Casemix funding in rural NSW:Exploring the effects of isolation and size

Abstract

The New South Wales Department of Health (NSW Health) wishes to makeappropriate use of casemix data as inputs to the determination of funding levels for small rural hospitals. However, other factors such as hospital size and degree of isolation might need to be taken into account.The study reported here involved correlation of actual expenditures with thosepredicted by use of a casemix model alone, across 105 small public hospitals in theState. We then explored the extent to which the correlation could be increased by theaddition of distance and isolation variables.It was found that actual costs were highly correlated with those predicted from thecasemix data alone, and that the correlation increased when both the distance andthe size variables were introduced. However, contrary to expectations, reduced sizewas associated with reduced costs, and reduced isolation was associated with increasedcosts.It was concluded that, while the predicted relationships may be present, they arelikely to be relatively weak and are probably being masked by other factors notpresent in the model. In particular, it seems likely that there are variations inseverity within the acute admitted patient category which are not fully explainedby the casemix instrument used in this study (the DRG classification). We suggestthat other terms be introduced to control for this possibility before any furtherattempt is made to test whether size and distance factors can be identified whichwork in the expected direction.