A diabetes education intervention that focused on individual telephone "coaching" failed to improve hemoglobin A_1c levels in a study of poor, uninsured, ethnic minority patients with poorly controlled diabetes that was reported online Oct. 10 in Archives of Internal Medicine.

"The intervention tested in this trial appeared to be insufficient to make a statistically significant difference compared with the control condition, despite anecdotal reports from participants that the intervention was valuable to them and provided support they had not previously received as part of their diabetes care," said Dominick L. Frosch, Ph.D., of the department of health services research, Palo Alto (Calif.) Medical Foundation Research Institute, and his associates.

The study findings were unexpected, given that "previous telephone intervention studies with different patient populations ... showed significant increases in self-care and in some cases significant improvements in glycemic control," the investigators said.

The intervention package they assessed included "a 24-minute video [for] behavior support ... [and] a workbook as well as five sessions of telephone coaching provided by a [bilingual] trained diabetes nurse." This was compared with a control condition: a 20-page brochure on diabetes self-care developed by the National Institutes of Health’s National Diabetes Education Program.

"We expected that the experimental group would be more likely to review the educational materials provided because a DVD might be perceived as more engaging than a printed brochure, thereby resulting in higher diabetes knowledge, and that the additional telephone support provided would be more likely to facilitate behavior change compared with the control condition," Dr. Frosch and his colleagues said.

The study subjects were patients aged 40 years and older who had type 2 diabetes and an HbA_1c value of 8% or higher. Most were obese. More than 70% had annual incomes of $15,000 or less. They were recruited from two internal medicine practices, one family medicine practice, and a community-based "safety net" clinic in the Los Angeles area.

"The participants in this trial had considerable deficits in their understanding of diabetes and what successful management of the condition requires."

The study subjects were randomly assigned to either the coaching intervention (100 patients) or the control condition (101 patients) and were followed up at 1 month and 6 months. Three-fourths of the intervention group completed all five possible telephone coaching sessions.

Nearly 90% of the subjects in both groups rated the clarity of the information they received as good, very good, or excellent. A total of 93% of the intervention group and 90% of the control group said they felt somewhat or very positive about other patients using the DVD or the brochure to learn about managing their diabetes.

The primary end point was change in HbA_1c value at 6 months, as compared with the baseline value. Both groups showed a significant decline, from a mean of 9.6% to a mean of 9.1%. However, the difference between the groups was not significant, the investigators said (Arch. Intern. Med. 2011 Oct. 10 [doi:10.1001/archinternmed.2011.497]).

Both groups of patients demonstrated significant increases in knowledge about their disease, as measured by the 23-item Diabetes Knowledge Test and the 25-item Summary of Diabetes Self-Care Activities, which assess diet, exercise, blood glucose testing, foot care, and smoking. However, the difference between the intervention group and the control group was not significant.

The researchers noted that the study results may have been influenced by certain unique circumstances. "Enrollment began just as it was becoming clear that the global economy was entering a severe recession," and the study subjects clearly were adversely affected by the economic downturn.

"Reports of job loss were common, leading many participants to have to struggle for basic survival. Making wise nutritious choices became impossible when the first priority was making sure that one could maintain shelter and have any food to eat at all," Dr. Frosch and his associates said.

In addition, the coaching intervention in this study differed from coaching interventions in previous studies that reported more positive results in that the "dose" of coaching was a maximum of five calls totaling 150 minutes of telephone contact. Previous studies assessed more frequent telephone contacts (up to 18 calls) as well as more total minutes of contact (up to 420 minutes).

"Our results at least suggest that five sessions lasting 150 minutes is not sufficient for the most vulnerable and disadvantaged patients with diabetes," they noted.

"The participants in this trial had considerable deficits in their understanding of diabetes and what successful management of the condition requires. The combination of this with the severe economic disadvantage and stress experienced by this population could not be overcome by the limited intervention tested in this trial," the researchers added.

This study was supported by the Robert Wood Johnson Foundation, the Foundation for Informed Medical Decision Making, the National Institute on Aging, and the National Institutes of Health. Dr. Frosch reported receiving honoraria from the Foundation for Informed Medical Decision Making.

'Coaching' Fails to Improve Diabetes in Disadvantaged Patients