Social Acceptability of Quarantine in Public Health Emergencies: A Systematic Review

Background: Quarantine and isolation is an effective method of controlling outbreaks of emerging and re-emerging infectious diseases. However, the effectiveness of these interventions depends on a high compliance rate, which is often compromised by multiple reasons to break quarantine or refuse isolation. In this systematic review, we highlight public attitudes and reactions towards quarantine including factors that may hinder quarantine measures during public health epidemics, public preferences for using quarantine during epidemics or infectious disease outbreaks and key considerations for the use of quarantine in public health epidemics. Methods: We searched ve databases for publications on quarantine and isolation, and screened for empirical studies on social acceptability. Results: We found 17 articles that met the inclusion criteria. A review of the articles showed some factors could impede compliance with quarantine and isolation. These include the feelings of guilt and social distress, concern about loss of income, and self and social stigma and/or discrimination. On the other hand, compliance with quarantine and isolation was positively associated with perceptions of being a civic duty or fears of infecting loved ones. The articles concluded that, quarantine compliance can be enhanced through assurance of income and promoting safe interaction with loved ones during isolation/quarantine. Conclusions: This review provides public health experts, emergency planners, and policy makers with key considerations to improve public compliance with isolation and quarantine measures during public health emergencies.


Introduction
In the current COVID-19 pandemic context, quarantine and isolation (Q&I) are at the center of public debates regarding 'lockdowns', individual liberty, and economic crisis. Q&I are age-old public health strategies dating at least as far back as the 14th century bubonic plague (Matovinovic, 1969, Paliga, 2020, Conti, 2008. However, the emergence of Severe Acute Respiratory Syndrome (SARS), the Middle East Respiratory Symptom (MERS) and the H1NI in uenza epidemics started a new wave of the use of Q&I in the 21st century. More recently, Q&I have been employed in response to outbreaks of other emerging and re-emerging infectious diseases, such as Ebola virus disease (EVD), Lassa fever and more recently COVID-19.
The words quarantine and isolation are sometimes used interchangeably and may have several meanings (Barbisch et al., 2015). Isolation is the separation of a person who has been diagnosed with an infectious disease from otherwise healthy people, while quarantine, is the confinement of an individual who has been exposed to an infectious disease but is not a con rmed case (WHO, 2020). There are also subtle differences in the objectives of both strategies. Quarantine can facilitate active monitoring, detection, and diagnosis of cases, while the main goal of isolation is to prevent the further spread of infection. Both strategies involve restriction of movement and con nement to a particular/speci c space.
Quarantine and isolation are effective in saving lives during disease outbreaks, especially when there is a high risk of person-to-person transmission and when there are no effective vaccines or therapeutic options (Day et al., 2006, Tang et al., 2020. However, the use of these two infectious disease control strategies raises a myriad of issues, including political, ethical, legal, socioeconomic public health and human rights, that require policy makers to nd a balance between public interest and individual rights (Tognotti, 2013, Upshur, 2003, Koch, 2016. In some instances, quarantine has been perceived as intrusive and stigmatizing (Barbisch et al., 2015, Miles, 2015 leading to reluctance to self-isolate in the case of exposure to an infected person, or to report symptomatic cases. Quarantine and isolation have also been associated with stigma and discrimination, resulting to some people perceiving Q&I as sometimes being worse than the health concerns associated with the disease (Newman, 2012). Poor compliance with Q&I can frustrate public health responses to epidemics or disease outbreaks and, in some instances, like the case of the Ebola outbraks in West Africa, lead government authorities to use autocratic approaches such as engaging the military to enforce quarantine and isolation. (Koch, 2016, Thompson, 2016, Ambe and Kombe, 2019).
Quarantine and isolation will likely remain essential public health strategies in the face of the increasing risk of epidemics of emerging and re-emerging infectious diseases. . It is therefore, important to identify and understand factors that promote or hamper compliance to Q&I, and improve the design and implementation of public health emergency measures. In this systematic review we highlight public attitudes and reactions towards Q&I speci cally in relation to: : 1) factors that may hinder Q&I measures during public health emergencies (PHE); 2) Public preferences for implementing Q&I; and 3) factors and activities that may promote/impede compliance with Q&I.

Methods
A systematic review was conducted following a protocol which was designed and registered ex ante in PROSPERO (Registration ID: #CRD42020175476). The goal of the sytematic review was to identify published literature and synthesise ndings on all published studies that have been conducted globally on the social acceptability of Q&I as a public health strategy during epidemics or public health emergencies. The review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher et al., 2009).

Search strategy
We used a three-stage search approach. The rst was a comprehensive search of the following literature databases: PubMed, PyschInfo, Scopus, CINAHL and social sciences citation Index (Web of Science).
These databases were queried using the search string: (Quarantine OR Isolation OR Social distancing) AND (Outbreaks OR Epidemics Or Pandemics OR Public health emergencies). The search results were exported to EndNote X9, and duplicates removed for screening titles and abstracts by two reviewers (see next section for details). Secondly, we conducted a search of cited references in the included articles to identify any that meet the includion critera but were not yet captured (after the assessment of risk of bias). The resulting list was then screened for relevance. The initial database search was performed in April 2020, with update searches performed at the end of July 2021 A study was considered eligible for inclusion in the review if it was: a) a primary research on public perceptions on Q&I; b) published in peer-reviewed journal; c) included outcomes on the acceptability of Q&I as a public health measure during disease outbreaks. Only studies published in English and French were included in the review (based on the language pro ciency of the review team).

Risk of bias (quality) assessment
To assess the methodological quality of the included articles, we used the Joanna Briggs Institute Critical Appraisal tools for each respective study design. (Aromataris and Munn, 2017). Two reviewers (NM and AM) independently assessed the methodological quality for each article and discrepancies were resolved by consensus following discussions with a third (PCC) reviewer.

Data synthesis
This review was initially designed to use a mixed methods approach. However, due to paucity of quantitative studies on the social acceptability of Q&I, the few quantitative studies that were identi ed were thematically analysed together with the qualitative studies. Full text of all eligible articles was imported into NVIVO 12 and analysed using the meta-ethnography approach for synthesising qualitative studies (Noblit and Hare, 1999). This involved two stages. The rst requires extracting descriptive details of the published article such as authors, study design, year of publication, outbreak and aim of the study. The second stage was the thematic analysis (Vaismoradi et al., 2016). This required repeated reading of the full text of each article to identify themes on acceptability of Q&I and to extract verbatim quotes from the papers to support the identi ed themes. We used the inductive thematic approach (Joffe, 2012) to develop a coding framework (Gibbs, 2007). The coding framework was then applied to two randomly selected articles in the dataset by two researchers in the review team who were not involved in the development of the initial coding framework.

Results
The original search yielded 5777 articles: PubMed (111), PyschInfo (408), Scopus (2000), CINAHL (1253) and Web of Science (2005). Of these, 5761references were excluded either due to inappropriate outcome evaluation and/or inadequate study design or population. Sixteen references were included for full-text screening. Citation tracking and hand searching of the 16 articles yielded an additional 19 titles that may be appropriate. The nal screening of full texts gave a total of 15 articles that met the inclusion criteria ( gure 1). The last search was done in July 2021 with the goal of identifying new articles that may have been published since the initial database search. This led to the identi cation of two additional articles that met the inclusion criteria.
The following diseases were covered in the articles: Ebola Virus Disease, Methicillin-Resistant Staphylococcus aureus, Sever Acute Respiratory Syndrome, COVID-19, Middle-Eastern Respiratory Syndrome and In uenza. The articles reported studies conducted in Canada, the Democratic Republic of Congo, Hong Kong, Iran, Mainland China, Singapore, South Korea, Sweden, Taiwan, the United States and Israel. Some of these studies were multi country studies ( Table 1).
The themes that emerged from the analysis can be grouped into ve categories: Emotional and psychological dimensions of Q&I during a PHE; reasons for compliance with Q&I measures; reasons for non-compliance with Q&I measures; measures that could make Q&I more socially acceptable; and life after Q&I (Table 2).

Emotional and psychological dimensions of quarantine/isolation during a PHE
The rst theme that emerged from our analysis was a view of the personal challenges that arose from the application of Q&I measures during a PHE. These were not explicitly referred to as 'barriers to compliance' to Q&I measures, but offered an insight into the individual emotional and psychological experience of Q&I. .
Five studies reported that individuals facing quarantine had feelings of frustration after having been exposed to an infected individual (Cava et al., 2005a, DiGiovanni et al., 2004, Dodgson et al., 2010, Pellecchia, 2017. This was mainly due to fears that they will be blamed by family members, colleagues or the general public for having, and possibly spreading, the disease. You don't know if you're going to be blamed. There's so much unknown … personally, from a social aspect, from a wellness, illness perspective. It just affects you, and it's very … it's unknown, but it's also just you (Cava et al., 2005a).
Thus, when they received information that they were to be quarantined, they immediately panicked and experienced guilt and shame. This guilt and shame can be both an inhibitor to compliance and a disincentive to testing as individuals may wish to avoid this guilt by not taking a test.
In some cases, guilt and fear was exacerbated by a perception that Q&I was a punishment. This was partly informed by media coverage of Q&I during epidemics, which shaped how quarantine was It seemed from reading international media reports that putting people into quarantine was the only means of cutting the transmission of the virus, yet Montserrado's citizens were not witnessing a decrease in the number of Ebola cases: on the contrary, cases continued to be identified despite the harsh measures taken, and the perception of being quarantined as a means of punishment fed people's already growing fears (Pellecchia, 2017) In other cases when Q&I was considered important,study participants mentioned that they felt like they were in prison and "stuck away", and that their independence was restricted as they had to rely on others for basic daily activities like taking a shower or having a drink (Barratt et al., 2010, Jacobs, 2007. This lack of freedom and independence, along with prevalent media coverage projecting Q&I as a punishment for spreading disease, exacerbated the guilt and shame, making Q&I less tolerable. It was difficult. I was isolated and locked in a room with double dividing walls and I did not get to go out, so it was boring. It was so dull, because they only came when they were going to clean or bring food. It was beautiful to look out onto the field and the greenery. (Skyman et al., 2010).
However in some cases, particularly where participants were provided private rooms, they mentioned that although quarantine may have reduced their opportuinity to socialize, it afforded some degree of privacy and solace that was valued at the time. (Barratt et al., 2010). There were also diverse views on the preference for home-quarantine to address the concerns about boredom or lack of socialisation and balance this with the concerns about infecting others. . These different factors were often interlinked. For example, lack of credibility or inconsistency in information, as well as differences in application of quarantine regulations could lead to breakdown of trust in the public health system making the public not to follow public health advice on Q&I.
That's why I didn't even ... need public health ... I could probably tell them, you know, I knew more information than they did ... I would listen to my dad more than the public health (department). It was kind of weird because they told us that I had to be quarantined. But it didn't make sense because my roommate didn't have to be quarantined as well because ... if I had the virus then most likely she had it as well. So it would only make sense to quarantine both of us, but I'm not sure what went on there. (Cava et al., 2005b) Emphasis added Not only does this quote show a lack of trust -"I would listen to my dad more than the public health (department)" -but also highlighted that an inconsistency in the application of rules eroded trust further.
Past experiences (either personally or of a family member) of poor Q&I conditions also led to increased mistrust in the public health system. Some participants explained how they were at greater risk of contracting an infection while they were in quarantine because they were not provided with basic personal protective equipment such as face masks (Cava et al., 2005b). In some cases, Poor Q&I conditions were a disincentive for the general population to report symptoms to public health o cials because of the fear of being held in poor conditions. A participant described how the conditions during quarantine "killed" her mother. "I hated my mother's quarantine. It was a bad practice. People always get sick and we do not take them to the ECC. They kept my mother for 3 days and then they killed her. My father already died two weeks before, it was me who touched him and washed his clothes when he was sick. I did not get vaccinated but I'm still alive 3 months later." (Mutombo et  An elderly woman in the village was mother to a young woman who had died of Ebola ten days earlier. She had been taking care of her daughter and her two children, so the authorities put all of them under quarantine in the same house. The old woman was helped by community members who provided her with food, water and firewood. From the beginning of her quarantine period, she only received two visits from the [Ministry of Health] staff member who was in charge of checking her temperature and received no support at all from other NGOs in terms of food. During her quarantine, her son-in-law came to visit and attempted to take the children away with him. The man was supposed to be in quarantine himself, but he was reported to have paid a bribe in order to be released. (Pellecchia, 2017) Here, the inability of health authorities to provide suitable amenities for Q&I led to all members of the household being held in Q&I in the same house thereby increasing their risk of contracting Ebola. Second, limited support from the Ministry of Health and other non-governmental organisations meant that community members had to step in, increasing the risk to community members. Third, this lack of support led to the son-in-law breaking his own quarantine, thereby endangering the community. Improving acceptability of quarantine and isolation Key approaches of making Q&I more socially acceptable were linked to the different reasons for (non-)compliance to Q&I measures. For example, in some studies, participants suggested that feelings of abandonment and boredom could be remedied by ensuring that persons in Q&I had access to communication systems (email, telephone, intranets or private internet chat rooms) which enable them to reach out to family members and healthcare workers (Barratt et al., 2010, DiGiovanni et al., 2004, Blendon et al., 2006. Despit being in quarantine, there was an expressed interest to be in physical contac with family members and friends through social visits (Barratt et al., 2010, Cava et al., 2005b, Pellecchia, 2017. Social visits from family members, community leaders, religious leaders and neighbours were considered important to reassure those in Q&I that they are not alone. (Pellecchia, 2017) Amongst adolescensts, access to study materials and e-entertainment was important for adolescents (DiGiovanni Secondly, compliance and social acceptability of Q&I can be improved if persons were nancially compensated or given paid leave during the time they are in Q&I. (Blendon et al., 2006, Bodas and Peleg, 2020b, DiGiovanni et al., 2004. This was particularly relevant for persons working in the informal sector as quarantine could have a negative impact on their household income. (Pellecchia, 2017). In cases where articipants were nancially compensated, there were no reports of nancial hardship although they reported that the process of compensation was slower and could be improved (Cava et al., 2005a).

Life after quarantine and isolation
Post-quarantine and isolation experiences had an impact on public perceptions about Q&I during epidemics ( Table 2). Stigma and discrimination were common experiences (Cava et al., 2005a, DiGiovanni et al., 2004, Blendon et al., 2006. Persons who had been in Q&I received unwanted attention, ridicule, avoidance, and withdrawn invitations from social events. This made it di cult to re-establish social relationships. There's a couple of girls there I go to lunch with daily. When I came back it was like ~Hi.~ One of them didn't talk to me for about 3 days. Wouldn't look at me .... and it's like [tentatively] ~Hi.~ . . . Stayed away. Stayed very far from me .... and didn't have lunch with me. They did not have lunch with me for a long time (Cava et al., 2005a) Sometimes the stigma and discrimination could go on for weeks after the end of the quarantine period, This could be extended to family members, such as children and spouses (DiGiovanni et al., 2004).
Despite these negative implications, positive impacts of Q&I were also reported (Wang et al., 2011, Cava et al., 2005b. For example Q&I offered an opportunity to become more conscious of public health measures, such as hand hygiene and avoiding crowded spaces during a PHE (Cava et al., 2005b); teach others about impacts and coping mechanisms Discussion Q&I have risen to the fore of various debates during the COVID-19 pandemic. It is therefore important to understand public perceptions of these issues to be able to improve public health response to epidemics. In this review, we examined some of these factors, including lack of trust in the public health system, inconsistency in the application of quarantine procedures, fear of loss of income, inability to access basic and social services, and self-perception of risk of transmission. Approaches that could be used to enhance compliance to Q&I measures include nancial compensation for those in Q&I (in case of loss of income); provision of basic amenities and social goods; permitting visits from loved ones (subject to appropriate safeguards); designing Q&I spaces to support privacy; and training non-health professionals who manned Q&I facilities on how to socially interact with persons in quarantine.
This review indicates that an exclusive focus on biological management to the exclusion of the psychosocial and economic management of patients have dire consequences for compliance. Compensation of lost wages led to more than 94% of public compliance to self-quarantine measures.
This dropped to less than 57% of public compliance when compensation was removed (Bodas and Peleg, 2020a). Quarantine and isolation could lead to severe economic hardship and by extension, shortages of food and other basic needs, especially for persons whose earnings are pro-rated or based on daily activities (Pellecchia, 2017). The feeling of boredom and neglect while in Q&I are also strong indicators for depression (Cava et al., 2005a;2005b). The need for mental health care during Q&I had been . We however, could not access any information on community-level strategies to reduce Q&I related stigma and discrimination. The ndings of this systematic review may provide public health experts, emergency planners, and policy makers with key considerations for not only improving compliance to Q&I during PHEs, but a strategic direction on eliminating Q&I related stigma and discrimination to improve compliance to and public trust in Q&I measures.

Declarations
Ethics approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials The dataset analysed during the current study is available from the corresponding author on reasonable request.

Competing interests
were responsible for revising intellectual content of subsequent versions of the manuscript. All authors read and approved the nal manuscript.   Figure 1 Flow Diagram