Background: Diabetes group visits (GVs) - shared appointments that incorporate medical care, diabetes self-management education, social support, and goal-setting - were recently examined in a network of U.S. community health centers (HCs) and found to improve glucose control and diabetes-related quality of life.

Objective: To evaluate the economic impact of group visits (GV) in adults with diabetes in HCs.

Methods: In a pilot prospective matched case-control trial, we implemented 6 monthly GV sessions in 5 HCs and compared intervention patients (n=49) to control patients receiving usual care within the same HCs (n=72). We conducted patient chart reviews to obtain healthcare utilization data for the prior 6 months, at baseline, 6, and 12 months and collected monthly logs of HC expenses and staff time spent on activities related to GVs, including training, planning, implementation, and evaluation. Total costs included HCs’ expenses and activities and patients’ healthcare use over one year. Wilcoxon rank-sum tests, generalized linear models, and modified Park tests were used to compare the healthcare use and cost data.

Results: Despite each HC spending $17,342 on average (or $1770 per-patient) for the GV preparation and implementation, the per-patient annual total costs were not different between the groups (GV: $3,982 vs. control: $4,287, p=0.73). Most of the benefits were due to GV patients having fewer hospitalizations (p=0.08) and fewer primary care visits that occurred outside the trial (p=0.01). After excluding the staff training costs (mean $7,962 per HC), the GV group had significantly lower total costs ($3,169) than the control group (p <0.01).

Conclusions: The diabetes group visits in health centers improved patients’ healthcare quality without increasing overall healthcare costs. In subsequent GV sessions, excluding the need for staff training, group visits may be cost saving. Future studies should assess the cost-effectiveness of group visits.


W. Wan: None. E.M. Staab: None. S.A. Ham: None. A. Campbell: None. C.T. Schaefer: None. M.T. Quinn: None. A. Baig: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at