Background
Novel respiratory virus outbreaks pose a significant threat to public health due to their ability to spread rapidly among populations with little or no prior immunity. A key feature of the global public health response to contain and slow the spread of COVID-19 has been community-based quarantine and self-isolation. According to the World Health Organization; quarantine separates “anyone who is a contact of someone infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes COVID-19, whether the infected person has symptoms or not” from others; isolation separates those “with COVID-19 symptoms or who have tested positive for the virus” from others [
1]. Quarantine and isolation for COVID-19 occurs for a period of approximately 14 days or until public health authorities advise it is safe to leave [
2]. Population level cooperation with quarantine and isolation guidelines can reduce the risk of virus transmission and assist contact tracing efforts. Therefore, barriers to initiation and completion can seriously hinder the public health response by accelerating community transmission.
Behavioural science frameworks identify capability, opportunity, and motivation behaviours (COM-B) as key factors which influence people’s response to changes in policy and guidelines such as the introduction of mandatory quarantine or isolation [
3]. According to the COM-B model, people require
capability in the form of knowledge of why a desired behaviour is important and the skills to plan, remember and act on an intention [
3]. They need
opportunity in the form of a supportive or conducive environment in which to practice the behaviours, this includes both their physical surrounds and the wider socio-political and economic context. Finally, behaviours can be
motivated by existing values or beliefs, a personal belief in the efficacy or importance of a behaviour, its impact on their identity and their ability to overcome existing habits or unconscious processes [
4]. Beyond this behaviour system, broader structural factors including the environmental, political, social and cultural contexts can also influence an individual’s experience of quarantine or isolation [
5]. Identifying, understanding and responding to barriers and enablers to quarantine and isolation behaviours is critical to developing strategies to enhance participation and reduce adverse health outcomes.
Studies conducted during previous respiratory outbreaks such as SARS in 2003 and H1N1 in 2009, have described barriers and enablers for individuals who are asked to comply with quarantine measures. They identify a lack of communication, confusion, mixed messages and inconsistent information from a number of sources of varying credibility to reducing capability and motivation to quarantine or self-isolate [
6‐
8]. Likewise, knowledge about the outbreak was found to be an important factor as demonstrated by Eastwood et al.where those with a basic level of knowledge of pandemic influenza were more likely to comply with restriction measures [
9]. Transparent, timely and evidence-based communication delivered by trust-worthy sources have therefore been identified as key ways to improve capability as well as public trust and allay anxiety [
10]. In addition, the level of self-assessed risk of contracting and transmitting the infection can influence an individual’s motivation to engage in subsequent behaviours. Cava et al.found that those who perceived a lesser risk questioned or ignored quarantine protocols compared to those who perceived a greater risk [
7]. Proximity to threat emerged as an important factor with some seeing geographical distance and low population density as protective factors, thus reducing motivation to act on public health emergency messages [
11].
Motivation to quarantine can be reduced by a fear or loss of employment and income if unable to attend work, concerns about inadequate supplies such as food, requiring medical attention, needing to visit family or to attend religious institutions [
12]. Negative psychosocial impacts of the quarantine period resulting in feelings of frustration, boredom, loneliness, fear of stigma and anxiety about contracting or transmitting the infection to others [
13] have also been shown to reduce motivation. Support services in the form of flexible psychosocial healthcare and employment options, government assisted financial support and leave entitlements, access to necessities such as medical assistance and groceries and social support groups have all been identified as potential facilitators to quarantine adherence [
13,
14]. Additionally, social factors have been found to influence motivation as described by the reciprocity principle or ‘social bargain’ whereby individuals expect their government and society’s assistance in exchange for the loss of liberties assumed during quarantine compliance [
15]. Assistance in this sense is not only functional support to create conducive conditions or the opportunity to successfully implement guidelines, but also a source of motivation by demonstrating shared responsibility for community wellbeing.
Quarantine and isolation have been shown to be effective public health strategies for the prevention of COVID-19, especially as there is no proven effective treatment [
16]. Particularly in the early stages of the pandemic, when the world was without access to a vaccine, non-pharmaceutical interventions such as quarantine and isolation were, and continue to be, key public health measures in combating the spread of the virus. Despite the development of a vaccine, there will continue to be a need for quarantine and isolation for COVID-19, given the lack of 100% efficacy of vaccines and the potential impact of new viral variants. While literature from previous infectious disease outbreaks has identified factors influencing quarantine and self-isolation behaviours, given the unprecedented scale of the COVID-19 pandemic, there is a need to understand how behavioural factors have influenced quarantine and isolation not only within the COVID-19 context, but also within the wider socio-political and economic contexts. New insights will strengthen our understanding of existing models and enable the advancement of more innovative, self-managed quarantine programs for current and future public health responses.
Methods
Setting
This study was conducted between 23–31 March 2020, during the early stage of the pandemic response in Australia. The Australian government closed its international borders on 20th March, 2020 with the exception of returning Australian citizens, residents and immediate family members. All travellers returning to Australia and identified close contacts were required to quarantine at home or in rented accommodation for approximately 14 days or as instructed by the state Department of Health. Diagnosed cases were also required to complete an isolation period of similar duration. A later implemented policy requiring all returned travellers to quarantine in a hotel or designated quarantine was introduced at the end of the study period.
Theoretical framework
This paper draws on the “COM-B model” of behaviour change [
4] to describe barriers and enablers to community-based quarantine and isolation in Australia during the COVID-19 pandemic.
Study design
This research was conducted as a preliminary phase of The Optimise Study: Optimising Isolation, Quarantine and Distancing for COVID-19 a research project led by Burnet Institute and Doherty Institute that aims to find out how people are experiencing COVID-19 and responding to the measures introduced to stop the spread of the virus. Semi-structured qualitative interviews (n = 25) were conducted with people living in Australia to understand their experiences of home-based quarantine or isolation as part of the COVID-19 pandemic response. This study (122/20) and The Optimise Study (333/20) were granted ethics approval by Alfred Hospital Ethics Committee.
Recruitment
Participants were eligible for the study if they were aged ≥ 18 years, were living or staying in Australia and self-identified as currently or having previously been required to undertake community-based quarantine or isolation before March 31st, 2020 for COVID-19. Recruitment was conducted using social media advertising on Facebook and through researcher networks, inviting eligible people to register their interest in the study. Purposive sampling was then used to select final interview participants based on age, gender, languages spoken, location, reason for quarantine and living situation. Invited participants gave informed written consent prior to the interview.
Sample characteristics
Over 300 people registered interest in the study. From these, 40 were screened using the purposive sampling criteria and invited via email and text message to participate, zero declined and five did not respond to messages, ten people responded after the data collection period had been completed. A total of 25 people participated in the study, each completing in a single interview. At the time of the interview, participants had completed between 2–14 days of quarantine or self-isolation at home while in Australia. Participants were aged between 18–73 years old at the time of the interview. In total 15 participants identified as female, eight as male, two as gender non-binary. Four participants spoke a language other than English at home. Reasons for quarantine or isolation in Australia included returned travel from overseas or interstate (17), diagnosed as a COVID-19 case (3), having close contact with a COVID-19 case (1) and awaiting test results (4). During quarantine or isolation participants lived alone (6), in shared residential accommodation (6) and with a partner or family (13). Two participants reported in the interview that they had completed quarantine in other countries prior to returning to Australia and one participant had been isolated in hospital before returning to complete the rest of their isolation period at home.
Data collection
A semi-structured interview guide was developed for this study based on a review of existing literature (see
Supplementary file). The interview guide was used to prompt and guide a conversation about the experience of quarantine or isolation across identified domains rather than as a questionnaire to be administered. The domains covered quarantine initiation, communication, service provision, daily experiences, understanding and adherence to guidelines and perceived impacts including physical, social, emotional, financial. Interviews were conducted by a female post-doctoral behavioural and social science researcher experienced in qualitative research (AD) who had no prior relationship with interview participants. Immediately prior to the interview, author AD introduced herself, provided information about her professional background and reiterated the reasons for conducting the research. Only the participant and researcher were present for the interviews which lasted approximately one hour and were conducted in English over the phone from the privacy of the author’s home. Interviews were audio recorded, field notes taken during the interview and recordings transcribed verbatim by a professional transcription service. Participants were given the option to review their transcript, though none elected to do so, and all were reimbursed for their time and effort with an AUD$50 voucher.
Data analysis
A public health framework analysis [
17] was conducted using NVivo software. Framework analysis is a thematic analysis process used to identify descriptive findings for rapid translation into policy and practice [
17]. First, data were coded by author AD in relation to the three framework nodes—capability, opportunity, and motivation—identified based on the COM-B behaviour change model [
3,
4]. Second, data under each node were coded by sub-themes as they arose by authors AD and SM until no new themes were identified. Coding and themes were discussed with the senior author throughout. Major and minor themes presented cover the breadth and depth of data collected with reference to the COM-B framework and quotations are provided to add richer descriptions. Every effort has been made to deidentify the contributions, including removing names and labelling the contributions with identification numbers.
Discussion
Our findings describe factors which influenced the capability, motivation and opportunity of people to quarantine or isolate at home in the early stages of the COVID-19 pandemic in Australia. As with evidence from previous pandemics, access to clear, trustworthy and timely information about guidelines [
18] and practical advice on how to apply these is required for an effective pandemic response. However, information alone is unlikely to enable the effective practice of quarantine or isolation without acknowledging the importance of motivation and opportunity [
3,
19]. Despite information about COVID-19 and quarantine and isolation requirements being available, participant awareness of its availability, ease of accessibility and the utility of the information to their situation was not always present. People with high levels of capability and motivation were unable to adhere to guidelines due to structural, social and environmental factors such as access to food and medication and poor mental health. International studies have demonstrated associations between quarantine and isolation and perceived stress and psychological impacts [
20,
21].
Importantly, our findings demonstrate the need to revisit recommendations [
22] that call for an extension of public health emergency planning beyond technical capability and institutional planning to ensure that communities have the capacity, opportunity and motivation to respond. It was perceived that at a community level, a lack of planning and ongoing uncertainty could result in individuals who had the desire and capacity being unable to follow them. Opportunities to plan for and practice strategies that people may need during their experience and linking them in with existing services and their own community networks may improve adherence and health outcomes. Consistent with natural disaster emergency planning and response, community-wide planning is a mechanism to engage community members, improve their capability to act and create a sense of self and community efficacy by establishing an environment that supports planned behaviours [
23]. In the case of rapid COVID-19 policy creation and responses, prior planning, preparedness and community engagement may have led to an already established sense of self-efficacy and hence improved resilience. This may not be useful or practical for all community members, however it will be for some. This could result in increased capacity for government to focus on providing intensive and targeted support to those who are less able to plan for and respond to the need to self-quarantine or isolate. The process may address some structural barriers experienced by people attempting quarantine and isolation and create a better sense of shared responsibility to motivate ongoing community wide participation in pandemic response measures [
15,
24].
Our findings suggest that supportive public messaging, which demonstrates the efficacy of quarantine and isolation and normalising the behaviour, would enhance a sense that community members are seen, supported and valued for their participation. While risk perception and belief in the efficacy of quarantine and isolation can be important motivators to follow guidelines [
7], this motivation may be reduced if community members feel as though their actions are in vain because of ongoing transmission or a perceived lack of shared responsibility. This highlights a sensitive balance required in the communication of risk and epidemiological trends during the COVID-19 pandemic. Clearly there is a need for governments to communicate the ongoing risk of transmission and the impacts of non-adherence to guidelines. However, other research suggests that focusing on calling out undesired behaviour is a less effective behavioural strategy than promoting, normalising and rewarding the desired behaviour to influence social norms [
19]. Positive and supportive messaging can also enhance the mental health of people during quarantine which has also been shown to impact motivation to follow guidelines [
13]. In addition to improved information on self-isolation, the provision of social support and clinical intervention to improve emotional wellbeing is likely to improve rates of adherence [
25].
Importantly our findings indicate that racism and discrimination impact the experience of quarantine and isolation. This is consistent with previous research highlighting the negative impacts of quarantine including reinforcing stigma against social minority groups [
26]. The fear and threat that results from discrimination can not only affect a person’s identity but can also affect attitudes to others, undermining empathy with those who are undertaking physical distancing measures [
27]. These impacts on identity and existing values and beliefs can play a significant role in the motivation to adhere to isolation and quarantine requirements. Further research and engagement with diverse community members is needed to identify and define key barriers and enablers to enhance the ability to follow guidelines and reduce negative impacts of the measures [
28]. This includes continuing to engage and support community members in testing and quarantine procedures, even if fully vaccinated against COVID-19.
Our findings clearly demonstrate that factors related to capability, opportunity, and motivation [
3] intersect to influence quarantine and isolation behaviour. They support an expanded focus beyond information and public messaging that only address capability and motivation, to enhance community planning and preparedness. This helps to ensure that policy and planning create enabling environments, addressing limitations of physical living circumstances; supporting provision of essential goods and services; and addressing mental health, stigma and discrimination that may result in population level challenges to effective quarantine and isolation [
18]. Understandably, it is difficult to influence the current physical environment of all community members who may need to quarantine and isolate in high density areas common to cities. However, this should influence the development and targeting of specific messaging and support services to people based on their physical environment and has potential long-term implications for urban planning. An example of this is the subsequent implementation of an emergency accommodation program in Victoria for community members who are unable to quarantine or isolate safely at home [
29]. Working with community members to develop and implement household level quarantine and isolation plans that connect them with services prior to their need arising, may enable people to access necessities during the experience, alert community organisations to people’s needs and promote an environment of shared responsibility.
Strengths and limitations
The in-depth data captured in these interviews from people in diverse circumstances provide valuable insights into experiences of quarantine and isolation. Alignment of the findings with the Capability Opportunity Motivation Behaviour model also contributes to consideration of the practical applications. However, this should be accompanied by caution in recognition of the study and sample limitations which include: recruitment conducted online and through researcher networks limiting the sample to technology confident groups, though in doing so, actually enhancing our insights into a group who were more likely to be impacted by the available information and communication; recruitment and data collection conducted only in English limiting the number of culturally and linguistically diverse (CALD) communities; single phone interviews for each participant, at various times during and after their quarantine or isolation experience; conducted during the early stages of Australia’s pandemic response when little was known or could be planned for in relation to the severity or scale of the pandemic. As such, the relevance of the application of our findings to groups not included should be noted with caution.
Acknowledgements
Optimise is a partnership between Burnet Institute and Peter Doherty Institute in collaboration with University of Melbourne, Swinburne University of Technology, Monash University, La Trobe University, Murdoch Children's Research Institute, the Centre for Culture Ethnicity and Health, and the Health Issues Centre. The authors gratefully acknowledge the generosity of the community members who participated in the study and the work of the Victorian Operational Infrastructure Support Program received by Burnet Institute.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.