DP6270 Comparing Subjective and Objective Measures of Health: Evidence from Hypertension for the Income/Health Gradient
|Author(s):||David W Johnston, Carol Propper, Michael Shields|
|Publication Date:||April 2007|
|Keyword(s):||hypertension, income, objective health, reporting error, self-reported health|
|JEL(s):||C42, I10, I18|
|Programme Areas:||Public Economics|
|Link to this Page:||cepr.org/active/publications/discussion_papers/dp.php?dpno=6270|
Economists rely heavily on self-reported measures of health status to examine the relationship between income and health. In this paper we directly compare survey responses to a self-reported measure of health that is commonly available in nationally-representative individual and household surveys, with objective measures of the same health condition. Our particular focus is on hypertension, which is the most prevalent health condition in Western countries. Using data from the Health Survey for England, we find that there is a substantial difference in the percentage of adult survey respondents reporting that they have hypertension as a chronic health condition compared to that from repeated measurements by a trained nurse. Around 85% of individuals measured as having hypertension do not report having it as a chronic illness. Importantly, we find no evidence of an income/health gradient using self-reported hypertension, but a large (about 14 times the size) gradient when using objectively measured hypertension. We also find that the probability of false negative reporting, that is an individual not reporting to have chronic hypertension when in fact they have it, is significantly higher for individuals living in low income households. Given the wide use of such self-reported chronic health conditions in applied research, and the asymptomatic nature of many major illnesses such as hypertension, diabetes, heart disease and cancer at moderate and sometimes very elevated levels, we show that using commonly available self-reported chronic health measures is likely to lead to an underestimate of true income-related inequalities in health. This has important implications for policy advice.