Specifically, we highlight the requirements for careful examination and interpretation of derived causal estimates and host (i.e., human) genetic effects themselves, triangulation across multiple study designs and inter-disciplinary collaborations. Here, we focus and discuss the utility of MR within the context of human gut microbiome research, review studies that have used this method and consider the strengths, limitations and challenges facing this research. MR is an established method that employs human genetic variation as natural “proxies” for clinically relevant (and ideally modifiable) traits to improve causality in observational associations between those traits and health outcomes. The integration of human genetics within population health sciences have proved successful in facilitating improved causal inference (e.g., with Mendelian randomization studies) and characterising inherited disease susceptibility. Therefore, there is a need for alternative approaches to interrogate causality in the context of gut microbiome research. This evidence that has not been translated between model systems impedes opportunities for harnessing the gut microbiome for improving population health. Whilst randomized controlled trials have made steps towards understanding the causal role played by the gut microbiome in disease, they are expensive and time-consuming.
Furthermore, epidemiological studies have been unconvincing in their ability to offer causal evidence due to their observational nature, where confounding by lifestyle and behavioural factors, reverse causation and bias are important limitations. In vitro models, few findings have been translated into an understanding of modifiable causal relationships. Evidence supports associations between human gut microbiome variation and multiple health outcomes and diseases.