Reliability and validity in hospital case-mix measurement.

摘要: There is widespread interest in the development of a measure hospital output. This paper describes problem measuring expected cost mix inpatient cases treated (hospital case-mix) and general approach to its solution. The solution based on set homogenous groups patients, defined by patient classification system, estimated relative weights corresponding categories. applied develop summary costliness Medicare patients 5,576 participating hospitals. The case-mix index evaluated estimating average function. provides direct test hypothesis that relationship between per case proportional. function analysis also means simulating effects error our estimate this relationship. Our results indicate valid robust hospital's case-mix.