Knowledge is the best medicine? An investigation into the role of lay health knowledge in the production of health

Presenter: Martin Karlsson, University of Oxford

Abstract

Lay health knowledge enjoys increasing interest as a key determinant of an individual's health. Over the last few decades we have witnessed gradual changes in the relationship between doctors and patients. Accordingly, patients are no longer passive recipients of expert advice from physicians, but tend to take on an active role in seeking out information concerning their condition and available treatments. Also in general, there seems to have been a surge in demand for health information. This trend is probably partly due to an increase in the level of education in the population, but the epidemiological transition has also, presumably, made it more worthwhile for the patient to build up a knowledge base. Furthermore, the emergence of new information technologies has significantly reduced the cost of acquiring health related information.

The aim of this paper is to investigate the role of knowledge in an individual’s production of health – while taking into account that health knowledge, by its very nature, is likely to suffer from the same type of endogeneity problems as other health investments, such as the demand for medical services. Furthermore, we investigate the extent to which this generally positive effect of knowledge manifests itself through changes in some health-related behaviours.

The dataset used is the Health and Lifestyles Survey, which was conducted in Britain in the early 1980s, with a follow-up survey a decade later. In addition, subsequent mortality data were used so as to achieve a health variable which is not self-reported. Also the knowledge variable is objective in the sense that it is based on the respondents’ actual knowledge.

Using maximum likelihood techniques, we estimate health and health knowledge as a joint normal distribution, using exclusion restrictions related to regional characteristics to achieve identification. Our results indicate that health knowledge has a substantial effect on health, which might have been overlooked in previous research due to the endogeneity problem, which tends to offset the direct effect (possibly reflecting reverse causality). We also find that the parameters of the model are significantly different for males and for females, thus giving empirical support to the common belief that males and females have different approaches to health knowledge.