Craig BM, Dusansky R, Koc C; Academy for Health Services Research and Health Policy. Meeting.
Abstr Acad Health Serv Res Health Policy Meet. 2000; 17: UNKNOWN.
Presented by: Benjamin Matthew Craig, M.S. Graduate Trainee, Department of Preventive Medicine, University of Wisconsin - Madison, WARF Building, Room 749, 610 Walnut Street, Madison, WI, 53705. Tel: 608-263-2820; Fax: 608-263-2820; E-mail: bmcraig@students.wisc.edu.
Research Objective: An individual's utilization of medical services is largely influenced by the presence of medical insurance. However, the effect of insurance on utilization, known as the moral hazard effect, is not uniform across medical services. This paper examines the differences in the moral hazard effect across medical services after removing the potentially confounding determinants of insurance status. Study Design: Using the 1993 National Health Interview Survey and its health insurance supplement, we estimate the utilization of medical services based on a modified version of the Michael Grossman's classic model of the demand for health and health capital. Eight medical services are studied separately; five physician services: physician visits, general practitioner visits, specialists visits, checkup visits, and condition-specific visits; two hospital services: nights in the hospital, and operations; and one emergency service: emergency room visits. To attain consistent parameter estimates, we utilized a series of parametric and semi-parametric techniques including: maximum likelihood estimation, generalized method of moments (GMM) estimation, and a recently developed GMM technique that incorporates an instrumental variable to remove the confounding effects of the private medical insurance choice, if required.Population Studied: Civilian noninstitutionalized adults between the ages of 18 and 64 in the United States.Principal Findings: For physician and hospital services, the presence of insurance (private or public) increases utilization. Private insurance increases the utilization of physician visits by 153%, while public insurance increases the utilization by about 122%. Insurance increases the utilization of specialist visits (Private: 152%, Public: 169%) more than general practitioner visits (Private: 56%, Public: 74%). For hospital services, private insurance increases utilization more than public insurance, including nights in the hospitals (Private: 92%, Public: 44%) and operations (Private: 78%, Public: 43%). This relationship is reversed in the utilization of emergency room visits (Private: - 55%, Public: 110%). The specification tests between the estimation techniques identified the choice of private medical insurance as a confounder in estimated utilization of physician services, but this confounder is not identified in the utilization of hospital or emergency services. Conclusions: The change in utilization associated with the presence of insurance depends on the service in question and whether the insurance is private or public. For example, private insurance decreases the utilization of emergency room visits by over 50%, while public insurance increases its utilization by over 100%. Also, the identification of private insurance determinants as confounders for the use of physician services suggests that privately insured individuals may account for future utilization of physician services, but not the utilization of hospital or emergency services.Implications for Policy, Delivery or Practice: If individuals choose private insurance based on expected utilization of physician services, as these results suggest, insurance policies that offer greater coverage for physician services may be more attractive. This produces a possible rationale for the widespread prevalence of insurance that promotes physician services, but not hospital or emergency services. Primary Funding Source: None.
Publication Types:
Keywords:
- Adult
- Health Services
- Health Services Accessibility
- Health Services Research
- Humans
- Insurance Coverage
- Insurance, Health
- Physicians
- Physicians, Family
- United States
- economics
- injuries
- methods
- utilization
- hsrmtgs
Other ID:
UI: 102272509
From Meeting Abstracts