2011 HSR&D National Meeting Abstract
3105 — From Exemplar Models of Care to Systemwide Quality Improvement: Ratings of Care with the VA Natural Experiment in Diffusion of Primary Care for Women
Washington DL (VA Greater Los Angeles Healthcare), Bean-Mayberry B
(VA Greater Los Angeles Healthcare), Riopelle D
(VA Greater Los Angeles Healthcare), Yano EM
(VA Greater Los Angeles Healthcare), Sun S
(VA Greater Los Angeles Healthcare), Mitchell MN
(VA Greater Los Angeles Healthcare)
Prior research found gaps in VA access and care quality for women veterans. Subsequent policy recommended VA sites adopt one of the primary care delivery models identified as optimal for women’s care (full-service women’s primary care clinics and/or teams providing comprehensive care, including basic gender-specific care). Our objective was to determine the impact of VA women’s health practice structure on women veterans’ ratings of their VA healthcare.
The 2008-09 National Survey of Women Veterans enrolled a population-based sample of women veterans. Measures included scales to rate VA healthcare in the prior 12-months on quality (CAHPS 0-to-10), gender-sensitive satisfaction (range 1-to-5), and access (range 1-to-4), with higher scores being better. The 2007 VHA Survey of Women Veterans Health Programs and Practices identified VA primary care practice arrangements for care delivery to women. VA healthcare users with matching information on women’s primary care practice structure comprised the analytic sample (n = 1,749 across 404 sites). We conducted multi-level random intercepts linear regression models, weighted by probability weights to represent the U.S. women veteran population. Statistical corrections for multiple comparisons were applied.
Sixty percent of large sites had implemented optimal women’s care models. Adjusting for variations in socio-demographics, health, and VA utilization, there was higher gender-sensitive satisfaction among women using sites that had both a women’s clinic and designated providers for women in the general primary care clinic, compared with women using sites with fully integrated primary care (3.6 versus 3.3, p = 0.0026). Adjusted ratings of care were similar between women veterans using sites with recommended care models and those using sites with other practice arrangements, though there was a non-statistically significant trend toward higher ratings from women using sites with recommended care models (8.1 versus 7.9 for quality; 3.4 versus 3.3 for access).
The VA natural experiment in diffusion of comprehensive primary care for women is not sufficient to assure availability of high quality gender-sensitive continuity care for women in all VA settings.
Rather, directive actions are required to assure systemwide uptake and complete implementation of comprehensive primary care for women. Our findings provide baseline data against which such transformative activities may be benchmarked.