This paper is concerned with the analysis of multivariate categorical data which are obtained from repeated measurement experiments. An expository discussion of pertinent hypotheses for such situations is given, and appropriate test statistics are developed through the application of weighted least squares methods. Special consideration is given to computational problems associated with the manipulation of large tables including the treatment of empty cells. Three applications of the methodology are provided.
BACKGROUND
National and population-based information on the increase in patient care by hospitalists in the United States is lacking.
METHODS
Using a 5% sample of Medicare beneficiaries in 1995, 1997, 1999, and the period from 2001 through 2006, we identified 120,226 physicians in general internal medicine who were providing care to older patients in 5800 U.S. hospitals. We defined hospitalists as general internists who derived 90% or more of their Medicare claims for evaluation-and-management services from the care of hospitalized patients. We then calculated the percentage of all inpatient Medicare services provided by hospitalists and identified patient and hospital characteristics associated with the receipt of hospitalist services.
RESULTS
The percentage of physicians in general internal medicine who were identified as hospitalists increased from 5.9% in 1995 to 19.0% in 2006, and the percentage of all claims for inpatient evaluation-and-management services by general internists that were attributed to hospitalists increased from 9.1% to 37.1% during this same period. Accompanying the increase in care by hospitalists was an increase in the percentage of all hospitalized Medicare patients who were treated by general internists (both hospitalists and traditional, non–hospital-based general internists), from 46.4% in 1995 to 61.0% in 2006. In a multilevel, multivariable analysis controlling for patient and hospital characteristics, the odds of receiving care from a hospitalist increased by 29.2% per year from 1997 through 2006. In 2006, there was marked geographic variation in the rates of care provided by hospitalists, with rates of more than 70% in some hospital-referral regions.
CONCLUSIONS
These analyses of data from Medicare claims showed a substantial increase in the care of hospitalized patients by hospitalist physicians from 1995 to 2006.
There is a high level of agreement between SEER and Medicare reporting of radiation treatments after a cancer diagnosis, suggesting that either source can be used to assess radiation-related treatment patterns. However, for the most complete ascertainment of radiation use, both SEER and Medicare sources should be used.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.