504 – Providor Profiling and Predicting Hospital Outcomes
Can Planned Visit Scores Predict Practice Improvements in Patient Experience Outcomes? A Meta-Regression Study
Gerald K. Arnold
American Board of Internal Medicine
Rebecca A. Baranowski
American Board of Internal Medicine
Lauren Duhigg
American Board of Internal Medicine
Planned visits protocols (PVP) help care teams treat chronic care patients: patients receive evidence-based care and self-management training. Practices completing PVP received better patient-experience scores (PROM) than matched-controls: PROM in PVP practices were higher in involving patients in care plans (p=.005) and helpful and courteous staff (p< .001). Study used 1,402 internists. Physicians completed ABIM Practice Improvement Module� on physician communication. Patients used CAHPS-based surveys. Patients:13,699, baseline and 12,646, follow-up. 201 physicians reported PVP; controls: 1,201 practices without PVP. PROM: physician communication (PC: Cronbach a=.87); shared decision-making (SDM: a = .65); staff helpfulness/courteousness (SHC: a=.84); lab results follow-up (LF: a= .86); and overall physician rating (OR: a=.89). Compared quality effect sizes (ES):practices with PVP or not. Dependent variable: Practice ES; predictors: planned-visit (PV) score, PVP dummy, and interaction. PV score predicts PC (p=.01), SHC (p=.03), LF (p=.03), and OR (p=.03); not predictive of SDM (p=.33). Results apply to minimally important differences in PROM at practice or physician levels.