We use a three-equation model to estimate a health production function for Australia using population data for 1996. Working at the level of the Statistical Local Area (n= 1335) we match (i) data on private practice medical services (derived from the Medicare scheme), (ii) mortality data derived from the de-identified unit records produced by the Australian Bureau of Statistics (ABS), and (iii) social and economic data from the Australian Census of Population and Housing (ABS 1997). We detect strong and statistically significant positive marginal effects of medical expenditure on health status as well as increasing returns. These findings are remarkably different from those of US health production studies, which have tended to lend support to the hypothesis of 'flat-of-the-curve (that is, zero marginal product) medicine'. Moreover, our conclusions are in stark contrast to those produced by Richardson and Peacock (2003, p. 6), who have argued that their econometric work indicates that 'an increase in the [Australian] doctor supply is associated with increasing mortality'.