Estimating mean hospital cost as a function of length of stay and patient characteristics.

ABSTRACT

Statistical models have been used to assess the influence of clinical and demographic factors on hospital charge and length of stay (LOS). Hospital costs constitute a significant proportion of overall expenditure in health care. With escalating costs, knowing the correlates of LOS and in-hospital cost is important for decisions on allocating resources. However, hospital charge and LOS are correlated. We describe two regression models that permit estimation of mean charges as a function of patient hospital stay and adjust for the influence of patient characteristics and treatment procedures on LOS and charge. In the first model, the mean charge over a specified duration is a weighted average of the expected cumulative charge, with weighting determined by the distribution of LOS. The second model for LOS and charge explicitly accounts for their correlation and yields estimates of the average charge per average LOS. The methods are applied to assess mean charges and mean charge per day by cardiac procedure in a cohort of patients hospitalized for acute myocardial infarction, while adjusting for the impact of patient demographic and clinical factors on LOS and charge. For relatively short hospital stays, and when only total hospital charges are available, these models provide a flexible approach to estimating summary measures on resource use while controlling for the effects of covariates on LOS and charge.