biostudies-literatureMillard LACIntegrative Epidemiology UnitThe UK Medical Research Council and WellcomeNIHRBritish Heart FoundationWellcome Trust’sElizabeth Blackwell Institute for ResearchUniversity of BristolStrategic Support Sciencethe Royal SocietyUniversity of Bristol for the Questionnaire 'COVID1'Wellcome Trust'sUniversity of Bristol and Medical Research CouncilLongitudinal Population Study Covid-19 Steering Group and SecretariatUniversity of Bristol Faculty Director's Discretionary Fund for the Questionnaire 'COVID2'University of Bristol for the Questionnaire ‘COVID1’Medical Research CouncilEuropean Union’s Horizon 2020Bristol BHF Accelerator AwardNational Institute for Health Research (NIHR)University of Bristol Faculty Director’s Discretionary Fund for the Questionnaire ‘COVID2’Wellcome TrustEuropean Union's Horizon 2020MRC44-57https://www.ebi.ac.uk/biostudies/studies/S-EPMC9908043biostudies-literatureUnknown52(1)<h4>Background</h4>Non-random selection of analytic subsamples could introduce selection bias in observational studies. We explored the potential presence and impact of selection in studies of SARS-CoV-2 infection and COVID-19 prognosis.<h4>Methods</h4>We tested the association of a broad range of characteristics with selection into COVID-19 analytic subsamples in the Avon Longitudinal Study of Parents and Children (ALSPAC) and UK Biobank (UKB). We then conducted empirical analyses and simulations to explore the potential presence, direction and magnitude of bias due to this selection (relative to our defined UK-based adult target populations) when estimating the association of body mass index (BMI) with SARS-CoV-2 infection and death-with-COVID-19.<h4>Results</h4>In both cohorts, a broad range of characteristics was related to selection, sometimes in opposite directions (e.g. more-educated people were more likely to have data on SARS-CoV-2 infection in ALSPAC, but less likely in UKB). Higher BMI was associated with higher odds of SARS-CoV-2 infection and death-with-COVID-19. We found non-negligible bias in many simulated scenarios.<h4>Conclusions</h4>Analyses using COVID-19 self-reported or national registry data may be biased due to selection. The magnitude and direction of this bias depend on the outcome definition, the true effect of the risk factor and the assumed selection mechanism; these are likely to differ between studies with different target populations. Bias due to sample selection is a key concern in COVID-19 research based on national registry data, especially as countries end free mass testing. The framework we have used can be applied by other researchers assessing the extent to which their results may be biased for their research question of interest.International journal of epidemiologyExploring the impact of selection bias in observational studies of COVID-19: a simulation study.PMC9908043733206MC_UU_00011/3MC_UU_12023/21NF-0616-10102MC_UU_00011/1AA/18/1/34219ES/T009101/1215408/Z/19/ZCH/F/20/90003MC_UU_12023/29217065/Z/19/ZMC_UU_00004/07221574/Z/20/ZNF-0616–10102MC_UU_00011/6Fernandez-Sanles AGriffith GJMajor-Smith DTilling KClayton GLBorges MCCarter ARMorris TPKawabata ELawlor DAHughes RADavey Smith GMillard LACfalseExploring the impact of selection bias in observational studies of COVID-19: a simulation study.<h4>Background</h4>Non-random selection of analytic subsamples could introduce selection bias in observational studies. We explored the potential presence and impact of selection in studies of SARS-CoV-2 infection and COVID-19 prognosis.<h4>Methods</h4>We tested the association of a broad range of characteristics with selection into COVID-19 analytic subsamples in the Avon Longitudinal Study of Parents and Children (ALSPAC) and UK Biobank (UKB). We then conducted empirical analyses and simulations to explore the potential presence, direction and magnitude of bias due to this selection (relative to our defined UK-based adult target populations) when estimating the association of body mass index (BMI) with SARS-CoV-2 infection and death-with-COVID-19.<h4>Results</h4>In both cohorts, a broad range of characteristics was related to selection, sometimes in opposite directions (e.g. more-educated people were more likely to have data on SARS-CoV-2 infection in ALSPAC, but less likely in UKB). Higher BMI was associated with higher odds of SARS-CoV-2 infection and death-with-COVID-19. We found non-negligible bias in many simulated scenarios.<h4>Conclusions</h4>Analyses using COVID-19 self-reported or national registry data may be biased due to selection. The magnitude and direction of this bias depend on the outcome definition, the true effect of the risk factor and the assumed selection mechanism; these are likely to differ between studies with different target populations. Bias due to sample selection is a key concern in COVID-19 research based on national registry data, especially as countries end free mass testing. The framework we have used can be applied by other researchers assessing the extent to which their results may be biased for their research question of interest.2023-01-01T00:00:00Z2023 Feb2025-04-04T12:55:55.529Z2025-04-04T12:55:55.529ZS-EPMC99080433647441410.1093/ije/dyac221