Full Program »
How Much Mortality Does Obesity Cause? Measuring Causality in PopulationsHernán and Taubman (2008) argue that seeking to estimate the effect of obesity on mortality in a population is meaningless because different ways of intervening to reduce obesity would result in different changes in mortality. They thus regard clearly specified interventions as a necessary condition on meaningful quantitative causal estimates. The methodological questions raised by the "Potential Outcomes Approach" have sparked considerable discussion. In this paper, I focus on the measurement question raised so effectively by Hernán and Taubman: whether measures of the proportion of an outcome attributable to a causal factor are meaningful, and what they mean.
University of Johannesburg
Public Health & Community Medicine
Tufts University School of Medicine