With two longitudinal Canadian health surveys cross-linked to tax returns, I am employing a causal modeling technique called marginal structural modeling which uses inverse probability weighting to examine causal effects with observational data to study the relationship between income composition and health status over the life course.

I previously checked a "monotonicity hypothesis" whereby constant increases in one's income were expected to be linked to superior self-rated health. The findings were null.

I am currently investigating whether income volatility can be linked to deteriorations in health, why there is massive volatility in the effective tax rate of the lowest decile of taxpayers, and whether there exists such a thing as an optimally diversified health portfolio when considering socio-economic inputs such as income, education, social support, and organizational involvement.