Hu Yi-Juan, Schmidt Amand F, Dudbridge Frank, Holmes Michael V, Brophy James M, Tragante Vinicius, Li Ziyi, Liao Peizhou, Quyyumi Arshed A, McCubrey Raymond O, Horne Benjamin D, Hingorani Aroon D, Asselbergs Folkert W, Patel Riyaz S, Long Qi
From the Department of Biostatistics and Bioinformatics, Emory University, Atlanta, GA (Y.-J.H., Z.L., P.L.); Groningen Research Institute of Pharmacy, University of Groningen, the Netherlands (A.F.S.); Institute of Cardiovascular Science and The Farr Institute, University College London, United Kingdom (A.F.S., A.D.H., F.W.A., R.P.); Department of Non-Communicable Disease Epidemiology, London School of Hygiene & Tropical Medicine, United Kingdom (F.D.); Department of Health Sciences, University of Leicester, United Kingdom (F.D.); Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, United Kingdom (M.V.H.); Medical Research Council Population Health Research Unit at the University of Oxford, United Kingdom (M.V.H.); Department of Medicine, McGill University, Montreal Quebec, Canada (J.M.B.); Division of Heart and Lungs, Department of Cardiology, University Medical Center Utrecht, The Netherlands (V.T., F.W.A.); Division of Cardiology, Department of Medicine, Emory Clinical Cardiovascular Research Institute, Emory University School of Medicine, Atlanta, GA (A.A.Q.); Intermountain Heart Institute, Intermountain Medical Center, Murray, UT (R.O.M., B.D.H.); Department of Biomedical Informatics, University of Utah, Salt Lake City (B.D.H.); and Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia (Q.L.).
Circ Cardiovasc Genet. 2017 Oct;10(5). doi: 10.1161/CIRCGENETICS.116.001616.
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.
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.
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.
复发性或后续疾病事件的研究可能容易受到因选择既经历了主要索引事件且存活下来的受试者而导致的偏倚影响。目前，任何选择偏倚的程度，尤其是基因关联研究中后续事件发生时间分析的偏倚程度尚不清楚。
我们采用基于经验启发的模拟研究，来探究选择偏倚对已确诊冠心病患者后续事件风险的边际风险比的影响。选择偏倚的程度由基因和非基因因素对索引（首次）冠心病事件的影响大小决定。除非基因风险比大到不切实际（每个等位基因>1.6），并且假设所有非基因风险比之和<10，否则偏倚通常<10%（朝着无效值方向下降）。尽管偏倚较小，但由于精度提高，随着样本量增加，置信区间包含真实效应的概率会降低（覆盖不足）。重要的是，假阳性率不受选择偏倚的影响。
在大多数实际情况下，预计选择偏倚对后续事件风险的基因效应估计影响有限。然而，由于覆盖不足随样本量增加，大多数置信区间会过于精确（不够宽）。当冠心病病史不存在效应修正时，关联检验的假阳性率将接近名义水平。
