UK’s official statisticians had concerns over covid-19 survey bias

0
9


By Adam Vaughan

The ONS study tracks the prevalence of covid-19 in parts of the UK

James Veysey / Shutterstock

The United Kingdom’s Office of National Statistics (ONS) discussed private concerns about the risk of its flagship covid-19 infection test, which provides a partial picture of the country’s epidemic, according to documents from between May and July.

The study is considered one of the best insights into the prevalence of the disease as it tests random households in Britain’s four nations. However, New scientist have previously reported that the response rate has decreased as the survey has expanded.

Of the households in England invited to participate, the response rate fell from 51 per cent in April to 5 per cent in October, fearing that if only a certain group of people were to carry out the survey, it could become less representative of covid- 19’s real prevalence in society. In October, ONS reported New scientist that the study was weighed and continued to offer reliable estimates of virus spread.

Advertising

Now documents released under the rules on freedom of information shows that the government body was sufficiently concerned about the risk of bias at declining response rates that it discussed action to address the issue.

“Locking has been eased and public attention has begun to shift to returning to work and ‘the new normal’. In this context, both the household response rates and the average number of respondents in a household have decreased, which increases the risk of non-response, ”says an appendix to the contract between ONS and IQVIA, its contractor for the study.

The annex discusses the need for action to reverse the declining response rate and reduce bias from household non-response. “Particularly interested in quick quick wins,” it says. Meetings between ONS and IQVIA were proposed in late July and August to discuss measures such as pushing letters and texts, given what officials called “the urgency” of the matter.

Contacted by New scientist about the documents, the ONS said that only initial work was done because the risks did not materialize, as evidenced by an October target of 150,000 unique tests every two weeks reached early. After reaching a low level of 5 per cent in October, the response rate has risen in the UK to 12 per cent by 11 December.

It is still unusually low for an ONS survey, as the response rate ideally hits more than 80 percent, says Sheila Bird at the University of Cambridge. She says the study is a “huge task”, but ONS should be “much clearer” on how it handles the risk of bias when the results are weighed, to make them reflect the wider population.

Elsewhere, the documents show key indicators for IQVIA, including “time to order appointments”, “termination of booked appointments” and missed calls returned within one business day.

But people participating in the survey deluged IQVIA with complaints in September and October, where participants are unable to get through to call centers, calls are not returned and absences to booked appointments. These problems appear to have persisted in November and December people take to social media to report ongoing issues with contacting IQVIA and missed appointments.

ONS said the investigation continued to provide reliable estimates of the epidemic. “The Coronavirus infection study has reached or exceeded all of its key performance targets and remains a vital source of timely data on Covid-19 infections and antibodies across the UK,” it said in a statement.

The documents also reveal when the ONS predicts that its investigation will become unnecessary when the country’s epidemic returns. Tables show that the body expects to have zero unique items in fourteen days by March 2022, two years after the start of the study.

Sign up for our free Health check newsletter for a summary of all the health and fitness news you need to know every Saturday

More on these topics:





Source link

LEAVE A REPLY

Please enter your comment!
Please enter your name here