Comparative Effectiveness of Two Approaches to Diabetes Management in the Uninsured

The results of the pre-pilot and pilot studies will be used to determine the feasibility of a fully powered trial that will test the effectiveness of two different approaches to diabetes management in an uninsured diabetic population who receives their care from a community based free clinic, namely nurse-directed interdisciplinary care management compared to peer-led self management. Effective programs for the uninsured diabetic will benefit both the individual and society by preventing serious illness, decreasing mortality and disability, decreasing medical debt, stimulating economic growth, improving business productivity, reducing job lock, decreasing health disparities, improving quality of life, and reducing cost shifting by decreasing uncompensated care. Findings will be generalizable to uninsured and insured patients across a wide spectrum of chronic conditions.

The randomized pre-pilot and pilot studies will inform the design of a fully powered trial that will test the effectiveness of a nurse-led interdisciplinary care management intervention (CM) for improving hemoglobin A1c in uninsured diabetic patients who receive their primary medical care from a community-based free clinic compared to similar patients who undergo peer-led self management (SM). The comparative effectiveness of these approaches has not been tested in a randomized trial.

To ensure the success of the pilot study the investigators will perform a 6 month pre-pilot study using 10 patients per group. This pre-pilot will be used to establish protocols covering enrollment, HIPAA, consent, interventions and outcome measurement. The subsequent 1 year pilot will be used to demonstrate feasibility, estimate recruitment, attendance and follow up, refine inclusion criteria, provide power analysis, optimize the interventions and determine appropriate secondary outcomes for a fully powered trial. For the pilot, thirty patients will be randomized to the CM intervention for the first 6 months. Major components of the nurse-led CM model include: 1) assessment and goal setting for self-management; 2) education for self efficacy and productive clinical encounters; 3) access to specialist care via an interdisciplinary team; 4) evidence-based treatment recommendations responsive to patients' goals; and 5) care coordination by a nurse care manager. From 6-12 months patients will return to usual clinic care.

The 30 patients randomized to the SM group will be invited to attend weekly peer-led SM training broadly based on the Stanford University Diabetes Self-management Program's Train the Trainer Model. The training will occur over 6 weeks and will cover topics common to most chronic disease SM programs including: goal setting, self-management of chronic conditions, making action plans, feedback and problem solving, handling emotions, symptom management techniques, communication with health care providers, medications and making treatment decisions. The peer advisors and a nurse will then lead 1 hour monthly group sessions for the remainder of the first 6 months for review/revision of action plans, review of self management techniques and ongoing social persuasion and modeling. From 6-12 months patients will return to usual clinic care.

For the pilot, outcome measures will be obtained at 6 and 12 months. For this pilot, the investigators will not test any hypotheses. Instead, effect sizes and confidence intervals will be calculated for each outcome. These results will be used for power analysis. The results will also allow us to see which interventions are most effective on which outcomes and allow us to optimize the interventions to have the greatest impact on the primary outcome, hemoglobin A1c. Cost effectiveness will be determined in a fully powered trial.