Impact of selection bias on estimation of subsequent event risk
BACKGROUND:Studies of recurrent or subsequent disease events may be susceptible to bias caused by selection of subjects who both experience and survive the primary indexing event. Currently, the magnitude of any selection bias, particularly for subsequent time-to-event analysis in genetic associatio...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
American Heart Association
2017
|
| _version_ | 1797092619012014080 |
|---|---|
| author | Hu, Y Schmidt, A Dudbridge, F Holmes, M Brophy, J Tragante, V Li, Z Liao, P Quyyumi, A McCubrey, R Horne, B Hingorani, A Asselbergs, F Patel, R Long, Q The Genius-Chd Consortium, |
| author_facet | Hu, Y Schmidt, A Dudbridge, F Holmes, M Brophy, J Tragante, V Li, Z Liao, P Quyyumi, A McCubrey, R Horne, B Hingorani, A Asselbergs, F Patel, R Long, Q The Genius-Chd Consortium, |
| author_sort | Hu, Y |
| collection | OXFORD |
| description | BACKGROUND:Studies of recurrent or subsequent disease events may be susceptible to bias caused by selection of subjects who both experience and survive the primary indexing event. Currently, the magnitude of any selection bias, particularly for subsequent time-to-event analysis in genetic association studies, is unknown. METHODS AND RESULTS:We used empirically inspired simulation studies to explore the impact of selection bias on the marginal hazard ratio for risk of subsequent events among those with established coronary heart disease. The extent of selection bias was determined by the magnitudes of genetic and nongenetic effects on the indexing (first) coronary heart disease event. Unless the genetic hazard ratio was unrealistically large (>1.6 per allele) and assuming the sum of all nongenetic hazard ratios was <10, bias was usually <10% (downward toward the null). Despite the low bias, the probability that a confidence interval included the true effect decreased (undercoverage) with increasing sample size because of increasing precision. Importantly, false-positive rates were not affected by selection bias. CONCLUSIONS:In most empirical settings, selection bias is expected to have a limited impact on genetic effect estimates of subsequent event risk. Nevertheless, because of undercoverage increasing with sample size, most confidence intervals will be over precise (not wide enough). When there is no effect modification by history of coronary heart disease, the false-positive rates of association tests will be close to nominal. |
| first_indexed | 2024-03-07T03:48:35Z |
| format | Journal article |
| id | oxford-uuid:c069a60c-c5c1-408c-a301-67fe6c4f5302 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T03:48:35Z |
| publishDate | 2017 |
| publisher | American Heart Association |
| record_format | dspace |
| spelling | oxford-uuid:c069a60c-c5c1-408c-a301-67fe6c4f53022022-03-27T05:54:13ZImpact of selection bias on estimation of subsequent event riskJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:c069a60c-c5c1-408c-a301-67fe6c4f5302EnglishSymplectic Elements at OxfordAmerican Heart Association2017Hu, YSchmidt, ADudbridge, FHolmes, MBrophy, JTragante, VLi, ZLiao, PQuyyumi, AMcCubrey, RHorne, BHingorani, AAsselbergs, FPatel, RLong, QThe Genius-Chd Consortium,BACKGROUND:Studies of recurrent or subsequent disease events may be susceptible to bias caused by selection of subjects who both experience and survive the primary indexing event. Currently, the magnitude of any selection bias, particularly for subsequent time-to-event analysis in genetic association studies, is unknown. METHODS AND RESULTS:We used empirically inspired simulation studies to explore the impact of selection bias on the marginal hazard ratio for risk of subsequent events among those with established coronary heart disease. The extent of selection bias was determined by the magnitudes of genetic and nongenetic effects on the indexing (first) coronary heart disease event. Unless the genetic hazard ratio was unrealistically large (>1.6 per allele) and assuming the sum of all nongenetic hazard ratios was <10, bias was usually <10% (downward toward the null). Despite the low bias, the probability that a confidence interval included the true effect decreased (undercoverage) with increasing sample size because of increasing precision. Importantly, false-positive rates were not affected by selection bias. CONCLUSIONS:In most empirical settings, selection bias is expected to have a limited impact on genetic effect estimates of subsequent event risk. Nevertheless, because of undercoverage increasing with sample size, most confidence intervals will be over precise (not wide enough). When there is no effect modification by history of coronary heart disease, the false-positive rates of association tests will be close to nominal. |
| spellingShingle | Hu, Y Schmidt, A Dudbridge, F Holmes, M Brophy, J Tragante, V Li, Z Liao, P Quyyumi, A McCubrey, R Horne, B Hingorani, A Asselbergs, F Patel, R Long, Q The Genius-Chd Consortium, Impact of selection bias on estimation of subsequent event risk |
| title | Impact of selection bias on estimation of subsequent event risk |
| title_full | Impact of selection bias on estimation of subsequent event risk |
| title_fullStr | Impact of selection bias on estimation of subsequent event risk |
| title_full_unstemmed | Impact of selection bias on estimation of subsequent event risk |
| title_short | Impact of selection bias on estimation of subsequent event risk |
| title_sort | impact of selection bias on estimation of subsequent event risk |
| work_keys_str_mv | AT huy impactofselectionbiasonestimationofsubsequenteventrisk AT schmidta impactofselectionbiasonestimationofsubsequenteventrisk AT dudbridgef impactofselectionbiasonestimationofsubsequenteventrisk AT holmesm impactofselectionbiasonestimationofsubsequenteventrisk AT brophyj impactofselectionbiasonestimationofsubsequenteventrisk AT tragantev impactofselectionbiasonestimationofsubsequenteventrisk AT liz impactofselectionbiasonestimationofsubsequenteventrisk AT liaop impactofselectionbiasonestimationofsubsequenteventrisk AT quyyumia impactofselectionbiasonestimationofsubsequenteventrisk AT mccubreyr impactofselectionbiasonestimationofsubsequenteventrisk AT horneb impactofselectionbiasonestimationofsubsequenteventrisk AT hingorania impactofselectionbiasonestimationofsubsequenteventrisk AT asselbergsf impactofselectionbiasonestimationofsubsequenteventrisk AT patelr impactofselectionbiasonestimationofsubsequenteventrisk AT longq impactofselectionbiasonestimationofsubsequenteventrisk AT thegeniuschdconsortium impactofselectionbiasonestimationofsubsequenteventrisk |