key: cord-0981553-pdkjj87a authors: Taylor, Melanie; Raphael, Beverley; Barr, Margo; Agho, Kingsley; Stevens, Garry; Jorm, Louisa title: Public health measures during an anticipated influenza pandemic: Factors influencing willingness to comply date: 2009-01-29 journal: Risk Manag Healthc Policy DOI: 10.2147/rmhp.s4810 sha: 14d04f36cb13550aa7769b61a079fa54031a21eb doc_id: 981553 cord_uid: pdkjj87a This research assessed factors associated with willingness to comply with vaccination, isolation, and face mask wearing during an anticipated influenza pandemic. Data were collected from 2081 adults (16+) using a module of questions incorporated into the NSW Health Adult Population Health Survey. High levels of willingness to comply were reported with 73% either very or extremely willing to receive vaccination, 67% willing to isolate themselves, 58% willing to wear a face mask, and 48% willing to comply with all three behaviors. Further analysis indicated concern for self and family and higher levels of education were associated with high levels of willingness to comply. Younger people (16–24) were the least willing to comply; especially with wearing a face mask. Those with children reported higher levels of willingness to receive vaccination, and respondents who speak a language other than English at home were less willing to isolate themselves or comply with all behaviors. These findings provide a baseline measure of anticipated public compliance with key public health behaviors in the event of an influenza pandemic in the Australian population, and help to identify groups that may be more resistant to individual measures and may require additional attention in terms of risk communication strategies or health education. Willingness of the general public to comply with protective public health measures in the event of pandemic infl uenza is necessary for success in disease response and containment. In addition, better estimates of population compliance with such measures is valuable when assessing their contributions in the mathematical modeling of pandemics and in estimating the effectiveness of pandemic control decisions and policies; both in terms of disease containment and cost. A recent systematic review of physical interventions to reduce the spread of respiratory diseases 1 has clearly demonstrated the signifi cant contribution of simple low-cost interventions, such as mask-wearing and hand washing. Data in this review, from severe acute respiratory syndrome (SARS)-related studies, indicated that mask-wearing reduced risk of spread by 68%; a fi gure that increased to 91% if N95 masks were worn. Review of the Australian Management Plan for Pandemic Infl uenza 2 and the NSW state health action plan 3 suggests that the need for the general population to wear face masks is likely to be a key component in ensuring continuity of essential services (eg, public needing to wear face masks in health care settings), and in supporting societal functioning (eg, people wearing face masks on public transport or when in common public areas such as shops/malls). Where possible, the continuation of daily activities is likely to be important for population mental health as well as commercial and economic resilience. Vaccination has been identifi ed in the Australian plan and by the World Health Organization (WHO) as a potential strategy in the early response to pandemic infl uenza, and hence, population willingness to receive vaccination is an important factor for consideration. State pandemic vaccination plans promote use of pneumococcal vaccination and inter-pandemic (or seasonal) infl uenza vaccine for priority and at-risk groups, and the WHO has recommended seasonal infl uenza vaccinations for health care workers to reduce risks of genetic shifts in the avian infl uenza virus and to minimize "background noise" in the event of an outbreak. 4 Pandemicstrain vaccination may take a number of months to supply in suffi cient quantities for population-level vaccination, however, if there was suffi cient lag in the global spread of infl uenza to Australia a widespread mass vaccination of the population could become a key preventative national strategy. Willingness to go into isolation or quarantine (either voluntarily or compulsorily) may also become a requirement of the containment strategy at state and national levels. A number of studies relating to public compliance with protective health behaviors have been published in recent years; including either reported levels of compliance during the outbreaks of SARS and H5N1/avian infl uenza, or in anticipation of an H5N1/avian infl uenza pandemic. Studies relating to behavior during or around SARS outbreaks in Hong Kong indicated that during SARS more than 90% of Hong Kong residents frequently wore face masks. 5, 6 Following SARS it was reported that 71% would wear face masks if there was a resurgence of SARS. 6 However, research in Singapore during SARS found that only 4% of respondents in a representative population sample had worn a face mask in the preceding three days. 7 SARS-related research from Toronto, Canada focused on compulsory quarantine and voluntary quarantine/isolation, mostly in health care workers rather than in the general population. 8, 9 Compliance with compulsory quarantine was reported as high. 9 In population research fear of job/income loss was one of the most common reasons for noncompliance or for not self-quarantining. 8 In addition, the importance of good social support (friends/family) was identifi ed, due to high reliance on others for groceries and routine supplies; a need which the Government was unable to meet. Studies of anticipated general population compliance with protective behaviors have been undertaken in Hong Kong, 10, 11 Norway, 12 Italy, 13 the USA, 14 and, comparatively, across a number of European and Asian countries. 15 In the small USA study 14 82% of participants reported that they would wear a face mask to protect others if they caught avian fl u, 78% would accept quarantine, and 84% reported that they would want vaccine to protect them from avian fl u if it were available. In the comparative Europe/Asia study, 15 45%-52% would limit contact with family and friends, and 24%-35% would stay indoors. In Hong Kong 73.8% of respondents reported that they would wear a mask in public venues and 88.3% reported that they would be fully compliant with any quarantine policies. 10 Factors infl uencing uptake of protective behaviors were reported in many studies of SARS, and avian infl uenza, with increased compliance often being associated with higher risk perception and anxiety, 16 concern for self/family/children, 10, 16 perceived susceptibility, 17, 18 greater knowledge regarding transmission, 13 and perceived effectiveness of protective measures. 10 With regard to demographic factors, increased adherence to protective behaviors is often associated with older age groups, female gender, higher levels of education, 10, 16 and being in full time employment. 10 Studies specifi cally concerning willingness to receive vaccination in the event of a future outbreak of H5N1/avian infl uenza or in relation to SARS have not been found in Australia or other countries not previously affected by SARS or H5N1, However, Australian data relating to population willingness to receive seasonal infl uenza exists, and suggests that for those in infl uenza risk groups (over 65s, and 40s-64s meeting established research council (NH&MRC) 'at risk' criteria) infl uenza vaccination uptake is infl uenced by perceptions of risk, self rated health status and beliefs about the effi cacy of the vaccination. 19 Data from Hong Kong indicate that rates of seasonal infl uenza vaccination have increased since SARS and H5N1 and vaccination uptake behavior was linked to higher perceived likelihood of a large local outbreak of H5N1 in the future and perceptions that any future outbreak would be worse than SARS had been. 17 The purpose of the current study was to gather the fi rst baseline Australian data on public levels of willingness to comply with vaccination, isolation, and face mask wearing in the event of an infl uenza pandemic in Australia, and to assess a selection of sociodemographic, health, and threat perception factors that might infl uence such decisions. A short six item pandemic infl uenza question module was developed, as the fi rst part of a larger module of questions on potential threats. These questions were fi eld tested, validated and subsequently administered within the NSW Population Health Survey, using the NSW Health Survey program CATI Anticipated public health compliance for pandemic infl uenza system, between January 22nd and March 31st, 2007. The NSW population health survey includes questions on health behaviors, health status (including psychological distress, using the Kessler 10 measure, 20 and self-rated health status), access to health services, as well as the demographics of the respondents and the households. The target sample was persons living in NSW stratifi ed by the state's eight area health services. Households were contacted using random digit dialing. Details of the sampling approach can be found in the 2007 NSW Health survey report. 21 The pandemic infl uenza question module comprised three questions addressing pandemic infl uenza threat perception and three questions addressing willingness to comply with requested protective health behaviors. The latter questions are the subject of this paper, and their wording was as follows: "In case of an emergency situation, government authorities might request co-operation from the public in a number of ways. Please indicate …" 1. How willing would you be to receive vaccination? 2. How willing would you be to isolate yourself from others if needed? 3. How willing would you be to wear a face mask? All responses were coded on a fi ve-point Likert-scale. Response options for all questions were 'not at all willing,' 'a little willing,' 'moderately willing,' 'very willing,' and 'extremely willing'. In addition, 'don't know' and 'refused' responses were coded. The remaining pandemic infl uenza questions on threat perception included a question on how likely respondents thought it was that there would be an infl uenza pandemic in Australia, how concerned they would be that they or their family would be affected by such a pandemic, and whether they had made changes to their life because of the possibility of an infl uenza pandemic. Prevalence data for these questions has been reported by Barr and colleagues. 22 In this study, the threat perception questions were used as independent variables in the analysis to assess whether aspects of threat perception infl uenced anticipated compliance. The survey data were weighted to adjust for probability of selection and for differing response rates among males and females and different age groups. 21 Data analysis was performed using the "SVY" commands of Stata version 9.2 (Stata Corp, College Station, TX, USA), which allowed for adjustments for sampling weights. The fi ve-point Likert-scale response used in the question module were dichotomized, such that responses of very/ extremely willing (high willingness) were coded as 1 and all other responses as 0. In addition a composite 'all' measure was calculated in which data from respondents who indicated that they would be very/extremely willing to comply with all three behaviors had their response coded as 1 and those with willingness to comply with two or fewer behaviors were coded as 0. This enabled identifi cation and analysis of a group in the sample that reported high universal willingness to comply. To assess the factors that infl uence willingness to comply with protective health behaviors, the dichotomized compliance question indicators and the 'all' indicator were used as outcome measures and these were investigated using the following set of independent variables: gender; age; marital status; have children; location (urban/rural) as defi ned by respondents' area health region; born in Australia; speak a language other than English at home; living alone; employed; highest level of formal education; household income; self-rated health status; psychological distress (as determined by the Kessler K10 measure, which is a composite measure with a range of 10-50 in which 'high psychological distress' has been categorized as a score ? 22 and 'low psychological distress' as a score Ͻ 22); 20 and the three pandemic infl uenza threat questions (concern for self/family, pandemic likely, and life changes). Multiple survey logistic regression using a stepwise backwards model was used in order to identify the factors signifi cantly associated with willingness to comply with health protective behaviors. All variables with statistical signifi cance of p Յ 0.05 were retained in the fi nal model. In total 2081 state residents aged 16 and over completed the module on pandemic infl uenza. The overall response rate was 65%. The key demographics of the weighted survey were comparable to Australian Bureau of Statistics (ABS) 2006 Australian population census data. 22 Overall 73% of the population indicated that they would be very or extremely willing to receive vaccination, 67% would be very/extremely willing to isolate from others if needed, 58% would be very/extremely willing to wear a face mask, and 48% would be very/extremely willing to do all three. Less than 8% of the population reported being 'not at all' willing to wear a face mask, and considerably lower proportions of the population were 'not at all' willing to comply with vaccination (3%) and isolation (5%). Tables 1A-D present results of the survey logistic regression modeling, showing the unadjusted and adjusted odds ratio (OR) for the associations between the three health protective behavior questions and the composite (All) indicator. The infl uence of age on willingness to comply with protective behaviors was found to be a key variable, and this effect is illustrated in Figure 1 . Generally, people reported high levels of willingness to comply with health protective behaviors in the event of pandemic infl uenza; with two thirds of the population reporting that they would be very/extremely willing to receive vaccination, two thirds reporting that they would be very/extremely willing to isolate if needed, and more than half reporting that they would be very/extremely willing to wear a face mask. Although these indications of high willingness to comply may appear lower than expected, from other research studies, it should be noted that these data were collected at a time when pandemic infl uenza was not regarded as a high threat by Australians; only 14% of the NSW population reported that they felt pandemic infl uenza was very or extremely likely to occur. Evidence from other countries, as well as our multivariate analysis, suggest that general compliance with these behaviors would increase substantially if general concern increased, 16 although other factors, such as access to face mask, social responsibility and social acceptance, perceived effectiveness, and communication strategies would also contribute. Multivariate analysis of the data indicated that the factors associated with high willingness to comply varied with each behavior, although generally, those reporting higher levels of concern that they or their families would be affected by pandemic infl uenza and those with higher levels of formal education were more likely to report high willingness to comply, and those who were younger (especially in the 16-24 year old age group) were generally less likely to report high willingness to comply. The relative importance of these factors in public health measure compliance is consistent with the fi ndings of others 10, 16 and provides helpful information to support those involved with risk communication and public health education. Addressing individual behaviors, and in addition to the factors associated with high willingness to comply noted above, those with children and those with higher incomes were likely to report high willingness to receive vaccination. Factors associated with lower levels of willingness to isolate oneself were speaking a language other than English at home, being younger and having never been married. These fi ndings are potentially important for those involved in disease or emergency response, and with further research to identify reasons for this lack of willingness to isolate, may suggest a need for tailored communication or support strategies. It is possible that these groups rely more on social contact outside of the household and, hence, would be less willing to forfeit this in the event of a pandemic. Indications in the data show that immigrants (those not born in Australia) and those who speak a language other than English at home may be less willing to isolate themselves. Response to mask wearing was associated with the general factors of age, education, and concern for self/ family. However, it is the only behaviour studied here that is also associated with respondents' perceived likelihood of pandemic infl uenza occurring. Mask wearing, overall, is the behavior that people reported being least willing to comply with and it is possible that, for this reason, higher threat perception (concern and likelihood of pandemic) is required to drive compliance with this behavior. Younger people (16-24 and 25-34) are generally less willing to wear face masks, and only those with university qualifi cations or equivalent are signifi cantly more likely to be willing to wear face masks. In the event of a pandemic it is likely that the public will be required to wear masks, especially in health care facilities, in public places, and in situations where individuals need to interface with the public, eg, certain critical occupations such as banking, welfare, post/delivery, fuel service, and shops. The mediating effect of increased threat perception in the event of actual pandemic is likely to raise compliance but data suggest that certain sectors of the general population are likely to remain resistant to mask wearing and require further encouragement, communication strategies, and education. Our analysis indicates a number of signifi cant differences between the levels of willingness to comply by respondents in differing marital status categories; most notably that those in the never married category report lower levels of willingness to comply. It should be noted, however, that there is a strong interaction with age, such that 52% of the never married category are in the 16-24 age range (76% in [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] and therefore indications of lower willingness to comply may more simply be related to age than marital status. Similarly, 73% of those in the widowed category are in the 65+ age ranges. Anticipated public health compliance for pandemic infl uenza Worthy of mention are those factors included in the study and analysis that were not associated with willingness to comply with protective behaviors. No statistically signifi cant effects were found for gender, those living alone, employment, or psychological distress. Gender and employment status have been identifi ed as determinants of health protective behaviors in some studies, 10,16 but were not found to be signifi cant here. Concern for self and family, has proven to be a signifi cant factor associated with willingness to comply with protective health behaviors. This fi nding suggests that the concomitant concern/risk that would come with an actual pandemic might be suffi cient to increase public health compliance to required levels, as noted in data from Hong Kong during SARS. 18 An alternative, more active approach might include the use of risk communication messages before a pandemic occurs, or in the pre-pandemic/increased alert stages. Although the adoption of such an approach might seem compelling, Middaugh in a recent paper 23 warns about the unintended consequences of raising concern over 'frightening' the public about germs, increasing concern about seasonal infl uenza (to increase take-up of vaccination) and emphasizing the role of social distancing. He argues that such approaches cause societal estrangement and frighten health care workers, fi rst responders, and those who would have contact with the public in the event of a pandemic. Finally, consideration should be given to the limitations of the current study. The main limitation is that the questions are based on a hypothetical, anticipated threat to an infl uenza pandemic-naive population and, hence, serve as only a general indication of the likely response of the public to such an event. In addition, the analysis has focused on the use of dichotomized responses and has not included those who were less willing, but still may have complied with the behaviors investigated in the study. As noted, reported willingness to comply with these behaviors is likely to be mediated by a number of factors, and it is probable that only a subset of these have been identifi ed in this study. The role of threat perception, anxiety, societal response (the compliance of others), media, and factors directly related to the course of the pandemic are all likely to infl uence public response. Data from this study provide the fi rst Australian population baseline in this area against which future response can be tracked and pandemic modeling can be informed. This study collected data regarding anticipated responses to an infl uenza pandemic and although in the event of an actual pandemic The infl uence of age on willingness to comply with health protective behaviors. Percentage shown is the proportion 'very'/'extremely' willing to comply. the overall level of compliance is likely to be infl uenced by a range of factors, it is probable that relative compliance levels within the data would be upheld and would be more robust for use in pandemic planning. Physical interventions to interrupt or reduce the spread of respiratory viruses: systematic review Australian Health Management Plan for Pandemic Infl uenza. Canberra: Australian Government, Department of Health and Ageing NSW Health interim infl uenza state action plan Guidelines for the use of seasonal infl uenza vaccine in humans at risk of H5N1 infection The psychological impact of SARS: A matter of heart and mind SARS-related perceptions in Hong Kong Crisis prevention and management during SARS outbreak Factors infl uencing compliance with quarantine in Toronto during the 2003 SARS outbreak Risk perception and compliance with quarantine during the SARS outbreak Griffi ths S. Anticipated and current preventative behaviors in response to an anticipated human-to-human H5N1 epidemic in Hong Kong Chinese general population Griffi ths S. Perceptions related to human avian infl uenza and their associations with anticipated psychological and behavioural responses at the onset of outbreak in the Hong Kong Chinese general population Infl uenza pandemic: perception of risk and individual precautions in a general population. Cross sectional study A survey of knowledge, attitudes and practices towards avian infl uenza in an adult population of Italy Patient knowledge and attitudes about avian infl uenza in an internal medicine clinic Precautionary behavior in response to perceived threat of pandemic infl uenza The impact of community psychological responses on outbreak control for severe acute respiratory syndrome in Hong Kong Griffi ths S. Perceptions related to birdto-human avian infl uenza, infl uenza vaccination, and use of face mask Monitoring community responses to the SARS epidemic in Hong Kong: from day 10 to day 62 Prevalence of infl uenza immunization in Australia and suggestions for future targeting of campaigns Short screening scales to monitor population prevalences and trends in non-specifi c psychological distress Report on Adult Health from the New South Wales Population Health Survey New South Wales Department of Health Pandemic infl uenza in Australia: perceptions of threat and preparedness to comply Pandemic influenza preparedness and community resiliency This study was funded by Emergency Management Australia. We would like to acknowledge the assistance of NSW Health staff in the Centre for Epidemiology and Research for their help with fi eld testing, data gathering, and statistical support. The authors report no confl icts of interest in this work.