Sales A, Conrad D, Liang SY, Tufano J, Maynard C; Academy for Health Services Research and Health Policy. Meeting.
Abstr Acad Health Serv Res Health Policy Meet. 2000; 17: UNKNOWN.
Presented by: Anne Sales, PhD, Research Scientist, HSR&D, VA Puget Sound Health Care System, 1660 S. Columbian Way, Seattle, WA 98108.
Research Objective: Data from large-scale surveys conducted by national membership organizations are often used in research to explore organizational behavior, particularly medical practice groups. Response rates to these surveys are often well below 50%. This paper examines factors that predict whether a medical group practice responds to either or both of two surveys from a national organization focused on medical group practices. This is particularly important when focusing on impact of managed care on markets.Study Design: We used a membership database furnished by the Medical Group Management Association (MGMA), a national medical group management organization, together with data on which group practices responded to either or both of two annual surveys conducted by MGMA. One of the surveys is the Production and Compensation (P&C) Survey; the other is the Cost Survey. Both survey instruments are long and require extensive amounts of data from the practices. We used logistic regression to predict response to either of the surveys separately. Variables included in the analysis were: group size, measured by number of FTE physicians in the practice; practice ownership; region of the country; and a categorical variable giving the proportion of practice revenue derived from at-risk contracts. A total of 5725 practices were included in the study.Population Studied: Medical group practices in the United States belonging to the MGMA.Principal Findings: 22.2% of practices responded to the Production and Compensation Survey, while 20.8% responded to the Cost Survey. Pseudo-R2 for the models was low, ranging from 1.8 to 2.8%, although the models were significant overall. The only variable predicting decreased likelihood of responding to either surveys was hospital ownership of the practice. For-profit ownership was a significant predictor of non-response to the P&C Survey. Larger practices were more likely to respond to either the P&C Survey or the Cost Survey. As proportion of revenues from at-risk contracts increased, response rates increased.Conclusions: We expected the findings that larger practices would be more likely to respond to the surveys. The finding that increased proportion of at-risk revenue predicts higher response rate was somewhat surprising, since our prior hypothesis was that practices with higher proportions of at-risk revenue may not have the administrative resources to respond to complex, long survey instruments. Implications for Policy, Delivery or Practice: The finding related to size of practice suggests that it might be worth developing shorter, more focused instruments targeted to small practices enabling the membership organization to gather essential information from all types of practices. The finding related to at-risk revenues suggests that if the information being requested in the survey instrument is viewed as responding to critical information needs, practices are more likely to respond to the surveys. This raises particular issues for researchers focusing on managed care and markets.
Publication Types:
Keywords:
- Data Collection
- Federal Government
- Group Practice
- Managed Care Programs
- Organizations
- Ownership
- United States
- economics
- organization & administration
- hsrmtgs
Other ID:
UI: 102272583
From Meeting Abstracts