Intended for healthcare professionals


How can we improve self-isolation and quarantine for covid-19?

BMJ 2021; 372 doi: (Published 10 March 2021) Cite this as: BMJ 2021;372:n625

Read our latest coverage of the coronavirus outbreak

Linked Editorial

Support for self-isolation is critical in covid-19 response

  1. Jay Patel, researcher12,
  2. Genevie Fernandes, postdoctoral research fellow2,
  3. Devi Sridhar, professor2
  1. 1Faculty of Medicine and Health, University of Leeds, Leeds, UK
  2. 2Global Health Governance Programme, Usher Institute, University of Edinburgh, Edinburgh, UK
  1. Correspondence to: J Patel

Jay Patel and colleagues examine international approaches and argue for comprehensive support initiatives driven by local government and community based teams

A consistent lesson from the covid-19 pandemic is the importance of a functional test, trace, isolate, and support system.12 The ability of people to isolate is foundational to this multipronged strategy, as it breaks chains of transmission and reduces infection rates in a population. Even the most effective mass testing and intense contact tracing systems have only marginal value if people who test positive and their close contacts are unable or unwilling to self-isolate. Given the global rise in cases of covid-19 and emergence of new variants effective isolation remains critical to controlling the pandemic. We examined the approaches to supporting and monitoring isolation in countries with available data to help understand what works. Data sources included government reports and websites, peer reviewed articles, preprints, and news media reports.

Self-isolation rates

The limited evidence available suggests adherence to self-isolation is generally low, and both financial and logistical factors determine an individual’s ability to isolate.3 A series of online surveys conducted in the UK with over 30 000 participants found that only 18% of those who had experienced symptoms in the past seven days had not left home since developing symptoms, and only 11% of close contacts quarantined.3 These figures are far from the stated public intention to self-isolate and quarantine, at around 70% and 65%, respectively.3 Commonly cited reasons for failure to isolate or quarantine include childcare responsibilities, experience of pandemic hardship, low awareness of covid-19 guidance, and working in a key sector.

Self-reported ability to isolate was lower among ethnic minority groups and those with annual household incomes below £20 000 (€23 000; $27 000) or savings of less than £100.4 Interim evaluation from the Liverpool covid-19 community testing pilot concluded that a major barrier to testing uptake––mostly in deprived communities––was the fear of not having adequate support to isolate.5 Similarly in Iran, although the ability to adhere did not follow the social gradient, people who considered themselves of lower social class were less likely to comply with social isolation measures because of a perceived lack of social support.6

The reason for the difference between intent and actual practice of self-isolation is relevant in predicting compliance. In particular, people who have symptoms or positive results are more likely to isolate than their contacts.78 In the Netherlands, public intention to isolate at home was around 95% if they were to receive a positive test result, reducing to 84% if a member of their household had tested positive, and 43% if a close contact had covid-19.7 A Norwegian study found that 65% of people required to self-isolate had not done so, but compliance was significantly higher among people with symptoms than among those who were asymptomatic.8 Public adherence to protective behaviours in the pandemic has been high, and for behaviours where it is lower––as in self-isolation––the intention to adhere is high, suggesting adequate support could help enable these behaviours.9

Public trust in institutions is a key determinant of compliance to public health guidelines, especially in times of crisis.10 The west African response to the 2014 Ebola epidemic showed that mobilising local leaders and promoting community engagement helped build trust and improved the success of public health measures.11 In the UK, longitudinal analysis confirmed a relation between trust in government and intention to follow covid-19 guidelines,12 although similar analyses on reported behaviours are currently lacking.

Support measures

We use “support” to refer to financial or other measures that enable people to follow self-isolation or quarantine guidance for the stipulated period. Most of the countries we studied have mandatory isolation periods for covid-19 of 10 to 14 days, with France (7 days and voluntary) and Vietnam (21 days and mandatory) notable exceptions (table 1). The support packages offered across countries can be divided into four types: financial support, employment benefits, practical support, and comprehensive services.

Table 1

Summary of support and enforcement for isolation and quarantine policies for covid-19 in 20 countries*

View this table:

Financial support—Some countries offer one-off payments to people who test positive for covid-19 and their contacts to self-isolate. Among the most generous are Australia (up to A$1500 (£840; €970; $1200)), UK (£500 for those receiving government income support), and South Korea (£270). Taiwan offers £25 for each day spent in isolation to cases and their care givers.13 The UK restricts payments to those who receive government benefits, whereas Singapore, South Korea, and Taiwan extend financial support to everyone required to isolate.

Employment benefits—These are commonly paid to those having to isolate, often alongside nominal allowances. Generally, these benefits can only be obtained by people who are employed or earn above a specified income level. In the UK, around two million low paid workers are not entitled to statutory sick pay of £95.85 a week.14

Practical support can take the form of home visits or help with food, medication, and alternative accommodation. The French government mobilised health teams to conduct home visits for confirmed cases, advising them to self-isolate, offering antigen (rapid) tests for household members, and providing extra practical support.15 In the Netherlands, those isolating can contact local municipalities and the Red Cross for practical help. Providing accommodation is particularly important given the heightened risk of household transmission and difficulties of isolating when living in large, crowded, and multigenerational households.16 In Denmark and Norway, local governments offer accommodation to anyone unable to isolate at home. In the US state of Vermont, housing policies enabling people to safely isolate from household members were considered central to its response.17 This involved strengthening existing infrastructure to provide comprehensive housing protection for vulnerable communities.

Comprehensive support—Local authorities in South Korea, Taiwan, New York (box 1), and San Francisco (box 2) offer comprehensive support for self-isolation. In South Korea, as well as one-off payments, people in quarantine are provided with daily necessities and sanitary kits worth around £40 and quarantine facilities if they cannot isolate at home.21 In Taiwan, local government centres offer transport arrangements, food delivery, medical care and household services, including accommodation for people without a residence.13 San Francisco’s “right to recover” programme provides eligible workers with two weeks of salary reimbursement at the minimum wage ($1285; £930; €1000), practical support, and alternative accommodation if required.22

Box 1

New York City’s “take care” initiative

The take care initiative aims to provide any resources an individual requires to safely observe their full self-isolation period either in a hotel or at home.18 This initiative is coordinated locally, with “resource navigators” from community organisations delivering a wide range of services, including financial help, food delivery, health insurance, medical kits, pet care, and mental health support across every neighbourhood. The support package has high acceptance, with only 2% refusing assistance. Preliminary findings showed that local contact tracers were able to locate 82-87% of people at home during random monitoring visits (personal communication). Even among those who left their homes multiple times a day before symptom onset or a positive test result, around 90% reported not leaving their home during the self-isolation period. This figure is increasing as the initiative continues to strengthen, and adherence may be as high as 95%.19

Box 2

San Francisco’s “test-to-care” initiative

The test-to-care model involves engaging with community members and local public health leaders in a densely populated and predominantly Latin American neighbourhood of San Francisco, California. This model, designed to support vulnerable, low income populations, has three support strands: information services, practical services (such as groceries, medication, hygiene products, and other necessities) and ongoing medical, social, and emotional support. Support is delivered by healthcare providers and community health workers. Although its evaluation did not directly assess rates of isolation and quarantine, 65% of participants received community support for the duration of the self-isolation period.20 In addition, around 1 in 10 participants disclosed more contacts than at the initial contact tracing interview.20


Monitoring self-isolation

Broadly, two mechanisms for monitoring isolation have been used with varying degrees of stringency: regular or random checks conducted in person or by telephone, and digital surveillance technologies. Checks are coordinated by local public health authorities or private sector staff and supported by the police. To be allowed to isolate at home in Slovakia, people must install a mobile phone app that carries out random facial recognition checks and provides tracking information. Australia, Singapore, South Korea, and Taiwan also conduct digital surveillance of people who are quarantined through mobile phone apps, location based software, drones, video calls, and close circuit television as well daily monitoring calls by local health teams. Violation can result in heavy fines and even prosecution. Other countries may find it harder to use digital surveillance measures. Countries in the Asia-Pacific region have a strong culture of surveillance combined with increased public trust in the government, whereas privacy laws in European countries and public attitudes towards governance and liberty may not support such measures.23

Most countries fine people who are found violating isolation guidelines, and flouting the rules can result in imprisonment in Australia, Germany, Italy, Finland, and Norway. In the absence of support, penalties alone are unlikely to encourage desirable behaviours during the pandemic.24 Given the material threat that covid-19 poses to individuals’ health, social support with a firm belief in collective responsibility is more likely to achieve constructive actions across communities.25

A key feature of monitoring compliance in countries that have been comparatively successful in controlling transmission is that their systems are driven by local public health authorities with linked testing, contact tracing, and supported isolation efforts. In the UK, contact tracing under the national Test and Trace service is divided between outsourced private companies and local health protection teams, while financial support for isolation is managed by local councils. Since the launch of Test and Trace, 98% of all contacts allocated to local health protection teams have been successfully reached, compared with 68% of those coordinated by the national call centre.26 Without locally delivered solutions, individuals are only loosely instructed to self-isolate without support or monitoring. Improved coordination between local health protection teams, councils, and community organisations for test-trace-isolate efforts could lead to improved public trust, reporting of contacts, and adherence.

Do support interventions work?

Despite the scarce data on the effectiveness of isolation support measures, financial and comprehensive support seems beneficial. The Families First Coronavirus Response Act allowed some US employees (subject to eligibility criteria) to receive 14 days of emergency sick leave at full pay (limited by an upper threshold).27 This measure is estimated to have reduced the number of confirmed daily cases by 400 per state, or 1 case for every 1300 workers.27 In Israel, 94% of adults said they would comply with self-quarantine if they got financial compensation, dropping to below 57% in the absence of financial support.28

While financial resources are important to enable self-isolation, wider support models are necessary to elicit high rates of adherence. Places that offer comprehensive support packages had high rates of adherence to self-isolation guidelines and few violations. In South Korea, the median number of people who quarantined was 36 561 a day, with around six violations recorded each day––a rate of 1.6 violations/10 000 self-quarantined people.21 Since the Asia-Pacific countries usually manage isolation through designated quarantine facilities, stringently monitored by healthcare workers, compliance is likely to be high. However, preliminary data from New York City showing high levels of isolation (box 1) indicate the effectiveness of comprehensive support.

Mutual aid groups––rapidly and widely developed to support vulnerable and shielded people to isolate––have helped protect community health and wellbeing.29 The support requests and activities of such groups show the needs of those in self-isolation and could be used to build effective isolation support policies, particularly through collaboration between local government bodies and community organisations.

Doing better

Policies around self-isolation should be supportive and compassionate in acknowledging the challenges that people face. Although strategies centred around strict monitoring and penalties for violations have not been thoroughly evaluated, they may even be counter-productive, compromising testing uptake and honest reporting during contact tracing and eroding public trust.19 Regular reporting of self-isolation behaviours is also needed to monitor the effectiveness of test-trace-isolate systems.

Local government driven efforts are central to successful crisis management but remain a largely overlooked and ignored tool.30 Being led by local health protection teams is an important, perhaps defining feature, of effective test-trace-isolate systems. The covid-19 pandemic presents many opportunities to improve links between local public bodies and community based organisations and to empower and mobilise community stakeholders to contribute to multiple aspects of the covid-19 response, including supportive strategies to encourage and facilitate self-isolation and quarantine.

Public knowledge and perceptions are varied and influence personal choices.31 The reason that someone needs to self-isolate is relevant in determining whether they fully comply. Particular emphasis should be placed on explaining the rationale for self-isolation. Clear public health messaging, with information accessible in a range of languages and to communities with varying degrees of health literacy is therefore important.

Finally, everyone instructed to self-isolate or quarantine should be entitled to adequate comprehensive support to do so safely. Sufficient immediate support should be offered to make isolation feasible, with particular consideration for those unable to safely separate at home and requiring designated quarantine facilities.

Our findings add strength to the call for urgent action around isolation measures, endorsing locally delivered, comprehensive support models.32 Without effective policies enabling people to safely self-isolate and quarantine, the success of test and trace infrastructures is jeopardised.

Key messages

  • Inadequate financial support is a commonly cited factor for not following self-isolation or quarantine rules

  • Comprehensive support models are required to make self-isolation or quarantine feasible

  • Alternative accommodation should be made available for people who cannot safely isolate at home

  • Locally delivered solutions and community engagement are highly effective, particularly for vulnerable or low income populations


We acknowledge support from the Wellcome Trust (106635/Z/14/Z).


  • Contributors and sources: This analysis was drawn from experiences in 20 countries, which were purposively selected based on available information on isolation related government schemes. JP and GF jointly collated the data. JP drafted the manuscript. GF and DS critically revised the draft. All authors approved the final version of the manuscript.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have the following interests to declare: DS is on the Scottish Government covid-19 advisory group, on the Royal Society DELVE group that feeds into SAGE, and a member of the UK Cabinet Office’s International Joint Comparisons Unit.

  • Provenance and peer review: Not commissioned; externally peer reviewed.