Telephone Targeting of Urban Poor With Diabetes Drops HbA1c

June 22, 2013

CHICAGO — Residents of the Bronx with type 2 diabetes who received telephone advice regarding medication adherence, healthy eating, and physical activity from health educators from their own community, in their own language, saw reductions in HbA1c comparable to that produced by some pharmaceuticals, a new study in almost 1000 individuals shows.

The findings, which showed the intervention was most effective in those with baseline HbA1c of greater than 9%, were presented yesterday by Elizabeth A. Walker PhD, RN, CDE, from Albert Einstein College of Medicine, New York, here at the American Diabetes Association 2013 Scientific Sessions.

The key, Dr. Walker told Medscape Medical News, is "that it's behavioral counseling, by telephone, in a voice that sounded familiar, like their neighbor's. It's the problem solving, it's the goal setting — helping people learn behavioral techniques — that's what was delivered to them, and the group that got that lowered their HbA1c, adjusted, by 0.6%."

Physicians should be interested in this finding, Dr. Walker stressed, because "this is something that could be done out of their office or out of a clinic. It's a low-cost intervention that can be done by telephone."

Asked to compare these results with others, she said: "At Einstein, we've had a string of studies over 15 years, and the HbA1c drop in this study was better than any of the other telephonic studies that we have been doing." And with regard to the literature, Dr. Walker noted that this is one of the largest such studies ever performed and the outcome is "comparable" to other similar published interventions.

The critical thing here is that our participants were very poor…representing the hardest-to-reach people. That's why it's so significant. Dr. Elizabeth A. Walker

"But the critical thing here is that our participants were very poor, 70% were foreign born, and 56% were Spanish speaking. The population that we have access to in the Bronx represents the hardest-to-reach people. That's why it's so significant."

Bronx HbA1c Study Participants From Urban Registry

Dr. Walker explained that almost 1 in 10 people (9.5%) in New York City have diabetes, and at the time of this study, the South Bronx had one of the highest prevalence rates in the city: 14.2%.

Participants were recruited from the New York City Health Department's HbA1c registry, an electronic reporting system that logs all HbA1c test results for New York City residents. Reporting of a patient's details, plus their provider, testing facilities, and results, for HbA1c have been mandatory requirements since 2006.

Elizabeth A. Walker

The registry tracks both type 1 and type 2 diabetes, but the participants in this intervention would have been "almost exclusively" type 2 diabetes patients, Dr. Walker said.

The specific aims of the Bronx HbA1c study were: to evaluate the additional effect of a telephonic behavioral intervention on mean HbA1c over and above that achieved by the HbA1c registry plus a print intervention alone; to determine what demographic, psychological, or behavioral factors mediated the effect of the intervention; and to provide estimates of implementation and maintenance costs.

Dr. Walker, a behavioral scientist and nurse, explained to Medscape Medical News that the health educators they trained to make the telephone calls "do not have to be healthcare professionals. We trained people who had bachelor's degrees over a 2-week period." The key, however, was "that they were from the community: Latinos and African Americans. And I only hire people who want to be helpful. That's the sort of person who was on the telephone.

"We had case-management meetings on a weekly basis and the health educators would say, 'This is what so and so said on the telephone,' and I would say, 'Well, why don't you try this next time you talk to them?' So it's not my time as a licensed healthcare professional on the telephone. It's a health educator and so the cost is kept low," she noted.

Although she didn't formally present the economic findings — they are still being analyzed — she said the costs are "modest."

Is HbA1c Harder to Move When It's Closer to Goal?

For the study, 941 adult diabetic patients from the HbA1c registry who lived in the South Bronx and who had a recent HbA1c level of greater than 7% were recruited. Of the recruits, 68% were Latino and 28% were black. Both the active telephone group (n = 443) and the control group (n = 498) were sent 4 self-management mailings a year, with all study interventions in English or Spanish, as preferred.

In addition, the telephone-intervention group received up to 8 telephone calls in 1 year if their HbA1c was greater than 9% or up to 4 calls a year if their HbA1c was greater than 7% to 9%. The mean baseline HbA1c was 9.2%.

During the calls, the participants were given problem-solving skills, goal setting, and advice regarding medication adherence, healthy eating, and physical activity. The primary outcome of the study was mean HbA1c at 1 year.

Complete outcome data were available on almost 75% of the sample. After adjustment for predictors, including age, body mass index (BMI), and sex, those in the active telephone-intervention group had a 0.6% lower HbA1c than controls (P = .006).

Further investigation revealed that the baseline HbA1c was extremely important when it came to predicting the intervention effect, Dr. Walker said. Those in the greater-than-9% tier in the intervention group had an average HbA1c of 11.3% at baseline, and this was reduced to a mean of 10.8% afterward (compared with 11% and 10.9%, respectively, in the control group with HbA1c greater than 9% at baseline).

Those in the greater than 7% to 9% HbA1c tier saw little change in their mean HbA1c in the intervention group, and HbA1c went up slightly in the control group for that tier (from 7.8% prestudy to 8% afterward).

"It's possible that the greater than 7% to 9% tier needed a greater number of phone calls to improve HbA1c, or it may be that it is simply harder to move HbA1c when it is closer to goal," Dr. Walker suggested.

Further analysis also revealed that the intervention effect was not mediated by medication adherence — the main aim of the study had been to improve this — nor was it seemingly affected by diabetes distress, depression, self-care behavior, or well-being, Dr. Walker said.

"None of the parameters that we measured, including medication adherence, changed. But do I actually believe that they probably became more adherent? I do believe that. We could see that everybody had changes in their medications, self-reported to us, but it's just that it happened in both groups," she commented.

Neither Dr. Walker nor her colleagues reported any relevant financial relationships.

American Diabetes Association 2013 Scientific Sessions. Abstract 2-OR, presented June 21, 2013.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: