The conclusion was
Factors associated with multiple outcomes
^Patterns of results were similar for all outcomes. Lower adherence was associated with being male, younger age, having a dependent child in the household, lower socio-economic grade, greater hardship, and being less informed about COVID-19 and guidance to prevent the
spread of the virus (e.g. not being able to identify key symptoms of COVID-19, not knowing government guidance if you were to develop symptoms of COVID-19, and disagreeing that someone can spread COVID-19 even if they are asymptomatic).^
DISCUSSION
As in other countries, the test, trace and isolate system is the cornerstone of the UK’s public health strategy for coping with the COVID-19 pandemic. Its success relies on adherence to multiple behaviours at multiple stages. Our data suggest that self-reported rates of adherence to isolating, testing and quarantining are currently low, as are rates of recognition of the main symptoms of COVID-19 (see Figure 6). Rates of intended behaviour are much higher than rates of self-reported behaviour. This is unsurprising. The percentage of people who share details of close contacts after receiving a positive COVID-19 antigen test is therefore likely much lower than the percentage of intended sharing of contacts reported here.
^Our observed rates of adherence were largely stable over time with the notable exceptions of symptom recognition, where recognition of ‘new’ symptoms increased over the first one or two weeks after they were introduced, and intended (but not actual) requests for an antigen test, which have been steadily increasing. These findings are in line with other research finding low rates of symptom identification and low adherence to self-isolation in the UK. Our estimate of the percentage of people requesting a test (11.9%) is lower than the
estimate that can be derived by dividing the number of cases per day identified in the community by NHS Test and Trace by the estimated daily incidence recorded by the Office of National Statistics (35% for 6-13 August). This discrepancy might be accounted for by different sample biases, the probable inclusion of people in our sample with
an obvious, non-COVID explanation for their symptoms, and the probable inclusion of asymptomatic cases in the NHS Test and Trace data.^
^Stability of the outcomes indicates that changes to messaging between March and early August have had little effect on behaviours relevant to test, trace and isolate. This suggests either that the changes introduced so far have been ineffective, the budget allocated to messaging about the test, trace and isolate system has been insufficient to allow changes to have an impact, or that the factors preventing people from engaging with behaviours are not
amenable to messaging alone.^
^Our results suggest that financial constraints and caring responsibilities impeded adherence to self-isolation, intending to share details of close contacts, and quarantining of contacts. The
disproportionate impact of the pandemic on people from lower socio-economic backgrounds and with caring responsibilities has been well-documented. Previous research has shown that people who have received help from others outside their household because of
COVID-19 were more likely to adhere to self-isolation. To encourage adherence, policies must ensure that people are adequately reimbursed for any potential losses that may arise from needing to self-isolate and facilitate practical considerations, such as shopping for groceries and medicines during self-isolation.^
In terms of capability, it appears that higher knowledge in general was associated with greater uptake of protective behaviours. It is impossible to disentangle causality here. People who are better informed may simply be more engaged generally in attempting to understand and tackle the pandemic, with the latter promoting adherence. Nevertheless, disseminating clear and easily understood information about the virus and how it spreads is likely to increase adherence to protective behaviours, especially where understanding is low.
Motivational factors, such as perceiving measures to be effective and being confident about returning a testing kit, were associated with intending to share details of close contacts and requesting a test respectively. This is in line with research findings from the H1N1 influenza pandemic. Making antigen testing as easy as possible, for example by introducing local testing sites in areas with high infection rates, may increase adherence to testing.
Messaging that highlights the effectiveness of adhering to each stage of the test, trace and isolate system, and that emphasises that adhering is straightforward and easy to do may further improve adherence.
^With regard to demographic differences, men and younger people were less likely to adhere to steps along the test, trace and isolate pathway. Similar findings emerged during the H1N1 pandemic. Gender and age differences in adherence may be caused by differences such as poorer health literacy in men and a greater desire to be active and to have contact with peer groups amongst younger age groups. Working in a key sector was associated with not
self-isolating. This may be because key workers have a greater financial need or feel a greater social pressure to attend work and are less likely to be able to work from home. Key workers and people from minority ethnic backgrounds were less likely to identify common symptoms of COVID-19. Targeted communications to these groups may help improve adherence and increase knowledge of common symptoms of COVID-19.^
^We found no evidence for associations between perceived risk to oneself or to people in the UK and adherence to test, trace and isolate behaviours. We also found no evidence for an association between self-isolation and concern about spreading the virus to those at risk of complications, thinking that your personal behaviour has an impact on the spread of COVID19, or perceived effectiveness of self-isolation. However, these factors were associated with intention to share details of close contacts. Taken together, these findings suggest either that concern about the risk of giving COVID-19 to others increases intention to adhere to protective behaviours but not actual behaviour, or that such concerns are mainly important
where any inconvenience associated with the behaviour is imposed on someone else (as is the case of quarantine) rather than on you (as is the case in self-isolation).^
^We found an association between perceiving information from the Government to be more credible and being less likely to self-isolate and to identify COVID-19 symptoms. It may be that this association is confounded by, for example, the political orientation of participants.
Strengths of this study include large sample sizes allowing us to investigate infrequent behaviours. Inclusion of survey items in multiple waves of data collection has enabled us to track uptake of protective behaviours and knowledge over time. This study also has
limitations. First, we used quota sampling to ensure that participant characteristics were representative of the UK adult population. While we cannot be sure that survey respondents are representative of the general population, online quota sampling is a pragmatic
approach when a large, demographically representative sample must be obtained in a very short time frame during a crisis. Second, data were self-reported, and so could have been influenced by social desirability and recall gaps and bias. Although self-reported adherence to protective measures for COVID-19 such as social distancing is associated with real-world behaviour, it is likely that rates reported here are overestimates of adherence. Third, data are cross-sectional, therefore we cannot infer causality. Fourth, although we asked participants if they had left home at all since developing COVID-19 symptoms, technically it is permissible to leave home under some circumstances, including to attend a medical
appointment, get a test or if you receive a negative test result. Given that only 12% of people with symptoms reported requesting a test, we do not believe this explanation accounts for more than a small fraction of the non-adherence that we observed. Fifth, while we had a large overall sample size, numbers of participants included in analyses investigating requesting an antigen test and quarantining after being alerted were smaller, and skewed outcome responses
resulted in small cell counts. We have presented these analyses for completeness, but these results should be treated with caution.^
In order for the test, trace and isolate system in the UK to succeed, people must recognise the key symptoms of COVID-19, self-isolate, request a test, share details of their close contacts and quarantine if contacted. Our results indicate that approximately half of people know the symptoms of COVID-19, and that adherence to each stage of the test, trace and isolate journey is low. Policies which support people financially and practically, and providing and communicating about a testing system that is both effective and easy to access, will be key to increasing uptake. Targeted communications especially to men, younger age groups and key workers are likely to further increase uptake.
And that is why the UK has had more of a problem than other places in Europe, because we have greater levels of inequality and why the North now has a bigger problem than the South.
If you live in a poor area (or close to many poor areas) you are going to have a long hard winter.
And the whole thing will have an effect of making the most economically deprived areas even more deprived because of the inability of the government to get these key messages across and the inability of poor people to isolate.
Nothing new then...