Defense Date


Document Type


Degree Name

Doctor of Philosophy


Health Administration

First Advisor

James W. Begun


The study purpose is to link hospital structure, represented by each hospital’s professional contingent, service mix, and inpatient capacity; and its environment, characterized by the penetration of managed care enrollees. The secondary purpose is to test the relationship between hospital structural change and subsequent hospital performance.

The study employs a non-experimental panel design, with a sample of 1882 community hospitals (service type: general medical and surgical). Environmental variables are measured for the base year 1989. Hospital structural variables are measured for 1989 and 1994, with change variables computed. Performance variables are measured for 1989 and 1995, with change computed for cost measures. Hospital structural change is viewed as a dependent variable related to the environment, as well as an independent variable related to performance.

Descriptive data are extracted from the American Hospital Association Annual Survey of Hospitals. Hospital cost performance data are from the Health Care Financing Administration Prospective Payment System Minimum Data Sets. Hospital mortality data for 1989 are from Medicare Hospital Mortality Information.

HMO enrollment data are extracted from the Interstudy Edge and aggregated to metropolitan statistical area (MSA) level. Market competition data are from the 1989 Area Resource File. A Herfindahl-Hirschman index (HHI) is calculated for each hospital’s MSA.

Analytical hypotheses are tested using ordinary least squares (OLS) technique. Results from Part 1 suggest that where HMO penetration was relatively high, sample hospitals tended to contain growth in their registered nurse (RN) staff between 1989 and 1994. Higher HMO penetration is also associated with more stabilization in occupancy rates, preventive services, and ambulatory workload. In contrast, market competition is associated with changes to a higher Medicare case-mix index (CMI), and increase in ambulatory visits.

Results from Part 2 indicate positive associations between increased RN staff and hospital cost growth between 1989 and 1995. Hospitals which did not experience an increased CMI are similarly linked with cost growth. Alternatively, reduction in hospital bedsize is associated with more controlled growth in hospital cost per patient day. Several control variables display noteworthy associations with the variables of interest. Theoretical and management implications for community hospitals are discussed.


Scanned, with permission from the author, from the original print version, which resides in University Archives.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission