Note: The risk of bias by domain corresponds to the highest risk of bias among outcomes by domain.
The overall risk of bias corresponds to the overall highest risk of bias assessed among outcomes.
| Bias | Author's judgement | Support for judgement |
| Confounding |
Serious |
The analysis adjusted for sex, age, calendar month, ethnicity, staff occupational role and patient contact. The study was limited to healthcare workers from Oxford University Hospitals, therefore geographical location was taken account of by study design. The extracted results were based on individuals participating in fortnightly testing, so also took account of health seeking behaviour. There was no adjustment for symptoms at time of potential vaccination and comorbidities. Symptoms at the time of potential vaccination is likely to be an important confounder. Also, adjustment for calendar month could potentially result in residual confounding by time, since infection rates changed substantially from one month to the next. |
| Selection of participants into the study |
Low |
No concerns in this domain. |
| Clasification of interventions |
Low |
No concerns in this domain - vaccination status is likely to be adequately measured. |
| Deviations from intervention |
Low |
No concerns in this domain - the study is observational. |
| Missing outcome data |
Low |
No concerns in this domain. |
| Measurement of the outcome |
Moderate |
There was evidence that testing rates differed after vaccination. |
| Selection of the reported results |
Moderate |
There is no evidence of a protocol or statistical analysis plan, and we have concerns that results could have been selected for reporting because of the findings. |
| Overall risk of bias |
Serious |
|
| Overall comment | In addition to the published report, the supplementary material were used in data extraction and risk of bias assessment. The prevalence reported for the Alpha variant was based on sequencing since December 1, 2020; however, results are for data from September 1, 2020.
The main concern is the potential for uncontrolled confounding. Although most confounders were adjusted for, the analysis did not control for symptoms at the time of potential vaccination, which is likely to be an important confounder. |