Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Previous tests of Grossman's model of the demand for health have been based on Grossman's own empirical formulation. This paper argues that this formulation fails to capture the dynamic character of the model. It proposes an alternative formulation, which appears to be more consistent with Grossman's theoretical model and which may also explain the apparent rejections of the model by the data in the author's earlier empirical work. The paper also presents some empirical results obtained using the new formulation, which are, on the whole, consistent with the predictions of Grossman's theoretical model.