Graduation Year

2011

Document Type

Dissertation

Degree

Ph.D.

Degree Granting Department

Public Health

Major Professor

Russell S. Kirby

Keywords

Healthcare Markets, Health outcomes, Health Plans, Hospital market share, Managed Care, Surgical Volume

Abstract

Abstract

Purpose

This study is designed to determine whether managed care has had an influence on the number and distribution of procedures with demonstrated volume and outcome endogeneity in Florida healthcare markets; in addition, methods are developed to determine which measures of managed care activity best predict the impact of managed care on surgical procedure utiliation.

Rationale

A shift in surgical procedure volume on the basis of preferred provider arrangements has the potential to redistribute surgical procedures within hospital markets. The surgical procedures for which such a distribution could have the greatest impact on population health are those for which the volume of cases performed has a strong inverse influence on the outcomes observed. A shift in high risk surgical procedures to low volume hospitals could potentially reduce the number of cases performed at high volume centers and increase cases at low volume centers, adversely impacting quality in both.

Methods

A retrospective population based cohort design is used to capitalize on the variability among Florida metropolitan statistical areas between 1995 and 1999, a period which captured the full business life cycle of managed care plans in Florida. Multiple regression models are used to measure the impact of changes in managed care activity as measured by penetration, index of competition and consolidation on the change in the number and distribution of seven procedures for which volume is associated with patient outcome, controlling for socio-demographic and market factors known to influence surgical procedure utilization. Difference scores derived for each of the model variables were used to measure change from the baseline in 1995 to 1999. Post hoc analysis of the count data models was performed using the cases from all study years in a log linear generalized estimating equation to provide validation of the difference score approach.

Key Findings

Study procedure volume increased over the period, and remained a consistent proportion of the total inpatient surgical procedure volume. Procedure rate remained stable over the study period with substantial small area variation. Change in managed care concentration was consistently and negatively associated with procedure volume at both the MSA (&betaâ&beta = -19.67; p = 0.0489) and hospital level (&betaâ&beta = -4.088; p = 0.0027).Change in the total population and the number of specialty surgeons had a substantial, consistent and positive relationship to change in procedure volume at both the market and hospital level. The change in the index of competition was positively associated with change in hospital market share (&betaâ&beta = 0.1005; p = 0.05); whereas, neither change in managed care penetration nor change in managed care index of competition was predictive of change in procedure volume at the market level. The managed care variables were not correlated when difference scores were tested providing evidence that the managed care variables measure different constructs and behave differently.

Implications

As markets for managed care became more concentrated, the number of surgical procedures with volume and outcome endogeneity declined; the specific reasons for the observed decline require additional study. Competitive managed care markets have a favorable impact on hospital market share for these high risk, high margin procedures. Studies of managed care require consideration of the stage of managed care development in order to understand its influence and the use of difference scores as a method to measure change over time has substantial potential for the study of health care markets.

Share

COinS