Econometric issues in testing the age neutrality of health care expenditure

A recent study by Zweifel et al. (Zweifel P, Felder S, Meiers M. Ageing of the population and health care expenditure: a red herring? Health Economics 1999; 8: 485-496) suggests that age is not related to health care expenditure among the elderly once 'closeness to death' is controlled for. If correct, this finding has major policy implications, but flaws in the econometric analysis undermine its credibility. We highlight two in particular, and propose methods to deal with them.