Descriptive statistics.

Using individual-level panel data from Understanding Society I estimate the response to a health risk realization on a healthy lifestyle index. To overcome the endogeneity of a diagnosis, I match on initial health risks. I find individuals improve their overall lifestyle healthiness when faced with a large negative health event such as a heart attack or diabetes diagnosis, interpreted as a precise signal about their health status, whereas they do not respond to a noisier signal through solely receiving information about certain health risk factors, such as a diagnosis of high blood pressure or angina (chest pain). The drivers of the overall effect are a decrease in the number of cigarettes smoked and an increase in not drinking alcohol; there is no significant effect found for either diet or exercise. I find some heterogeneity by sex, but only when looking at individual lifestyle behaviours. Overall, the findings suggest that the realization of a health risk leads individuals to improve their lifestyle behaviours, while only a noisier signal about their health risks leads to no such change.