Abstract
In this paper we estimate a 'Grossman' model of demand for health based on Swedish micro data. The data set consists of a random sample of over 5000 individuals taken from the Swedish adult population. Health capital is measured by a categorical measure of overall health status, and an ordered probit model is used to econometrically estimate the demand for health equation. The results are consistent with the theoretical predictions and show that the demand for health increases with income and education and decreases with age, male gender, overweight, living in big cities and being single.