Susan M. C. Payne

SMALL AREA VARIATIONS IN HOSPITALIZATION RATES : HOW MUCH YOU SEE DEPENDS ON HOW YOU LOOK

This research investigates the degree that estimates of the magnitude of small area variations in hospitalization rates depend on both the estimation method and the number of years of data used. Hospital discharge abstracts for patients 65 and older from acute care hospitals in Massachusetts from 1982 to 1987 were analyzed. The SCV statistic, the approach used in many current small area variation studies, and empirical Bayes (EB), an approach that adjusts more fully for the effect of random variation, were compared. EB estimates based on 3 years of data were best able to predict future area-specific hospitalization rates. Compared to EB estimates using 3 years of data, the SCV statistic with 1 year of data overestimated the median amount of systematic variation by over 70% for the 68 conditions studied; with 3 years of data, the SCV overestimated the median by 55%. Regardless of method, the same conditions were identified as relatively more variable and the same geographic areas were found to have higher than expected hospitalization rates. The magnitude of differences in hospitalization rates depends on how the data are analyzed and how many years of data are used. Hospitalization rates across small geographic areas may vary substantially less than reported previously.

Does more "appropriateness" explain higher rates of cardiac procedures among patients hospitalized with coronary heart disease

Background. There have been few studies of the extent to which differences in the pool of patients being managed might account for geographic variations in treatment rates. Objective. For two cardiac procedures, cardiac catheterization and revascularization, we evaluate the hypothesis that differences in “the percentage of patients for whom the procedure is appropriate” is a factor explaining variations in use rates among those hospitalized with coronary heart disease (CHD). Research Design. Based on hospital utilization patterns in Massachusetts in 1990, we created 70 small geographic areas. Using 1992 Massachusetts Peer Review Organization data, areas were ranked from highest to lowest based on (empirical-Bayes-adjusted) hospitalization rates for each procedure. One thousand seven hundred four cases from 43 hospitals were sampled, roughly half each from high and low use areas. Half had a procedure and half were candidates for the same procedure but did not have it. For each procedure, medical records were reviewed to determine whether the procedure was (or, for those not having it, would have been) appropriate, based on criteria developed using a modified Delphi approach. Results. Among those having either procedure, appropriateness rates were similar in high and low rate areas (P = 0.59 for catheterization and P = 0.30 for revascularization). However, among candidates for either procedure who did not have it, appropriateness for performing the procedure was greater in high-rate areas (41.4% vs. 32.1%, P = 0.05 for catheterization; 71.2% vs. 57.2%, P = 0.003, for revascularization). Conclusion. Among those hospitalized with CHD, appropriateness rates for two cardiac procedures are higher in areas with higher use rates.

The Self-Adapting Focused Review System. Probability sampling of medical records to monitor utilization and quality of care

Medical record review is increasing in importance as the need to identify and monitor utilization and quality of care problems grow. To conserve resources, reviews are usually performed on a subset of cases. If judgment is used to identify subgroups for review, this raises the following questions: How should subgroups be determined, particularly since the locus of problems can change over time? What standard of comparison should be used in interpreting rates of problems found in subgroups? How can population problem rates be estimated from observed subgroup rates? How can the bias be avoided that arises because reviewers know that selected cases are suspected of having problems? How can changes in problem rates over time be interpreted when evaluating intervention programs? Simple random sampling, an alternative to subgroup review, overcomes the problems implied by these questions but is inefficient. The Self-Adapting Focused Review System (SAFRS), introduced and described here, provides an adaptive approach to record selection that is based upon model-weighted probability sampling. It retains the desirable inferential properties of random sampling while allowing reviews to be concentrated on cases currently thought most likely to be problematic. Model development and evaluation are illustrated using hospital data to predict inappropriate admissions.

A primer: Health care databases, diagnostic coding, severity adjustment systems and improved parameter estimation

Data on outcomes of medical care are becoming much more available in health care organizations and systems of care. This will create new opportunities for operations researchers to make contributions to health care policy and management. To provide some background to those new to the health care area, in this article we do the following: (1) provide a brief exposure to major administrative databases that are available and useful for analyzing outcomes data; (2) discuss the strengths and limitations of the diagnostic information contained in these databases; (3) describe several systems that use this information, or in some cases information from the medical record, to determine patient severity, thus providing a basis for severity-adjustment before considering outcomes; and (4) finally, provide an overview of some recent advances in obtaining improved parameter estimates from large databases.