Shih YC, Konrad TR; Academy for Health Services Research and Health Policy. Meeting.
Abstr Acad Health Serv Res Health Policy Meet. 2000; 17: UNKNOWN.
Presented by: Ya-Chen Tina Shih, Ph.D. Research Scientist, MEDTAP International, 7101 Wisconsin Avenue, Suite 600, Bethesda, MD 20854, Tel: 301-664-7273, Fax: 301-654-9864, E-mail: shih@medtap.com.
Research Objective: The objective of this study is to explore factors which affect physician income at various points of the income distribution. In addition to the mean, this study will also examines the median, 10th, 25th, 75th, and 90th percentile levels in order to provide a more comprehensive understanding of the income disparities among physicians.Study Design: This study uses the Physician Worklife Survey data, which is a 1996 national representative mail survey of 2,325 physicians in generalist and subspecialist disciplines drawn from the AMA master file. Information included in the survey were categorized into five factors. Specifically, sociodemographic factors include age, gender, ethnicity, seniority, training specialty, and working hours. Institutional factors refer to the type of practice, ownership status, and numbers of HMOs involved in practice. Management factors measure physicians' attitude toward clinical guidelines, gatekeeping, and business aspects of the practice. Compensation factors are physicians' sources of revenue and the degree of association between compensation and individual productivity. Market factors encompass availability of physicians workforce, geographic, population and economic characteristics of the local market. Weighted least square models are used to examine the relationship between these five factors and physician income at the mean level, and quantile regression models are used to study this relationship at various percentile levels.Principal Findings: Our findings suggest that income differentials between white and non-white physicians were significant at the mean level, but not at the median, 10th, and 90th percentiles. Geographical variations in income were only significant at the mean, 50th, and 75th percentile levels. The associations between compensation factors and physician income were only observed at the median and 75th percentile level, whereas management factors had strongest impact on the 75th percentile. Associations between practice type and income were most significant at the 25th and 75th percentiles. Income differentials between male/female, specialist/non-specialist, owner/non-owner persisted throughout the income distribution. A consistent pattern of positive and significant association between working hours was also found.Population Studied: Practicing physicians in generalist and subspecialist disciplines.Conclusions: We found that factors associated with physician income at the mean level may not necessarily affect income at other points of the distribution, and vice versa. Policies that targeted at the "right" group will achieve the desired impacts in a more effective manner.Implications for Policy, Delivery or Practice: Examinations of physician income beyond the mean level provides more complete information for: (1) projecting impacts of regulatory and/or market-based policy initiatives on different segments of the physician labor market which might in turn impact on their political reaction to or support for such initiatives; (2) identifying effective policy parameters when designing manpower policies targeted at improving geographic distribution of physicians or income inequality among physicians.Primary Funding Source: Robert Wood Johnson Foundation
Publication Types:
Keywords:
- Data Collection
- Efficiency
- Ethnic Groups
- Female
- Gatekeeping
- Health Services
- Income
- Male
- Physicians
- Physicians, Family
- Regression (Psychology)
- Specialties, Medical
- manpower
- supply & distribution
- hsrmtgs
Other ID:
UI: 102272802
From Meeting Abstracts