Are smokers too optimistic about their health status: Ex ante perception versus ex post observation

As the world has moved toward the era of non-communicable diseases, whether the individuals are in a capable position to accurately evaluate their own health status has an important implication on disease prevention in particularly and population health outcome in general. In this paper, we address four important questions surrounding the accuracy of health perception: (1) to what extent that individuals can make an accurate evaluation on their own health status; (2) what are the major factors influencing health misperception if any; (3) what are the causal directions between health behavior and health perception; and (4) whether individuals can learn and update their self-evaluation on health status over time and whether such learning is productive in that it mitigates the health misperception. Specifically, we use a longitudinal data set obtained from Taiwan that covers six waves of survey over about twenty-year period to compare the ex ante subjective perception on health and the ex post mortality hazards. Our results suggest that over one third of the survey respondents are not performing well in the evaluation of their own health status. We also find that smokers are more likely to have an optimistic bias on their own health assessment as compared to nonsmokers. After controlling for the simultaneous causality problem, we find a causal effect of individuals' misperceptions on continuing smoking, but not vice versa. In addition, our results show that individuals update their subjective perception on health over time through the learning from personal health shocks and the provision of public information on smoking hazards. Although the learning process tends to be overshooting among smokers, it is beneficial to mitigate the optimistic bias. We also find the evidence that personal health shock has a stronger impact on updating behavior than public information, indicating that personal experience is a more effective channel through which to correct the bias in health perception, compared to the provision of public information, such as anti-smoking campaign.