Major Finding: At 6 months, the mean hemoglobin A1c of patients in the intervention group decreased from 8.02% to 7.73%, while the mean hemoglobin A1c of the controls increased from 7.93% to 8.22%.
Data Source: Randomized, controlled trial comparing a peer-support intervention with conventional nurse-led case management in 244 men who had poor glycemic control and were treated in a Veterans Affairs program.
Disclosures: Dr. Heisler had no financial conflicts to disclose.
MINNEAPOLIS — A peer-support intervention was associated with better diabetes control, compared with conventional nurse-led case management, in a 6-month Veterans Affairs study of men with poor glycemic control.
In the randomized prospective study, hemoglobin A1c (HbA1c) levels, insulin starts, and self-reported social support significantly improved in the 125 men with diabetes and HbA1c levels higher than 7.5% who were enrolled in a peer-support intervention. The outcome measures did not improve in 119 matched patients who were randomized to usual care and conventional nurse-led case management, Dr. Michele Heisler reported.
Additionally, peer support was far less time intensive from a staff and resource perspective than other tested programs that have shown similar or less-significant improvements, Dr. Heisler said.
Blood pressure changes during the study were not significantly different for the two groups. Levels of diabetes distress and diabetes social support were assessed based on patient interviews, and new insulin starts were documented from patients' medical records.
For the study, all participants attended an initial session led by a Veterans Affairs (VA) nurse case manager, during which their baseline HbA1c and blood pressure measures were reviewed and their questions were addressed, explained Dr. Heisler of the University of Michigan in Ann Arbor. After the initial meeting, patients assigned to the intervention arm participated in a group session designed to facilitate communication skills and help them set short-term goals for behavioral changes. Those assigned to usual care received nurse-led case management.
The demographics and baseline patient characteristics were similar in both groups. “The mean age of the predominantly white [82%] male veterans participating in the study was 62 years, and the majority [63%] had an annual income less than $30,000,” Dr. Heisler said. At baseline, the mean HbA1c levels for the intervention and control groups, respectively, were 8.03% and 7.93%.
Age-matched patients were paired within the same cohort to serve as peer partners, Dr. Heisler said. “Patients were encouraged to call their peer partners at least weekly to provide mutual support and encouragement,” she noted. “We developed a computer platform that enabled them to use their own phones to make calls without exchanging personal phone numbers, and it let us monitor and record the initiation, frequency, and duration of the calls. If patients hadn't made contact with each other within a week, they received reminders.”
Intervention participants also were offered three optional 1.5-hour group sessions at months 1, 3, and 6. “Although these were nurse-led programs, they were completely patient driven and served as a forum for sharing concerns, questions, and strategies and for discussing progress on their action plans,” Dr. Heisler said. In the control arm of the study, patients attended an educational session on nurse-led case management and were offered the services of a nurse case manager.
“At 6 months, the mean A1c of the intervention patients decreased from 8.02% to 7.73%, while the mean A1c of the control arm participants increased from 7.93% to 8.22%,” Dr. Heisler reported. “We were especially concerned about patients at high risk, so we did a stratified analysis, looking specifically at the change in A1c at 6 months for those patients with a baseline A1c higher than 9.0% and the differences remained significant.” Specifically, in the latter analysis, the mean HbA1c decrease for intervention arm participants with a baseline HbA1c higher than 9.0% was 0.88%, compared with a decrease of 0.07% in the control group, she said.
Regarding secondary outcomes, “we did see a 3.4% reduction in blood pressure results for the intervention group, but the differences compared with the control group were not statistically significant,” Dr. Heisler said. “Also, there were eight new insulin starts in the intervention group and only one in the control group, and the diabetes social support outcomes were significantly higher for the intervention group as well.”
An evaluation of intervention participation showed that more than 90% of the peer partners made computer-facilitated weekly calls. Also, 40% of the intervention patients attended all three of the optional group sessions, while 25% attended two sessions and 12% went to one session, she said.
“This model is far less time and resource intensive than other tested programs that have led to similar improvements in A1c,” Dr. Heisler said.