Effects of Implementing Two Models of Diabetes Shared Medical Appointments on Patient Centered Care and Team Coordination

Context: The Invested in Diabetes study is a cluster randomized pragmatic trial comparing two models of diabetes shared medical appointments (SMAs).

Objective: To examine effects of condition on changes in patient centered chronic illness care and team coordination over time among practices randomized to one of two diabetes SMA models.

Study Design and Analysis: Practices were randomly assigned to implement either patient-driven (PTD) or standardized (STD) diabetes SMAs. Diabetes care team members completed individual self-report surveys at baseline (T0), midpoint (T1; pre-COVID), and final follow-up (T2; during COVID). Descriptive statistics were used to assess mean scores over time. A difference-in-difference analysis using linear mixed effects models was used to test effects of condition on changes in outcomes.

Setting or Dataset: 22 primary care practices in CO and MO.

Population Studied: Diabetes care team members (e.g., health educators, providers).

Intervention: The STD model involves delivery of an evidence-based curriculum by a health educator (HE). The PTD model involves a multidisciplinary care team approach including HEs, peer mentors, and behavioral health providers (BHP). Patients in both conditions have 1:1 visits by a prescribing provider (PP).

Outcome Measures: Patient centered chronic illness care was measured using two domains of the Assessment of Chronic Illness Care (ACIC): Self-Management (SM) and Integration of Chronic Care (ICC); 11-point scale. Team coordination was measured using the Relational Coordination Survey (RCS); 5-point scale.

Results: Across both conditions, mean ACIC-SM scores improved between T0 (M = 7.07) and T1 (M = 7.30) but worsened by T2 (M = 6.51). Similarly, ACIC-ICC scores improved between T0 (M = 6.36) and T1 (M = 6.63) but worsened by T2 (M = 5.83). Mean RCS scores somewhat improved between T0 to T1 across conditions (coordination with the SMAC: 3.20 to 4.29; HE: 3.81 to 4.17; BHP: 3.54 to 4.06; PP: 3.88 to 4.29) but worsened at T2 (SMAC: 3.85; HE: 4.01; BHP: 3.73; PP: 4.08). There was no significant effect of condition on changes in ACIC-SM, ACIC-ICC, or RCS scores at any time point (all p = ns).

Conclusions: Implementing diabetes SMAs was generally associated with improved perceived patient centered care and team coordination, regardless of SMA model. Improvements were not sustained at final follow-up (during COVID), suggesting negative effects of the pandemic on quality of diabetes care.