Does low participation in cohort studies induce bias?
Does low participation in cohort studies induce bias?
复制标题
DOI：
10.1097/01.ede.0000220549.14177.60
复制
发表时间：
2006-07-01
期刊：
影响因子：
5.4
通讯作者：
Olsen, J
中科院分区：
文献类型：
--
作者：
Nohr, EA;Frydenberg, M;Olsen, J
背景：在过去20年中，大型队列研究的参与率有所下降。这一趋势对相对风险估计的影响尚不清楚。方法：在来源人口中的49,751名妇女中，包括15,373名队列研究参与者，研究了低参与率(30%)对丹麦国家出生队列的影响。在独立数据收集的基础上，我们估计了来源人群和参与者中3种暴露风险关联的优势比(OR)：(A)体外受精和早产；(B)怀孕期间吸烟和小于胎龄婴儿的出生；以及(C)孕前体重指数和产前死产。不参与的影响通过相对优势比(ROR)来描述，即OR(参与者)/OR(来源人群)。两种相对优势比可信区间的计算方法也进行了评估。结果：无参与对选择的ORs影响较小。尽管一些可信区间很宽，但相对优势比接近于1，偏差从未超过16%。两种方法计算可信区间的结果非常相似，一个小型的模拟研究表明，覆盖概率接近95%的名义水平。结论：对于所选择的3个协会，OR不会因为没有参与而产生偏差。这一结果对于基于丹麦队列和类似孕妇队列的研究来说是令人放心的。用于计算相对优势比的可信区间的方法在所考虑的情景中表现良好。
Background: Participation rates in large cohort studies have decreased during the last 2 decades. The consequences of this trend for relative risk estimation are unknown.Methods: The impact of a low participation rate (30%) on the Danish National Birth Cohort was examined among 49,751 women from the source population, including 15,373 participants in the cohort study. On the basis of independent data collection, we estimated odds ratios (ORs) in the source population and among participants for 3 exposure-risk associations: (a) in vitro fertilization and preterm birth, (b) smoking during pregnancy and birth of a small-for-gestational-age infant, and (c) prepregnancy body mass index and antepartum stillbirth. The effect of nonparticipation was described by a relative odds ratio (ROR), calculated as the OR (participants)/OR (source population). Two methods for calculation of confidence intervals for the relative odds ratio also were assessed.Results: The effect of nonparticipation on the selected ORs was small. The relative ORs were close to one and the bias was never larger than 16%, although some of the confidence intervals were wide. The 2 methods for calculation of confidence intervals gave very similar results and a small simulation study showed that the coverage probabilities were close to the 95% nominal level.Conclusion: For the 3 chosen associations, the ORs were not biased by nonparticipation. The results are reassuring for studies based on the Danish cohort and similar cohorts of pregnant women. The methodology used to compute confidence intervals for the relative odds ratios performed well in the scenarios considered.