Ahern BM, Molinari C; Academy for Health Services Research and Health Policy. Meeting.

Abstr Acad Health Serv Res Health Policy Meet. 2000; 17: UNKNOWN.

Presented by: Melissa Ahern, Ph.D., and Carol Molinari, Ph.D., Washington State University-Spokane, 668 N Riverpoint Blvd, Box B, Spokane, WA 99202-1662, 509-358-7984, FAX 509-358-7900, E-mail: ahernm@wsu.edu, molinari@wsu.edu.

Research Objective: This study estimates (1) the relationship between structural characteristics of HMOs and utilization outcomes; and (2) the relationship between structural characteristics and the management processes that may affect utilization. In this study, we are particularly interested in the impact of ownership on utilization of HMO services and on an HMO's choice of management processes to control utilization. Study Design: Data are from the American Association of Health Plans national 1995 survey. We estimate the relationship between structural characteristics and utilization through ordinary least squares. Utilization is operationalized across three utilization variables (dependent variables): discharges per 1000 members; days per 100 members; average length of stay; and cardiac catheterization procedures per 1000 members. Structural characteristics (independent variables) include HMO size, for-profit status, model type, geographic region, length of time in business, proportion of members across payers (Medicare, Medicaid, commercial, and other government), and predominant ownership type (insurance-owned, hospital-owned, physician-owned, and owned by a national managed care company). Using a second set of logistic regression models, we estimate the probability that HMOs with particular structural characteristics use particular medical management processes. Process measures (dependent variables) are operationalized using dummy variables representing (1) risk-bearing by providers (e.g., capitation payment); (2) risk-bearing by consumers (e.g., copays for hospitalization and emergency room use); and (3) other medical management strategies (gatekeeping, provider profiling, catastrophic case management, provider discounts, physician recredentialing, physician peer review, and quality assurance program. The independent structural variables are operationalized as in the first set of regression models.Population Studied: Organizations in the United States that have identified themselves as predominately HMOs, even though many of them have other lines of business. Principal Findings: We find that hospital utilization (discharges and days per thousand) is significantly negatively associated with insurance ownership. In contrast, insurance-owned HMOs have longer lengths of stay. Physician-owned HMOs are associated with greater use of cardiac catheterization procedures. Other factors significantly related to utilization include the proportion of Medicare and commercial membership, geographic region of operation, and model type. In particular, Medicare membership is associated with higher levels of hospital discharges, hospital days per thousand members, and longer lengths of stay. Region 1 (New England and Mid-Atlantic) is associated with more hospital discharges, days per thousand, and average length of stay, as well as higher levels of cardiac catheterization procedures. Staff model HMOs have the lowest lengths of hospital stays, and network models also have lower lengths of stay than group and IPA models. To further understand why particular structural characteristics are associated with utilization, we model the relationship between structural characteristics and medical management processes. We find that insurance-owned HMOs are more likely than all other HMOs to have hospital risk-sharing, provider capitation, and provider discounts. In contrast, national MCO-owned HMOs are less likely to use provider risk-sharing, and out-of-pocket payments for hospital use. In addition, hospital-owned HMOs are more likely to use hospital risk-sharing. In addition, for-profit HMOs are more likely to use provider financial incentives. Network models are more likely to use both provider and hospital risk-sharing. Staff model HMOs are less likely than IPA HMOs to use physician risk-sharing and out-of-pocket payment for ER use. Finally, national MCO-owned HMOs are less likely to use gatekeeping and catastrophic case-management. Conclusions: We find that HMO ownership makes a difference for both utilization outcomes and management strategies chosen to impact utilization. Primarily, insurance-owned HMOs is associated with lower hospital utilization for both discharges and days per thousand members. Further, insurance-owned HMOs are substantially more likely than other HMOs to use hospital risk-sharing, provider capitation, and provider discounts. These results suggest that insurance-owned HMOs have lower utilization because they are more likely to use financial provider incentives to control utilization. We also find that provider-owned HMOs (physician-owned and physician-hospital-owned) have a higher propensity to use their own services (more cardiac catheterization procedures). Implications for Policy, Delivery, or Practice: As the health care delivery system consolidates through mergers and acquisitions, the structure of integrated delivery systems like HMOs are evolving into more adaptive organizational configurations. For example, HMOs owned by insurance companies appear to be more efficient delivery systems. This suggests that as health care becomes more competitive, those HMO delivery systems whose owners have expertise in managing risk fare better in controlling costs than other HMOs. In addition, as HMO markets become more competitive, HMOs may have less leeway to indulge in ownership biases, and HMOs may become more homogeneous in their approaches to managing utilization.Primary Funding Source: WSU-Spokane, Health Policy and Administration program.

Publication Types:

• Meeting Abstracts

Keywords:

• Case Management
• Delivery of Health Care
• Delivery of Health Care, Integrated
• Health Facilities, Proprietary
• Health Maintenance Organizations
• Hospitalization
• Hospitals
• Insurance, Health
• Managed Care Programs
• Medicaid
• Medicare
• New England
• Ownership
• Physicians
• United States
• economics
• organization & administration
• utilization
• hsrmtgs

Other ID:

• GWHSR0000900

UI: 102272574

From Meeting Abstracts