|
BACKGROUND/RATIONALE:
It is recognized that medical practice varies widely across the United States. Extensive work using Medicare claims has shown many-fold variations in the use of specific services and procedures between different, even neighboring, communities. Underlying the wide variation is a poorly understood characteristic called clinical practice intensity - the tendency of individual providers to order tests, referrals and treatments. Practice intensity is important not only because it is a powerful determinant of health care costs, but also of health care outcomes. The VA in particular needs to understand practice intensity. Because VA providers experience different pressures than their counterparts in the private sector, there is reason to believe they may behave differently.
OBJECTIVE(S):
1. To describe VA primary care physicians' practice intensity and assessment of their VA practice environment.
2. To compare practice intensity and assessment of practice environment between VA physicians and their private sector counterparts.
METHODS:
STUDY DESIGN: National mail survey of VA primary care physicians and non-VA primary care physicians.
SURVEY INSTRUMENT: The survey will include clinical management questions in the form of clinical vignettes as well as questions about physicians' perceptions of their practice environment. Selected VA-specific questions will be developed, field-tested and included in the survey.
Process: VA primary care directors and recognized VA experts in physician survey research will collaborate with us in the development of new questions. Questions will be field-tested using live pilot surveys during scheduled medical conferences, and in focus groups conducted by a contracted survey research firm.
Content: New vignette development will focus on clinical problems common among VA patients. Response options will represent the array of choices available in practice. Vignettes will be graded according to the level of evidence supporting (or opposing) the proposed intervention. Practice environment questions will be developed to assess physicians' opinions of specific aspects of outpatient practice.
SAMPLE: Primary care physicians who practice at least 20 hours per week will be eligible to participate. We will survey a national sample of 500 physicians, stratified by region of the United States (Northeast, South, Midwest, West) in each of two groups:
VA National Physician Sample. A stratified random sample of VA primary care physicians will be identified from the VA PAID database with the assistance of the Workforce Information Systems Team (WIST), and from the PCMM (Primary Care Management Module).
Private Sector National Sample. A stratified random sample of non-VA adult primary care physicians will be identified from the master files of the American Medical Association and the American Osteopathic Association.
ADMINISTRATION: The survey will be administered by mail by Harris Interactive, an international survey research firm. Private sector participants will receive a financial incentive of $100 as an honorarium. We project a 40-50% response rate.
ANALYSIS: We will perform simple descriptive statistics (objective 1) and analyses comparing VA with private sector physicians (objective 2), employing sampling weights equal to the reciprocal of each physician's probability of selection. We will explore how sensitive our results are to the use of poststratification weights. Secondary analyses will control for characteristics of physicians, patients, and practice environment.
FINDINGS/RESULTS:
Focus groups have been completed in Denver and Philadelphia. Survey development is underway; survey questions and clinical vignettes are being revised based on focus group session feedback, which has been invaluable. Sample frames for the VA and non-VA samples are being generated. The survey is anticipated to go into the field during summer or fall 2008.
IMPACT:
By identifying common clinical decisions for which evidence is lacking and there is high variability among physicians, our study will provide important insight into the range of "resource-intensive gray areas" within VA practice. By benchmarking providers and facilities against VA and community practice norms, we wil invite discourse about and target interventions related to improved clinical decision-making in the absence of evidence.
PUBLICATIONS:
None at this time.
DRA:
Health Services and Systems
DRE:
Treatment, Resource Use and Cost
Keywords:
VA/non-VA comparisons
MeSH Terms:
none
|
