Information regarding COVID-19 infection survey statistics

You asked

Please provide the answers to these ten questions.

1. Does the ONS model include data on reported number of cases in order to make predictions?
2. What factors are included in the model that mean raw positivity is always modelled up before lockdowns and then down after lockdowns?
3. What number of or percentage of positive swabs each fortnight had a Ct value less than or equal to 25 and more than two positive genes identified?
4. Have positive test results in the ONS survey ever been calibrated using confirmatory testing e.g. viral culture, genomic sequencing?
5. What percentage of positive test results were a single positive, with the same individual testing negative the week before and the week after?
6. What proportion of participants testing positive on ONS testing have had a follow up antibody test and what proportion of them have tested positive for antibodies?
7. What proportion of participants have tested positive at any one time? Of these what percentage had symptoms at the time of their positive test?
8. What proportion of participants have tested positive on at least two genes and with a Ct value of 25 or less? Of these what percentage had symptoms at the time of their positive test?
9. Of the participants in the household transmission study how many had symptoms and how were they distributed? i.e. how many households had both members of a household pair with symptoms; how many households had a single individual with symptoms and how many were both positives symptomless?
10. The number of positive people; the number of people swabbed and their symptomatic status by region for each report.

We said

Thank you for your request. Please see the following answers to your questions:

1. Our current headline figure is the positivity rate which is the percentage of people who have tested positive for COVID-19 at a point in time. The latest positivity results can be found in our weekly bulletin. We do not use this data to make predictions or produce forecasts.
2. Dynamic Bayesian multi-level regression post-stratification (MRP) is used to produce our modelled estimates. The models adjust for age, sex and region. More information can be found in our methods article.
3. (And 8) In the dataset accompanying our weekly bulletin, you can find percentage of positive tests by gene and Cycle threshold (Ct) values for all genes on tab 6a. You can also find the number of people testing positive and the sample size for the latest two and six week periods in England (tab 1c), Wales (tab 3c), Northern Ireland (tab 4c), and Scotland (tab 5c).
4. Full genome sequencing on our swab tests is undertaken by the Sanger Institute.
5. To provide information on the percentage of tests that were a single positive, with the same individual testing negative the week before and the week after, we would need to create bespoke analysis. Under the Freedom of Information Act 2000, Public Authorities are not obligated to create information in order to respond to requests. We therefore consider this to be information not held.
6. 20% of our sample are also invited to provide a blood sample. This is irrespective of whether they have had a positive swab test. We publish the results from the antibody testing in our fortnightly antibodies article.
7. Our latest symptoms analysis can be found in our characteristics article published on 9 February 2021.
8. Please see the Ct analysis described in number 3 above.
9. To provide information on symptoms of the participants in the household transmission analysis, again this would require bespoke analysis, therefore we consider this to be information not held.
10. Analysis by region is published in our weekly bulletin and analysis by symptoms was published as outlined in question 7. Due to small sample sizes, we cannot produce this symptoms analysis by region.