Skip Navigation
Service Delivery Innovation Profile

Telephone-Based Mentoring From Demographically Similar Peers Improves Diabetes Control in African-American Veterans

Tab for The Profile Tab for Expert Comments



At the Philadelphia Veterans Affairs Medical Center, African-American veterans with diabetes who had their blood glucose under control served as mentors to diabetes patients with a similar racial background who did not. After completing a 1-hour training session, mentors telephoned (and in some cases met with) their assigned patient on a regular basis over a 6-month period to help them address diabetes-related challenges, such as diet, exercise, and insulin use. The program significantly reduced blood glucose levels and generated a positive reaction from many participants.

Evidence Rating (What is this?)

Strong: The evidence consists of a randomized trial comparing trends in blood glucose levels among diabetes patients receiving peer mentoring to a control group of similar patients who did not. Additional evidence includes post-implementation interviews with participants about their perceptions of the program.
begin do

Developing Organizations

Philadelphia VA Medical Center
end do

Use By Other Organizations

Program leaders have received multiple inquiries from staff at other hospitals and clinics about setting up a similar peer mentoring program.

Date First Implemented


Problem Addressed

Compared to whites, African Americans disproportionately suffer from type 2 diabetes, struggle more with glucose control, and experience higher rates of diabetes-related complications. Peer support can be effective in helping patients keep diabetes under control, but the resource-intensive nature of such programs often limits their use.
  • Higher incidence among African Americans: About a quarter of African Americans between the ages of 65 and 74 suffer from diabetes, and overall, African Americans are 1.8 times more likely than non-Hispanic whites to have the disease.1 At the Philadelphia Veterans Affairs (VA) Medical Center, 45 percent of patients with diabetes are African American.
  • Poorer control, more complications: Compared with whites, African Americans with diabetes tend to have worse glucose control, which leads to more disease-related complications, including eye disease, amputations, and kidney disease. For example, African Americans with diabetes are roughly 1.5 times more likely to develop retinopathy and 2.7 times more likely to have a lower limb amputated than are whites with the disease.1
  • Unrealized potential of peer mentoring: Studies suggest that peers who are successfully managing diabetes can be effective in helping those with poor control improve self-management of the disease. However, these programs are not common due to their resource intensity, since mentors may require significant upfront training, substantial ongoing incentives, and/or periodic assistance from professional support staff, particularly when programs involve inperson group meetings. For example, most programs require mentors to attend multiple training sessions and in some cases college courses to prepare them for patient interactions.2

What They Did

Back to Top

Description of the Innovative Activity

At the Philadelphia VA Medical Center, African-American veterans with diabetes who had their blood glucose under control served as mentors to diabetes patients with a similar background who did not. After completing a brief training session, mentors telephoned (and in some cases met with) their assigned patient regularly over a 6-month period to help them address diabetes-related challenges, such as diet, exercise, and insulin use. A detailed description follows:
  • Marketing and enrollment: The program served African Americans with diabetes between the ages of 50 and 70 who used the medical center's primary care clinic and whose last two hemoglobin tests were above 8 percent, indicating a persistent problem in controlling glucose levels and a high risk of diabetes-related complications. Eligible patients received a letter and followup phone call. Those interested could sign up at the hospital.
  • Initial measurement and goal-setting: Those who signed up had their glucose level tested. Program staff called them a day later to provide the results and to suggest a 6-month goal for bringing glucose levels under control. (In the trial of this program, a control group also received this service as part of "usual care.")
  • Match to demographically similar peer: Typically within 2 weeks of enrolling, participants were assigned a mentor with the same general demographic background—including being within 10 years of the participant's age along with being the same race and gender. Each mentor had already been successful in getting their glucose levels under control (defined as falling from above 8 percent sometime in the past 3 years to below 7.5 percent in the past 3 months). Like the participants, mentors signed up for the program after learning about it in a letter and/or through a followup phone call. While the program was open to both males and females, all the mentors and their assigned patients were men. Matching peers by sex, race, and gender made the program innately culturally competent in that mentors and patients came from the same cultural background. This made it likely that the patients would quickly build rapport with their mentors and have someone to talk with about commonly experienced problems.
  • Brief mentor training: A research assistant provided an hour of one-on-one training to each mentor, focusing on use of motivational interviewing techniques. Techniques included asking about the participant's background and then using this information to understand his motivations; helping him identify the differences between behaviors and goals; and guiding him toward a realistic plan for achieving these goals. The trainer also provided tips on making followup calls and assessing whether the patient was heeding the advice. The trainer provided sample questions that could be asked of patients, but encouraged mentors to draw on their own experiences as well. Mentors received $25 for completing the training session.
  • Nominal incentives for weekly contact: Mentors were encouraged to call their patients at least once a week, and/or to meet face-to-face if desired, and they could have more frequent contacts if both parties wanted to. To encourage regular contact, at the end of each month mentors whose patients confirmed that they had spoken at least weekly during that month received $20. On average, mentors and participants talked the most during the first month (an average of four calls) and then tended to communicate somewhat less often over time, with an average of two calls by the sixth month. Some pairs met regularly in person at the VA Medical Center (supplemented by phone calls), while most used only phone conversations.
  • Practical advice during sessions: While conversations varied, mentors often helped participants overcome challenges related to insulin injections and gave them specific suggestions on how to maintain a healthy diet and/or find time to exercise. Pairs who had frequent contact tended to develop a strong rapport, in some cases leading to participants revealing things to their mentors that they did not feel comfortable sharing with their doctor. Participants' shared cultural background often proved to be helpful. For example, one participant told his mentor that he had a hard time resisting home-made foods at family gatherings, such as biscuits and gravy and sweet potato pie. Having been through the same experience, the mentor offered practical, real-world strategies for resisting tempting foods.

Context of the Innovation

The Philadelphia VA Medical Center provides health care to veterans living in America's fifth largest metropolitan area, including the city of Philadelphia and six surrounding counties in southeastern Pennsylvania and southern New Jersey. More than 90,000 veterans receive care at the medical center, with nearly 60,000 visiting the center at least once a year. As noted, roughly 45 percent of the medical center's patients with diabetes are African American.

The peer mentoring program came out of ongoing research by a staff physician into strategies for improving glucose control and reducing disease-related complications in African Americans with diabetes. The researcher wanted to test whether the community health worker model (in which lay people help patients with similar cultural backgrounds) could be effective with this population, particularly if the mentors also had diabetes and had been successful in getting the condition under control. The researcher also wanted to test whether a largely telephone-based initiative could work, thus making it less resource-intensive and easier to replicate.

Did It Work?

Back to Top


The program significantly reduced blood glucose levels (by an average of 1.1 percent) and generated a positive reaction from many participants.
  • Significantly better glucose control: The average blood glucose level in a group of 38 patients matched with a peer mentor fell from 9.8 percent at baseline to to 8.7 percent at the end of the program, a drop of 1.1 percentage points. By contrast, average blood glucose levels changed little (from 9.9 to 9.8 percent) in a control group of 39 nonparticipants who were only told their baseline level and given a 6-month target for dropping it. The 1.1-point decline is considered a major improvement and significantly lowers an individual's chances of diabetes-related complications. (The study also included a third group of individuals who were told their baseline level and offered monetary incentives for reducing it to target levels. More information on this financial incentive program can be found in a related profile: Financial Incentives Do Not Improve Glucose Control in African-American Veterans With Diabetes.)
  • Positive reaction from participants: Many participants reported positive feelings about the program. Half of the 28 participants who completed the exit interview found the mentoring to be educational, and nearly two-thirds (20) felt it was important that the mentor also had diabetes.

Evidence Rating (What is this?)

Strong: The evidence consists of a randomized trial comparing trends in blood glucose levels among diabetes patients receiving peer mentoring to a control group of similar patients who did not. Additional evidence includes post-implementation interviews with participants about their perceptions of the program.

How They Did It

Back to Top

Planning and Development Process

Key steps included the following:
  • Developing program: To inform program development, the physician and her research team reviewed recent studies on community health workers and diabetes interventions. They also held focus groups with diabetes patients at the VA who had been successful in bringing their glucose level under control. These conversations helped provide a sense of what distinguishes patients who manage diabetes effectively from those unable to do so.
  • Securing funding and hiring staff: In September 2009, Dr. Long secured an $80,000 grant that allowed her to hire a research assistant to manage the program on a day to day basis.
  • Training the trainer: A member of the research team with expertise in training spent several hours teaching the research assistant how to train the mentors. This individual sat in on the initial training sessions with mentors to provide additional guidance.
  • Program trial: The researchers matched patients with trained mentors on an ongoing basis from October 2009 to April 2010, with followup being completed by October 2010. The original version of the program ended at this time, as the researchers' intent was to test its effectiveness in a time-limited trial. The researchers are currently running a separate trial of a modified version of the program; see the Adoption Considerations section for more details.

Resources Used and Skills Needed

  • Staffing: A full-time research assistant handled most of the day-to-day work involved in the program, such as recruiting mentors and patients, training mentors, and monitoring the frequency of participant-mentor contacts. During the trial, the physician leading the study spent 2 to 5 hours a week on program implementation, supervision, and evaluation. Three other researchers, including a statistician, worked part-time on program oversight and evaluation. During the trial, 38 mentors participated.
  • Costs: Program costs totaled $80,000. Major expenses included staff salaries, blood tests for patients, incentives distributed to mentors, and incentives for patients that were part of the related trial alluded to earlier.
begin fsxml

Funding Sources

National Institute on Aging
A grant from the National Institute on Aging Roybal Center covered the costs of the pilot study.end fs

Adoption Considerations

Back to Top

Getting Started with This Innovation

  • Begin mentorship quickly after enrollment: To maintain patients' enthusiasm, match them with a mentor as soon as possible after enrollment (ideally within 1 or 2 weeks). If it proves difficult to find an age-appropriate mentor, consider relaxing the age requirement (e.g., to within 15 years of the participant's age rather than 10).
  • Carefully consider merits of face-to-face meetings: Inperson sessions have both benefits and drawbacks. Face-to-face sessions may help the pairs develop a rapport more quickly and foster a stronger relationship over time. However, such meetings require more planning (e.g., to come up with a meeting place) and time (e.g., to drive to meetings) and some people may be less likely to participate if they have to leave home to do so, especially if they live far from the proposed meeting site. In making this decision, program leaders should consider demographic and geographic issues likely to influence participant preferences.
  • Consider insulin use when matching: After the program began, researchers realized that in addition to cultural background, gender, and age, it is helpful to assign mentors who have experience with insulin injections to patients who use the drug. Injecting oneself is a unique (and often challenging) experience, and injection-related issues frequently contribute to patients having trouble managing glucose levels. A mentor who has familiarity with insulin use can likely be helpful in addressing any challenges these patients may be having. In fact, in a subsequent study, members of the research team will be matching patients who use insulin with mentors who have experience with self-injections.

Sustaining This Innovation

  • Make patients into mentors: While the Philadelphia VA program was a time-limited trial intended to last only 6 months, those interested in a permanent program should consider recruiting participants who have been successful in bringing their blood glucose under control to serve as mentors to others who need assistance in doing so.
  • Consider duration of disease when matching: Once the program has grown to include many mentors, it may become possible to match individuals who have had the disease for a similar amount of time. This approach could help to strengthen rapport between mentor and patient, as the issues facing those who have had diabetes for a few years will be different than those who have struggled with it for a decade or more.
  • Sell potential benefits to mentors: Although it was not formally evaluated for this program, other studies have found that peer support programs can benefit mentors' health by motivating them to maintain disease control as an example to those they are mentoring. Clinicians can use this potential side benefit to help make the case to leaders for maintaining peer mentoring programs as a way to benefit all patients with diabetes.

Use By Other Organizations

Program leaders have received multiple inquiries from staff at other hospitals and clinics about setting up a similar peer mentoring program.

More Information

Back to Top

Contact the Innovator

Judith Long, MD
Penn Medicine
Division of General Internal Medicine
1201 Blockley Hall
423 Guardian Drive
Philadelphia, PA 19104-6021
(215) 898-4311

Innovator Disclosures

Dr. Long reported having no financial interests or business/professional affiliations relevant to the work described in this profile.

References/Related Articles

English T. Veterans with diabetes "battle" to stay healthy in Philadelphia. website. April 16, 2012. Available at:

Long JA, Jahnle EC, Richardson DM, et al. Peer mentoring and financial incentives to improve glucose control in African American veterans: a randomized trial. Ann Intern Med. 2012;156:416-24. [PubMed]

Long, JA. "Buddy system" of peer mentors may help control diabetes. Leonard Davis Institute of Health Economics: Issue Brief. 2012 Mar;17(6):1-4. Available at:
(If you don't have the software to open this PDF, download free Adobe Acrobat ReaderĀ® software External Web Site Policy.).

Shacter HE, Shea JA, Akhabue E, et al. A qualitative evaluation of racial disparities in glucose control. Ethn Dis. 2009;19:121-7. [PubMed]


1 American Diabetes Association Web site. African Americans and complications. Available at:
2 Heisler M. Different models to mobilize peer support to improve diabetes self-management and clinical outcomes: evidence, logistics, evaluation considerations and needs for future research. Fam Pract. 2010;27 Suppl 1:i23-32. [PubMed]
Comment on this Innovation

Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.

Original publication: July 18, 2012.
Original publication indicates the date the profile was first posted to the Innovations Exchange.

Last updated: July 30, 2014.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.

Date verified by innovator: July 16, 2014.
Date verified by innovator indicates the most recent date the innovator provided feedback during the annual review process. The innovator is invited to review, update, and verify the profile annually.

Look for Similar Items by Subject