Long-Term Impact of Evidence-Based Quality Improvement for Facilitating Medical Home Implementation on Primary Care Health Professional Morale Lisa S. Meredith, Ph.D. Benjamin Batorsky, M.A. Doctoral Fellow Matthew Cefalu, Ph.D. Jill Darling, M.S.H.S. Susan Stockdale, Ph.D. Elizabeth M. Yano, Ph.D., M.S.P.H. Lisa V. Rubenstein, M.D., M.S.P.H. AcademyHealth June 26, 2017
New models of care hold promise • New patient-centered primary care models can potentially improve primary care provider (PCP) and staff morale – Patient-Centered Medical Home (PCMH) – Including the VA’s Patient Aligned Care Team (PACT) initiative (launched in 2010) • They also may improve efficiency by reducing unnecessary utilization and costs, and ultimately improve patient care • And can be associated with positive experiences 2
But implementation is challenging • May increase PCP/staff burnout due to high levels of transformational change required • Large literature on prevalence/causes of burnout but little on how healthcare organizations can work to reduce it. • Use of evidence-based quality improvement (EBQI)* to facilitate change is promising – Multi-level strategy to promote regional and local primary care practice engagement in innovation – Ease potential PCP/staff burnout during system- wide transformation to PCMH 3 * Rubenstein et al., 2010; 2014
Veterans Assessment Improvement Laboratory (VAIL) for Patient-Centered Care • 1 of 5 diverse VHA PACT Demonstration Laboratories funded by the VHA Office of Patient Care Services* in 2010 • To develop and implement systematic methods for supporting and evaluating the VHA’s transition to the PCMH model (PACT) • Used EBQI as in improvement intervention in one regions to engage practices in innovations and easy potential burnout *to Drs. Rubenstein, Yano, and Altman. 4
EBQI Intervention • Promotes cross-discipline, data-driven problem solving in local primary care practices • Aligns local practices with organizational priorities to sustain successful QI innovations over time and spread them across teams/clinics • Focuses on engaging and empowering front-line primary care teams with multi-level, interdisciplinary stakeholders in structured EBQI 5
EBQI Proposal Review and Approval Process • Solicited brief innovation proposals (n=71) submitted through – The EBQI practice’s quality council (supported by a quality council coordinator) OR – An across-EBQI site VAIL workgroup • Additional support for approved projects – Limited release time for leaders (per prior MoU) – Priority setting by regional leaders (administrators, quality, medical care, information technology, patient advocacy, & pharmacy experts) who reviewed/rated proposals – Annual collaborative learning sessions for EBQI practices – Local primary care site audit and feedback (practice level data on patients, providers/staff) for quality councils 6
Approved EBQI Projects (n=26) • Received additional support: – Rapid innovation evidence review – Budget of $12,000 – QI facilitation for project management and measures • Successful projects generated toolkits (n=12) if the innovation spread to at least 1 other site • There were 6 approved and completed projects plus 6 additional volunteer projects completed during the reported time period that addressed burnout 7
Study Design and Intervention Evaluation • Quasi-experiment to assess change over time in attitudes/experiences with PACT implementation across 42 months • Cohort of 356 PCPs/staff • Compared the impact of PACT transformation alone to PACT + EBQI on burnout (emotional exhaustion) – 3 Early EBQI intervention clinics (August 2010) – 3 Late EBQI intervention clinics (May 2012) – 17 Comparison clinics 8
Data: 3 Waves of Surveys Wave 1 Wave 2 Wave 3 (11/30/11 – 3/13/12) (8/1/13 – 10/11/13) (9/10/15 – 1/8/16) 9
Survey Response Rates by Staff Type and Wave 10
Survey Response Rates for EBQI and Comparison Practices by Wave EBQI Comparison 70 59 59 60 55 49 48 50 38 40 % Complete 30 20 10 0 Wave 1 Wave 2 Wave 3 11
Demographic Characteristics EBQI Comparison All Characteristic (n=181) (n=175) (n=356) Female, n (%) 124 (67) 121 (70) 245 (69) Latino, n (%) 20 (11) 15 (9) 35 (10) Non-white Non-Latino, n (%) 87 (47) 71 (41) 158 (44) Age, mean years (SD) 47.4 (10) 47.6(11) 46.8 (10.9) Years in clinic, mean (SD) 8.0 (8.1) 5.2 (7.1) 7.0 (7.7)* *p<01, where EBQI and comparison employees differ significantly for these variables. 12
Professional Characteristics EBQI Comparison All Characteristic (n=181) (n=175) (n=356) Physician* _ 75 (21) Gen Practice/Family Med 3 (2) 7 (4) 10 (3) Internal Medicine 39 (22) 20 (11) 59 (17) Other Specialty 3 (2) 3 (2) 6 (2) Nurse Practitioner 12 (7) 16 (9) 28 (8) Physician Assistant 2 (1) 2 (1) 4 (1) Registered Nurse 48 (27) 49 (28) 97 (27) Licensed Practical/Voc.Nurse 37 (20) 41 (23) 78 (22) Mental Health Professional 5 (3) 9 (5) 14 (4) Dietician or Nutritionist 5 (3) 3 (2) 8 (2) Pharmacist 11 (6) 12 (7) 23 (6) MedicalTech/Assistant/Clerk 16 (8) 13 (7) 28 (8) *Other specialty includes rheumatology, geriatrics, and infectious disease. Data are missing for 32 physicians. 13
Change in Burnout Scores (Emotional Exhaustion) Across Wave by Intervention Group for PCPs and Staff PCPs Staff 30 30 25 25 20 20 15 15 Wave 1 Wave 2 Wave 3 Wave 1 Wave 2 Wave 3 14
Change in Job Satisfaction Across Wave by Intervention Group for Primary Care Providers PCPs Staff 4 4.1 3.9 4 3.8 3.9 3.7 3.8 3.6 3.7 3.5 3.6 3.4 3.5 3.3 3.4 3.2 3.1 3.3 Wave 1 Wave 2 Wave 3 Wave 1 Wave 2 Wave 3 15
Summary of Findings and Implications • EBQI was effective in reducing burnout for PCPs (but not staff) relative to PACT alone – Effect sizes of 0.40-0.50 of a standard deviation on the 0-54 point EE dimension of burnout • EBQI to support PCMH transformation may alleviate burnout and reduce variation in implementation outcomes across clinics during early implementation • EBQI approach is consistent with recommendations from West et al. (2016) review or interventions to reduce burnout – Used both individual and organizational strategies to engage providers and leadership 16
Questions? lisa_meredith@rand.org 17
Results from Regression Models for Change in Burnout PCPs Staff Effect EE Change (CI) EE Change (CI) Difference in Differences (Wave 3 – Wave 1) Early EBQI-PACT Intervention vs. Comparison Group -1.42 (-8.74, 5.90) -1.44 (-7.02, 4.15) Late EBQI-PACT Intervention vs. Comparison Group -1.30 (-6.72, 4.11) -6.82 (-13.29, -0.35)* Change within Group (Wave 3 – Wave 1) Comparison Group 4.96 (0.66, 9.25)* 0.84 (-2.28, 3.96) Early EBQI-PACT Intervention 3.54 (-2.53, 9.60) -0.60 (-5.28, 4.08) Late EBQI-PACT Intervention -1.86 (-6.84, 3.11) -0.46 (-4.98, 4.06) Change from Wave 1 (Comparison Group) Wave 1 (Reference group) Wave 2 1.75 (-2.18, 5.69) 0.03 (-2.77, 2.83) Wave 3 4.96 (0.66, 9.25)* 0.84 (-2.28, 3.96) Difference between Comparison Group and Early EBQI Wave 1 -1.66 (-8.69, 5.36) 0.41 (-4.80, 5.63) Wave 2 0.52 (-6.07, 7.11) -0.12 (-4.58, 4.33) Wave 3 -3.08 (-10.28, 4.12) -1.02 (-5.88, 3.84) Difference between Comparison Group and Late EBQI Wave 1 0.59 (-5.38, 6.56) 0.10 (-4.29, 4.48) Wave 2 -0.64 (-6.88, 5.60) 0.06 (-4.18, 4.30) Wave 3 -6.23 (-13.04, 0.59) -1.21 (-6.39, 3.98) 18 *p<.05; **p<.01.
Recommend
More recommend
