Skip to main content

Adherence to the test, trace and isolate system: results from a time series of 21 nationally representative surveys in the UK (the COVID-19 Rapid Survey of Adherence to Interventions and Responses study)

Our take —

Test, trace, and isolate protocols are a crucial part of the UK COVID-19 response. In this large survey, available as a preprint and thus not yet peer reviewed, knowledge of COVID-19 symptoms was fairly poor, and there was a large gap between stated intentions and actual behaviors with respect to test, trace, and isolate guidelines. This gap has remained relatively unchanged from March to August, 2020. Lower adherence was associated with male sex, younger age, lower economic status, and less COVID-19 knowledge, but not with risk perception. The authors suggest that financial assistance might improve outcomes, but this conclusion is speculative.

Study design

Cross-Sectional, Other

Study population and setting

This study used polling services to conduct a series of nationally representative surveys regarding COVID-19 knowledge, behaviors, and intentions of 31,787 adults (16 years or older) living in the UK between March 2 and August 5, 2020. Surveys were conducted on a weekly or every other weekly basis, with ~2,000 participants per survey round (21 rounds in all). Some people were surveyed in multiple rounds; there were a total of 42,127 responses. Most questions remained the same throughout the survey, while some questions were added in later survey rounds. Participants were asked to identify the most common COVID-19 symptoms, and were deemed to have done so correctly if they identified fever, cough, and loss of taste or smell. They were also asked whether they had experienced any of those symptoms. Those reporting symptoms within the past 7 days were asked if they had self-quarantined and/or requested an antigen test. In addition to self-reported behaviors, the survey also asked about intentions to self-quarantine, to request an antigen test, and to disclose symptoms to close contacts if participants should become symptomatic. Participants were also asked about their knowledge and beliefs regarding a range of COVID-19 topics, including protective measures, isolation protocols, testing eligibility, government guidance, perceived risk, perceived efficacy of protective measures, and government credibility. Finally, demographic characteristics were obtained via self-report. The study used logistic regression to examine differences in responses by demographic categories (e.g. age, gender, region, employment, etc.).

Summary of Main Findings

Among all respondents, only 49% identified cough, high temperature/fever and loss of sense of smell or taste as the most common symptoms of COVID-19. There was a large gap between intended and self-reported behaviors. While ~70% of people indicated that they intended to self-isolate should they have symptoms in the future, only 18% of people with symptoms within the past 7 days indicated that they had self-isolated. The most common stated reasons for not self-isolating were going to the grocery store or pharmacy (18%), improvement of symptoms (16%), and attending to non-COVID-19 medical needs (15%). There were similar differences between intended and actual behavior for seeking a SARS-CoV-2 test upon experiencing symptoms (~40% vs. 12%) and self-quarantining after being notified by the National Health Service contact tracing program (~65% vs. 11%). Responses to these questions, and the gaps between intended and reported behaviors, remained relatively constant over time. Those who were male, younger, of lower economic status, and less knowledgeable about COVID-19 were, on average, less likely to be adherent to the test, trace, and isolate guidelines. Perceived risk was not associated with adherence behaviors.

Study Strengths

The study included a large and representative population, using well-established polling services to conduct the research. The main survey methodology was straightforward and well-designed, and data were reported relatively clearly. The repetition over time using the same methodologies allows examination of changing knowledge and behaviors.


Data were almost entirely self-reported; both self-reported behaviors and intentions may be overestimated due to social desirability and related biases. Response rates were not reported, nor were details of how participants were selected and how they were contacted. Knowledge of COVID-19 symptoms was not disaggregated by individual symptoms, and the binary outcome variable incorporating three common symptoms may not be a good proxy for general awareness of COVID-19 symptoms. Associations between COVID-19 knowledge and behaviors are cross-sectional, and no causality can be inferred. The claim that financial support would improve outcomes is largely speculative and not meaningfully tested in the study itself.

Value added

Little evidence has emerged to date about adherence to test, trace, and isolate procedures. This study adds relatively robust, nationally representative data from the UK regarding knowledge, intentions, and behaviors with respect to test, trace, and isolate guidelines over time.

This review was posted on: 13 October 2020