Skip to navigation Skip to main content Skip to footer

Approved Research

Mendelian Randomization (MR) analysis of the causal effect of phenotypes on socioeconomic outcomes: A kinship fixed effects approach

Principal Investigator: Dr Kamhon Kan
Approved Research ID: 72222
Approval date: August 25th 2022

Lay summary


This research plan to obtain causal effects of an individual's physical features or social attitudes (exposures) on his/her economic and health outcomes. Specifically, we seek answers to the following questions:

Q1: Does body height and BMI cause higher socioeconomic outcomes?

Q2: Is happiness in the genes? Does happiness cause better health?  An MR investigation.

The causal factorss that we look at include height, body weight, and happiness etc. The outcomes that we examine include income (to answer Q1) and several health outcomes (to answer Q2; e.g., self-reported health, dementia, mental health, etc.).


To obtain causal relations between exposures and outcomes, we use an individual's genetic features which are correlated with physical features or happiness such that we rule out the possibility that the estimate represents causation in the opposite. Unlike other causal studies using variations in exposures generated by genetic features, we take into account the fact that an individual's genetic features were inherited from his/her parents. Without accounting for the inheritability of genetic features, the estimate may  not represent causal effect. We do this by relating siblings' genetic features and the differences in their exposures.

If we do find a causal effect of body height or BMI on income, then it is likely that there is labor market discrimination based on stature or appearance because the effect does not represent productivity differences among individuals of different stature.

Moreover, our analysis of the causal effect of happiness on health. This is among the first causal analysis. If we find a causal effect, then the promotion of happiness should be an important part of the government's public health program to promote population health. A byproduct of our analysis is an understanding of the extent to which happiness is determined genetically (inherited) and to what extent it can be modified by modifiable factors (neighborhood peers, education, marriage, etc).

Data and Duration:

For this research, we plan to include the full database from all 500,000 individuals in the UK Biobank, with their medical history, and information on their genotypes, kinship and demographics.

This project will last for 3 years.