Using Country-Level Data and Panel Econometric Methods to Estimate the Effect of Population Average Milk Consumption on Prostate and Ovarian Cancer Mortality Rates

Prostate and ovarian cancers have been related to milk consumption. However, existing ecological studies using country-level data do not attempt to identify causal effects since they are only based on simple cross-sectional analyses. Only under very restrictive assumptions the cross-sectional approach is an appropriate way of identifying a causal effect since it is almost impossible to control for all confounding factors. We do not use the cross-sectional (between-country) variation of the mortality rates and the nutrition variables, but time-series (within-country) variation to identify the effect of average milk consumption in countries on their mortality rates of cancer. We estimate whether a change in milk consumption within up to 50 countries is associated with a change in mortality rates over the period 1990 to 2008, after controlling for other factors. The results indicate fairly well that an increase in average milk consumption increases both the average mortality rate of prostate cancer as well as the mortality rate of ovarian cancer