An extensive literature of small area analysis attempted to test the existence of the variations in physicians' practice patterns by using a chi-squared test; for example, Mantel and Haenszel (1959), Cain and Diehr (1991), and Diehr et al. (1992). However, a chi-squared test is deficient in isolating the variations due to differences in practice patterns if policy characteristics play an important role in determining the variations. Based on the regression framework, this study, hence, derives a weighted Lagrange multiplier statistic for testing the variations which allows a researcher to control for policy characteristics such as DRG relative weights. It demonstrates this approach by testing the existence of the variations in hospital admission rates for atherosclerosis in New York State and shows that the medical variations are indeed real.
Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to: 1. become familiar with the development of small area analysis; 2. test medical variations when policy characteristics may become important
Keywords: Public Policy, Statistics
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.