2006 HSR&D National Meeting Abstract
3024 — Provider Efficacy and Attitudes and Their Relationship to Treatment Outcomes
Litaker DG (Louis Stokes Cleveland DVA Medical Ctr)
Watts B (Louis Stokes Cleveland DVA Medical Ctr)
Lawrence R (Louis Stokes Cleveland DVA Medical Ctr)
Samaan R (Louis Stokes Cleveland DVA Medical Ctr)
Aron DC (Louis Stokes Cleveland DVA Medical Ctr)
Interventions to enhance health care quality are often directed at modifying physician behavior and are based on assumptions that greater efficacy and perceived importance of specific forms of care will result in better treatment outcomes. This study evaluated the extent to which primary care providers’ (PCP) perceived efficacy and attitudes toward cardiovascular prevention (CVP) correlated with intermediate treatment outcomes of their patients with diabetes mellitus (DM), a high-risk subgroup for whom the benefits of CVP are unequivocal.
PCPs (n=71 faculty physicians, residents, and nurse practitioners) in outpatient clinics of a large VA facility were approached. Subscales from validated instruments assessed: 1) efficacy in treating to recommended goals (4 items); 2) perceived effectiveness at changing CVP-related patient behavior (3 items); 3) whether achieving risk factor control close to goal was sufficient (6 items); and 4) urgency in attaining recommended CVP-related treatment goals (4 items). Intermediate treatment outcomes assessed in diabetic patients seen in our primary care clinics between 7/1 – 9/30/05 included the Framingham Risk Score (FRS: a prediction of 10-year incidence of cardiovascular events), systolic blood pressure (SBP), and the most recent (within the last 6-months) low-density lipoprotein-cholesterol (LDL-c) value. Multivariable multilevel models assessed associations between providers’ attitudes and perceived efficacy with CVP outcomes measured at the patient level, after accounting for patient clustering within provider practices and adjusting for provider and patient characteristics.
Sixty providers (84.5%) responding to the survey provided care to 829 veterans (mean age = 67.1 (+/- 11.3) years). 70.9% were being treated for dyslipidemia and 89.7% received anti-hypertensive medications. 33.6% had a SBP < 130; 88.7% had an LDL-c < 120 (of those with a recent lipid profile); mean FRS was 16.4 (+/- 7.8)%. We observed no significant relationships between CVP outcomes and measures of provider efficacy and attitudes toward CVP despite adequate power to detect small effect sizes.
The absence of important correlations potentially explains the marginal benefit of quality improvement interventions based on continuing medical education.
These findings challenge us to address factors other than provider attitudes in designing interventions to improve the quality of care veterans receive.