Bonilla Carolina, Baccarini Lara Novaes
Departamento de Medicina Preventiva, Faculdade de Medicina, Universidade de São Paulo, São Paulo 01246, Brazil.
Faculdade de Saúde Pública, Universidade de São Paulo, São Paulo 01246, Brazil.
Genes (Basel). 2020 May 4;11(5):507. doi: 10.3390/genes11050507.
Epidemiology seeks to determine the causal effects of exposures on outcomes related to the health and wellbeing of populations. Observational studies, one of the most commonly used designs in epidemiology, can be biased due to confounding and reverse causation, which makes it difficult to establish causal relationships. In recent times, genetically informed methods, like Mendelian randomization (MR), have been developed in an attempt to overcome these disadvantages. MR relies on the association of genetic variants with outcomes of interest, where the genetic variants are proxies or instruments for modifiable exposures. Because genotypes are sorted independently and at random at the time of conception, they are less prone to confounding and reverse causation. Implementation of MR depends on, among other things, a strong association of the genetic variants with the exposure, which has usually been defined via genome-wide association studies (GWAS). Because GWAS have been most often carried out in European populations, the limited identification of strong instruments in other populations poses a major problem for the application of MR in Latin America. We suggest potential solutions that can be realized with the resources at hand and others that will have to wait for increased funding and access to technology.
流行病学旨在确定暴露因素对与人群健康和福祉相关结局的因果效应。观察性研究是流行病学中最常用的设计之一，由于存在混杂因素和反向因果关系，可能会产生偏差，这使得建立因果关系变得困难。近年来，为克服这些缺点，人们开发了诸如孟德尔随机化（MR）等基于基因信息的方法。MR依赖于基因变异与感兴趣结局之间的关联，其中基因变异是可改变暴露因素的代理或工具。由于基因型在受孕时是独立且随机分配的，因此它们不太容易受到混杂因素和反向因果关系的影响。MR的实施除其他外，取决于基因变异与暴露因素之间的强关联，这通常通过全基因组关联研究（GWAS）来定义。由于GWAS大多在欧洲人群中进行，因此在其他人群中有限地识别强工具变量给MR在拉丁美洲的应用带来了重大问题。我们提出了一些利用现有资源可以实现的潜在解决方案，以及其他一些必须等待资金增加和技术获取机会改善才能实现的方案。
