Genetic instrumental variable studies of maternal risk behaviors for oral clefts

描述（由申请人提供）：唇裂和/或腭裂（CL/P）是常见且造成负担的出生缺陷，其病因学复杂，涉及遗传、行为和环境因素。 CL/P 给受影响个人和家庭的健康、生活质量和经济带来了巨大负担。多项研究表明，孕期母亲的危险行为，如吸烟、饮酒和肥胖，在 CL/P 的病因学中发挥着重要作用。然而，这些研究并未考虑可能影响母亲自我选择风险行为的未观察或未测量因素，而是与 CL/P 风险相关，例如母亲对健康和风险承受能力的偏好以及母亲对健康风险的看法。这些因素引起了几种孕产妇风险和健康行为之间的相关性，并且它们本身与生物、环境和孕产妇健康风险因素相关。实际上不可能观察可用数据样本中的所有相关行为和风险因素。忽略这些未观察到的相关因素可能会混淆孕产妇危险行为对 CL/P 影响的估计，包括平均效应和与遗传风险因素的交互效应的估计。为制定有效的 CL/P 咨询和预防计划，需要准确估计孕产妇危险行为的影响。本研究的目的是使用考虑了未观察到的混杂因素的遗传工具变量 (IV) 模型来估计孕期母亲危险行为（包括吸烟、饮酒和肥胖）对 CL/P 风险的影响。 IV 模型将使用风险行为的变化来估计这些影响，这些风险行为是由遗传变异（工具）预测的，这些变异与这些风险行为显着相关，但通过未观察到的混杂因素与 CL/P 无关。使用适当的工具，IV 模型将提供这些风险行为对 CL/P 的“因果”影响的一致（渐近无偏）估计。该研究将包括一个独特的大型国际群体样本，其中包括约 8917 名 CL/P 儿童的母亲和未受影响儿童的母亲。该研究将选择和基因分型作为研究孕产妇风险行为工具的有力候选基因变异，评估这些变异作为工具的效用，并使用 IV 模型和适合 IV 条件的遗传工具估计孕产妇风险行为对 CL/P 的影响。采用的 IV 模型将估计这些风险行为对 CL/P 的平均影响，以及它们与 CL/P 遗传风险因素的相互作用。该研究将对主要孕产妇危险行为对 CL/P 的“因果”影响提供一致的估计，这对于制定有效且具有成本效益的预防 CL/P 计划和为高危妇女提供咨询是非常需要的。此外，该研究采用了一种新颖的分析方法，可用于研究孕产妇危险行为对其他出生缺陷和围产期结局的影响，并有可能显着推进该研究领域。 公共卫生相关性：该项目将有助于确定怀孕期间几种孕产妇危险行为（包括吸烟、酗酒和肥胖）对唇裂和/或腭裂风险的影响，唇裂和/或腭裂是常见且造成负担的出生缺陷。需要准确估计这些风险行为的影响，以制定有效的干预措施和政策来预防唇裂和/或腭裂。

DESCRIPTION (provided by applicant): Cleft lip and/or palate (CL/P) are common and burdensome birth defects with a complex etiology of genetic, behavioral and environmental contributors. CL/P impose significant burden on the health, quality of life and economics of affected individuals individual and families. Several studies have suggested that maternal risk behaviors during pregnancy such as smoking, alcohol use, and obesity play a role in the etiology of CL/P. However, these studies that do not account for unobserved or unmeasured factors that may influence maternal self-selection into risk behaviors but are related to CL/P risks, such as maternal preferences for health and risk tolerance and maternal perceptions of health risks. These factors induce correlations between several maternal risk and health behaviors and are themselves correlated with biologic, environmental and maternal health risk factors. It is virtually impossible to observe all relevant behaviors and risk factors in available data samples. Ignoring these unobserved yet relevant factors may confound the estimates of the effects of maternal risk behaviors on CL/P, including estimates of average effects and interactive effects with genetic risk factors. Accurate estimation of the effects of maternal risk behaviors is needed for developing effective counseling and prevention programs for CL/P. The objective of this study is to estimate the effects of maternal risk behaviors during pregnancy including smoking, alcohol, and obesity on CL/P risks using a genetic instrumental variable (IV) model that accounts for unobserved confounders. The IV model will estimate these effects using variation in the risk behaviors that is predicted by genetic variants (instruments) that are significantly correlated with these risk behaviors but are otherwise unrelated to CL/P through unobserved confounders. Using appropriate instruments, the IV model will provide consistent (asymptotically unbiased) estimates of the "causal" effects of these risk behaviors on CL/P. The study will include a unique and large international consortium of population-level samples of about 8917 mothers of children with CL/P and mothers of unaffected children. The study will select and genotype genetic variants that are strong candidates for use as instruments for the study maternal risk behaviors, evaluate the utility of these variants as instruments, and estimate the effects of the maternal risk behaviors on CL/P using an IV model with the genetic instruments that fit the IV conditions. The employed IV model will estimate the average effects of these risk behaviors on CL/P as well as their interactive effects with genetic risk factors for CL/P. The study will provide consistent estimates of the "causal" effects of major maternal risk behaviors on CL/P which are highly needed for developing effective and cost-effective programs for prevention of CL/P and counseling at risk women. Further, the study utilizes a novel analytical approach that can be applied to study the effects of maternal risk behaviors on other birth defects and perinatal outcomes and has the potential for advancing this research field significantly. PUBLIC HEALTH RELEVANCE: This project will help to identify the effects of several maternal risk behaviors during pregnancy including smoking, alcohol and obesity on the risks of cleft lip and/or palate, which are common and burdensome birth defects. Accurate estimates of the effects of these risk behaviors are needed for developing effective interventions and policies to prevent cleft lip and/or palate.