Secondary Logo

Journal Logo

Articles

A comprehensive systematic review of healthcare workers' perceptions of risk from exposure to emerging acute respiratory infectious diseases and the perceived effectiveness of strategies used to facilitate healthy coping in acute hospital and community healthcare settings

Yiwen, Koh1; Hegney, Desley1; Drury, Vicki1

Author Information
JBI Library of Systematic Reviews: Volume 8 - Issue 23 - p 917-971
doi: 10.11124/jbisrir-2010-150
  • Free

Abstract

Background

Emerging infectious diseases, defined as diseases that have "newly appeared in a population or have existed previously but are rapidly increasing in incidence or geographic range" (p. 242),1 have always been a threat to nations and are the second leading cause of death worldwide.2 More specifically, the World Health Organization indicated that emerging acute respiratory infectious diseases (EARIDs) have the highest mortality rate worldwide compared to other emerging infectious diseases.2 Significant EARIDs which have emerged in the 21st century include - Severe Acute Respiratory Syndrome (SARS) in 2003, the Avian Influenza A/H5N1 virus in early 2004 and the Influenza A/H1N1 virus in 2009. SARS and the Influenza A viruses share the same modes of transmission, which is via respiratory droplets and person-to-person contact.3 They are contagious and can spread rapidly within populations worldwide, leading quickly to a global pandemic. Additionally, these viruses, especially the SARS and Avian influenza/H5N1 viruses, are virulent with high morbidity and mortality rates.4 They are also predisposed to rapid and efficient mutations which may increase their virulence and resistance to current drug regimens.5, 6

These features of EARIDs pose a problem for health authorities and especially health care workers (HCWs),7 as they are in constant close contact with affected patients and their body fluids - primarily respiratory droplets - through which respiratory viruses are transmitted.4 According to the World Health Organization (2010),8 HCWs may be defined as personnel who are involved in providing health services (i.e. doctors, nurses and pharmacists) as well as management and support workers such as hospital cleaners and clerks. For example, the attack rates on HCWs during nosocomial influenza outbreaks were estimated to be as high as 60%.4 Apart from personal health risks9 from EARIDs, HCWs are also at risk of social isolation and stigmatization because of the social basis of disease transmission.10 In addition, they face risks of spreading the diseases to friends and family, resulting in emotional conflict and guilt.10

According to the Health Belief Model, an individual's risk perceptions are instrumental in influencing their behavior.9 A key review conducted by Leppin and Aro (2009),11 which examined empirical studies focusing on individuals' risk perceptions towards EARIDs, found that few of these studies had explicitly defined the meaning of 'risk perception'. The authors went on to distil these study findings and identify several key constructs of HCWs' risks perceptions within the organization which were used to guide this literature review. These were: personal health risks; health risks to others; social isolation; and acceptance of risks. Very high risk perceptions may hinder the retention of HCWs within the workforce12 as they fear for the health and safety of themselves and their loved ones. Some may also be less devoted in the care provided to patients due to an internal conflict between professional conduct in caring for patients and the innate sense of self-preservation.13 On the other end of the spectrum, HCWs with very low risk perceptions may be non-compliant to protective behaviour such as vaccinations4 and facial protection,14 increasing both their risk and propagation of nosocomial transmission within the hospital and community. In short, it is evident that HCWs' workplace behaviours in response to their risk perceptions have a major impact on themselves, the workforce, individual patient care as well as the wider community.

Many factors play a part in one's risk perceptions and may vary between individuals. Hence there needs to be a greater understanding on how individual nurses perceive such risks, including how these are influenced by associated socio-cultural, psychological, attitudinal factors, and how such perceptions affect their workplace strategies. This knowledge can inform the creation of other strategies to acknowledge and manage the personal risks to HCWs from these EARIDs and maximize the quality of care delivered. Some strategies suggested in the literature include the implementation of social, emotional and psychological coping mechanisms to address nurses' distress and fears in relation to their exposure (i.e. the provision of psychosocial support by nursing administrators and educational interventions to keep nurses up-to-date with current developments).15, 16 Another suggested strategy is the use of verbal approval or monetary rewards to reinforce proper behaviour (i.e. adoption of facial protection and vaccinations) and attitudes.15 Examination and evaluation of this wide repertoire of strategies in primary research, namely, its effectiveness and applicability to the hospital and community settings will inform policy development that will hopefully not only protect nurses but also minimize the disruption of healthcare provision during such crises.

Prior to the commencement of this systematic review, the Cochrane Library, Joanna Briggs Institute (JBI) database and CINAHL database were searched and no previous systematic reviews on this specific topic were identified.

Objectives

The overall aim of this systematic review was to critically appraise, synthesize and present the best available evidence in relation to the risk perceptions and workplace strategies of health care workers (HCWs) to emerging acute respiratory infectious diseases (EARIDs) in acute hospital and community healthcare settings; and to make recommendation for practice that will protect them and their patients/clients.

More specifically, the qualitative component of the review sought to determine:

  • How do HCWs practising in hospitals and community healthcare settings perceive the risks of being exposed to recently EARIDs?
  • What are the socio-cultural, psychological, attitudinal and environmental factors influencing the HCWs' risk perceptions?
  • How do HCWs perceive the effectiveness of the implemented organizational strategies?

The quantitative component of the review seeks to determine:

  • What are the individual strategies adopted by HCWs to manage their risk perceptions when exposed to EARIDs?
  • What are the key organizational, environmental and individual factors influencing their use of the strategies?
  • What are the organizational strategies implemented by the organization?

There is a slight difference between the review questions indicated in the approved protocol and in this SR as changes to the review questions were made after the protocol was approved. This was primarily because the included primary research papers did not have, or had little, content pertaining specifically to the original review questions. As such, the following quantitative review question from the approved protocol was removed:

  • How effective are current directives in ensuring the adherence of nurses to recommended hospital and government protocols?

In addition, the following quantitative review questions were modified to focus more on: the strategies used by HCWs and the hospital to manage their risk perceptions; and HCWs' perceived effectiveness of the organizational strategies. As their perceived effectiveness of the organizational strategies was deemed to be a qualitative research question in nature, it was moved to the qualitative component of this SR.

  • What are the behaviours evident in nurses when exposed to emerging acute respiratory infectious diseases?
  • What are the other key organizational, environmental and individual factors influencing their behaviour?
  • What is the effectiveness of the strategies used to manage the nurses' risk perceptions and therefore their perceptions of workplace safety and quality of life?

Criteria for considering studies for this review

Types of participants

The review considered publications that include male and female nurses practicing in acute hospital and community health care settings. If there were insufficient studies which specifically focus on nurses, the review also considered studies that address all health personnel. This included both personnel involved in providing health services (i.e. doctors, nurses and pharmacists) as well as management and support workers such as hospital cleaners and clerks.

Types of interventions/phenomena of interest

The qualitative component of this review considered studies that investigated HCWs' risk perceptions and their perceived meaning and effectiveness of the strategies that were used to protect them and their patients.

Specifically, HCWs' risk perceptions may be considered to be comprised of the following constructs - perceived health risks, social risks (i.e. stigmatization and risks to others) and acceptance of risks - as asserted in a key review conducted by Leppin and Aro11 which examined empirical studies focusing on individuals' risk perceptions towards EARIDs.

The quantitative component of this review considered studies that examined the individual and organizational strategies that were implemented in response to EARIDs, and the associated factors affecting HCWs' use of such strategies.

Context

This review considered studies which focused on acute hospital and community health care settings in any socio-cultural context.

Types of outcome measures

The main focus of the qualitative component of the review was the meaning HCWs ascribed to the factors affecting their risk perceptions. The study also reported on their perceived effectiveness of the organizational strategies implemented to mitigate individual and social risk.

The outcomes of interest for the quantitative component of the review considered the objective or subjective measures of the following:

  • Compliance and adherence to hospital infection control policies and precautionary measures (i.e. use of personal protective equipment [PPE]);
  • Feelings of fear; and
  • Intention to provide care.

Secondary outcome measures of interest included:

  • Availability of PPE; and
  • Key organisational, environmental and individual factors influencing their risk perception and adoption of work place strategies.

There is a slight difference between the outcomes of interest indicated in the approved protocol and in this SR. HCWs' intentions to provide care was added into this SR as it was found that the majority of the primary research studies included in this SR focused extensively on this aspect.

Types of studies

The qualitative component of the review considered any interpretive studies that drew on the perceptions of risks of being exposed to EARIDs in male and female HCWs practicing in acute hospital and community healthcare settings including, but not limited to, designs such as phenomenology, grounded theory, ethnography, action research and feminist research.

The quantitative component of the review considered any meta-analyses and randomised controlled trials (RCTs); in the absence of meta-analyses and RCTs, other research designs of a quantitative nature, such as non-randomised controlled trials, before and after studies, cohort studies, case control studies, descriptive studies, case series/reports were considered for inclusion in a narrative summary to enable the identification of current best evidence regarding male and female nurses' exposure to emerging acute respiratory infectious diseases in acute hospital and community healthcare settings. Due to the absence of higher level studies such as meta-analyses and randomised controlled studies (RCTs), other research designs of a quantitative nature, specifically descriptive studies, were considered for inclusion in a narrative summary.

Exclusion criteria

This review excluded the following:

  • Studies involving expert opinions & secondary research papers
  • Studies written in any languages other than English
  • Studies which were conducted before the year 1997. This time frame was chosen first, to ensure that the findings were contemporary and relevant and second, because the year 1997 marked the emergence of a novel respiratory communicable disease of pandemic potential - Avian Influenza A/H5N1 virus - in Hong Kong, the first in two decades since the 1977 Russian Flu (H1N1) pandemic.17

Search strategy

The search strategy aimed to find both published and unpublished primary research studies published during the last 12 years (between years 1997-2009) in the English Language. A three-step search strategy was utilised in each component of this review. An initial limited search of PubMed (MEDLINE) and CINAHL was undertaken followed by analysis of the text words contained in the title and abstract, and of the index terms used to describe article. A second extensive search using all identified keywords and index terms was then undertaken across all included databases. Third, the reference list and bibliographies of all identified reports and articles were searched for additional studies. A hand search of relevant key journals not indexed in the databases was not conducted.

Due to time constraints, only the following electronic databases were searched:

  • CINAHL
  • PubMed (MEDLINE)
  • SCOPUS
  • ScienceDirect
  • Sociological Abstracts
  • PsycINFO
  • Web of Science (Social Sciences Citation Index)

A list of initial keywords used in the search strategy can be found in Appendix I. Figure 1 outlines the selection and evaluation process for this review.

Methods of the review

Assessment of methodological quality

Qualitative papers selected for retrieval were assessed by two independent reviewers for methodological validity prior to inclusion in the review using standardised critical appraisal instruments from the Joanna Briggs Institute Qualitative Assessment and Review Instrument (JBI-QARI) (see Appendix II). Quantitative papers selected for retrieval were assessed by two independent reviewers for methodological validity prior to inclusion in the review using standardised critical appraisal instruments from the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) (see Appendix II).

Any disagreements that arose between the reviewers were resolved through discussion, and with a third reviewer.

Data extraction

Qualitative data were extracted from the paper included in the review using the standardised data extraction tool from the Joanna Briggs Institute Qualitative Assessment and Review Instrument JBI-QARI (see Appendix III). Quantitative data were extracted from papers included in the review using the standardised data extraction tool from JBI-MAStARI (see Appendix III). The data extracted included specific details about the interventions, populations, study methods and outcomes of significance to the review question and specific objectives.

Data synthesis

Meta-synthesis was performed for the two qualitative papers. Seven findings were obtained from analysis of the papers and these were then categorised and synthesized to form two meta-syntheses.

The descriptive-correlational quantitative papers that were included in the review used different outcome measures and this prevented statistical pooling. Hence the findings have been presented in descriptive narrative form.

Description of studies

A total of 16 primary research papers were included in the review and they were comprised of:

  • Fourteen descriptive-correlational survey studies; and
  • Two grounded theory qualitative studies.

The table of excluded studies, including the reason(s) for exclusion is found in Appendix IV.

Description and methodological quality of quantitative studies

Fourteen quantitative studies were considered to be suitable for inclusion in this review. Figure 1 outlines the selection and evaluation process involved in identifying the qualitative papers for inclusion in this study.

Figure 1: Selection and evaluation process for quantitative studies
Figure 1: Selection and evaluation process for quantitative studies

Nine of the studies focused on HCWs' perceptions towards SARS.15, 16, 18-24 No studies were found which examined HCWs' perceptions during or following a pandemic influenza outbreak. Instead, the five included studies25-29 which focused on HCWs' pertaining to HCWs' perceptions of a pandemic influenza were based on a possible pandemic influenza outbreak. Of these, eight studies focused on all HCWs, including physicians, nurses and non-medical workers; two studies focused only on physicians; and four focused solely on nurses.

The studies were conducted from across an international spectrum with studies originating in the United States of America, Singapore, Taiwan, Canada, Japan and Hong Kong.

All the quantitative studies included in the review were descriptive-correlational design studies which utilized self-administered questionnaires. The content of the questionnaires that were relevant to the outcomes of this review were developed based on: literature15, 19, 20, 23, 27; well-established questionnaires that were already validated with satisfactory construct validity, such as the General Health Questionnaire,16 Impact of Event Scale,18, 21, 23 Chinese Health Questionnaire,18 and SF-8 Health Survey;24 and/or questionnaires used in other survey studies which had been established for validity.27 In several studies, the internal and content validity of the questionnaires were also established through pilot-tests with HCWs such as nurses, clinicians and health care service researchers15, 16, 22 as well as through the use of an expert panel from the field of infectious diseases such as nosocomial infection control nurse experts15, 22, 23 prior to commencing the study. Based on the pilot study and expert opinions, some changes were made to the questionnaire to increase the clarity of the items. Several studies21, 24-26, 29 did not indicate if one or several components of their questionnaires were valid or reliable.

Correlational design studies were used as higher level studies (i.e. randomised controlled trial designs) could not be found. As such, based on JBI's level of evidence, the level of quantitative evidence in this systematic review would be considered to be level three (Refer to appendix VI). The paucity of higher level studies found for review may be due to the nature of the study which explores HCWs' perceptions and would not be suited for higher level designs

A summary of the characteristics of the included quantitative (i.e. participants, population sample, measures and outcomes) and included qualitative papers (i.e. methodology, method, setting, participants and data analysis) can be found in Appendix V.

Description and methodological quality of qualitative studies

Two qualitative studies30, 31 were considered to be suitable for inclusion in this review. Figure 2 outlines the selection and evaluation process involved in identifying the qualitative papers for inclusion in this study.

Figure 2: Selection and evaluation process for qualitative studies
Figure 2: Selection and evaluation process for qualitative studies

Both studies sought to examine the impact of SARS on physicians working in university hospitals in Canada during the SARS outbreak. Semi-structured individual telephone interviews with open-ended questions were conducted with the participants. Both studies also utilized the grounded theory methodology in developing the findings. Four of the findings obtained from the study were considered "credible" and three of the findings were considered "unequivocal" according to the JBI's three levels of qualitative evidence. A summary of the characteristics of the included qualitative papers (i.e. methodology, method, setting, participants and data analysis) can be found in Appendix V.

Results for the quantitative component of the review

Some papers addressed more than one of the review objectives, therefore the study findings have been presented in relation to the respective objectives leading the review, which were:

  1. How do health care workers practising in hospitals and community health care settings perceive the risks of being exposed to recently emerging acute respiratory infectious diseases?
  2. What are the socio-cultural, psychological, attitudinal and environmental factors influencing the health care workers' risk perceptions?
  3. What are the individual strategies adopted by health care workers when exposed to emerging acute respiratory infectious diseases?
  4. What are the key organizational, environmental and individual factors influencing their use of the strategies?
  5. What are the organizational strategies implemented and how do HCWs perceive the effectiveness of the implemented organizational strategies?

How do health care workers practicing in hospitals and community healthcare settings perceive the risks of being exposed to recently emerging acute respiratory infectious diseases?

Analysis of HCWs' risk perceptions from the included papers resulted in three main categories - perceptions of (i) health risks, (ii) social risks, and (iii) acceptance of risk. These categories have been used as sub-headings to compile the results that concern this question.

Health risks

Ten papers reported findings pertaining to HCWs' perceptions of their exposure to EARIDs as a source of health risks.16, 18-21, 25-29 All the findings were from the descriptive arm of these papers, which utilized surveys to ascertain HCWs' perceptions.

Chong et al (2004)18 conducted a cross sectional survey study with 1310 HCWs who worked in a tertiary hospital in Taiwan during the SARS outbreak over two study periods - the reaction phase (i.e. during the initial stages of infection with increasing infection rates) and the repair phase (i.e. during the latter stages when there were no new infections). Survey results showed that HCWs surveyed in both phases perceived a high risk of personal infection and low survival chance if infected. They were also fearful of falling ill and felt that they had little control over whether they would be infected.

Similar findings were found by Koh et al (2005)21 who examined the risk perceptions of HCWs (N=10511) working in nine major health care institutions in Singapore during the late stages of the SARS epidemic through the use of a self-administered descriptive survey. It was found that 66% of the HCWs felt that they were at great risk of exposure to SARS and 76% felt fearful of being infected.

Nickell et al (2004)16 likewise identified that 65% of respondents (N=2001) were concerned about their health in the study conducted to explore the psychosocial impact of SARS on HCWs working in a tertiary care institution in Toronto. This study similarly entailed the use of a self-administered descriptive questionnaire.

Another study (Imai et al, 2005)20 was conducted to examine HCWs' perceptions of risks in seven tertiary-level hospitals in Japan (which did not encounter any patients during the international SARS outbreak) through the use of a self-administered questionnaire (N=7463). Results indicated that 55.3% of surveyed HCWs felt fearful of falling ill and around 60% felt that they had little personal control over whether they would fall ill.

Contrary results, however, were found in Grace et al's (2004)19 cross sectional survey study which examined the degree of perceived occupational risk of contracting SARS among physicians (N=193) during the SARS outbreak in Toronto university hospitals. Physicians were asked to rate their likelihood of contracting SARS on a scale from one ("not likely") to five ("extremely likely") based on a dependent variable of perceived risk. Their mean personal perceived risk of contracting SARS was 1.74, which indicated that they perceived themselves unlikely to contract SARS. The physicians' low risk perceptions were attributed to their high self-rated health status as well as their low contact with the infectious diseases.

Another study (Balicer et al, 2006),25 was conducted to assess HCWs' risk perceptions towards a possible local outbreak of pandemic influenza and its influencing factors in county health departments in Maryland, United States of America. Self-administered anonymous surveys were sent to all HCWs and completed by 308 respondents. Results showed that 66% of the respondents perceived themselves to be at personal risk as a result of their job.

Similarly, Cheong et al (2007)26 conducted a study to determine HCWs' concerns and preparedness for an avian influenza pandemic. Anonymous self-administered questionnaires were given to HCWs (N=1234) working at community hospitals (CH) and tertiary hospitals (TH) in Singapore. Results indicated that the majority of TH HCWs (78.3%) and CH HCWs (67.5%) felt that they were at great risk of exposure to avian influenza as they perceived themselves to be at risk of exposure to the pandemic. Chi-square analysis further found that this proportion of TH HCWs was significantly larger than the proportion of CH HCWs who reported similar perceptions (p=0.012). That is, TH HCWs perceived themselves to face greater health risks as compared to CH HCWs. This was because the patients CH HCWs cared for were more likely to be free from infection as they were usually screened in TH before their transfer to the CH, which served as an intermediate care facility for patients who were discharged from TH. These findings underscore the relationship between the HCWs' nature of work and their perceptions of risks.

Another study (Wong et al, 2008)29 which was conducted in Singapore examined the concerns of primary care physicians working in both private (n=136) and public primary care (n=149) outpatient clinics towards a possible avian influenza pandemic. A self-administered descriptive questionnaire was used. It was found that the majority of primary HCWs in both groups felt that they were at risk of infection with avian influenza (95%) and were afraid of falling ill with the disease (90%).

In Imai et al's (2008) study,27 questionnaires were administered to HCWs working in seven tertiary-level hospitals in Japan in order to determine their perceptions towards a hypothetical avian influenza pandemic. Results showed that 65% of the HCWs surveyed felt fearful towards the pandemic influenza.

Tzeng and Yin (2006)28 likewise conducted a cross-sectional study to survey nurses' fears and related factors when faced with possible avian flu pandemic with human-to-human transmission in Taiwan. The respondents (N=225) were conveniently sampled from nurses who were attending a two-year Bachelor degree nursing program in a private university. The study found that 41% of the participants felt fearful of infection.

Social risks

Six papers found that HCWs perceived social risks as a result of their exposure. Social risks included stigmatization to family and self, as well as risk of infection to others (i.e. family, friends and colleagues). Specifically, one paper focused only on stigmatization to themselves and their family;23 and five papers focused on both stigmatization and risks of infection to others.16, 18, 21, 26, 29 All studies focused on HCWs' experience of SARS and the results were from the descriptive arm of the studies.

Styra et al's (2008)23 conducted a cross-sectional study which utilized self-report questionnaires to examine the psychological effects of the Toronto SARS outbreak on HCWs (N=248) working in clinical units which were considered high risk for contracting SARS (i.e. Intensive Care Unit, special SARS unit, and the Emergency Department) in tertiary care health care institutions. In the study, 60% of the respondents noted that their friends and neighbours were avoiding them. Additionally, 36% reported that people were avoiding their family members because of concerns of contracting SARS.

Koh et al (2005)21 found that around half of the HCWs surveyed (49%) perceived that they were being avoided by others because of their job. Thirty-one percent also thought that their family members were being avoided by others. Additionally, the majority were concerned about inadvertent SARS transmission to family, friends and colleagues (89%) and believed that their loved ones were likewise worried about being infected with the disease from contact with the HCW (69%).

In contrast, perceived stigmatization was found to be low amongst employees surveyed by Nickell et al (2004)16 where only 28% reported being treated differently by people because of their work in the hospital. In addition, only 38% reported changes to personal and familial lifestyle such as avoiding public spaces (e.g. restaurants and shopping centres) and avoiding interaction with friends and family as a result of the SARS outbreak. However the majority (63%) of employees were concerned about infection of family members.

In Chong et al's (2004)18 study, HCWs reported fears of being stigmatised and rejected by others, however, no mention was made of whether they had indeed suffered stigmatization. HCWs were also afraid of inadvertent SARS transmission to others (i.e. their friends, family and colleagues) and perceived that their friends and family were worried about being infected by them.

The study conducted by Wong et al (2008)29 found that the majority (81.4%) of the primary care physicians surveyed from both private and public health care settings expressed concerns about their family members being at risk of infection with avian influenza as a result of their jobs. The majority of the physicians (75%) also perceived that others would avoid them and their family members during a possible avian influenza pandemic.

Cheong et al (2007)26 found that TH HCWs were more significantly concerned about inadvertent transmission of avian influenza to their loved ones as compared to CH HCWs (p=0.001). A significantly higher proportion of TH HCWs also perceived that people would avoid them (p=0.002) and their family members (p=0.034) because of their occupation as compared to CH HCWs. In short, these study findings suggests that TH HCWs perceived themselves and their families to face greater health risks and social risks (i.e. stigmatization) compared to CH HCWs. This difference was suggested by the authors to be due to the TH HCWs' perceived greater exposure to avian influenza due to the nature of the patients they saw. Hence, this further emphasizes the importance of the nature of the patients they are exposed to in influencing their perceptions.

Acceptance of risks

Four studies20, 21, 27, 29 which were described in the preceding sections found that despite their fears, HCWs were accepting of the risks of caring for SARS-infected or avian influenza-infected patients. Koh et al (2005)21 indicated that the majority of HCWs surveyed (70%, n=7384) accepted that the risk came with their job. Imai et al (2005)20 reported that around 64% of its respondents were accepting of the risks. Wong et al (2008)29 similarly reported that 82.5% of its respondents accepted the risk of contracting avian influenza as part of their jobs and Imai et al (2008)27 reported that 75% of its respondents were accepting of the risks of being infected with avian influenza.

What are the socio-cultural, psychological, attitudinal, and environmental factors influencing HCWs' risk perceptions?

Several factors affecting HCWs' risk perceptions were identified in the included studies.16, 18-21, 25, 28 As the constructs of risk perceptions (i.e. health risks, social risks and acceptance of risks) were inter-related, the factors influencing their perceptions have been discussed together.

Chong et al (2004)18 compared the risk perceptions of two groups of HCWs during the reaction phase and repair phase of the SARS outbreak using chi-square tests and reported that: HCWs in the reaction phase perceived themselves to be at significantly higher risk of infection (χχ=120.1; p<0.001); had a greater fear of falling ill (χχ=150.03; p<0.001); perceived little control over whether they would be infected (χχ=124.91; p<0.001); were more afraid of transmitting SARS to others (i.e. their friends, family, colleagues and patients) (χχ=291.42; p<0.001); and perceived that their friends and family were worried about being infected by them (χχ=155.31; p<0.001). However, there were no significant differences between both groups' perceived chances of survival if infected with SARS (χχ=1.76) and acceptance of risks (χχ=0.73). In summary, HCWs in the reaction phase perceived significantly higher health risks (i.e. infection; falling ill) to themselves and their families compared to HCWs in the repair phase. However, both groups perceived equally that they had low chances of survival if infected. The HCWs' higher risk perceptions in the reaction phase was postulated by the authors to be due to the level of uncertainty and lack of preparedness and safeguards that characterized the reaction phase in contrast to the repair phase. These findings clearly highlight the importance of pandemic preparedness and its influence on risk perceptions.

In another study (Grace et al, 2004),19 a univariate analysis of co-variance (ANCOVA) analysis was used to examine demographic, logistic, and attitudinal factors affecting physicians' risk perceptions. The following results were found: the perceived likelihood of infection was significantly higher in those who personally knew someone who had contracted SARS (F=6.08; p<0.02) and had previous exposure to an infectious agent (F=5.51; p<0.02) when compared to those who perceived lesser risk. The authors postulated that these results were due to the phenomenon of availability heuristic. That is, individuals were more likely to perceive a higher likelihood of an event occurring (i.e. infection) if they were able to easily recall instances (i.e. infected patients and colleagues) associated with it.

In Imai et al's (2005)20 study, student T-tests were used to compare the differences in risk perceptions between different demographic groups (i.e. age, profession, sex and workplace) of HCWs surveyed. Specifically, perceptions of fear was found to be significantly higher in nurses (p<0.001), female HCWs (p<0.001) and those who were working in non-university hospitals (p<0.001) compared to physicians, male HCWs and those who were working in university hospitals respectively. Similarly, nurses, female HCWs, and younger HCWs had lower acceptance of risks compared to physicians, male HCWs and older HCWs (p<0.001). These results indicate that HCWs who were nurses, female and worked in non-university hospitals were significantly more likely to perceive fear and have low acceptance of risks compared to HCWs who were physicians, male and worked in university hospitals. Logistic regression analyses further found that HCWs' acceptance of risk was positively and significantly correlated with HCWs' perceived availability of institutional measures (OR 1.18; CI 1.12-1.24), and in nurses who were above 35 years of age (OR 1.27; CI 1.06-1.51). That is, individuals who perceived institutional measures to be available were more likely to be accepting of risk and authors concluded that this was because these measures were perceived to be protective against infection.

Multiple logistic regressions were also used in Koh et al's (2005)21 study to determine factors that the following factors that were significant in increasing HCWs' perceptions of health risks. Results showed that those who: were working in a SARS-affected hospital; were exposed on a daily basis to SARS patients; were a doctor, nurse, or ambulance crew as compared to other occupations; and had a high Impact of Event Scale (IES) score (OR 1.88; 95% CI 1.68-2.09), indicating higher emotional distress, had higher perceptions of health risks. HCWs with higher IES scores were also more likely to feel stigmatized (OR 2.00; 95% CI = 1.80-2.21). From these findings, it is evident that HCWs who were in greater exposure to infection (i.e. working in SARS-affected hospitals; and working in a frontline position) were more likely to have higher risk perceptions. In addition, these findings also highlight the positive relationship between an individual's level of emotional distress and their risk perceptions. Occupation was found to be a significant predictor of an individual's acceptance of risk of being infected with SARS, with doctors being significantly more ready to accept the risk than all other categories of HCWs.

Nickell et al (2004)16 found that more nurses were concerned about their health during the SARS outbreak (76.3%, n=363) in contrast to doctors (60.1%, n=104). Logistic regression further found that concern for personal and family health was higher in HCWs who: perceived greater risk of death from SARS (OR 5.0; 95% CI 2.6-9.6); stayed with children (OR 1.8; 95% CI 1.5-2.3); suffered stigmatisation (OR 1.6; 95% CI 1.2-2.1); and/or whose personal or family lifestyle was affected (OR 3.3; 95% CI 2.5-4.3) as a result of the outbreak. In contrast, those who were working in a management or supervisory position (OR 0.6; 95% CI 0.4-0.8), perceived that the precautionary measures in the organization were sufficient (OR 0.4; 95% CI 0.3-0.5) and were aged 50 years and above were less concerned (OR 0.6; 95% CI 0.4-0.9). This study identified several significant factors which increased HCWs' perceived personal and family health risks and these were: their high perceived risk of death from infection; experience of stigmatization; and affected family lifestyle. Significant factors which lowered their perceived personal and family perceived health risks were nature of their work (i.e. management or supervisory position); their perceptions that the precautionary measures in the organization were sufficient; as well as being aged above 50 years old.

The following two studies focused on the factors affecting HCWs' perceived risks of exposure to pandemic influenza. Balicer et al25 (2006) conducted a multivariate logistic regression and found that lower risk perceptions among HCWs were associated with their perception of existing knowledge about public health impact of pandemic influenza (OR 4.1; CI 2.3-7.6); family preparation (OR 2.5; CI 1.4-4.3); perceived ability of health care organization to provide timely information (OR 5.4; CI 2.7-10.7); perception of their capacity to effectively communicate risk (OR 4.8; CI 2.6-9.0); and familiarity with their role-specific response requirements (OR 3.5; CI 1.8-6.2). These findings highlight the influence HCWs' perceived personal, familial and organizational preparedness has on lowering their risk perceptions.

Tzeng and Yin28 (2006), through the use of Pearson Chi-square tests, found that nurses' fears were significantly and negatively correlated with current employment in the health care industry (Phi value=-0.160; P=0.016), that is, nurses who were currently working in health care were less fearful about a possible influenza pandemic outbreak. On the contrary, nurses' fears were found to be significantly and positively correlated with their family's fears about the avian influenza epidemic (Phi value=0.564; P=0.000). Therefore, nurses whose families were fearful about the epidemic were highly likely to be fearful as well.

What are the individual strategies adopted by HCWs to manage their risk perceptions when exposed to emerging acute respiratory infectious diseases and what are the key organizational, environmental and individual factors influencing their use of these strategies?

Analysis of the included studies found two main categories of strategies that were used by the HCWs to manage their risk perceptions and they were: (a) behaviour towards their patients; and (b) compliance with personal protective equipment (PPE).

Five of these papers15, 18, 20, 22, 24 focused on HCWs' behaviour towards patients during SARS while four papers25, 26, 28, 29 focused on their intentions during a possible influenza pandemic outbreak. Only one paper was found which focused on HCWs' compliance with PPE.21 Papers regarding other individual strategies adopted by HCWs were found to be methodically unsound and hence were excluded from the narrative summary.

Behaviour towards patients

A study was conducted by Chong et al (2004)18 to examine the willingness of two groups of HCWs from the reaction and repair phases of the SARS outbreak respectively to care for infected patients. Both groups were compared through the use of chi-squared tests. Results indicated that a third of the HCWs from both groups were unwilling to care for infected patients (Reaction phase: 37.0%, n=164; Repair phase: 40.1%; n=129) and there was no significant difference between both groups' intentions to care (χχ=0.73). However, a significantly higher proportion of HCWs thought of resigning during the repair phase as compared to the reaction phase (χχ=73.61; p<0.001). Thus the authors concluded that although there was no significant difference between both groups of HCWs' unwillingness to care for infected patients, there was a significant difference between their intentions to resign. Specifically, more HCWs were inclined to resign during the repair phase.

Another paper (Ko et al, 2005)15 sought to examine nurses' intention and volunteering behaviour to care for SARS patients, and to determine the extent to which the following factors (i.e. attitudes, subjective norms and perceived control) influenced their behaviour. These factors were constructs of the Ajzen's Theory of Planned Behavior (TPB) which was selected as a theoretical framework for the study (Refer to Figure 2)

Figure 3: Ajzen's Theory of Planned Behavior (TPB)
Figure 3: Ajzen's Theory of Planned Behavior (TPB):
15

A cross-sectional correlational design was utilized involving 750 nurses who had provided direct patient care in a hospital in Southern Taiwan during the SARS epidemic. Results from the descriptive survey showed that around 75% of nurses perceived that it was their professional obligation to care for SARS patients. Around 42.7% of nurses surveyed indicated a positive intention to care for SARS patients and 25.4% were willing to volunteer to care for SARS patients. Pearson's correlation found that all the constructs of the TPB model were significantly and positively correlated with the nurses'intentions and volunteering to care for SARS patients (p ≤ 0.01, p< 0.05), with the exception of knowledge. Nurses' intention to care was likewise found to be highly correlated with their volunteering behaviour (r = 0.37; p< 0.01). In short, nurses who: had a positive attitude towards caring; were influenced by subjective norms to care for their patients; and perceived themselves to have the ability to care for infected patients were highly likely to have the intention to or volunteer to care for their patients, all of which were constructs of the TPB framework. Results further indicate that nurses who had the intentions to care for SARS patients were highly likely to volunteer to care for them.

A multiple regression analysis (i.e. Path analysis) further showed that several constructs of the TPB model were significant predictors of their behaviour. Four factors from the model were found to significantly make up 35% of the overall variance affecting nurses' intentions to care for SARS patients. These factors were: self-efficacy (25% of variance; β = 0.39; p ≤ 0.001); attitude (6% of variance; β = 0.25; p ≤ 0.001); number of working years in the participating hospital (2% of variance; β = -0.15; p ≤ 0.001); and receiving resources from the hospital (2% of variance; β = 0.13; p ≤ 0.001). In contrast, only two factors from the model were found to significantly affect nurses' volunteering behaviour, and these two factors made up only 15% of the total variance. These factors were nurses' intention (13% of variance; β = 0.31; p ≤ 0.001) and attitude (2% of variance; β = 0.15; p ≤ 0.001). Although earlier findings had indicated that most of the constructs of the TPB framework were highly correlated to nurses' intentions to care and their consequent volunteering behaviour, these findings suggest that four constructs had a more significant influence on the nurses' intentions than the rest. In particular, these four constructs were: perceived self-efficacy; attitude; number of working years in the participating hospital; and whether the nurses were receiving resources from the hospital. Having said so, these constructs were not completely predictive of nurses' intentions to care, according to the Path analysis, as they only made up 35% of the overall variance affecting nurses' intentions to care. Likewise, only two constructs from the TPB framework - nurses' intention and attitude towards caring - were found to significantly affect nurses' volunteering behaviour and these were similarly not completely predictive of nurses' intentions to care. In summary, these findings suggest that there are other factors, apart from the ones identified in the TPB framework that may affect nurses' intentions to care and their volunteering behaviour.

In addition, the descriptive arm of the same study also examined the benefits that nurses perceived to be motivating in their intention to care and these were, in order of descending importance: the availability of adequate protective facilities or equipment; a workplace with negative-pressure isolation rooms; receiving continuing education and updated information about the outbreak; and special vacations as well as compensation payments.

Another study (Tzeng et al, 2004)24 examined nurses' intentions and willingness to care and the associated factors through the use of a cross sectional survey design. Data were collected from a convenience sample of 172 nurses working in Taiwan hospitals during and after the SARS epidemic. T-tests comparing nurses surveyed during SARS ('During SARS' group) and nurses surveyed after SARS ('Post-SARS' group) found that those in the 'post-SARS' group were more willing to provide care for infected SARS patients compared to those in the 'during SARS' group. Results further indicated that the 'post-SARS' group of nurses had significantly higher perceived professional obligation (t=2.032; P=0.044), were younger (t=3.033; P=0.003) and had less professional experience (t=3.108; P=0.002). That is, the 'post-SARS' group of nurses were more willing to care for infected patients, and they shared similar characteristics such as higher perceived professional obligation, younger age, and less professional experience, suggesting that these factors may influence their risk perceptions. Logistic ordinal regressions further showed that, in general, nurses' levels of agreement with general SARS infection control measures and the need for quarantine after providing care for infected patients were significant predictors of their willingness to care. The same factors were predictors of the nurses' willingness to care for the 'post-SARS' group. For the 'during-SARS group', only the nurses' level of agreement with general infection control measures was the significant predictor of their willingness to care for patients with SARS.

In Imai et al's (2005)20 study, which examined perceptions of risks amongst HCWs who did not encounter SARS in their health care system, a high proportion (91.7%) of these HCWs expressed intentions to avoid infected SARS patients. Twenty-seven percent also considered changing jobs in order to avoid infected patients and it was found that a higher proportion of nurses than physicians considered avoiding patients (93.4% vs. 69.5%) or changing jobs (34.1% vs. 14.3%). Student T-tests further showed that the number of female HCWs, younger HCWs as well as those who worked in non-university hospitals who expressed intentions of avoiding patients or changing jobs were significantly higher than the number of male HCWs, older HCWs and those working in university hospitals respectively (P<0.001). A logistic regression model further found that avoidance was significantly higher in fearful HCWs (OR 2.31; CI 1.91-2.80) and female HCWs (OR 1.93; CI 1.59-2.33). On the contrary, avoidance was significantly lower in those who were aged 35 years and above (OR 0.81; 0.67-0.97) and who perceived institutional measures to be sufficient (OR 0.81; CI 0.75-0.88). In short, HCWs who: were female, were younger, worked in non-university hospitals; were fearful; and perceived that the institutional measures were insufficient were more likely to avoid their patients or change jobs. Hence, the authors suggested that there was a need to address this group of HCWs in order to ensure that they are well-prepared to face the challenges associated with such pandemics.

Shiao et al (2007)22 examined how nurses perceived their risk from SARS, their intentions to leave the workforce and the factors influencing their intentions, including their perceived risks from SARS. A self-administered anonymous questionnaire was given to nurses working in community hospitals, secondary and tertiary referral hospitals during the SARS outbreak in Taiwan (N=753). Results showed that a minority of nurses felt that they should not be caring for patients with SARS (12.2%) while others stated that they intended to resign or were looking for another job as a result of their perceived risks (25.9%). A small percentage (7.6%) of nurses surveyed expressed both their reluctance to care for SARS patients and their intentions to resign or look for another job. The factors influencing their willingness to care for SARS-infected patients were examined through the use of a multiple logistic regression. It was found that nurses: with shorter tenure (<8.3 years); who perceived that they had increased workload (49.9%; OR 3.73, CI 1.82-8.24, p=0.0006); who perceived that they were being avoided by others because of their job (32.4%; OR 2.67; CI 1.37-5.30; p=0.0043); who perceived a high chance of dying from SARS than cancer were significantly more likely to manifest such intentions (18.6%; OR 4.70; CI 2.43-9.16; p<0.0001). In contrast, increased contact with SARS patient was found to be significantly associated with a lower likelihood of nurses leaving the workforce (14.7%; OR 0.32; CI 0.09-0.89; p=0.0471). Nurses' age; marital status; perceptions of organizational support (i.e. the implementation of clear policies to follow); perceived risk of contracting SARS; perceived emotional support (i.e. appreciation from the hospital, clinic and employers); and perceived availability and effectiveness of personal protective equipment were not significant in influencing their intentions to avoid the patients or leave the workforce. These findings highlight the factors which were significant in determining nurses' willingness to care for infected patients and these were: length of tenure; workload; perceived stigmatization; perceived chances of death; and amount of contact with SARS patient.

Similar findings were found in Balicer et al's (2006)25 study which examined public health employees' risk perceptions and likelihood of reporting to duty during a influenza pandemic outbreak, and the variables that might affect these outcomes. Results of the survey indicated that around half (53.8%) of the HCWs who participated were likely to report to work. A multivariate analysis of the survey results further indicated that clinical staff were significantly more likely to report to duty compared to technical and support staff (OR 2.5; CI 1.3-4.7). Another multivariate analysis found that HCWs' willingness to work was significantly associated with the following factors: their perceived importance of their work role in the outbreak (OR 9.5; CI 4.6-19.9); their familiarity with their work role requirements during outbreaks (OR 7.6; CI 3.4-16.9); and their perceived capacity to effectively communicate risk as part of their work role (OR 6.6; CI 3.2-13.5). Therefore, HCWs who: perceived their work role to be important; were familiar with and perceived themselves to be capable of handling the increased work requirements associated with the outbreak were more willing to work.

Tzeng and Yin's (2006)28 study likewise found that the majority (56.9%) of the 225 nurses surveyed were willing to care for infected patients in an influenza pandemic outbreak. Factors significantly associated with nurses' willingness to care were obtained using Pearson's Chi square tests. Perceived sufficiency of infection control measures and equipment (Phi-value = 0.239; P<0.01) and engaging in religious activities (i.e. praying for help at a temple when they or their loved ones were sick) (Phi-value=0.159; P<0.05) were positively correlated with willingness to care. On the contrary, fear about the influenza pandemic outbreak was negatively correlated with willingness to care for patients (Phi-value = -0.152; P<0.05). In short, nurses who: perceived that infection control measures and equipment were sufficient; engaged in religious activities; and did not express fear about the influenza pandemic outbreak were more willing to care for patients.

Similarly, only 26.3% of the TH HCWs and 16.0% CH HCWs surveyed in the study conducted by Cheong et al (2007)26 felt that they should not be looking after patients infected with avian influenza. Chi-square analysis showed a significant difference between the proportion of TH and CH HCWs (p=0.012). Likewise, a greater proportion of TH HCWs reported that they were considering resigning or looking for another job because of the risks (16.5% vs 10.7%; p=0.001). It is evident from these findings that TH HCWs were less willing to care for infected patients and more inclined to resigning as compared to CH HCWs. This was attributed to their perceived level of risks which were higher than those perceived by CH HCWs as they were in contact with more acutely ill patients compared to CH HCWs.

In Wong et al's (2008) study,29 a minority of the physicians surveyed (27.7%) were also unwilling to care for patients infected with avian influenza and 10% considered looking for another job as a result of the risks. Specifically, 36.2% of the primary physicians in the private settings felt unwilling to care for infected patient and this proportion was significantly greater than the proportion of primary physicians in the public settings who had the same perceptions (20.1%; p=0.003). These findings underscore the private primary care physicians' unwillingness to care for infected patients as compared to public primary care physicians. The authors suggested that this difference was likely because public primary care physicians perceived the public health care organizations to be better able to take care of their medical needs in contrast to private organizations.

Compliance to preventive measures

The only study examining HCWs' compliance to institutional preventive measures was Koh et al's (2005) study.21 The study found that the majority of the respondents reported experiencing little difficulty complying with the measures (72%) and most of the staff were compliant to the measures (92%). Multiple logistic regression reported that nurses perceived significantly greater difficulty in adhering to preventive measures compared to other occupational groups (i.e. doctors) (OR 0.49; 95% CI 0.39-0.61). On the contrary, HCWs who had longer work experience perceived significantly less difficulty in adhering to preventive measures compared to those with shorter work experience (OR 1.21; 95% CI 1.13-1.29). In short, HCWs who were nurses and had shorter work experience were more likely to experience difficulty in adhering to preventive measures as compared to other HCWs.

What are the organizational strategies implemented and how do HCWs perceive the effectiveness of these strategies?

Three papers examined the organizational strategies implemented by the organizations during the SARS outbreak and HCWs' perceived effectiveness of these strategies.16, 19, 21 Four papers focused on HCWs' perceived effectiveness of hypothesized organizational strategies during a possible influenza pandemic outbreak.25, 26, 28, 29

In Grace et al's (2004)19 study, descriptive survey results indicate that the majority (88.1%, n=170) of the physicians surveyed thought they had been given sufficient and appropriate information regarding proper infection control precautions to be taken against SARS. These infection control precautions were also perceived as being effective in limiting the spread of SARS (mean: 3.79 based on a Likert scale of one to five with five being most effective).

Another study (Nickell et al, 2004)16 found that infection control measures were utilized in the hospital and 74.1% of the respondents (n=1460) perceived the measures to be adequate in preventing SARS infection. Respondents' perceived adequacy of the measures was, in turn, significantly associated with lower perceived personal and family health risk (OR 0.4; 95% CI 0.3-0.4) according to a logistic regression analysis. That is, respondents who perceived these measures to be adequate had lower personal and family risk perceptions. However, respondents also reported that these precautionary measures were bothersome. Mask-wearing, in particular, was cited by around 85.4% of the respondents (n=1710) to be the most bothersome precaution. Specifically, the majority of these respondents (92.9%) reported that the masks were a source of physical discomfort and around half of them (47%) reported that they had difficulties communicating with the masks on.

Descriptive survey results in Koh et al's (2005)21 study likewise indicated that the majority of the HCWs: perceived the protective measures implemented at work to be generally effective (96%); were satisfied with the hospital's explanation of the necessity and importance of implementing the strategies (95%); agreed that the policies and protocols were clear enough for everyone to follow (93%); and thought that the policies/protocols were implemented quickly enough (90%).

In Balicer et al's (2006)25 study examining HCWs' perceptions towards organizational strategies, which may be implemented during an influenza pandemic outbreak, 83% of the respondents felt that additional training activities were important. Additionally, 57.1% of the respondents perceived psychological support during the outbreak to be important and around 60% perceived psychological support after the event to be important. Multivariate analysis results indicated that respondents who were willing to report to work during the outbreak significantly perceived psychological support during and after the event to be important (OR 2.4; CI 1.4-4.2 and OR 2.8; CI 1.6-4.8, respectively). These findings indicate that the majority of the HCWs interviewed perceived additional training activities and psychological support during and after the outbreak to be important. A significant positive relationship was also found between HCWs' perceived availability of psychological support and willingness to work, further highlighting the importance of the provision of psychological support.

Wong et al (2008)29 examined the preparedness of primary care physicians (PCPs) in Singapore towards avian influenza and found that 78.7% of the physicians surveyed felt that they were personally prepared for an avian influenza outbreak. Specifically, three-quarters of the PCPs surveyed had taken the influenza vaccination, as recommended, and 84.4% reported being trained adequately in the use of personal protective equipment (PPE). Chi-square analysis further found that PCPs working in the public settings perceived themselves to have more infection control training compared to the PCPs in private health care settings (82.8% vs. 52.8%, P <0.001). The public PCPs also perceived that they someone to turn to for help in use of PPE (92.6% vs. 61.0%, P <0.001) than private PCPs. Public PCPs were also more likely to feel that their workplace was prepared (93.2% vs. 65.5%, P <0.05) and to have a clinic preparedness plan (98.3% vs. 63.6%, P = 0.001). However, private PCPs were more likely to have bought anti-bird flu medications (60.7% vs. 1.4%, P <0.001) and N95 masks for themselves. These findings highlight the difference in the perceived level of preparedness between public and private PCPs. In short, public PCPs perceived a greater availability of the following compared to private PCPs: amount of infection control training; technical support in the use of PPE; work place preparedness and availability of a clinic preparedness plan. However, the findings also indicate that private PCPs were more likely to have made personal preparedness plans such as buying influenza medication and N95 masks for themselves as compared to public PCPs who were more reliant on their organization for such support.

In Cheong et al's (2007) study,26 the majority of HCWs had received training for infection control (87.9%), were aware of the hospital's outbreak preparedness plan (87.2%) and had been informed about the plan (77.8%). As such, the majority of the HCWs (71.6%) felt that they were personally prepared and 82.8% felt that their institutions were prepared. Chi square analysis showed that HCWs working in tertiary hospitals were more likely to feel personally prepared (74.1% vs 64.7%, p=0.045) and perceive that their hospital was prepared for the Avian influenza pandemic compared to HCWs from community hospitals (86.8% vs 71.8, p=0.000). Specifically, a significantly greater percentage of HCWs from tertiary institutions reported having had training (90.0% vs. 82.2%, p=0.014), were aware of the institution's preparedness plan (91.4% vs. 75.5%, p=0.000) and were informed about the plan (82.3% vs. 65.3%) compared to their counterparts in CH. There was a significantly higher proportion of HCWs from tertiary institutions who attended infection control related meetings (59.8% vs. 50.6%, p=0.039), received influenza vaccination (82.5% vs 63.8%, p=0.000) as well as underwent adequate training in use of personal protective equipment (81.5% vs. 71.8%, p=0.032). In short, these findings indicate that HCWs in tertiary hospitals significantly perceived themselves to be more prepared against avian influenza in contrast to the HCWs in community hospitals, who perceived a lower organizational preparedness.

Finally, in Tzeng and Yin's (2006)28 study, 58.7% of the nurses surveyed perceived that their hospitals would not have sufficient infection control measures and equipment to prevent nosocomial infection in an outbreak of avian influenza. Pearson's chi square tests further found that these perceptions were significantly and positively correlated with their fear towards the pandemic (Phi-value=0.239; P<0.01). That is, nurses who perceived that the hospitals were unprepared for the pandemic were more fearful, thereby highlighting the relationship between nurses' perceived organizational preparedness and their risk perceptions.

Results for the qualitative component of the review

Two qualitative papers30, 31 were used in the meta-synthesis.

Straus et al (2005)30 conducted semi-structured, individual telephone interviews with 14 physicians working in specialties involved with the care of SARS inpatients (i.e infectious diseases, critical care medicine and general internal medicine) from hospitals in Toronto, Canada to explore their risk perceptions towards SARS. Data were analysed using the grounded theory methodology.

Rambaldini et al (2005)31 likewise sought to examine the risk perceptions and attitudes of physicians towards caring for SARS-infected patients in hospitals in Toronto, Canada through use of a grounded theory methodology. Semi-structured open-ended telephone interviews were conducted individually with each participant, and 17 participants were interviewed before data saturation was reached. Grounded theory approach was likewise used in data analysis.

A total of seven findings were included from the two qualitative papers.30, 31 The findings and illustrations to support those findings are presented below:

Findings

Finding 1: Balance between care of patients and accepted personal risk [C]

Illustration 1: SARS has made everybody think about would I participate in a high risk procedure with a SARS patient? And I think most of us have come to the conclusion that yes we would as long as we were well informed about what the risk was and as long as we were provided with the appropriate protection… But I'm sure everybody has thought about where the line is now that they would draw (p. 2).30

You see somebody for example with SARS who desaturates, and you're not allowed to enter the room… until we've assumed appropriate precautions. It takes quite a while to get dressed… the whole time you're looking at somebody through a video screen or through the window and you're praying that they're still alive by the time you get in the room (p.2).30

Finding 2: Physicians felt that it was their professional obligation to continue caring for their patients [U]

Illustration 2: I don't think it's appropriate for healthcare workers to refuse to look after SARS patients or any other patient. As healthcare professionals we chose this field and that's what we do (p.2).30

A sense of "9-to-5ism" that medicine never was that's been slowly emerging over the past few years… in the past where doctors would stay all hours, it's not like that anymore, there's a sense of more that this is a job and less of a profession (p.2).30

Finding 3: Communication between physicians and patients' relatives were impeded as a result of the precautionary measures [U]

Illustration 3: To support somebody over the phone was less than ideal and very difficult. I remember telling, I don't know how many families, that their loved one was going to die and [to] do that over the phone and with SARS, [and] having them die alone, that was even worse (p.2).30

Finding 4: The physicians' work was affected as a result of the precautionary measures [U]

Illustration 4: You are toiling under the most stressful clinical time in your professional career. You have a headache, the mask hurts, you're sweating, and it's impossible to establish any of the usual non-verbal clues with patients. You can't feel things through those gloves, you can't tap things out (p.2).30

Finding 5: The physicians felt isolated from others (including their HCWs and their family members) as a result of the social restrictions implemented by the hospitals and by themselves. [C]

Illustration 5: …Pretty much stopped doing anything outside of work [as] I was concerned about going to the movies, the gym, and restaurants (p. 383).31

Finding 6: Physicians were worried about the potential risk of acquiring SARS and the risk they posed to their loved ones. [C]

Illustration 6: There was an element of fear or worry especially at the beginning when information was so limited and changing by the hour (p. 382).31

Finding 7: Some residents felt obliged to care for their patients as they perceived it as their responsibility. However, some felt resentful. [U]

Illustration 7: A lot of the primary care is left to us residents… it's not unusual… I did feel pressured, we really didn't have the opportunity to say no… I do feel a lot of resentment (p. 383).31

I think as part of the medical profession this is a very principle of why people are in medicine, whether you are a nurse or a doctor or a front desk clerk, you know, you choose this profession for a reason, and to be tested like this with SARS, it really rings true why medicine was once or even now considered by some people to be a noble profession… these are the sacrifices that you make and I take it as fundamental rather than an option (p. 383).31

These findings were analysed to produce four categories. The categories were further analysed to produce two synthesises.

Synthesis 1

A total of five findings were grouped into three categories from which a synthesis was derived regarding the perceptions of physicians towards caring for SARS infected patients.

Figure 4: QARI view - Synthesized Finding 1
Figure 4: QARI view - Synthesized Finding 1

The first category related to the physicians' perceptions of health risks to themselves and their family members and the second category related to their perceptions of social risks such as social isolation. The third category related to physicians' perceptions of professional obligation to care for their patients, regardless of their risk perceptions.

These three categories were developed into the first synthesized finding:

Synthesis 1: Although physicians perceive personal health risks and social risks to themselves from caring for patients with SARS, they also recognize that it was their responsibility and professional obligation to care for their patients.

Therefore, there is a need for the implementation of educational and support strategies at national level to enhance physicians' and other HCWs' sense of responsibility and professional obligation, as well as to address and minimise their fear for social and health risks. This will help to ensure that they will continue to care for their patients in the face of personal health and social risks in future pandemics.

Synthesis 2

A total of two findings were grouped into one category relating to the negative effect the precautionary measures implemented during the SARS infection period had on the physicians and their daily practices.

Figure 5: QARI view - Synthesized Finding 2
Figure 5: QARI view - Synthesized Finding 2

This resulted in the following synthesis:

Synthesis 2: The implementation of precautionary measures during the SARS epidemic period affected the daily practices of physicians negatively.

Therefore, there is a need for a strategy, such as education or incentives, to instil the importance of adherence to precautionary measures against EARIDs regardless of the negative impact these measures had on their daily practices.

Discussion

Often pandemic disease outbreaks are found to affect HCWs disproportionately due to the nature of their work and close interaction with infected patients. During the SARS outbreak, 21% of the SARS victims worldwide were found to be HCWs.32 This review reveals how HCWs perceive their exposure to these EARIDs such as SARS and potential pandemic influenza. It further illustrates the relationship between their perceptions of risks, the individual and organizational strategies which were implemented in response to their exposure, and the influencing factors that governed both. From this review, general observations may be made which may form the basis for new practices within health care organisations and grounds for new research.

Discussion of quantitative results

Perceived risks

Specifically, HCWs' risk perceptions were found to be comprised of the following constructs: risks to health,16, 18-21, 25-29 social risks (i.e. stigmatization and health risks to family and self)18, 21, 23, 26, 29 as well as risk acceptance.20, 21 Their perceptions of risks to their health were not unexpected as such pandemic outbreaks were often associated with high morbidity and mortality rates.21 Perceptions of social risks, that is, stigmatization and fear of transmission to family members were also found to be of key concern to HCWs. These findings were likewise not unexpected as such diseases, unlike any other disease, were transmitted via social contact among populations. As such, their risk perceptions would undoubtedly entail a social dimension.16 Despite these risks, the majority of HCWs in the evaluated studies were found to be accepting of these risks as part of their job roles.20, 21, 27, 29 This suggests that their perceived professional responsibility as HCWs was capable of overriding their perceived risks to themselves.21

The findings go on to show that individual factors, organisational factors and factors at the broader societal level play key roles in HCWs' risk perceptions. For example, HCWs who:

  • Were female;20
  • Were younger;16, 20
  • Were married with children;16
  • Were nurses compared to other HCWs (i.e. physicians);16, 20, 21
  • Were working in private primary health care settings;29
  • Were working in tertiary hospitals;26
  • Had higher Impact of Event Scale (IES) scores, indicating emotional distress;21
  • Perceived families to be fearful and unprepared;25, 28
  • Personally knew someone, especially colleagues, who had contracted SARS and previous exposure to infectious agents;19 and/or
  • Perceived great risk of death from SARS16 were more likely to have higher risk perceptions and lower acceptance of risks.

Additionally, different forms of risk perceptions were found to be inter-related. That is, HCWs who perceived that they were stigmatized and/or whose lifestyles were affected as a result of the SARS outbreak were more likely to be more concerned for their personal and family's health.16 All of these suggest a need to identify HCWs who possess these characteristics, predisposing them to higher risk perceptions, and provide them with relevant support to help mitigate their risk perceptions.

The findings also suggest that HCWs' perceived level of control over the situation at the macro-and micro-level was crucial in affecting their risk perceptions. For example, HCWs were found to have higher risks perceptions during the early phases of the SARS outbreak,18 as it was marked by uncertainty. Those who felt that they had little information about the public health impact of influenza pandemic outbreak similarly had increased risk perceptions.25 In other words, they deemed themselves to have little control at the wider societal level, that is, the macro-level, which in turn increased their risk perceptions.

Similarly, HCWs who perceived that they were in control at their work places, that is, at the micro-level were found to report lower risk perceptions. For example, those who:

  • Worked in management or supervisory positions;16
  • Were older with more work experience;16
  • Were more familiar with their job role requirements (i.e. being able to effectively communicate risk);25 and/or
  • Perceived that the organisation was able to provide appropriate and adequate updates and sufficient training activities were less likely to perceive risks during both SARS and influenza pandemic.25

This was because they were better able to cope with the emergency situations associated with outbreaks. As such, HCWs who were not currently working in the health care industry were more likely to perceive risks to themselves in the event of an influenza pandemic outbreak.28 Hence, there clearly is a need to empower HCWs during such crises both at the macro-and micro-level, where possible, to ensure that they are fully equipped with the knowledge and ability to cope with the uncertainties as well as their additional roles and responsibilities during such situations.

Finally, organisational factors (i.e. nature of work and organisational safeguards) were also crucial in influencing their risk perceptions as they were predictive of the extent to which HCWs were exposed to infected patients. For example, HCWs who were exposed to SARS patients on a daily basis such as those working in SARS-affected hospitals and/or as nurses or physicians were more likely to perceive increased risks to themselves.21 In contrast, those who perceived organisational safeguards (i.e. protective equipment and other implemented institutional measures) were available and effective in protecting them were found to have lower risk perceptions.20 These findings suggest that health care organizations should ensure that their HCWs, especially those who are frequently exposed to infected patients, are sufficiently protected with institutional measures and protective equipment.

Risk-mitigating strategies

This review found three main categories of strategies that were identified by HCWs as risk-mitigating and these were strategies which were individually implemented by HCWs (i.e. behaviour towards patients and compliance with protective measures) and those implemented by the organization.

Behaviour towards patients

Findings pertaining to HCWs' behaviour towards their patients suggest that HCWs' willingness/intentions to care, volunteering to care for patients and intentions to change their jobs were distinct concepts. For example, in the studies focusing solely on the SARS outbreak, it was found that the majority of the HCWs were unwilling to care for or avoided infected patients.15, 18, 20 However, the proportion who considered resigning was lower than the proportion of those who had expressed unwillingness to care in the same study,20 indicating that these two concepts were distinct and possibly affected by different factors.

This was likewise for HCWs' volunteering behaviour where there was a lower proportion of HCWs who volunteered to care for infected patients in comparison to those who had expressed their intention to care for infected patients.15 This was further affirmed by a logistic regression analysis done in the same study which found that HCWs' expressed intentions to care only explained 13% of the variance in their volunteering behaviour to care for infected patients.15 Hence, these two issues suggest that HCWs' intentions do not necessarily translate into actual behaviour.

In addition, these findings indicating the negative behavioural intentions of HCWs towards their patients also contradict the findings in the preceding section which indicated that HCWs were mostly accepting of their risks as health care professionals.20, 21 However, it has to be acknowledged that only one study20 examined both HCWs' acceptance of risks and willingness to care in the same survey. Regardless, these results suggest that there is a complex psychology linking these concepts, each of which impacts the health care organization and work force differently. Hence there is a need for more research to be carried out in order to determine whether HCWs' willingness to care, volunteering behaviour, intentions to resign and their acceptance of risks were indeed distinct concepts and if so, to ascertain the factors influencing these differences.

In studies examining HCWs' willingness to care during a possible influenza pandemic outbreak, it was found that the majority were willing to care for infected patients. This is in contrast to aforementioned findings pertaining to the SARS outbreak. As HCWs' perceptions towards an influenza pandemic outbreak and SARS were examined in separate studies, it is not known if their perceptions were significantly different. As such, there is also clearly a need for research to be done which directly compares the differences between HCWs' behaviour during SARS and a possible outbreak of influenza pandemic in the same study, and its influencing factors.

Within this review, several factors were found to decrease HCWs' willingness to care for their patients and these included demographic factors (i.e. nurses, female HCWs, younger HCWs and less professional experience)20, 24 as well as HCWs' perceived:

  • Insufficiency of infection control measures, facilities (i.e. negative pressure isolation rooms) and equipment;15
  • Inability to handle additional job requirements (i.e. communicate risk);25
  • Uncertainty within the organization and the environment;18 and
  • Unavailability of continuing education and information about the outbreak.15

It is evident that these factors are similar to those affecting their risk perceptions, suggesting that these factors may have affected HCWs' willingness to care through indirectly increasing their risk perceptions. However, this relationship is not clearly stipulated in the studies and this signifies the need for more studies to be conducted for better understanding of the relationship between these factors, risk perceptions and HCWs' behaviour towards their patients.

Apart from demographic and organizational factors, other factors were also found to be important in increasing HCWs' willingness to work and these were their:

  • Perceived importance of their job role within the organization;25
  • Perceived social and organisational norms, that is, HCWs' perceived obligation as a HCW and their perceived necessity to comply with expectations of others (i.e. superiors);15
  • Perceived availability of organisational incentives (i.e. compensation payments, special vacations, and not having to be quarantined after caring for infected patients);15 and
  • Personal practice of coping measures such as engaging in religious activities.28

These findings indicate that there are many variables which influence an individual's behavioural intentions or behaviour besides their risk perceptions. The relationship between HCWs' perceived norms and organizational incentives in influencing HCWs' willingness to care was examined as part of a theoretical model - Theory of Planned Behaviour (TPB) - by Ko et al (2004).15 Through the use of Pearson's correlation, it was shown that these two constructs were positively correlated with nurses' intentions to care and their volunteering behaviour. However, these factors were not significant in a multivariate analysis which was further conducted. Although the influence of the aforementioned factors cannot be discounted, it is necessary to carry out further studies which are based on such theoretical models in order to ascertain the extent of influence norms and incentives may have. The studies also provided little insight into how HCWs' personal practice of religious activities may affect their willingness to care for patients, hence warranting the need for further studies to explore this relationship.

Compliance with preventive measures

Only one methodically sound paper21 was found to examine HCWs' compliance to preventive measures within the health care organisation. Findings indicated that the majority of HCWs surveyed were compliant. However, findings also indicate that nurses, in contrast to physicians, and HCWs who were younger with less work experience were more likely to perceive difficulty in adhering to these measures.21 The precise relationship between an individual's demographic factors and compliance to preventive measures was not explained hence suggesting a need for more methodologically sound studies to be conducted to fulfil this research gap.

Organisational strategies

This review also found that infection control measures were implemented within health care organizations, and these were perceived by the majority of the HCWs to be: adequate; clearly and sufficiently explained; implemented quickly enough; and effective in limiting the spread of SARS.16, 19, 21 These perceptions of the organizational strategies were found to be risk-mitigating.16 In studies examining HCWs' perceptions of a possible influenza pandemic outbreak, the following organizational factors were identified as important. They were the provision of: training activities; psychological support; sufficient infection control measures and equipment. These were also significant in lowering their perceptions of fear and increase their likelihood of working during the outbreak.25, 28 Clearly, the same organizational factors were perceived to be important in influencing HCWs' perceptions of risks during both SARS and pandemic influenza. The importance of organizational strategies was also clearly illustrated in preceding sections, hence emphasizing the need for organizations to ensure that these measures are carried out during outbreaks in order to mitigate HCWs' risk perceptions.

However, the precautionary measures, specifically mask-wearing, were also reported to be a source of discomfort16 and a form of communication barrier.16 It is not known if their perceptions had any influence on their levels of compliance and neither was this relationship examined in the preceding section. Therefore, there may be a need for studies to be conducted to examine the factors influencing their compliance behaviour, which will allow targeted strategies to be carried out to encourage HCWs' adherence to protective measures.

Discussion of qualitative results

Findings from the qualitative arm were similar to those in the quantitative arm of this review. The findings indicate that HCWS perceived personal health risks31 and social risks31 to themselves from caring for patients with SARS. In spite of these risks, HCWs were accepting of the risks and willing to continue caring for their patients as they recognized that it was their responsibility and professional obligation to do so.30 These findings highlight the important role of the HCWs' perceived sense of professional obligation in overriding their risk perceptions during such pandemics. Hence, this suggests the need for organizations to implement a strategy to encourage and strengthen this sense of responsibility, in order to ensure that they will continue to care for their patients. Findings also show that the implementation of precautionary measures during the pandemic had negatively affected HCWs daily practices and communication with the patients' family members.30 Although it is not clear whether this had affected HCWs' compliance to the measures, there is a need for the organization to implement strategies to enforce the importance of adhering to these infection-control strategies.

Limitations of the review

There are several limitations in this review. First, only papers that were published in English Language were included due to resource limitations hence key studies in this area which were published in other languages may have been omitted. Second, due to the limitations in the number of databases which were searched as a result of time and resource constraints, key studies may have likewise been omitted. Third, due to the emergent nature of the subject area, more studies may have been conducted since the search strategy was undertaken which were not included in this review.

Conclusions

In conclusion, HCWs perceived many risks to themselves as a result of their exposure to EARIDs. Several key organizational, individual and demographic factors which influenced their perceptions of risks were identified in this review. These factors may serve to form the implications for future practice, which will hopefully mitigate their risk perceptions. The relationship between HCWs' risk perceptions, perceptions towards and use of organizational and individual strategies and influencing factors governing both was far less clear, with nebulous distinctions between these categories. Hence, there is a need for future research to examine the nuances of this relationship.

Implications for practice

Several implications for practice may be derived from the quantitative findings. Each of these recommendations is assigned a level of evidence according to JBI criteria (see Appendix VI). They are:

  • HCWs who possess individual characteristics which place them at risk of negative perceptions (i.e. female, married with children, being a nurse, with higher IES scores, whose families and lifestyles are affected, personally knowing someone who was infected, perceiving greater risk of death and stigmatization) should be identified and employers/policy makers should put into place relevant support to these HCWs in order to reduce their risk perceptions and increase their willingness to care. (Level 3 evidence)
  • HCWs within their work places should be provided with tools (such as education and training) to increase their perceived level of control in coping with the uncertainties and additional responsibilities that are required of them during potentially stressful pandemic outbreaks. This will thereby reduce their perceived risks and increase their willingness to care (Level 3 evidence)
  • Institutions should ensure that appropriate and adequate institutional measures and protective resources such as personal protective equipment are implemented and provided to safeguard HCWs during pandemic outbreaks (Level 3 evidence)
  • Institutions should provide HCWs with organisational incentives such as compensation payments, special vacations and do not quarantine them following their exposure to infected patients. These strategies will increase their willingness to work (Level 3 evidence)

Likewise, several implications for practice may be derived from the qualitative findings. They are:

  • Institutions and the government should provide educational and support strategies at the institutional and national level respectively to enhance physicians' and other HCWs' sense of responsibility and professional obligation, as well as to address and minimise their fear for social and health risks. This will help to ensure that they will continue to care for their patients in the face of personal health and social risks in future pandemics. (Level 3 evidence)
  • Institutions and the government should also provide education or incentives, to instil the importance of adherence to precautionary measures against EARIDs regardless of the negative impact these measures had on their daily practices. (Level 3 evidence)

Implications for research

Future methodically sound cross-sectional quantitative studies should be undertaken to:

  • Compare the differences between HCWs' behavioural intentions to care for patients during SARS and influenza outbreaks, and their influencing factors;
  • Compare and/or correlate the concepts of HCWs' intentions to care or resign, volunteering behaviour, and their acceptance of risks in order to ascertain if they are distinct concepts and if so, determine the factors influencing these differences;
  • Establish the relationship between HCWs' risk perceptions and behavioural intentions and the respective influencing factors in order to establish if the factors affecting their willingness to care directly or indirectly;
  • Ascertain the factors correlated with HCWs' intentions to care and volunteering behaviour based on theoretical models, such as the Theory of Planned Behaviour, in order to determine the extent of influence these factors may have;
  • Ascertain the relationship between HCWs' personal practices of religious activities and how this affects their willingness to care for patients; and
  • Determine the factors influencing HCWs' compliance to personal protective equipment and other infection control measures.

Conflict of interests

No conflict of interest.

References

1. Morens DM, Folkers GK, Fauci AS. The challenge of emerging and re-emerging infectious diseases. Nature. 2004 Jul 8;430(6996):242-9.
2. Fauci AS. Infectious diseases: considerations for the 21st century. Clinical Infectious Diseases. 2001 Mar 1;32(5):675-85.
3. Cutter J. Preparing for an influenza pandemic in Singapore. ANNALS Academy of Medicine Singapore. 2008 Jun;37(6):497-503.
4. Low JG, Wilder-Smith A. Infectious respiratory illnesses and their impact on healthcare workers: a review. ANNALS Academy of Medicine Singapore. 2005 Jan;34(1):105-10.
5. Maines TR, Chen LM, Matsuoka Y, Chen H, Rowe T, Ortin J, et al. Lack of transmission of H5N1 avian-human reassortant influenza viruses in a ferret model. Proceedings of the National Academy of Sciences U S A. 2006 Aug 8;103(32):12121-6.
6. Moscona A. Medical management of influenza infection. Annual Review of Medicine. 2008;59:397-413.
7. Blendon RJ, DesRoches CM, Cetron MS, Benson JM, Meinhardt T, Pollard W. Attitudes toward the use of quarantine in a public health emergency in four countries. Health Affairs (Millwood). 2006 Mar-Apr;25(2):w15-25.
8. The global shortage of health workers and its impact. World Health Organization Media Center; 2010; Available from: http://www.who.int/mediacentre/factsheets/fs302/en/index.html.
9. Leppin A, Aro AR. Risk perceptions related to SARS and avian influenza: theoretical foundations of current empirical research. International Journal of Behavioral Medicine. 2009;16(1):7-29.
10. Maunder R, Hunter J, Vincent L, Bennett J, Peladeau N, Leszcz M, et al. The immediate psychological and occupational impact of the 2003 SARS outbreak in a teaching hospital. Canadian Medical Association Journal. 2003 May 13;168(10):1245-51.
11. Leppin A, Aro AR. Risk perceptions related to SARS and avian influenza: theoretical foundations of current empirical research. Int J Behav Med. 2009;16(1):7-29.
12. Stone PW, Clarke SP, Cimiotti J, Correa-de-Araujo R. Nurses' working conditions: implications for infectious disease. Emerging Infectious Diseases. 2004 Nov;10(11):1984-9.
13. Masur H, Emanuel E, Lane HC. Severe acute respiratory syndrome: providing care in the face of uncertainty. Journal of the American Medical Association. 2003 Jun 4;289(21):2861-3.
14. Nichol K, Bigelow P, O'Brien-Pallas L, McGeer A, Manno M, Holness DL. 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. 2008 Sep;36(7):481-7.
15. Ko N-Y, Feng M-C, Chiu D-Y, Wu M-H, Feng J-Y, Pan S-M. Applying Theory of Planned Behavior to Predict Nurses' Intention and Volunteering to Care for Sars Patients in Southern Taiwan. The Kaohsiung Journal of Medical Sciences. 2004;20(8):389-98.
16. 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. CMAJ. 2004 Mar 2;170(5):793-8.
17. Snacken R, Kendal AP, Haaheim LR, Wood JM. The next influenza pandemic: lessons from Hong Kong, 1997. Emerging Infectious Diseases. 1999 Mar-Apr;5(2):195-203.
18. Chong MY, Wang WC, Hsieh WC, Lee CY, Chiu NM, Yeh WC, et al. Psychological impact of severe acute respiratory syndrome on health workers in a tertiary hospital. Br J Psychiatry. 2004 Aug;185:127-33.
19. Grace SL, Hershenfield K, Robertson E, Stewart DE. Factors affecting perceived risk of contracting severe acute respiratory syndrome among academic physicians. Infect Control Hosp Epidemiol. 2004 Dec;25(12):1111-3.
20. Imai T, Takahashi K, Hasegawa N, Lim MK, Koh D. SARS risk perceptions in healthcare workers, Japan. Emerg Infect Dis. 2005 Mar;11(3):404-10.
21. Koh D, Lim MK, Chia SE, Ko SM, Qian F, Ng V, et al. Risk perception and impact of Severe Acute Respiratory Syndrome (SARS) on work and personal lives of healthcare workers in Singapore: what can we learn? Med Care. 2005 Jul;43(7):676-82.
22. Shiao JS, Koh D, Lo LH, Lim MK, Guo YL. Factors predicting nurses' consideration of leaving their job during the SARS outbreak. Nurs Ethics. 2007 Jan;14(1):5-17.
23. Styra R, Hawryluck L, Robinson S, Kasapinovic S, Fones C, Gold WL. Impact on health care workers employed in high-risk areas during the Toronto SARS outbreak. J Psychosom Res. 2008 Feb;64(2):177-83.
24. Tzeng HM. Nurses' professional care obligation and their attitudes towards SARS infection control measures in Taiwan during and after the 2003 epidemic. Nurs Ethics. 2004;11(3):277-89.
25. Balicer R, Omer S, Barnett D, Everly G. Local public health workers' perceptions toward responding to an influenza pandemic. BMC Public Health. 2006;6:99.
26. Cheong SK, Wong TY, Lee HY, Fong YT, Tan BY, Koh G, et al. Concerns and preparedness for an avian influenza pandemic: a comparison between community hospital and tertiary hospital healthcare workers. Ind Health. 2007 Oct;45(5):653-61.
27. Imai T, Takahashi K, Todoroki M, Kunishima H, Hoshuyama T, Ide R, et al. Perception in relation to a potential influenza pandemic among healthcare workers in Japan: implications for preparedness. J Occup Health. 2008;50:13 - 23.
28. Tzeng HM, Yin CY. Nurses' fears and professional obligations concerning possible human-to-human avian flu. Nurs Ethics. 2006 Sep;13(5):455-70.
29. Wong TY, Koh GC, Cheong SK, Sundram M, Koh K, Chia SE, et al. A cross-sectional study of primary-care physicians in Singapore on their concerns and preparedness for an avian influenza outbreak. Ann Acad Med Singapore. 2008 Jun;37(6):458-64.
30. Straus SE, Wilson K, Rambaldini G, Rath D, Lin Y, Gold WL, et al. Severe acute respiratory syndrome and its impact on professionalism: qualitative study of physicians' behaviour during an emerging healthcare crisis. BMJ. 2004 Jul 10;329(7457):83.
31. Rambaldini G, Wilson K, Rath D, Lin Y, Gold WL, Kapral MK, et al. The impact of severe acute respiratory syndrome on medical house staff: a qualitative study. J Gen Intern Med. 2005 May;20(5):381-5.
32. Tai DY. SARS plague: Duty of care or medical heroism. Ann Acad Med Singapore. 2006;35:374-8.

Appendix I: Initial keywords or terms used in search strategy

Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.

Appendix II: Critical appraisal tools for MASTARI & QARI

JBI Critical Appraisal Checklist for Experimental Studies

Table
Table:
No Caption available.

JBI critical appraisal checklist for cohort and case control studies

Table
Table:
No Caption available.

JBI critical appraisal checklist for descriptive/case series studies

Table
Table:
No Caption available.
Table
Table:
No Caption available.

Appendix III: Data extraction tools for MASTARI & QARI

Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.

Appendix IV: Excluded studies

Alexander, G. C., & Wynia, M. K. (2003). Ready and willing? Physicians' sense of preparedness for bioterrorism. Health Aff (Millwood), 22(5), 189-197.

Reason for exclusion: Does not specifically focus on EARID.

Al-Tawfiq, J. A., Antony, A., & Abed, M. S. (2009). Attitudes towards influenza vaccination of multi-nationality health-care workers in Saudi Arabia. Vaccine, 27(40), 5538-5541.

Reason for exclusion: Focusing on seasonal influenza not pandemic influenza (different outcome)

Bai, Y., Lin, C. C., Lin, C. Y., Chen, J. Y., Chue, C. M., & Chou, P. (2004). Survey of stress reactions among health care workers involved with the SARS outbreak. Psychiatr Serv, 55(9), 1055-1057.

Reason for exclusion: Focusing on psychological reactions and not risk perceptions per se

Bryce, E., Copes, R., Gamage, B., Lockhart, K., & Yassi, A. (2008). Staff perception and institutional reporting: two views of infection control compliance in British Columbia and Ontario three years after an outbreak of severe acute respiratory syndrome. Journal of Hospital Infection, 69 (2), 169-176.

Reason for exclusion: Focusing on resource accessibility and allocation and not HCWs' perceptions of risks of EARIDs

Bryce, E., Forrester, L., Scharf, S., & Eshghpour, M. (2008). What do healthcare workers think? A survey of facial protection equipment user preferences. Journal of Hospital Infection, 68(3), 241-247.

Reason for exclusion: Focus on preferences of facial protection equipment and not their risk perceptions per se

Caputo, K. M., Byrick, R., Chapman, M. G., Orser, B. J., & Orser, B. A. (2006). Intubation of SARS patients: infection and perspectives of healthcare workers. Can J Anaesth, 53(2), 122-129.

Reason for exclusion: Qualitative arm of the paper not presented in a methodically sound qualitative way

Cassidy, I. (2006). Student nurses' experiences of caring for infectious patients in source isolation. A hermeneutic phenomenological study. Journal of Clinical Nursing, 15(10), 1247-1256.

Reason for exclusion: Not focusing specifically on EARID

Caves, N. D., & Irwin, M. G. (2006). Attitudes to basic life support among medical students following the 2003 SARS outbreak in Hong Kong. Resuscitation, 68(1), 93-100.

Reason for exclusion: Focus on basic life support and not risk perceptions per se towards EARID and also based on outside the hospital, not within inclusion criteria.

Chan, A. O., & Huak, C. Y. (2004). Psychological impact of the 2003 severe acute respiratory syndrome outbreak on health care workers in a medium size regional general hospital in Singapore. Occup Med (Lond), 54(3), 190-196.

Reason for exclusion: Focusing more on psychological reaction and not risk perceptions

Chau, J. P., Thompson, D. R., Twinn, S., Lee, D. T., Lopez, V., & Ho, L. S. (2008). An evaluation of SARS and droplet infection control practices in acute and rehabilitation hospitals in Hong Kong. Hong Kong Med J, 14 Suppl 4, 44-47.

Reason for exclusion: Methodically unsound with low response rates. Data analysis, survey process and limitations were not properly described.

Ehrenstein, B. P., Hanses, F., & Salzberger, B. (2006). Influenza pandemic and professional duty: family or patients first? A survey of hospital employees. BMC Public Health, 6, 311.

Reason for exclusion: Little focus on risk perceptions, but focusing more on duty and obligation to care (different outcome)

Godin, G., Naccache, H., & Fortin, C. (1998). Understanding physicians' intention to use a simple infection control measure: Wearing gloves. American Journal of Infection Control, 26(4), 413-417.

Reason for exclusion: Not focusing on EARID specifically.

Gullion, J. S. (2004). School nurses as volunteers in a bioterrorism event. Biosecur Bioterror, 2(2), 112-117.

Reason for exclusion: Not focusing on EARID specifically.

Herman, B., Rosychuk, R. J., Bailey, T., Lake, R., Yonge, O., & Marrie, T. J. (2007). Medical students and pandemic influenza. Emerg Infect Dis, 13(11), 1781-1783.

Reason for exclusion: Methodically unsound with no mention of inclusion or exclusion criteria, data analysis process, IRB approval. Does not really focus on risk perceptions but more on influenza knowledge.

Ho, S. M., Kwong-Lo, R. S., Mak, C. W., & Wong, J. S. (2005). Fear of severe acute respiratory syndrome (SARS) among health care workers. J Consult Clin Psychol, 73(2), 344-349.

Reason for exclusion: Comparing risk perceptions of HCWs who had taken care of SARS patients and those who were infected with SARS themselves thus the population not relevant.

Jiménez-García, R., Hernández-Barrera, V., Carrasco-Garrido, P., Sierra-Moros, M. J., Martinez-Hernandez, D., & De Miguel, A. G. (2006). Influenza vaccination coverages among Spanish children, adults and health care workers. Infection, 34(3), 135-141.

Reason for exclusion: Focus on seasonal influenza and not pandemic influenza.

Koh, D., Takahashi, K., Lim, M. K., Imai, T., Chia, S. E., Qian, F., et al. (2005). SARS risk perception and preventive measures, Singapore and Japan. Emerg Infect Dis, 11(4), 641-642.

Reason for exclusion: The literature review, results and methods of analysis are too brief and there is no mention of ethical perspectives etc.

Lau, P. Y., & Chan, C. W. H. (2005). SARS (Severe Acute Respiratory Syndrome): Reflective practice of a nurse manager. Journal of Clinical Nursing, 14(1), 28-34.

Reason for exclusion: No mention of perceptions of risks.

Maunder, R. G., Lancee, W. J., Rourke, S., Hunter, J. J., Goldbloom, D., Balderson, K., et al. (2004). Factors associated with the psychological impact of severe acute respiratory syndrome on nurses and other hospital workers in Toronto. Psychosom Med, 66(6), 938-942.

Reason for exclusion: Focusing on psychological impact and not risk perceptions

McAlonan, G. M., Lee, A. M., Cheung, V., Cheung, C., Tsang, K. W., Sham, P. C., et al. (2007). Immediate and sustained psychological impact of an emerging infectious disease outbreak on health care workers. Can J Psychiatry, 52(4), 241-247.

Reason for exclusion: Focusing on psychological impact, not risk perceptions

Mok, E., Chung, B. P., Chung, J. W., & Wong, T. K. (2005). An exploratory study of nurses suffering from severe acute respiratory syndrome (SARS). Int J Nurs Pract, 11(4), 150-160.

Reason for exclusion: Focus on nurses who had been infected with SARS not their risk perceptions towards caring for patient with SARS

Moore, D. M., 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. AAOHN J, 53(6), 257-266.

Reason for exclusion: Not focusing on risk perceptions but more on the infection control practices

Nichol, K., Bigelow, P., O'Brien-Pallas, L., McGeer, A., Manno, M., & Holness, D. 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), 481-487.

Reason for exclusion: Focusing more on PPE and the factors affecting it and not the risk perceptions

Noula, M., Raftopoulos, V., Gesouli, E., Tsaprounis, T., & Deltsidou, A. (2008). Greek nursing students' immunization coverage: Data from central continental Greece. Nursing and Health Sciences, 10(3), 169-174.

Reason for exclusion: Focusing on seasonal immunization not pandemic influenza

Phua, D. H., Tang, H. K., & Tham, K. Y. (2005). Coping responses of emergency physicians and nurses to the 2003 severe acute respiratory syndrome outbreak. Acad Emerg Med, 12(4), 322-328.

Reason for exclusion: Focusing more on coping and not risk perceptions per se

Qureshi, K., Gershon, R. R., Sherman, M. F., Straub, T., Gebbie, E., McCollum, M., et al. (2005). Health care workers' ability and willingness to report to duty during catastrophic disasters. J Urban Health, 82(3), 378-388

Reason for exclusion: Focusing on general emergencies and not EARID specifically

Sarikaya, O., & Erbaydar, T. (2007). Avian influenza outbreak in Turkey through health personnel's views: A qualitative study. BMC Public Health, 7.

Reason for exclusion: Not methodically sound with no method or methodology mentioned

Seto, W. H., Tsang, D., Yung, R. W., Ching, T. Y., Ng, T. K., Ho, M., et al. (2003). Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). Lancet, 361(9368), 1519-1520.

Reason for exclusion: Focusing solely on perceived effectiveness of strategies and not on risk perceptions primarily

Shabanowitz, R. B., & Reardon, J. E. (2009). Avian flu pandemic - Flight of the healthcare worker? HEC Forum, 21(4), 365-385.

Reason for exclusion: Little focus on risk perceptions and low return rate

Shih, F.-J., Gau, M.-L., Kao, C.-C., Yang, C.-Y., Lin, Y.-S., Liao, Y.-C., et al. (2007). Dying and caring on the edge: Taiwan's surviving nurses' reflections on taking care of patients with severe acute respiratory syndrome. Applied Nursing Research, 20(4), 171-180.

Reason for exclusion: The representation of the participants' voices do not match the results, that is, the quotes are irrelevant

Smedley, J., Poole, J., Waclawski, E., Stevens, A., Harrison, J., Watson, J., et al. (2007). Influenza immunisation: attitudes and beliefs of UK healthcare workers. Occupational and Environmental Medicine, 64(4), 223-227.

Reason for exclusion: Focusing on seasonal influenza and not pandemic influenza immunisation

Stephenson, I., Roper, J.P., & Nicholson, K.G. (2002). Healthcare workers and their attitudes to influenza vaccination. Commun Dis Public Health, 5(3), 247-252

Reason for exclusion: Focusing on seasonal influenza vaccinations and not pandemic influenza

Watkins, R. E., Wynaden, D., Hart, L., Landsborough, I., McGowan, S., Speed, G., et al. (2006). Perceptions of infection control practices among health professionals. Contemporary Nurse: A Journal for the Australian Nursing Profession, 22(1), 109-119.

Reason for exclusion: Does not focus specifically on risk perceptions of EARID

Wong, T. W., Yau, J. K., Chan, C. L., Kwong, R. S., Ho, S. M., Lau, C. C., et al. (2005). The psychological impact of severe acute respiratory syndrome outbreak on healthcare workers in emergency departments and how they cope. Eur J Emerg Med, 12(1), 13-18.

Reason for exclusion: Focusing on psychological impact, and not risk perceptions

Wong, T. W., & Tam, W. W.-S. (2005). Handwashing practice and the use of personal protective equipment among medical students after the SARS epidemic in Hong Kong. American Journal of Infection Control, 33(10), 580-586.

Reason for exclusion: Focusing more on practices rather than risk perceptions

Appendix V: Included studies

Quantitative studies (Correlational)

Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.
Table
Table:
No Caption available.

Qualitative studies

Table
Table:
No Caption available.

Appendix VI: Level of evidence (quantitative and qualitative)

Table
Table:
No Caption available.

Appendix VII: Level of credibility (Qualitative)

Joanna Briggs Institute Levels of Evidence for Qualitative Studies

Table
Table:
No Caption available.
Keywords:

Health care workers; emerging acute respiratory infectious diseases; risk perceptions; perceived effectiveness; risk-coping strategies

© 2010 by Lippincott Williams & Wilkins, Inc.