What is the effectiveness of financial support schemes for individuals requested to self-isolate following a positive Covid test or positive contact: A rapid review

1.1 Purpose of this review

Testing for COVID-19 has been deployed globally as a tool to interrupt transmission through isolating positive contacts from the broader population. Financial support systems have been deployed to increase isolation compliance, there is uncertainty as to the effectiveness of these measures.

2.1 Overview of the Evidence Base

Preliminary scoping searches identified three reviews that were relevant to this review question: two rapid reviews (Cardwell et al. 2022, Patel et al. 2021) and one systematic review (Bahji et al. 2021). A summary of the included reviews is provided in Table 1. Cardwell et al. (2022) identified and assessed the effectiveness of measures to support people in isolation or quarantine during the COVID-19 pandemic, and reported on two primary studies (Kavanagh et al. 2011, Kavanagh et al. 2012). Patel et al. (2021) evaluated the effectiveness of financial support interventions and reported on two studies (Bodas & Peleg 2020, Pichler et al. 2020). The third review by Bahji et al. (2021) aimed to evaluate strategies to aid self-isolation and quarantine for individuals with severe and persistent mental illness during the COVID-19 pandemic, and included one relevant study (Fuchs et al. 2021). All five studies identified by these reviews were unique, and no studies overlapped between the reviews. Due to the lack of primary evidence identified for this report, each of these five studies have been extracted and reported individually alongside the other primary evidence in Table 2.

We identified a further four studies published after the review searches were undertaken (Albani et al. 2021, Bourdeaux et al. 2021, Homaie Rad et al. 2021, Johnson et al. 2022). Alongside the five studies identified from reviews, a total of nine primary studies are included in this report. Most of the included studies focused on the COVID-19 pandemic (n=7), with two publications set in the context of the H1N1 pandemic. The study types included: epidemiological modelling studies (n=2), an economic modelling study (n=1), a questionnaire-based publication (n=1), discrete choice experiments (n=2), a retrospective observational study (n=1), and household surveys (both H1N1 studies, n=2). A summary of all included primary studies is provided in Table 2 and their findings summarised in Section 2.

We also identified one ongoing systematic review by Mendonca et al. (2020), which aims to identify barriers and facilitators to population level adherence to COVID-19 prevention measures. The review focuses on qualitative methods and synthesis. The inclusion criteria are adults who have received protective behaviour recommendations related to a pandemic. A summary of the review protocol is provided in Table 3. No results have yet been reported for this review.

2.2 Effectiveness of financial support for isolation

2.3 Bottom line results for effectiveness of support for isolation

A limited evidence base suggests that financial support for isolation is associated with an increase in isolation compliance, lower social engagement, and a reduction in infection levels. The hypothesised causal pathway between the implementation of financial support for isolation and a reduction in COVID-19 cases is well documented, however, the extent to which the components interact is uncertain. There is evidence suggesting individuals are more likely to comply with isolation when they are financially supported to do so (Albani et al. 2021, Bodas & Peleg 2020, Homaie Rad et al. 2021, Johnson et al. 2022, Kavanagh et al. 2012, Pichler et al. 2020). Evidence suggesting a positive relationship with longer isolation duration and higher financial requirements (Homaie Rad et al. 2021). Increased levels of social isolation are associated with reduced transmission of COVID-19 in epidemiological models (Albani et al. 2021, Pichler et al. 2020). Financial support programmes have the potential to be a cost-effective intervention (Bourdeaux et al. 2021). In addition to the evidence on epidemiological and cost-effectiveness, Fuchs et al. (2021) offered evidence supporting the viability of delivering medically assisted isolation hotels for people unable to isolate at home. Further to the COVID-19 literature, two studies found that a financial support and improved social restriction information was associated with compliance with H1N1 isolation (Kavanagh et al. 2011, Kavanagh et al. 2012). However, direct real-world evidence on the implementation of financial support for COVID-19 and its effectiveness on self-isolation compliance and subsequent reduction in SARS-CoV-2 transmission, is limited.

3.1 Summary of the findings

This rapid evidence review aimed to identify evidence on the effectiveness of financial support for people who are required to isolate. Multiple questionnaire-based publications have reported higher isolation compliance because of financial support. Two epidemiological modelling studies have observed higher levels of social isolation in locations which made finance for isolation/sick leave accessible compared to those without support. A major limitation with this literature is that the studies either rely on reported behaviours or, where observed levels are assessed, there is a range of confounders not accounted for. Whilst there are multiple publications concluding that financial support is associated with a higher compliance to isolate, there remains considerable uncertainty as to its effectiveness on transmission.

There is limited evidence to suggest that financial support improves isolation compliance and reduces COVID-19 transmission levels. The two epidemiological studies, which observed higher social isolation, both suffer from short data collection periods, uncontrolled confounders, generalisability concerns and issues with seasonality. As with support for isolation compliance, there is uncertainty as to the extent to which financial support impacts transmission.

The nature of assessing the effectiveness of an intervention during a dynamic pandemic has meant that the certainty with which we can conclude effectiveness is relatively low. For studies that utilise observed trends and outcomes there is a consistent lack of appropriate comparator and the risk of unobserved confounders. Studies which undertook a questionnaire approach must accept the uncertainty related to behavioural adaptation to the constantly moving pandemic setting.

There was no evidence related to effectiveness of financial support for disadvantaged populations who are required to isolate or any insight to the impact of financial support on equality.

3.2 Limitations of the available evidence

The evidence base consists of a small collection of studies with no randomised control trials (RCTs). The range of limitations inherent in the current literature can be broadly broken up into structural study limitations and poor generalisability of findings. The structural limitation heading includes the factors associated with the undertaking of the research, such as, inappropriate comparator, unobserved confounders, and hypothetical scenarios.

The pandemic setting offers an ever-changing landscape for researchers to try to isolate causal inference. The lack of well controlled studies means that the study designs risk having inherent bias. Whilst there were efforts taken to offer appropriate comparators within the comparative studies, it is difficult to control for potential differences across timelines or geographies. Findings were adjusted according to the relative characteristics of the comparator, however, there, was a high likelihood of unobserved confounding. Confounding effects which are not accounted for may influence the outcomes of interest and introduce bias.

Most primary studies in this review are based on questionnaire response assessments. The central issue with questionnaire studies is that the outcomes are reported as opposed to observed. Participants may signal a proposed behaviour which doesn’t correlate with observed action. In addition to this central limitation, questionnaire responses may suffer from responder agenda, in this case, wanting to support a policy to offer financial support.

The second limitation area is that of generalisability. There is uncertainty as to the likelihood that the conclusions drawn from the literature would be observed in another time period, location or population. The pandemic is dynamic, whether that is current infection levels, dominant COVID-19 variant, season, vaccination coverage or non-pharmaceutical intervention status. Findings based on data collected on a sub sample of the pandemic duration may find that their conclusions are not observed when the study is repeated at a later date.

Location generalisability is a major limitation when drawing conclusions from a geographically disparate literature like the one included in this review. The variability of financial and social behaviours, social policy, and earnings is highly heterogeneous across the country settings for the studies. When assessing the impact of financial support, the country’s financial situation, distribution of finance within the country, behaviours towards finance, and current social support measures are important considerations. There may be population generalisability limitations in the included studies. The sample sizes reported are relatively small and may not reflect the true population preference.

3.3 Implications for policy and practice

Findings corroborate the use of financial support for isolation to increase isolation compliance, however there are caveats. Whilst the general finding that financial support is associated with a reduction in COVID-19 transmission there are significant limitations with the evidence base. There is no generalisable evidence to inform the optimal scale of financial support. Current evidence is insufficient to offer an insight into the scale of effectiveness of financial support for isolation may achieve. The most informative current evidence makes use of large routine data collections, it is possible that further research will offer more definitive and instructive conclusions.

3.4 Strengths and limitations of this Rapid Review

The studies included in this rapid review were identified using a systematic literature review of a range of carefully selected publication databases. The research question and study protocol were developed with significant input from experts in the field. The abstract and full text screening was conducted by a single researcher, uncertainty was checked by a second reviewer. The data extraction was performed by a single reviewer and checked for consistency by a second researcher.

Whilst the review methods undertaken have been pragmatically robust, there remains the possibility that additional eligible texts have been missed. The search was limited to English language publications which may have induced bias. We developed our search strategy and implemented a date restriction informed by the two rapid reviews that established the basis of this review. There may be eligible literature that was not identified in either review or was published prior to our date restriction.

The literature includes studies with serious limitations, whilst much of this is due to the difficulty of hypothesis testing within the COVID-19 context, this reduces the strength of conclusions. Efforts have been made to summarise the limitations of each study; however, this review undertakes no formal risk of bias assessments, therefore we are unable to say with appropriate context the strength of conclusions.