Diabetes Support Program with Peers versus Community Health Workers

San Francisco—Patients with type 2 diabetes who participated in a 6-month diabetes self-management education program followed by a 12-month peer-led or community health worker-led diabetes self-management support program maintained their goals throughout the 18 months, according to a randomized, controlled trial.

At 18 months, participants in the peer-led group maintained their improvements in hemoglobin A1c (HbA1c) and blood pressure (BP), but the community health worker group did not sustain the improvements. Patients were recruited from the Community Health and Social Services (CHASS) Center in Detroit, Michigan.

Tricia S. Tang, PhD, the study’s lead author, discussed the results during an oral presentation at the ADA meeting. Peers for Progress, a program of the American Academy of Family Physicians Foundation, funded the study.

“I think these results are very promising,” Dr. Tang said. “What it suggests is that to maintain long-term gains in low-resource settings, we can actually rely on the resources in our community.”

Comparison of Outcomes
In the peer-led group, the mean HbA1c was 8.2% at baseline, 7.5% at 6 months, 7.6% at 12 months, and 7.6% at 18 months. In the community health worker group, the mean HbA1c was 7.8% at baseline, 7.3% at 6 months, 7.4% at 12 months, and 7.5% at 18 months. There were no significant differences between the groups at any of the time points.

The mean low-density lipoprotein (LDL) cholesterol in the peer-led group was 102.1 mg/dL at baseline, 104.2 mg/dL at 6 months, 102.9 mg/dL at 12 months, and 87.8 mg/dL at 18 months. In the community health worker group, the mean LDL cholesterol was 95.5 mg/dL at baseline, 99.4 mg/dL at 6 months, 96.7 mg/dL at 12 months, and 87.1 mg/dL at 18 months. There were no significant differences between the groups at any of the time points, although the community health worker group had a significant reduction at 18 months.

The peer-led group had significant reductions in systolic BP at 6 months, 12 months, and 18 months. Both groups had significant reductions in waist circumference at all 3 time points.

Self-Education
Dr. Tang said self-education diabetes management programs have shown to improve glycemic control and other health outcomes, although the results may not last if there is no continued follow-up and
support. She added that the models for long-term support should be flexible, sustainable, low-cost, and not reliant on paid healthcare professionals.

In this study, the authors used a computerized database to identify potential participants. Individuals were eligible if they had physician-diagnosed diabetes, were ≥21 years of age, had a regular health provider, and self-identified as Latino.

The study included 116 participants: 60 in the peer-led group and 56 in the community health worker group. At 18 months, 37 and 38 participants, respectively, remained in the trial. At baseline, the groups were well-balanced. The mean age was 49.3 years, 41.4% of participants were male, 77.4% had some high school education or less, and 94.4% had a yearly income <$20,000.

The 6-month self-management program included 11 education classes in a group setting that lasted 2 hours each, 2 one-on-one support visits at the patients’ homes by a community health worker, and at least 1 clinic visit accompanied by a community health worker. Previous studies found similar programs led to a 0.8% reduction in HbA1c compared with a control group.

Participants in the peer-led group received 12 months of a diabetes self-management support program that included weekly, group-based support sessions with 2 peer leaders. Participants were encouraged to attend the sessions as frequently as they needed to or as frequently as they could, according to Dr. Tang. During the sessions, 5 core components were covered: (1) discussing self-management challenges; (2) sharing emotions and feelings; (3) engaging in problem solving; (4) asking self-management questions; and (5) setting goals and developing action plans. Peer leaders were also instructed to call participants who did not attend 3 consecutive support sessions.

The peer leaders were ≥21 years of age, had type 2 diabetes, were residents of southwest Detroit, were bilingual, and had participated in the 6-month diabetes self-management education program. They were also required to participate in a 46-hour training program that was conducted over 12 weeks and pass a test that evaluated their diabetes knowledge, active listening, empowerment facilitation, and self-efficacy.

Participants in the community health worker group received 12 months of a diabetes self-management support program that included monthly telephone outreach delivered by a community health worker. The goals of the conversations were the same as with the peer-led group.

The community health workers were CHASS employees, were Latino and bilingual, lived in southwest Detroit, had a high school diploma or general education development, and had experience in facilitating the 6-month diabetes self-management education program. They were also required to have undergone 240 hours of training that included community outreach, diabetes education, home visits, ethics, behavior modification, motivational interviewing, and basic computer skills. They were not required to have diabetes.

Although the findings were promising, Dr. Tang said the results needed to be interpreted with caution because the peer-led group had weekly interventions and the community health leader group had monthly interventions. She added that there was no significant difference in the frequency of contacts between the patients and the peer leaders or community health workers.

Dr. Tang said most of the contacts between patients and peer leaders or community health workers occurred via the telephone instead of in-person, which the authors did not expect. She said a possible explanation as to why the peer-led group did well was because the peer leaders also had type 2 diabetes. “It is completely possible that participants identified more closely with peer leaders,
and [thus] made greater changes,” she said.

Peer leaders cost $1000 per year, while community health workers cost $26,000 per year,
according to Dr. Tang. She recommended future studies examine the cost-effectiveness of these models.

“I am not saying that we should replace community health workers with peer leaders,” she said. “What I think this means is that each model has its [own] unique purpose. To be most cost-effective, we might have community health workers deliver the short-term education in the clinics and [have] peer leaders deliver the long-term support at home.”—Tim Casey