'Mendelian randomization': an approach for exploring causal relations in epidemiology. (April 2017)

• Record Type: Journal Article
• Authors: Gupta, V.; Walia, G.K.; Sachdeva, M.P.
• Is Part Of: Public health. Volume 145(2017)
• Page Start: 113
• Page End: 119
• Publication Date: 2017-04
• Subjects: Mendelian randomization -- Genome-wide association studies -- Epidemiology -- India; Public health -- Periodicals; Public health -- Periodicals; Electronic journals; 362.1
• DOI: 10.1016/j.puhe.2016.12.033
• Languages: English
• ISSNs: 0033-3506

Abstract

Abstract: Objectives: To assess the current status of Mendelian randomization (MR) approach in effectively influencing the observational epidemiology for examining causal relationships. Methods: Narrative review on studies related to principle, strengths, limitations, and achievements of MR approach. Results: Observational epidemiological studies have repeatedly produced several beneficiary associations which were discarded when tested by standard randomized controlled trials (RCTs). The technique which is more feasible, highly similar to RCTs, and has the potential to establish a causal relationship between modifiable exposures and disease outcomes is known as MR. The technique uses genetic variants related to modifiable traits/exposures as instruments for detecting causal and directional associations with outcomes. Conclusions: In the last decade, the approach of MR has methodologically developed and progressed to a stage of high acceptance among the epidemiologists and is gradually expanding the landscape of causal relationships in non-communicable chronic diseases. Highlights: Findings of large observational studies are failing the litmus test of randomized controlled trials (RCTs). Genome-wide association studies have detected genetic variants related to number of modifiable exposures or risk factors. Genes are randomly fixed, resistant to unmeasured confounders and genotyped using robust techniques at population level. Mendelian randomization (MR) uses genes as proxy for changeable exposures to detect causal associations with disease outcomes. MR approach is highly effective in exploring and selecting the promising interventions to be evaluated for future RCTs.