Search | Search by Center | Search by Source | Keywords in Title
Nelson KM, Sylling PW, Wong ES, Taylor LL, Helfrich CD, Curtis I, Schectman G, Stark R, Fihn SD. Implementation of the Patient Centered Medical Home (PCMH) in the Veterans Health Administration (VHA): Associations with Clinical Outcomes, Patient Satisfaction, Provider Burnout and Health Care Use. Poster session presented at: Society of General Internal Medicine Annual Meeting; 2015 Apr 22; Toronto, Canada.
BACKGROUND: In 2010, the Veterans Health Administration (VHA) began implementing the patient centered medical home (PCMH). To evaluate progress and outcomes system wide, we examined the association of PCMH implementation in 2012 and 2013 with important outcomes including patient satisfaction, staff burnout, health care use and clinical quality. METHODS: We conducted an observational study using: (1) VHA clinical and administrative data for > 5.6 million Veterans who received primary care at 923 VHA hospital- and community-based outpatient clinics (CBOCs); (2) n = 279,841 responses to the Consumer Assessment of Health Plans PCMH module administered to a weighted, random sample of Veterans who received outpatient care in 2013; and (3) 6,959 responses to a survey distributed to all VHA primary care providers and staff in 2013. Measures: To measure PCMH implementation, we utilized a previously validated instrument, the PACT implementation progress index (PI2) comprised of 53 items from 8 core PCMH domains: access, continuity, care coordination, comprehensiveness, self-management support, patient-centered care and communication, shared decision-making and team-based care. In order to assess site PI2 performance across years, we used 2012 as a reference year, and standardized 2013 items in the PI2 index by the 2012 average and standard deviation of items across sites. A 2013 overall PI2 score was assigned to each clinic based on the number of domains in the top and bottom 2012 quartiles for the domain scores, ranging from 8 (all domain scores in the top 2012 quartile) to -8 (all domain scores in the bottom 2012 quartile). Our main outcome measures were patient satisfaction (defined by a provider rating from 0 - 10), staff burnout (measured by a single item question), health care use (hospitalization and emergency room use), and VHA clinical quality indicators. Statistical analysis: We used a non-parametric test of trend for the ranks across ordered groups to test for trends in patient satisfaction and provider burnout by PI2 scores. We examined 2013 utilization of emergency department and urgent care visits and total hospitalizations for high versus low implementation sites, adjusting for age, CBOC clinics and co-morbidity. To assess how trends in hospitalizations changed following the PACT initiative in 2010, we estimated time-series models of ambulatory care sensitive condition (ACSC) hospitalizations and all-cause hospitalizations from 10/1/2003 through 9/2013 for each facility. Changes in hospitalizations were calculated as the difference between the observed rate of admissions and the predicted rate of admissions that would have occurred had PACT not been implemented, during the period after PACT initiation. We tested differences in proportions of eligible patients at each VHA clinic fulfilling each of 44 quality indicators by the degree to which implementation of PACT was successful, as measured by the PI2. We calculated rates of services at the facility level by dividing the number of patients who satisfied the quality measure by the number of patients who met inclusion criteria for each quality measure. RESULTS: PCMH areas with the greatest gains from 2012 to 2013 included the percentage of Veterans using secure email messaging with their providers, the percentage of patients enrolled in home tele-monitoring, increased use of telephone clinics, and increased clinic staffing to recommended ratios. Among sites achieving the most effective implementation, we found improvements in patient satisfaction, provider burnout, and emergency room utilization. Patient satisfaction was meaningfully higher among sites that had effectively implemented PACT than those that had not (mean rating for satisfaction with provider 9.24 vs. 7.74, p < 0.001). A similarly favorable pattern was observed for staff burnout (34% vs. 44% of staff reporting burnout, p = 0.011). Emergency department and urgent care encounters were marginally lower at sites with more rather than less effective implementation (205 vs. 222 encounters per 1000 patients, p = 0.091). No difference was noted in all-cause hospitalization. However, in the interrupted time series analysis, sites with 2013 PI2 scores in the highest decile had higher rates of ACSCs and all-cause hospitalization. 17 of 44 clinical performance measures were significantly associated with PI2 scores, with more effective PCMH implementation associated with higher quality, including LDL cholesterol of < 100 for patients with IHD (64.1% vs. 72.8%, p < 0.001) or diabetes (64.1% vs. 70.8%, p < 0.001), influenza immunization (74.8% vs. 78.6%, p = 0.021), and offering medications for tobacco cessation (94.1 vs. 96.4%, p = 0.044). Overall, 38 of the 44 measures were higher at sites with the highest PI2 scores. CONCLUSIONS: Sites with sustained levels of PCMH implementation had higher patient satisfaction, lower provider burnout and better clinical quality.