysis Software (SAS) Version 8.0 (1999; SAS Institute Inc., Cary, NC). Descriptive statistics consisted primarily of frequency distributions and means. Reliability of scales was assessed using Cronbach coefficient a. x2 tests for categorical variables and t tests for continuous variables were conducted to examine the direct effects of each factor on compliance. Variables with a univariate P value , .15 were selected for inclusion in a stepwise logistic regression model to determine those factors with independent predictive value.17 Interpretation was based on odds ratios (OR) where an OR of 1.0 refers to the equal probability of exposure to the factor in nurses classified as compliant and those classified as noncompliant.

RESULTS

Demographics

A total of 177 completed surveys were returned for a response rate of 44%. The sample was predominantly female and young (one third of respondents reported being less than 30 years of age) (Table 1). Most of the respondents were from an intensive care unit (44%), whereas the rest represented inpatient units or outpatient clinics (30%) and emergency (26%).

Compliance

Results showed that 42% of respondents met our definition of compliance and answered always or mostly to at least 7 of the 8 items within the compliance scale. Twenty-one percent of respondents answered always or mostly to all 8 items.

145

Reliability testing

Cronbach coefficient a estimates the reliability of a scale. It is suggested that a score $0.70 is an acceptable reliability coefficient.18 Cronbach a raw scores were calculated for each measure, and 3 scales (knowledge, effectiveness of preventive actions, and perception of risk) had scores less than 0.70. The scale to measure knowledge demonstrated a reliability coefficient of 0.32. Although this low score might be considered problematic for scales measuring a single trait or content domain, a high a statistic cannot be expected for a multi-dimensional knowledge scale such as the one used. The scale to measure perception of risk demonstrated a reliability coefficient of 0.64. When the item measuring risk was separated from those measuring impact, the raw score increased to 0.79, suggesting that risk and impact, need to be measured separately. The scale to measure perception of effectiveness of preventive actions demonstrated a reliability coefficient of 0.50. This could be due to the discrepancy between nurses‘ perception of effectiveness of respirators and hand hygiene as opposed to surgical masks and eye protection. Equipment may need to be evaluated on an individual basis.

Table 1. Demographic characteristics of participating nurses

Variable N = 177, n (%) Sex Female 154 (87.5) Age, yr 20-30 59 (34) 31-40 3 (30) 41-50 47 (27) 511 16 (9) Education Diploma 74 (42) Baccalaureate 89 (51) Master‘s 13 (7) Work status Full-time 148 (84) Supervisory status Yes 75 (43) Tenure as a nurse 11.7 6 9.6 (sample mean, yr) Current job tenure 6.9 6 7.4 (sample mean, yr)

Univariate analysis

Table 2 presents the results of the univariate analysis of associations between explanatory variables and compliance. One demographic variable was found to be significantly associated with compliance: nurses with more tenure reported better compliance with recommended use of facial protection. No individual factors demonstrated a significant association. Of the 4 environmental factors measured, 3 showed a significant association with enhanced compliance: cleanliness, availability of facial protection, and media coverage. Two of the 4 organizational factors assessed in this study expressed a significant association with compliance: organizational support for health and safety and an absence of job hindrances.

146

Table 2. Relationship between compliance with the use of facial protection and each demographic, individual, environmental, and organization factor Noncompliant Compliant Total Variable Description Level n (%) n (%) n (%) Demographic Sex Female 87 (87) 65 (89) 152 (88) Age, yr 20-30 41 (41) 18 (25) 59 (34) 31-40 28 (28) 24 (33) 52 (30) 41-50 24 (24) 22 (31) 46 (27) 51+ 7 (7) 8 (11) 15 (9) Education Diploma 53 (53) 34 (47) 87 (50) Bachelor‘s 40 (40) 33 (45) 73 (42) Master‘s 7 (7) 6 (8) 13 (8) Work status Full-time 81 (81) 65 (89) 146 (84) Supervisory status Yes 43 (44) 30 (42) 73 (43) Tenure as a nurse (yr)* Mean (SD) 9.8 (8.8) 13.8 (9.9) 11.6 (9.5) Job tenure (yr) Mean (SD) 6.1 (7.0) 7.9 (7.8) 7.0 (7.5) Individual Knowledge Knowledgeable 46 (46) 41 (56) 87 (50) Effectiveness of preventive actions Perceived effectiveness 34 (34) 28 (38) 62 (36) Exposure: frequency of use of facial protection ≥Monthly 78 (79) 64 (88) 142 (83) Rarely/never 21 (21) 9 (12) 30 (17) Exposure: personal Reported personal experience experience with exposure 41 (42) 38 (52) 79 (46) Perception of risk Perceived occupational risk 86 (86) 60 (82) 146 (84) Personal barriers to use Reported personal barriers 81 (93) 58 (89) 139 (91) Environmental Cleanliness/orderliness* Reported clean/orderly unit 19 (19) 24 (33) 43 (25) Availability of facial Reported facial protection protection* to be available 39 (39) 40 (56) 79 (46) Training Reported being trained and fit tested in last year 28 (28) 22 (30) 50 (29)

Media coveragey Reported media influenced risk perception and work 23 (23) 35 (48) 58 (34) Organizational Organizational support* Reported organizational support for hlth and safety 27 (28) 29 (41) 56 (34) Absence of job hindrances* Reported absence of job hindrances 34 (35) 38 (53) 72 (43) Feedback Reported peer/supervisor feedback re: hlth and safety 27 (28) 27 (40) 54 (33) Conflict/communication Reported good employee relations and communication 47 (48) 34 (50) 81 (49)

*P ≤.05. yP ≤.001.

147

Multivariate analysis

Significant variables and variables of interest were entered into a stepwise logistic regression model. Five significant predictors of compliance were revealed (Table 3).

Table 3. Adjusted odds ratios for compliance with the use of facial protection Variable Description Level Odds ratio (95% CI) Demographic Work status Part-time vs full-time 0.34 (0.11-0.999) Tenure as a nurse 5-10 yr vs < 5 yr 2.97 (1.05-8.39) 10-20 yr vs <5 yr 3.92 (1.46-10.53) 20+ yr vs <5 yr 4.83 (1.72-13.58) Individual Frequency of use of facial protection ≥Monthly vs rarely/never 2.66 (1.05-6.75) Environmental Media coverage Media coverage important vs not 2.8 (1.26-6.2) Organizational Organizational support Organizational support present vs not 2.37 (1.13-4.97)

DISCUSSION

Demographics

According to the 2005 Canadian Nurses Association RN Workforce Profile by Area of Responsibility, our sample is fairly representative of the population of nurses working in medicine/surgery, critical care, and emergency.19 The Canadian Nurses Association profile indicates that the average age of this population was approximately 41 years, and our sample showed the average age to be 45 years. Regarding sex, the profile showed that approximately 7% of the population was male, and our sample was 4% male. Approximately 55% of the population worked full-time, and 67% of our sample reported full-time work status. Finally, a higher percentage of the population in the profile was trained at the diploma level (approximately 72%), compared with our sample (42% diploma trained).

Compliance

Health care workers‘ compliance with safe work behaviors to prevent the spread of infectious disease is historically poor.3,9 Although adherence with hand hygiene, glove use, and immunization has been well studied, compliance with the use of facial protection to prevent the spread of communicable respiratory disease has not been as well studied. One study of 3 US hospitals over 3 years found that health care workers wore appropriate respiratory protection with tuberculosis patients 44% to 97% of the time.20 Another study showed compliance with respirator use to be 57% when the diagnosis of tuberculosis was unconfirmed and 84% when it was confirmed.21 A third study retrospectively reviewed health care workers who worked during the SARS outbreaks and their compliance with respirator use.22 Sixty-six health care workers reported exposure to a patient who was coughing and later found to be SARS positive, yet 40% of these workers did not use a respirator. 22 Our study of acute care nurses in 2 hospitals in Toronto described a comparable rate, with 42% of responding nurses reporting compliance with the recommended use of facial protection.

Demographic characteristics influencing compliance

Demographic factors such as sex, education level, and occupation have not been found to be consistently associated with compliance with infection control procedures. 7 Our study was consistent with this finding and did not show a relationship between age, education, job tenure, or supervisory status and compliance. Work status was shown to be significantly associated with compliance. Full-time nurses were 3 times more likely to report compliance with the recommended use of facial protection than part-time nurses. This finding is important because part-time nurses may need more intervention than full-time nurses to enhance their level of compliance, and interventions with this group may be logistically more difficult because they spend less time at the workplace. We also found that tenured nurses were more likely to report compliance than new nurses. The literature provides conflicting reports. In a study of correctional workers, it was found that young workers were more likely to be compliant with universal precautions than older workers.8 On the other hand, another study showed that longer job tenure was related to health care worker compliance with universal precautions and suggests that health care workers with more time on the job have

148 had the opportunity to incorporate experience and judgment into their clinical practice, which could promote the use of appropriate preventive behaviors.4 This finding is important because managers could enhance compliance by placing nurses with greater tenure into supervisory and mentoring roles.

Individual factors influencing compliance

Our study found that nurses who reported using facial protection at least monthly were more likely to report compliance than nurses who reported using facial protection rarely or never. Although this relationship may have some inherent contributory bias, it is important to note that nurses who report daily, weekly, or monthly use of facial protection should be the ones to care for patients with communicable respiratory disease or supervise those who are providing the care.

Environmental factors influencing compliance

Cleanliness and orderliness of the workplace were shown to be associated with compliance at the univariate level. In a study on adherence to universal precautions, it was found that cleanliness and orderliness of the worksite were significantly associated with enhanced compliance.15 Ensuring that there are resources to keep the workplace clean, allocating adequate storage space, and ensuring that this storage space is used are important strategies to improve compliance on a unit.

Availability of facial protection was also significantly associated with enhanced compliance at the univariate level. One hospital study in the United States found availability of protective equipment to be associated with compliance with universal precautions. 15 In a study on SARS transmission in health care workers in Hong Kong, it was found that workers who perceived the amount of available personal protective equipment to be poor were more likely to have developed SARS.23 It is possible that, in our study, availability of facial protection was not found to be significant at the multivariate level because of a correlation with the frequency of use of facial protection. Having dedicated and convenient areas to store facial protection and the resources to keep these areas well stocked may assist with compliance.

Media coveragewas shown to be very strongly associated with compliance at both the univariate and multivariate level. Nurses who thought that media coverage of communicable diseases made them more aware of their risk at work and work more carefully were more likely to report compliance with the recommended use of facial protection. One qualitative study found that media coverage of HIVand hepatitis B was an important factor in health careworkers‘ compliance with universal precautions.10 The author comments that, although first- and secondhand experience of events are more emotionally loaded, mass media coverage of bloodborne disease has resulted in a sharpened attention to the problems. This finding provides opportunities for further research. Using the media to communicate risk and affect healthy work behaviors to prevent occupational transmission of communicable respiratory illness is an area that has not been well studied.

Organizational factors influencing compliance

Several studies have found organizational factors to be the most significant predictor of safe work behaviors, specifically compliance with universal precautions. 8,9,15,16,24 Our study found that nurses who felt they had organizational support for health and safety were significantly more likely to report compliance with the recommended use of facial protection. This finding was significant at the univariate and multivariate levels of analysis. Measures for organizational support included management making health and safety a high priority, taking all reasonable steps to minimize hazards, encouraging employees‘ involvement in health and safety matters, and actively working to protect employees. This finding shows how important it is that nurses think their health and safety are valued by their employer. Those who think that they are valued will engage in healthier work behaviors. Interventions targeted at improving how supportive the organization is of employee health and safety can result in enhanced compliance. This finding is supported in the literature related to universal precautions.4,25

An absence of job hindrances was shown to be linked to enhanced compliance at the univariate level. Proper use of facial protection can be hindered by the type of duty carried out (eg, carrying out a long procedure while using a tight-fitting respirator), a lack of time to properly use and dispose of the equipment (eg, busy unit or emergency situation), and the desire to provide good quality patient care (eg, having the patient unable to see your facial

149 expression while explaining a risky procedure). One study found that a worker‘s perception that the use of universal precautions interfered with their work (job hindrance) was a strong predictor of failure to comply with universal precautions.16 Reducing or eliminating these or other identified job hindrances is a good strategy to enhance compliance. The literature has also shown that a lack of available time and a heavy workload negatively influenced compliance with hand hygiene precautions.26-29

Study limitations

Our study used a new data collection tool. Reliability testing for 3 of the explanatory measures fell below the acceptable range and require further evaluation. The cross-sectional study design precludes the determination of causality. Study results may not be generalized to all professions, geographic location, or type of health care facility because these findings are from a convenience sample from 2 acute care, urban hospitals with a 44% response rate. Self-selection bias may be a limitation because participation in the study was voluntary and a decision to participate may be correlated with traits that affect the study, making the participants a non-representative sample. For example, people who have strong opinions or substantial knowledge may be more willing to spend time answering the survey than those who do not. Subject recall and social desirability bias may be a problem given the reliance on self-report data for this study. Self-reported compliance has been found to be higher than actual compliance,30-32 and it is possible that our data might be an overestimate of our respondents‘ true compliance.

CONCLUSION

The objectives of this study were to describe the factors that influence nurses‘ decisions to use facial protection and to determine their relative importance in predicting compliance. Multivariate analysis showed 5 factors to be key predictors of nurses‘ compliance with the recommended use of facial protection. These factors include full-time work status, greater than 5 years tenure as a nurse, at least monthly use of facial protection, a belief that media coverage of infectious diseases impacts risk perception and work practices, and organizational support for health and safety. Strategies and interventions based on these findings should result in enhanced compliance with facial protection and, ultimately, a reduction in occupational transmission of communicable respiratory illness.

References 1. Moore D, Gamage B, Bryce E, Copes R, Yassi A. Protecting health care workers from SARS and other respiratory pathogens: organizational and individual factors that affect adherence to infection control guidelines. Am J Infect Control 2005;33:88-96. 2. Nickell LA, Crighton EJ, Tracy CS, Al-Enazy H, Bolaji Y, Hanjrah S,et al. Psychosocial effects of SARS on hospital staff: survey of a large tertiary care institution. J Can Med Assoc 2004;170:793-8. 3. Evanoff B, Kim L, Mutha S, Jeffe D, Haase C, Andereck D, et al. Compliance with universal precautions among emergency department personnel caring for trauma patients. Ann Emerg Med 1999;33:160-5. 4. McGovern PM, Vesley D, Kochevar L, Gershon RRM, Rhame FS, Anderson E. Factors affecting universal precautions compliance. J Business Psychol 2000;15:149-61. 5. Green LW, KreuterMW. Health promotion planning: an educational and environmental approach. 2nd ed. Mountain View, CA: Mayfield; 1991. 6. DeJoy DM. Theoretical models of health behavior and workplace selfprotective behavior. J Safety Res 1996;27:61-72. 7. Gershon R, Vlahov D, Felknor S, Vesley D, Johnson P, Delclos G, et al. Compliance with universal precautions among health care workers at three regional hospitals. Am J Infect Control 1995;23:225-36. 8. Gershon RM, Karkashian CD, Vlahov D, Kummer L, Kasting C, Green-Mckenzie J, et al. Compliance with universal precautions in correctional health care facilities. J Occup Environ Med 1999;41:181-9. 9. Godin G, Naccache H, Morel S, Ebacher M-F. Determinants of nurses‘ adherence to universal precautions for venipunctures. Am J Infect Control 2000;28:359-64. 10. Lymer UB, Richt B, Isaksson B. Blood exposure: factors promoting health care workers‘ compliance with guidelines in connection with risk. J Clin Nurs 2004;13:547-54. 11. Provincial Infectious Diseases Advisory Committee. Preventing febrile respiratory illness: protecting patients and staff. Toronto, Ontario: Ontario Ministry of Health and Long-Term Care; 2005. Available at: http://www.health.gov.on.ca. Accessed August 15, 2007. 12. DeJoy D, Gershon R, Schaffer B. Safety climate—assessing management and organizational influences on safety. Professional Safety 2004;49:50-7. 13. Osborne S. Influences on compliance with standard precautions among operating room nurses. Am J Infect Control 2003;31:415-23. 14. Chia S, Koh D, Fones C, Qian F, Ng V, Tan BH, et al. Appropriate use of personal protective equipment among health care workers in public sector hospitals and primary health care polyclinics during the SARS outbreak in Singapore. J Occup Environ Med 2005;62:473-7. 15. Gershon RRM, Karkashian CD, Grosch JW, Murphy LR, Escamilla- Cejudo A, Flanagan PA, et al. Hospital safety climate and its relationship with safe work practices and workplace exposure incidents. Am J Infect Control 2000;28:211-21. 16. DeJoy DM, Murphy LR, Gershon RM. The influence of employee, job/task, and organizational factors on adherence to universal precautions among nurses. Int J Ind Ergonomics 1995;16:43-55. 17. Shtatland E, Kleinman K, Cain E. Stepwise methods in using SAS proc logistic and SAS enterpriseminer for prediction; SAS SUGI proceedings: statistics, data analysis and data mining, April 16-19, 2007, Orlando, Florida. Available at: http://www2.sas.com/proceedings/

150

sugi28/258-28.pdf. Accessed August 15, 2007. 18. Nunnally J, Bernstein I. Psychometric theory. 2nd ed. New York: McGraw-Hill, Inc; 1994. 19. Canadian Nurses Association. RN workforce profiles by area of responsibility year 2005. October 2006. CNA Department of Public Policy. Available at: http://www.cna-nurses.ca/CNA/documents/pdf/publications/RN-Specialty-Profiles-2005-e.pdf. Accessed August 15, 2007. 20. Tokars J, McKinley G, Otten J, Woodley C, Sordillo EM, Caldwell J, et al. Use and efficacy of tuberculosis infection control practices at hospitals with previous outbreaks of multidrug-resistant tuberculosis. Infect Control Hosp Epidemiol 2001;2:449-55. 21. Kellerman S, Saiman L, San Gabriel P, Besser R, Jarvis W. Observational study of the use of infection control interventions for Mycobacterium tuberculosis in pediatric facilities. Pediatr Infect Dis J 2001;20:566-73. 22. Park B, Peck A, Kuehnert M, Newbern C, Smelser C, Comer JA, et al. Lack of SARS transmission among health care workers, United States. Emerg Infect Dis 2004;10:244-8. 23. Lau J, Fung K, Wong TW, Kim J, Wong E, Chung S, et al. SARS transmission among hospital workers in Hong Kong. Emerg Infect Dis 2004;10:280-6. 24. Kretzer EK, Larson E. Behavioral intentions to improve infection control practices. Am J Infect Control 1998;26:245-53. 25. Rivers D, Aday L, Frankowski R, Felknor S, White D, Nichols B. Predictors of nurses‘ acceptance of an intravenous catheter safety device. Nurs Res 2003;52:249-55. 26. Larson E. Compliance with isolation technique. Am J Infect Control 1983;11:221-5. 27. Pittet D, Mourouga P, Perneger TV. Compliance with handwashing in a teaching hospital. Ann Intern Med 1999;130:126-30. 28. Stone PW, Clarke SP, Cimiotti J, Correa-de-Araujo R. Nurses‘ working conditions: implications for infectious diseases. Emerg Infect Dis 2004;10:1984-9. 29. Ferguson KJ, Waitkin H, Beekmann SE, Doebbeling BN. Critical incidents of nonadherence with standard precautions guidelines among community hospital-based health care workers. J Gen Intern Med 2004;19:726-31. 30. Saturno P, Palmer R, Gascon J. Physician attitudes, self-estimated performance and actual compliance with locally peer-defined quality evaluation criteria. Int J Qual Health Care 1999;11:487-96. 31. Henry K, Campbell S, Maki M. A comparison of observed and selfreported compliance with universal precautions among emergency department personnel at a Minnesota public teaching hospital: implications for assessing infection control programs. Ann Emerg Med 1992; 21:940-6. 32. Simmons B, Bryant J, Neiman K, et al. The role of handwashing in prevention of endemic intensive care unit infections. Infect Control Hosp Epidemiol 1990;11:589-94.

151

Appendix 3 - Two Page Proposal Summary THE ORGANIZATIONAL, ENVIRONMENTAL AND INDIVIDUAL FACTORS THAT INFLUENCE NURSES‘ DECISIONS TO USE FACIAL PROTECTION TO PREVENT OCCUPATIONAL TRANSMISSION OF COMMUNICABLE RESPIRATORY ILLNESS IN ACUTE CARE HOSPITALS

Kathryn Nichol, RN, BScN, MHSc - PhD Student, University of Toronto Supervisor - Dr. D. Linn Holness PAC Members - Dr. Linda O‘Brien-Pallas, Dr. Allison McGeer, Dr. Phil Bigelow Funding – Centre for Research Expertise in Occupational Disease, Workplace Safety and Insurance Board

BACKGROUND:

One of the key reasons for occupational transmission of communicable respiratory disease is the failure to implement appropriate barrier precautions. Compliance with safe work practices to protect against transmission of blood borne disease (universal precautions) has been found to be substandard. Workplace organizational factors have been shown to be the main influence on nurses‘ decisions to follow universal precautions. Compliance with safe work practices to prevent the spread of communicable respiratory disease, such as immunization, have also been found to be poor. Facial protection is identified as the personal protective equipment that is least complied with by nurses yet it is an important barrier precaution against respiratory illness. Understanding why nurses fail to appropriately use facial protection has not been well researched.

STUDY OBJECTIVES:

a) To measure nurses‘ reported compliance with the recommended use of facial protection in acute care hospitals. b) To compare self-report compliance data from the survey with observations made on the unit. c) To describe the reported health outcomes of nurses as a result of occupational exposure to communicable respiratory illness. d) To describe the organizational, environmental and individual factors that influence nurses‘ compliance with facial protection. e) To describe the organizational, environmental and individual factors that are associated with adverse health outcomes of nurses as a result of occupational exposure to communicable respiratory illness. f) To determine the relative importance of organizational, environmental and individual factors in predicting nurses‘ compliance with the use of facial protection and nurses‘ health outcomes.

MAIN HYPOTHESIS:

Organizational and environmental factors will be the primary influence on nurses‘ decisions to use facial protection to prevent occupational transmission of communicable respiratory illness in the acute care setting.

152

METHODS:

The proposed study will use a 3-phased approach. Phase 1 of the study will be a focus group involving 6 – 12 nurses to evaluate the survey tool on its‘ length, comprehensiveness and format. The focus group will be held at the first hospital that agrees to participate in the study and participants will be recruited from the units where data collection will occur. An information sheet describing the study and outlining that participation is voluntary will be distributed and nurses will sign a form consenting to participate in the focus group.

Phase 2 of the study will be a cross-sectional survey of 300 nurses at 4-6 acute care hospitals (approximately 50 nurses per hospital) who regularly use facial protection to prevent occupational transmission of communicable respiratory disease. Examples of units where nurses regularly use facial protection include Intensive Care Units (ICU), Emergency Departments, Step-Down Units, Respirology units or clinics and any other in-patient units that regularly receive patients with respiratory symptoms. In all units except the ICU, the survey will be anonymous. In the ICU, identifying numbers will be put on completed surveys. This will be done to facilitate matching observed data from Phase 3 of the study with self-report data from the survey. At the same time the surveys are being distributed, information on the cleanliness and orderliness of the unit and the ready availability of facial protection will be collected by the researcher on a unit-by-unit basis. This information will be compared with self-report data from the survey.

Phase 3 will be an observational study of 100 nurses in the ICU (approximately 20 from each hospital). This setting was chosen as it is feasible to unobtrusively observe nurses‘ work practices outside patient rooms from the nursing station. Attempts will be made to observe ICU nurses who completed a survey. Observations regarding compliance with each hospital‘s requirements for donning, use, doffing and disposal of facial protection will be conducted by a trained observer approximately 1 - 3 months after the collection of survey data. Only one observation will be recorded per individual to ensure data is independent. Individuals‘ observed measures of compliance will be matched with their self-reported compliance with recommended use of facial protection. This comparison will assist to validate the outcome measure of compliance used in the study.

Four to six acute-care hospitals in the GTA will be recruited to participate in this study. Based on the pilot study, it is estimated that each hospital will have 4 or 5 units where nurses regularly use facial protection. It is anticipated that data collection will be carried out between August, 2007 and March, 2008. The Vice President and/or Chief Nursing Officer (VP/CNO) of the hospitals will be approached to discuss recruitment of the hospital as site for study. The VP/CNO will be able to provide the investigators with a list of units where facial protection is regularly worn by nursing staff. Ethics approval will be sought for each participating hospital through their Ethics Review Boards.

For the anonymous cross-sectional survey, convenience sampling will be used and participants will be recruited on their unit. Several days will be spent on each unit distributing surveys. Potential participants will be given an information sheet describing the study and provided with an opportunity to ask any questions. They will be informed that participation is voluntary and individual responses will not be identified in any presentation or publication arising from the

153 study results. Consent will be implied when participants agree to complete the 20-minute questionnaire.

For the cross-sectional survey and observational study in the ICU, convenience sampling will also be used and participants will be recruited on their unit. An information sheet will be provided and participants will sign a form consenting to complete the survey and participate in observational study. Participation will be voluntary and individual responses will not be identified in any presentation or publication arising from the study results.

Knowledge transfer and exchange of study results will be offered to all participating units and hospitals.

154

Appendix 4 - Letter of Information Participation in a Research Study - Survey Only

Introduction Before agreeing to participate in this research study, it is important that you read and understand this research consent form. This form provides all the information we think you will need to know in order to decide whether you wish to participate in the study. If you have any questions after you read through this form, ask one of the investigators. You should not sign this form until you are sure you understand everything on this form. You may also wish to discuss your participation in this study with your supervisor or colleagues.

Title of Research Study: The organizational, environmental and individual factors that influence nurses‘ decisions to use facial protection to prevent occupational transmission of communicable respiratory illness in acute care hospitals.

Principal Investigator Dr. D. Linn Holness, MD, MHSc. Director, Occupational Health Services, St. Michael‘s Hospital, 416-864-5074.

Co-Investigator Kathryn Nichol, BScN, MHSc. Graduate Student, University of Toronto, 416-360-4000 x3334. Kathryn is conducting this as student research in partial fulfillment of her doctoral studies.

Purpose of the Research: The purpose of this study is to examine the factors that affect nurses‘ decisions to use facial protection to prevent occupational transmission of communicable respiratory illness. Facial protection is personal protective equipment worn on the face. In health care, it includes surgical or procedure masks, respirators such as N95, face shields and eye protection including goggles and safety glasses. Occupational transmission is the spread of a communicable disease in the workplace such as a hospital. Communicable respiratory illness is any communicable illness that affects the respiratory system such as influenza, tuberculosis, severe acute respiratory syndrome, and respiratory syncytial virus (RSV). Once the most influential factors are identified, strategies can be developed to enhance compliance with facial protection and reduce the prevalence of worker illness and disease.

Description of the Research: There are two parts to this study. The first part of this study will involve the completion of a questionnaire to collect information on:  current knowledge of droplet and airborne spread communicable diseases,  use of facial protection at work and  the factors that influence your use of facial protection. The questionnaire will take approximately 20 minutes to fill out. We expect to conduct this study in four to six large acute care hospitals in the Greater Toronto Area and hope to recruit 300 nurses to complete the survey in total.

155

The second part of this study will involve nurses working in the Intensive Care Unit (ICU) only. In addition to completing the questionnaire, ICU nurses will be asked to participate in an observational study. Approximately one to three months after the survey is administered, a trained observer will return to the workplace to watch nurses put on, use, take off and dispose of their facial protective equipment and record compliance with hospital protocols.

Your Participation Your participation is being requested for the first part of this study only – the completion of a 20- minute questionnaire. Your consent to participate will be implied from the completion and return of the survey.

Potential Harms (Injury, Discomforts or Inconvenience): There are no known harms associated with participation in this study. The work that you do will in no way be affected by your decision to participate.

Potential Benefits: While there are no direct benefits to your participation in this study, the information we gather from this research project will be used to provide important information about the factors that influence nurses‘ use of facial protection in acute care hospitals. Our study might potentially lead to programs and strategies to enhance the use of facial protection in your hospital and in other hospitals as well.

Confidentiality and Privacy: Confidentiality will be respected and no information that discloses your identity will be released or published without consent unless required by law. Your identity will be known only to the study researchers.

Publication of results: The results of this study may be presented at conferences, seminars and other public forums. The results may also be published. Any presentation or publication will not identify individuals and will only present group results.

Compensation for Injury: In no way does signing this form waive your legal rights nor relieve the investigator, sponsors or involved institutions from their legal and professional responsibility.

Reimbursement: $5 coffee card/certificate

Participation and Withdrawal: Participation in research is voluntary. The work that you do at the hospital will in no way be affected by your decision to participate. Once the study is complete, the research findings will be available to you upon request.

Research Ethics Board If you have any questions regarding your rights as a research participant, you may contact Dr. Julie Spence, Chair, Research Ethics Board, St. Michael‘s Hospital, at 416-864-6060 ext. 2557.

156

Appendix 5 - Letter of Information and Consent Participation in a Research Study – Survey and Observations

Principal Investigator Dr. D. Linn Holness, MD, MHSc. Director, Occupational Health Services, St. Michael‘s Hospital, 416-864-5074.

Co-Investigator Kathryn Nichol, BScN, MHSc. Graduate Student, University of Toronto, 416-360-4000 x3334. Kathryn is conducting this as student research in partial fulfillment of her doctoral studies.

Description of the Research: There are two parts to this study.

1) The first part of this study will involve the completion of a questionnaire to collect information on:  current knowledge of droplet and airborne spread communicable diseases,  use of facial protection at work and  the factors that influence your use of facial protection.

157

The questionnaire will take approximately 20 minutes to fill out. We expect to conduct this study in four to six large acute care hospitals in the Greater Toronto Area and hope to recruit 300 nurses to complete the survey in total.

2) The second part of this study will involve nurses working in the Intensive Care Unit (ICU) only. In addition to completing the questionnaire, ICU nurses will be asked to participate in an observational study. Approximately one to three months after the survey is administered, a trained observer will return to the workplace to observe nurses‘ that consented to participate in the observational study. This observer will not announce their presence to these nurses and will attempt to remain anonymous and unobtrusive. The observer will watch these nurses put on, use, take off and dispose of their facial protective equipment and record compliance with hospital protocols. To match the information collected from the survey with the observations, number identifiers will be placed on completed surveys and observation records. Names will not be recorded on surveys or observation records. Supervisors and/or unit managers will not know which nurses are being observed. We hope to recruit 100 ICU nurses from four to six hospitals to complete the survey and participate in the observational study.

Your Participation As you work in the ICU, your participation is being requested for both parts of the study – the survey and the observations. We will ask for your written consent to participate.

Compensation for Injury: In no way does signing this form waive your legal rights nor relieve the investigator, sponsors or involved institutions from their legal and professional responsibility.

158

Reimbursement: $5 coffee card/certificate

Participation and Withdrawal: Participation in research is voluntary. If you change your mind and wish to withdraw from the observational part of the study, please contact the research team. If the observation has not already taken place, it will be cancelled. The work that you do at the hospital will in no way be affected by your decision to participate or by your decision to withdraw. Once the study is complete, the research findings will be available to you upon request.

159

THE ORGANIZATIONAL, ENVIRONMENTAL AND INDIVIDUAL FACTORS THAT INFLUENCE NURSES’ DECISIONS TO USE FACIAL PROTECTION TO PREVENT OCCUPATIONAL TRANSMISSION OF COMMUNICABLE RESPIRATORY ILLNESS IN ACUTE CARE HOSPITALS.

CONSENT TO PARTICIPATE IN A RESEARCH STUDY

I acknowledge that the research study described above has been explained to me and that any questions that I have asked have been answered to my satisfaction. I have been informed of the alternatives to participation in this study, including the right not to participate and the right to withdraw. I understand that if I don‘t participate or if I withdraw from the study, the work that I do at the hospital will in no way be affected. As well, the potential risks, harms and discomforts have been explained to me and I also understand the benefits of participating in the research study.

I understand that I have not waived my legal rights nor released the investigators, sponsors, or involved institutions from their legal and professional duties. I know that I may ask now, or in the future, any questions I have about the study or the research procedures. I have been assured that records relating to me will be kept confidential and that no information will be released or printed that would disclose personal identity without my permission unless required by law. I have been given sufficient time to read and understand the above information.

I hereby consent to participate, and will be given a copy of this consent form.

Research Subject:

Name (Printed) Signature Date

Study Investigator:

Name (Printed) Signature Date

160 Appendix 6 - Facial Protection Questionnaire

Thank you for completing this confidential questionnaire. Please read the directions at the beginning of each section before answering the questions.

A Background Information

Please fill in the blanks or check the box that most accurately reflects your current situation. Please check only one box per question.

1. My gender is ⁪ Female ⁪ Male

2. My age is __ __ years

3. I am a ⁪ RN ⁪ RPN

4. My education is ⁪ Certificate ⁪ Diploma ⁪ Degree ⁪ Master‘s Degree

5. I have been a nurse for __ __ years

6. I have worked as a nurse on this unit for __ __ years

7. I work ⁪ Full time ⁪ Part time

8. In the last 12 months I have been in charge of a shift or held a supervisory role on the unit. ⁪ Yes ⁪ No

9. In the last year, the frequency with which I have had to ⁪ daily ⁪ weekly wear facial protection (respirator, surgical mask, eye ⁪ monthly ⁪ rarely protection) at work to protect myself is ⁪ never ______

B-I Droplet and Airborne Spread Disease – Using Facial Protection

Please check the box that most accurately reflects what you know about using facial protection to prevent transmission of droplet and airborne spread communicable respiratory disease. Please circle only one number per question. 1 2 3 4 5 6 Strongly Agree Neutral Disagree Strongly Don‘t Agree Disagree Know

10. Fit testing of N95 respirators is done to ensure the make, 1 2 3 4 5 6 model and size of the respirator seals to my face.

11. Wearing a disposable N95 respirator will help protect 1 2 3 4 5 6 me from catching an airborne communicable disease.

12. An N95 respirator filters out a minimum of 95% of 1 2 3 4 5 6 particles or droplets that may be communicable.

161 13. There is no recommended way to remove an N95 1 2 3 4 5 6 respirator.

14. Wearing a surgical mask will help protect me from 1 2 3 4 5 6 catching a droplet spread communicable disease.

15. It is good practice to reuse a surgical mask several times 1 2 3 4 5 6 as protection from a communicable disease.

16. Acceptable eye protection includes regular reading glasses. 1 2 3 4 5 6

17. Reusable eye shields or goggles used as protection from a 1 2 3 4 5 6 communicable disease should be disinfected after each use. ______

B-II Droplet and Airborne Spread Disease – Transmission of Influenza

Please check the box that most accurately reflects what you know about the transmission of influenza. Please circle only one number per question. 1 2 3 4 5 6 Strongly Agree Neutral Disagree Strongly Don‘t Agree Disagree Know

18. Influenza is thought to be primarily spread through 1 2 3 4 5 6 large droplets (droplet transmission) that directly contact the nose, mouth or eyes.

19. These droplets are produced when infected people cough, 1 2 3 4 5 6 sneeze or talk, sending the relatively large infectious droplets and very small sprays (aerosols) into the nearby air and into contact with other people.

20. Large droplets can only travel a limited range; therefore, 1 2 3 4 5 6 people should limit close contact (within 6 feet) with others when possible.

21. To a lesser degree, human influenza is spread by touching 1 2 3 4 5 6 objects contaminated with influenza viruses and then transferring the infected material from the hands to the nose, mouth or eyes.

22. Influenza may also be spread by very small infectious 1 2 3 4 5 6 particles (aerosols) traveling in the air.

23. The contribution of each route of exposure to influenza 1 2 3 4 5 6 transmission is uncertain at this time and may vary based upon the characteristics of the influenza strain. ______

162 C Effectiveness of Preventive Actions

Please circle the number that most accurately reflects your beliefs regarding the effectiveness of these preventive actions. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

24. I can reduce my risk of catching a communicable 1 2 3 4 5 respiratory disease by using an N95 respirator.

25. I can reduce my risk of catching a communicable 1 2 3 4 5 respiratory disease by using a surgical mask.

26. I can reduce my risk of catching a communicable 1 2 3 4 5 respiratory disease by wearing eye protection. ______

D Exposure

Please circle the number that most accurately reflects your current situation regarding exposure to communicable respiratory illness. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

27. I have been exposed to a communicable respiratory illness at 1 2 3 4 5 work.

28. I have contracted a communicable respiratory illness through 1 2 3 4 5 work.

29. I have suffered adverse physical health effects as a result of 1 2 3 4 5 exposure to a communicable respiratory illness at work.

30. I have suffered adverse mental health effects as a result of 1 2 3 4 5 exposure to a communicable respiratory illness at work.

31. People I know who have been exposed to a communicable ⁪ family member ⁪ close friend respiratory illness at work which resulted in an undesirable ⁪ colleague ⁪ acquaintance physical or mental health outcome include: ⁪ not applicable

If you have experienced any adverse health outcomes, please use the space below to provide further information about the nature of these illnesses. ______

163 E Perception of Risk

Please circle the number that most accurately reflects your risk of becoming infected at work. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

32. I am at risk of becoming infected with a communicable 1 2 3 4 5 respiratory illness through work.

33. I am concerned I may catch a communicable respiratory illness 1 2 3 4 5 at work.

34. I am concerned my colleagues may catch a communicable 1 2 3 4 5 respiratory illness at work. ______

F Personal Barriers to Use

Please circle the number that most accurately reflects how the following protective measures affect your ability to do your job. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

35. Wearing a surgical mask interferes with my ability to do my job due to: a) physical discomfort 1 2 3 4 5 b) difficulty breathing 1 2 3 4 5 c) adverse skin reactions 1 2 3 4 5 d) anxiety due to having my face covered 1 2 3 4 5 e) communication breakdown 1 2 3 4 5

36. Wearing a respirator interferes with my ability to do my job due to: a) physical discomfort 1 2 3 4 5 b) difficulty breathing 1 2 3 4 5 c) adverse skin reactions 1 2 3 4 5 d) anxiety due to having my face covered 1 2 3 4 5 e) communication breakdown 1 2 3 4 5

37. Wearing eye protection interferes with my ability to do my job due to: a) physical discomfort 1 2 3 4 5 b) anxiety 1 2 3 4 5 c) difficulty seeing 1 2 3 4 5 ______

164 G – Use of Facial Protective Equipment

Please circle the number that most accurately reflects how you currently use facial protection at work. Please circle only one number per question. 1 2 3 4 5 Always Mostly Sometimes Rarely Never

38. I wear an N95 respirator when caring for a patient that I 1 2 3 4 5 suspect may have an airborne spread disease.

39. I wear an N95 respirator when caring for a patient who is in 1 2 3 4 5 airborne precautions even if I don‘t believe that they have an airborne spread disease.

40. I wear an N95 respirator when caring for a patient that has 1 2 3 4 5 been diagnosed with an airborne spread disease.

41 I conduct a seal check each time I put on an N95 respirator. 1 2 3 4 5

42. I wear a surgical mask or an N95 respirator when within 1 metre 1 2 3 4 5 of a patient that I suspect may have a droplet spread disease.

43. I wear a surgical mask or an N95 respirator when within 1 metre 1 2 3 4 5 of a patient that has a diagnosed droplet spread disease.

44. I wear eye protection (face shield, goggles, fitted eye 1 2 3 4 5 protection) when within 1 metre of a coughing patient that I suspect may have a droplet spread disease.

45. I wear eye protection when within 1 metre of a coughing patient 1 2 3 4 5 that has a diagnosed droplet spread disease. ______

H Cleanliness/Orderliness of the Workplace

Please circle the number that most accurately reflects the cleanliness and orderliness of your unit or department. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

46. My work area is kept clean. 1 2 3 4 5

47. My work area is not cluttered. 1 2 3 4 5

48. My work area is not crowded. 1 2 3 4 5 ______

165 I Availability of Facial Protection

Please circle the number that most accurately reflects the availability of facial protection on your unit or in your department. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

49. N95 respirators that fit me are readily accessible in my work area. 1 2 3 4 5

50. Surgical masks are readily accessible in my work area. 1 2 3 4 5

51. Eye protection is readily accessible in my work area. 1 2 3 4 5 ______

J Training

Please circle the number, fill in the blanks or choose the box that most accurately reflects the training you have received at work. Please provide only one answer per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree 52. I have received training on: a) the different communicable respiratory diseases that I 1 2 3 4 5 may be exposed to at work b) how they are transmitted 1 2 3 4 5 c) what I can do to protect myself 1 2 3 4 5 d) how to properly use a surgical mask 1 2 3 4 5 e) how to properly use eye protection 1 2 3 4 5 f) how to properly use an N95 respirator (including how to conduct a seal check) 1 2 3 4 5

53. The year that I last received this training was ______

54. In the last two years, I have been fit tested for an N95 respirator.  Yes  No

55. The N95 respirator that I was successfully fit tested for is: Make/Brand ______Model/Number ______

K Media Coverage of Communicable Diseases

Please circle the number that most accurately reflects how media coverage of communicable diseases affects your work practices. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

56. Media coverage of communicable diseases makes me more 1 2 3 4 5 aware of my risk of catching a respiratory disease at work.

166

57. As a result of media coverage of communicable diseases, I am 1 2 3 4 5 more careful at work. ______

L Organizational Support

Please circle the number that most accurately reflects your current work situation. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

58. The protection of workers from occupational exposure to 1 2 3 4 5 communicable respiratory disease is a high priority where I work.

59. My workplace has a functioning joint health and safety 1 2 3 4 5 committee.

60. On my unit, all reasonable steps are taken to minimize hazardous 1 2 3 4 5 job tasks and procedures.

61. On my unit, employees are encouraged to become involved in 1 2 3 4 5 health and safety matters.

62. Managers on my unit do their part to ensure employees‘ 1 2 3 4 5 protection from occupational exposure to communicable respiratory disease.

63. In my current work area, written policies for eye and 1 2 3 4 5 respiratory protection are readily accessible.

64. My job duties do not interfere with my being able to use 1 2 3 4 5 facial protection.

65. I have enough time in my work to use facial protection properly. 1 2 3 4 5

66. I can provide good quality care to my patients while wearing 1 2 3 4 5 facial protection.

67. On my unit, employees are encouraged to identify unsafe work 1 2 3 4 5 practices amongst themselves.

68. On my unit, unsafe work practices that aren‘t resolved are 1 2 3 4 5 corrected by supervisors.

69. In my current work area, my coworkers support me in following 1 2 3 4 5 safe work practices.

70. In my current work area, supervisors regularly discuss safety at 1 2 3 4 5

167 staff meetings.

71. There are good employee relations on my unit. 1 2 3 4 5

72. Management and staff on my unit support one another. 1 2 3 4 5

73. On my unit, there is open communication between supervisors 1 2 3 4 5 and staff.

74. My supervisor enforces compliance with legislation and workplace policy regarding proper use of facial protection. 1 2 3 4 5 ______

M Patient Safety Climate

Please circle the number that most accurately reflects how you feel about patient safety in your current work situation. Please circle only one number per question. 1 2 3 4 5 Strongly Agree Neutral Disagree Strongly Agree Disagree

75. The senior leaders in my hospital listen to me and care 1 2 3 4 5 about my concerns.

76. The physician and clinical leaders in my area listen to me 1 2 3 4 5 and care about my concerns.

77. Leadership is driving us to be a safety centered institution. 1 2 3 4 5

78. My suggestions about safety would be acted upon if I 1 2 3 4 5 expressed them to management.

79. Management/leadership does not knowingly compromise 1 2 3 4 5 safety concerns for productivity.

80. I am encouraged by my colleagues to report any safety 1 2 3 4 5 concerns I may have.

81. I know the proper channels to direct questions regarding 1 2 3 4 5 patient safety.

82. I am satisfied with the availability of clinical 1 2 3 4 5 leadership.

83. I believe that most adverse events occur as a result of multiple 1 2 3 4 5 system failures and are not attributable to one individual‘s actions.

84. I am aware that patient safety has become a major area for 1 2 3 4 5 improvement in this institution.

168 THANK YOU!

If you have any information you would like to share regarding facial protection, feel free to make comments in the space below!

______

169

Appendix 7 - Unit Observation Record

Facial Protection Study Hospital ______Site ______Unit Observations Dept ______Date ______

Strongly Agree Inconsistent Disagree Strongly agree disagree 1 2 3 4 5 Cleanliness/orderliness of the unit

1. The unit is clean 1 2 3 4 5

______

2. The unit is orderly (not cluttered or crowded) 1 2 3 4 5

______

Availability of facial protective equipment

1. N95 respirators are available at point of use 1 2 3 4 5

______

2. Surgical masks are available at point of use 1 2 3 4 5

______

3. Eye protection is available at point of use 1 2 3 4 5

______

PLEASE PUT ANY OTHER OBSERVATIONS ON BACK OF THIS RECORD – THANKS.

170

Appendix 8 - Participant Observation Record

Identification # ______Observer ______Start Time ______End Time ______

Patient Assignment (room/bed #)______Patient on  Level 1  Level 2  Level 3 precautions Patient Assignment (room/bed #)______Patient on  Level 1  Level 2  Level 3 precautions

Covering Assignment (room/bed #)______Patient on  Level 1  Level 2  Level 3 precautions Covering Assignment (room/bed #)______Patient on  Level 1  Level 2  Level 3 precautions

Observations conducted  during care of assigned patient  while covering for another nurse  requested demonstration

Level 1 Level 2 Level 3 Contact Precautions Contact/Droplet Precautions Contact/Airborne Precautions

*Nurse uses surgical mask and eye *Nurse uses surgical mask when *Nurse uses N95 respirator when protection IF anticipates providing patient care. providing patient care splash/spray of blood or body fluids *Nurse uses eye protection IF *Nurse uses eye protection IF anticipates splash/spray of blood or anticipates splash/spray of blood or body fluids body fluids

Donning Yes No Donning Yes No Donning Yes No In hallway In hallway In hallway Surgical mask put on Surgical mask put on N95 respirator on Follows guidelines Follows guidelines Follows guidelines - ear loops fully over - ear loops fully over - holds respirator ears ears correctly - strings uncrossed - strings uncrossed - put face piece over - both sets of strings - both sets of strings mouth and nose tied securely tied securely - adjust straps - higher strap at crown - higher strap at crown - straps uncrossed - lower strap at nape - lower strap at nape - higher strap at crown - metal nose piece - metal nose piece - lower strap at nape adjusted adjusted - adjust nose piece - fully covers mouth - fully covers mouth - seal check and nose and nose - fully covers mouth and nose

Eye protection put on Eye protection put on Eye protection put on Surgical mask put on Surgical mask put on Respirator put on before eye protection before eye protection before eye protection

Use Yes No Use Yes No Use Yes No Leaves facial Leaves facial Leaves facial protection on while protection on while protection on while caring for patient caring for patient caring for patient

Refrains from Refrains from touching Refrains from touching touching front of mask front of mask or eye front of respirator or or eye protection while protection while caring eye protection while caring for patient for patient caring for patient

171

Level 1 Level 2 Level 3 Contact Precautions Contact/Droplet Precautions Contact/Airborne Precautions *Nurse uses surgical mask and eye *Nurse uses surgical mask when *Nurse uses N95 respirator when protection IF anticipates providing patient care. providing patient care splash/spray of blood or body fluids *Nurse uses eye protection IF *Nurse uses eye protection IF anticipates splash/spray of blood or anticipates splash/spray of blood or body fluids body fluids

Doffing/Disposal Yes No Doffing/Disposal Yes No Doffing/Disposal Yes No Near door in patient Near door in patient Near door in patient room room room In anteroom In anteroom In anteroom Wash hands or alcohol Wash hands or alcohol Wash hands or alcohol hand rub hand rub hand rub Removes eye protection Removes eye Removes eye - refrains from touching protection protection - refrains eye pieces - refrains from from touching eye - remove using arms touching eye pieces pieces - discard directly or - remove using arms - remove using arms place in plastic bag for - discard directly or - discard directly or disinfection place in plastic bag for place in plastic bag for disinfection disinfection Remove surgical mask Remove surgical mask Remove N95 respirator - refrains from touching - refrains from - refrains from touching face piece touching face piece face piece - untie strings or lift - untie strings or lift - lift neck strap over loops loops head first - undo neck string first - undo neck string first - lift crown strap over - pull mask away from - pull mask away from head next – mask face holding crown face holding crown should move away strings strings from face - discard directly - discard directly - discard directly Removes eye protection Removes eye Removes eye first then surgical mask protection first then protection first then surgical mask respirator Wash hands or use Wash hands or use Wash hands or use alcohol hand rub alcohol hand rub alcohol hand rub

Observer identified self to participant during observation?  Yes  No Was the nurse fit-tested for the respirator that they used?  Yes  No  n/a If yes, what was the brand (ie. 3M) and model (ie. 8210)? ______

Notes: ______

172

Appendix 9 - Participant Observation Guide Scheduling Observations 1. Make an appointment with the Unit Manager to see staff schedule. 2. Note at least two weeks of each participants schedule. 3. Do not share participants names with the Unit Manager 4. Book times to observe each participant. Attempt to book all observations within a two week period. 5. Suggested number of participants to observe during one shift – 3.

Conducting Observations 1. Call unit first to determine if any patients are on respiratory precautions. Names and phone numbers are on FP spreadsheet. 2. Upon entry to the unit, determine the best way to identify participants – unit clerk? Do not share participants name with this person. 3. Identify the type of patient each participant is assigned to as well as which patients they are assigned to cover on lunch and breaks. Identify patients by room number or bed number (not by name). 4. If participants are not assigned to or covering patients on precautions, end the visit. 5. If participants are assigned to or covering patients on precautions, identify an area where you can unobtrusively observe nurses. 6. Use a pencil to fill in the form and correct with an eraser. 7. If the answer to a question is ‗not applicable‘ just write n/a. 8. Conduct one full observation (donning, use, doffing and disposal) per participant. 9. If asked, introduce yourself to the participant and explain your role. Continue with the observation but make a note of it. 10. If you lose sight of the participant (ie. privacy curtains), make a note of it. Continue the observation when the participant reappears. 11. Do not put the participants name on the observation record – use their corresponding identification number only.

Rescheduling Observations 1. If, on the first visit, participants are not assigned to or covering patients on precautions, wait one week before your second observation. If, on this second visit, the participant is still not assigned to or covering patients on precautions, identify yourself to the participant, explain your role in the study and ask them to demonstrate the proper use of facial protective equipment required for a patient on airborne precautions. It would be ideal to wait until all other observations are completed during that visit before asking specific nurses to demonstrate.

173

Appendix 10 - Case Definitions for Explanatory Variables

Category Variable Case Definition Demographic Gender Female/Male Age Years* Nurse type Registered nurse/Registered practical nurse Education Degree (bachelor‘s/master‘s)/No Degree (certificate/diploma) Tenure as a nurse Years* Tenure on the unit Years* Work status Full time/Part time Supervisory/charge nurse In charge of shift in last year/Not in charge of shift in last year status Type of hospital Teaching/Community Type of unit Emergency Critical Care (critical care/intensive care) In patient medical unit Individual Frequency of use of facial Frequent (daily/weekly/monthly) protective equipment in last Not frequent (rarely/never) year Knowledge of Knowledgable (answered at least 7 of 8 items correctly) recommended use of facial Left blank or selected ―Don‘t Know‖ was classified as incorrect protection Knowledge of transmission Knowledgable (answered at least 5 of 6 items correctly) of influenza Left blank or selected ―Don‘t Know‖ was classified as incorrect Perception of effectiveness Perceived preventive actions to be effective (SA/A with all 3 items) of preventive actions Health effects – personal Reported contracting occupational illness or suffering adverse health experience (self) effects (SA/A with survey items 28, 29 or 30) Health effects – personal Reported exposure and adverse health outcome to family, close experience (others) friend or colleague (ticked off family member, close friend or colleague in survey item 31) Perception of risk Perceived occupational risk (SA/A with all 3 items) Personal barriers to use Reported personal barriers to using FPE (SA/A with at least 1 of 13 items Environmental Cleanliness/Orderliness of Reported unit to be clean/orderly (SA/A with all three items) the workplace Availability of facial Reported FPE to be readily available (SA/A with all 3 items) protective equipment Media coverage of Reported media influenced perception of risk and work practices communicable diseases (SA/A with both items) Organizational Training and fit testing Trained and fit tested (trained and fit tested within the last 2 years, SA/A with being trained on at least 5 of 6 items identified) Knowledge of N95 Reported knowing the N95 they were fit tested for (provided valid make/brand or model name/number of N95) Organizational support Perceived organizational support for health and safety (SA/A with all 6 items) Absence of job hindrances Reported absence of job hindrances to working safely (SA/A with all 3 items) Communication Reported good communication in workplace (SA/A with all 8 items) * Variable analyzed as a continuous variable in years SA/A – strongly agree/agree with item

174

1. October 15, 2009 permission received from Dr. David Moore to use "Figure 1 - Theoretical model to explain self-protective behaviour at work", found in the publication "Moore, D., Gilbert, M., Saunders, S., Bryce, E., & Yassi, A. (2005). Occupational health and infection control practices related to severe acute respiratory syndrome: Health care worker perceptions. Journal of the American Association of Occupational Health Nurses, 53 (6), 257-66".

2. October 15, 2009 permission received from Lori Magda for Dr. Robyn Gershon to use the "Hospital Safety Climate Scale" from "Gershon, R., Karkashian, C., Grosch, J., Murphy, L., Escamilla-Cejudo, A., Flanagan, P., et al. (2000). Hospital safety climate and its relationship with safe work practices and workplace exposure incidents. American Journal of Infection Control, 28 (3), 211-21."

3. October 15, 2009 permission received from Dr. Lawrence Green to use "The Precede- Proceed Model of Health Program Planning & Evaluation" figure found at http://www.lgreen.net/precede.htm.

4. October 15, 2009 permission received from Dr. David DeJoy to use "Figure I. Principal hypothesized relationships among the three diagnostic factors in the PRECEDE model" found in the publication "DeJoy, D., Searcy, C., Murphy, L., & Gershon, R. (2000). Behavioral-diagnostic analysis of compliance with universal precautions among nurses. Journal of Occupational Health Psychology, 5 (1), 127- 141".

5. October 16, 2009 permission received from Elsevier to use the publication ―The individual, environmental, and organizational factors that influence nurses' use of facial protection to prevent occupational transmission of communicable respiratory illness in acute care hospitals‖ in the American Journal of Infection Control in this thesis.

6. November 12, 2009 permission received from Dr. Peter Pronovost to use the "Patient Safety Climate Scale" from the article "Evaluation of the culture of safety: survey of clinicians and managers in an academic medical center. (Original Article). PJ Pronovost, B Weast, CG Holzmueller, BJ Rosenstein, RP Kidwell, KB Haller, ER Feroli, JB Sexton and HR Rubin. Quality and Safety in Health Care 12.6 (Dec 2003): p405(6)".

175

UTORmail: Inbox: Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis Date: Thu, 15 Oct 2009 10:41:38 -0700 [15/10/09 01:41:38 PM EST] From: David Moore To: [email protected] Cc: [email protected] Subject: Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis Hi Kathryn, Fine with me. Thanks for your interest in our paper. David [email protected] wrote: October 15, 2009 Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis Dear Dr. David Moore I am a University of Toronto graduate student completing my Doctoral / Master's thesis entitled "BEHIND THE MASK: Determinants of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals".My thesis will be available in full-text on the internet for reference, study and / or copy. Except in situations where a thesis is under embargo or restriction, the electronic version will be accessible through the U of T Libraries web pages, the Library's web catalogue, and also through web search engines. I will also be granting Library and Archives Canada and ProQuest/UMI a non-exclusive license to reproduce, loan, distribute, or sell single copies of my thesis by any means and in any form or format. These rights will in no way restrict republication of the material in any other form by you or by others authorized by you. I would like permission to allow inclusion of "Figure 1 - Theoretical model to explain self-protective behaviour at work", found in your publication "Moore, D., Gilbert, M., Saunders, S., Bryce, E., & Yassi, A. (2005). Occupational health and infection control practices related to severe acute respiratory syndrome: Health care worker perceptions. Journal of the American Association of Occupational Health Nurses, 53 (6), 257-66". The model provided the framework upon which I based my study. The figure will be attributed through a citation. Please confirm in writing or by email that these arrangements meet with your approval. Sincerely Kathryn Nichol

176

UTORmail: Inbox: Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis From: "Magda, Lori" To: Subject: RE: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis Date: Thursday, October 15, 2009 1:05 PM Hello Kathryn, I spoke to Dr. Gershon, and YES, you have permission to include the scale mentioned below in your thesis. Please let me know if there is anything else you need. Best, Lori

Lori Magda Data Manager for Dr. Robyn Gershon Department of Sociomedical Sciences Mailman School of Public Health Columbia University 722 W. 168th St., Room 939 New York, NY 10032 (p):212 305-1186 (f):212 305-8284

-----Original Message----- From: [email protected] [mailto:[email protected]] Sent: Thursday, October 15, 2009 12:36 PM To: [email protected] Cc: [email protected] Subject: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis October 15, 2009 Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis Dear Dr. Robyn Gershon: I am a University of Toronto graduate student completing my Doctoral / Master's thesis entitled "BEHIND THE MASK: Determinants of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals". My thesis will be available in full-text on the internet for reference, study and / or copy. Except in situations where a thesis is under embargo or restriction, the electronic version will be accessible through the U of T Libraries web pages, the Library's web catalogue, and also through web search engines. I will also be granting Library and Archives Canada and ProQuest/UMI a non-exclusive license to reproduce, loan, distribute, or sell single copies of my thesis by any means and in any form or format. These rights will in no way restrict republication of the material in any other form by you or by others authorized by you. I would like permission to allow inclusion of the following scale in my thesis: "Hospital Safety Climate Scale" from "Gershon, R., Karkashian, C., Grosch, J., Murphy, L., Escamilla-Cejudo, A., Flanagan, P., et al. (2000). Hospital safety climate and its relationship with safe work practices and workplace exposure incidents. American Journal of Infection Control , 28 (3), 211-21." The material will be attributed through a citation. Please confirm in writing or by email that these arrangements meet with your approval. Sincerely, Kathryn Nichol

177

UTORmail: Inbox: Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis From: To: Cc: "Dr Marshall Kreuter" Subject: Re: Permission to Use Copyrighted Material in a Doctoral/Master‘s Thesis Date: Thursday, October 15, 2009 2:41 PM

I can give you permission to reproduce the figure you mention as it appears on my website. Marshall Kreuter and I have also given freely of our permission to reproduce materials from the book, and a version of this figure appears in the 4th edition of the book, which is copyrighted by McGraw-Hill Higher Education. We leave to your own judgment whether you need to get their copyright permission, which can be done on their website. --Lawrence Green

------Original Message------From: [email protected] To: Lawrence W Green Cc: [email protected] Sent: Oct 15, 2009 8:54 AM Subject: Permission to Use Copyrighted Material in aDoctoral/Master‘s Thesis October 15, 2009 Re: Permission to Use Copyrighted Material in a Doctoral/Master‘s Thesis Dear Dr. Green: I am a University of Toronto graduate student completing my Doctoral / Master‘s thesis entitled ―BEHIND THE MASK: Determinants of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals‖. My thesis will be available in full-text on the internet for reference, study and / or copy. Except in situations where a thesis is under embargo or restriction, the electronic version will be accessible through the U of T Libraries web pages, the Library‘s web catalogue, and also through web search engines. I will also be granting Library and Archives Canada and ProQuest/UMI a non-exclusive license to reproduce, loan, distribute, or sell single copies of my thesis by any means and in any form or format. These rights will in no way restrict republication of the material in any other form by you or by others authorized by you. I would like permission to allow inclusion of the following figure in my thesis: "The Precede-Proceed Model of Health Program Planning & Evaluation" found at http://www.lgreen.net/precede.htm. The figure will be attributed through a citation. Please confirm in writing or by email that these arrangements meet with your approval. Sincerely Kathryn Nichol

178

UTORmail: Inbox: Re: Permission to Use Copyrighted Material in a Doctoral/Master's Thesis From: "Dave DeJoy" To: Subject: Re: Permission to Use Copyrighted Material in a Doctoral/Master‘s Thesis Date: Thursday, October 15, 2009 3:26 PM Kathryn: Permission is so granted. Best of luck with your work. Regards. Dave DeJoy Professor Department of Health Promotion and Behavior College of Public Health University of Georgia Athens, GA 30602-6522 706-542-4368 (Office) 706-542-3313 (Department) 706-542-4328 (Workplace Health Group) 706-542-4956 (FAX) [email protected] wrote: October 15, 2009 Re: Permission to Use Copyrighted Material in a Doctoral/Master‘s Thesis Dear Dr. David DeJoy: I am a University of Toronto graduate student completing my Doctoral / Master‘s thesis entitled ―BEHIND THE MASK: Determinants of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals‖. My thesis will be available in full-text on the internet for reference, study and / or copy. Except in situations where a thesis is under embargo or restriction, the electronic version will be accessible through the U of T Libraries web pages, the Library‘s web > catalogue, and also through web search engines. I will also be granting Library and Archives Canada and ProQuest/UMI a non-exclusive license to reproduce, loan, distribute, or sell single copies of my thesis by any means and in any form or format. These rights will in no way restrict republication of the material in any other form by you or by others authorized by you. I would like permission to allow inclusion of the following figure in my thesis: "Figure I. Principal hypothesized relationships among the three diagnostic factors in the PRECEDE model" found in your publication "DeJoy, D., Searcy, C., Murphy, L., & Gershon, R. (2000). Behavioral- diagnostic analysis of compliance with universal precautions among nurses. Journal of Occupational Health Psychology , 5 (1), 127-141". The figure will be attributed through a citation. Please confirm in writing or by email that these arrangements meet with your approval. Sincerely Kathryn Nichol

179

ELSEVIER LICENSE TERMS AND CONDITIONS Oct 16, 2009

This is a License Agreement between Kathryn A Nichol ("You") and Elsevier ("Elsevier") provided by Copyright Clearance Center ("CCC"). The license consists of your order details, the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

Supplier Elsevier Limited The Boulevard,Langford Lane Kidlington,Oxford,OX5 1GB,UK

Registered Company Number 1982084

Customer name Kathryn A Nichol

Customer address 293 Waverley Road Toronto, ON M4L 3T5

License Number 2290800207089

License date Oct 16, 2009

Licensed content publisher Elsevier

Licensed content publication American Journal of Infection Control

Licensed content title The individual, environmental, and organizational factors that influence nurses' use of facial protection to prevent occupational transmission of communicable respiratory illness in acute care hospitals

Licensed content author Kathryn Nichol, Philip Bigelow, Linda O'Brien-Pallas, Allison McGeer, Mike Manno and D. Linn Holness

Licensed content date September 2008

Volume number 36

Issue number 7

Pages 7

Type of Use Thesis / Dissertation Portion Full article Format Both print and electronic

You are an author of the Elsevier article Yes

Are you translating? No Order Reference Number Expected publication date Elsevier VAT number GB 494 6272 12 Permissions price 0.00 USD Value added tax 0.0% 0.00 USD

180

Total 0.00 USD Terms and Conditions INTRODUCTION

1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in connection with completing this licensing transaction, you agree that the following terms and conditions apply to this transaction (along with the Billing and Payment terms and conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you opened your Rightslink account and that are available at any time at http://myaccount.copyright.com).

GENERAL TERMS

2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to the terms and conditions indicated.

3. Acknowledgement: If any part of the material to be used (for example, figures) has appeared in our publication with credit or acknowledgement to another source, permission must also be sought from that source. If such permission is not obtained then that material may not be included in your publication/copies. Suitable acknowledgement to the source must be made, either as a footnote or in a reference list at the end of your publication, as follows:

―Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE SOCIETY COPYRIGHT OWNER].‖ Also Lancet special credit - ―Reprinted from The Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with permission from Elsevier.‖

4. Reproduction of this material is confined to the purpose and/or media for which permission is hereby given.

5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions and/or any other alterations shall be made only with prior written authorization of Elsevier Ltd. (Please contact Elsevier at [email protected])

6. If the permission fee for the requested use of our material is waived in this instance, please be advised that your future requests for Elsevier materials may attract a fee.

7. Reservation of Rights: Publisher reserves all rights not specifically granted in the combination of (i) the license details provided by you and accepted in the course of this licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment terms and conditions.

181

8. License Contingent Upon Payment: While you may exercise the rights licensed immediately upon issuance of the license at the end of the licensing process for the transaction, provided that you have disclosed complete and accurate details of your proposed use, no license is finally effective unless and until full payment is received from you (either by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If full payment is not received on a timely basis, then any license preliminarily granted shall be deemed automatically revoked and shall be void as if never granted. Further, in the event that you breach any of these terms and conditions or any of CCC's Billing and Payment terms and conditions, the license is automatically revoked and shall be void as if never granted. Use of materials as described in a revoked license, as well as any use of the materials beyond the scope of an unrevoked license, may constitute copyright infringement and publisher reserves the right to take any and all action to protect its copyright in the materials.

9. Warranties: Publisher makes no representations or warranties with respect to the licensed material.

10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and their respective officers, directors, employees and agents, from and against any and all claims arising out of your use of the licensed material other than as specifically authorized pursuant to this license.

11. No Transfer of License: This license is personal to you and may not be sublicensed, assigned, or transferred by you to any other person without publisher's written permission.

12. No Amendment Except in Writing: This license may not be amended except in a writing signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).

13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any purchase order, acknowledgment, check endorsement or other writing prepared by you, which terms are inconsistent with these terms and conditions or CCC's Billing and Payment terms and conditions. These terms and conditions, together with CCC's Billing and Payment terms and conditions (which are incorporated herein), comprise the entire agreement between you and publisher (and CCC) concerning this licensing transaction. In the event of any conflict between your obligations established by these terms and conditions and those established by CCC's Billing and Payment terms and conditions, these terms and conditions shall control.

14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described in this License at their sole discretion, for any reason or no reason, with a full refund payable to you. Notice of such denial will be made using the contact information provided by you. Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier or Copyright Clearance Center be responsible or liable for any costs, expenses or damage incurred by you as a result of a denial of your permission request, other than a refund of the amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied permissions.

182

LIMITED LICENSE

The following terms and conditions apply only to specific license types:

15. Translation: This permission is granted for non-exclusive world English rights only unless your license was granted for translation rights. If you licensed translation rights you may only translate this content into the languages you requested. A professional translator must perform all translations and reproduce the content word for word preserving the integrity of the article. If this license is to re-use 1 or 2 figures then permission is granted for non-exclusive world rights in all languages.

16. Website: The following terms and conditions apply to electronic reserve and author websites: Electronic reserve: If licensed material is to be posted to website, the web site is to be password-protected and made available only to bona fide students registered on a relevant course if: This license was made in connection with a course, This permission is granted for 1 year only. You may obtain a license for future website posting, All content posted to the web site must maintain the copyright information line on the bottom of each image, A hyper-text must be included to the Homepage of the journal from which you are licensing at http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at http://www.elsevier.com , and Central Storage: This license does not include permission for a scanned version of the material to be stored in a central repository such as that provided by Heron/XanEdu.

17. Author website for journals with the following additional clauses:

All content posted to the web site must maintain the copyright information line on the bottom of each image, and he permission granted is limited to the personal version of your paper. You are not allowed to download and post the published electronic version of your article (whether PDF or HTML, proof or final version), nor may you scan the printed edition to create an electronic version, A hyper-text must be included to the Homepage of the journal from which you are licensing at http://www.sciencedirect.com/science/journal/xxxxx , As part of our normal production process, you will receive an e-mail notice when your article appears on Elsevier‘s online service ScienceDirect (www.sciencedirect.com). That e-mail will include the article‘s Digital Object Identifier (DOI). This number provides the electronic link to the published article and should be included in the posting of your personal version. We ask that you wait until you receive this e-mail and have the DOI to do any posting. Central Storage: This license does not include permission for a scanned version of the material to be stored in a central repository such as that provided by Heron/XanEdu.

18. Author website for books with the following additional clauses:

183

Authors are permitted to place a brief summary of their work online only. A hyper-text must be included to the Elsevier homepage at http://www.elsevier.com

All content posted to the web site must maintain the copyright information line on the bottom of each image You are not allowed to download and post the published electronic version of your chapter, nor may you scan the printed edition to create an electronic version. Central Storage: This license does not include permission for a scanned version of the material to be stored in a central repository such as that provided by Heron/XanEdu.

19. Website (regular and for author): A hyper-text must be included to the Homepage of the journal from which you are licensing at http://www.sciencedirect.com/science/journal/xxxxx. or for books to the Elsevier homepage at http://www.elsevier.com

20. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be submitted to your institution in either print or electronic form. Should your thesis be published commercially, please reapply for permission. These requirements include permission for the Library and Archives of Canada to supply single copies, on demand, of the complete thesis and include permission for UMI to supply single copies, on demand, of the complete thesis. Should your thesis be published commercially, please reapply for permission.

21. Other ConditionsNone v1.6

Gratis licenses (referencing $0 in the Total field) are free. Please retain this printable license for your reference. No payment is required.

If you would like to pay for this license now, please remit this license along with your payment made payable to "COPYRIGHT CLEARANCE CENTER" otherwise you will be invoiced within 30 days of the license date. Payment should be in the form of a check or money order referencing your account number and this license number 2290800207089. If you would prefer to pay for this license by credit card, please go to http://www.copyright.com/creditcard to download our credit card payment authorization form.

Make Payment To: Copyright Clearance Center Dept 001 P.O. Box 843006 Boston, MA 02284-3006

If you find copyrighted material related to this license will not be used and wish to cancel, please contact us referencing this license number 2290800207089 and noting the reason for cancellation.

Questions? [email protected] or +1-877-622-5543 (toll free in the US) or +1-978-646- 2777.

184

From: Peter Pronovost [[email protected]] Sent: Friday, December 11, 2009 5:57 AM To: Kathryn Nichol Subject: Re: Message for Dr. Pronovost re: Copyright permission

That is fine. However you may want to check out the cultyre survey from ahrq and the full safety attitudes questionnaire.

From: Kathryn Nichol To: Linda Marcellino; [email protected] ; Peter Pronovost Cc: Kathryn Nichol ; [email protected] Sent: Thu Dec 10 17:37:00 2009 Subject: Message for Dr. Pronovost re: Copyright permission December 10, 2009

Re: Permission to Use Copyrighted Material in a Doctoral/Master‘s Thesis

Dear Dr. Pronovost:

I am a University of Toronto graduate student completing my Doctoral thesis entitled ―BEHIND THE MASK: Determinants of nurses‘ adherence to recommended use of facial protective equipment to prevent occupational transmission of communicable respiratory illness in acute care hospitals‖.

My thesis will be available in full-text on the internet for reference, study and / or copy. Except in situations where a thesis is under embargo or restriction, the electronic version will be accessible through the U of T Libraries web pages, the Library‘s web catalogue, and also through web search engines. I will also be granting Library and Archives Canada and ProQuest/UMI a non-exclusive license to reproduce, loan, distribute, or sell single copies of my thesis by any means and in any form or format. These rights will in no way restrict republication of the material in any other form by you or by others authorized by you.

I would like permission to allow inclusion of the following scale in my thesis: "Patient Safety Climate Scale" from your article "Evaluation of the culture of safety: survey of clinicians and managers in an academic medical center.(Original Article). PJ Pronovost, B Weast, CG Holzmueller, BJ Rosenstein, RP Kidwell, KB Haller, ER Feroli, JB Sexton and HR Rubin. Quality and Safety in Health Care 12.6 (Dec 2003): p405(6)".

The material will be attributed through a citation.

Please confirm in writing or by email that these arrangements meet with your approval. Thank you.

Sincerely Kathryn Nichol