RESUMO
This study uses longitudinal data of inpatient treatment from the Agency for Healthcare Research and Quality's (AHRQ's) Healthcare Cost and Utilization Project (HCUP) to examine the differences in historical trends and build future projections of charges, costs, and lengths of stay (LOS) for inpatient treatment of four of the most prevalent cancer types: breast, colon, lung, and prostate. We stratify our data by hospital ownership type and for the aforementioned four major cancer types. We use the Kruskal Wallis (nonparametric ANOVA) Test and time series models to analyze variance and build projections, respectively, for mean charges per discharge, mean costs per discharge, mean LOS per discharge, mean charges per day, and mean costs per day. We find that significant differences exist in both the mean charges per discharge and mean charges per day for breast, colon, lung, and prostate cancers and in the mean LOS per discharge for breast cancer. Additionally, we find that both mean charges and mean costs are forecast to continue increasing while mean LOS are forecast to continue decreasing over the forecast period 2008 to 2012. The methodologies we employ may be used by individual hospital systems, and by health care policy-makers, for various financial planning purposes. Future studies could examine additional financial and nonfinancial variables for these and other cancer types, test for geographic disparities, or focus on procedural-level hospital measures.
