Lifestyle Intervention Reveals a Truth: Maintenance Is the Hard Part

Published on: 

Data from CDC consistently show that low-income, minority groups are at higher risk for diabetes, for genetic but especially for environmental reasons.

A study that examined the effectiveness of an intense lifestyle intervention among a low-income, obese and mostly female group of African American persons with type 2 diabetes (T2D) found what many family doctors already know: getting on track with healthy living is the easy part. Staying there is something else.

The intervention involving 211 participants from outpatient clinics in the Cook County Hospital System, Chicago, was presented Sunday afternoon at the 76th Scientific Sessions of the American Diabetes Association, being held in New Orleans, Louisiana.

Lifestyle Intervention through Food and Exercise, or LIFE, compared 106 participants who were given 28 group sessions with dietitians and peer supporters over a year, compared with 105 in a control group given standard diabetes education. The study participants had an average age of 55 years, and a mean body mass index (BMI) of 35.68 kg/m2, which the CDC defines as Class 2 obesity. They had lived with diabetes an average of 11.39 years, and they earned no more than $20,000 a year.

Data from CDC consistently show that low-income, minority populations have higher rates of diabetes, attributable in part to genetics but also to a host of environmental factors, including lack of access to healthy food and adequate medical care. Duke University researchers Mark Feinglos, MD, and Richard Surwit, PhD, have explored the intertwined relationships among stress, diabetes, and the brain, and its relationship with the increased central adiposity among many African American women. (For an interview in Evidence-Based Diabetes Management, click here.)


While the counseling and dietary sessions in the LIFE study were designed to be culturally sensitive, they were also aimed at training the participants to eventually manage their own diet and exercise with less support. Based on the results, that did not work so well.

Counseling sessions took place weekly for the first 4 months and biweekly for the second 4 months, then monthly for the last 4 months of the study. Activities included group exercise sessions and discussions, and problem-solving.

At the six-month mark, the intervention group was doing well: this arm had average glycated hemoglobin (A1C) reductions of 0.76%, compared with 0.21 for the control arm. This arm also had a higher share that had at least 0.5% decline in blood glucose levels (63% vs 42% in the control arm).

By the one-year mark, however, that gap had disappeared. Patients in the control arm also saw a drop in A1C (-0.65% for LIFE vs -0.45 for control), and the share with a 0.5% blood glucose reduction was nearly equal, 53% for LIFE and 51% in the control arm.

The study’s lead author, Elizabeth B. Lynch, associate professor of Preventive Medicine at Rush University in Chicago, said maintaining healthy diet and exercise habits may be especially difficult for low-income persons with diabetes. She speculated that while there was backsliding in the intervention group after the number of counseling sessions declined, the more motivated participants in the control group showed improved medication adherence.

“While our results confirm positive results in a high-risk population with intensive educational and social support, they also suggest that, over time, motivated patients take the initiative and achieve glycemic control with standard of care education.”

But a presenter from Friday’s program at the Scientific Sessions, University of Michigan’s Robert Anderson, EdD, said that many who take part in diabetes education hit a plateau after approximately 2 months and then need a different type of support, so that gains are not reversed.


Lynch EB, Mack L, Avery E, et al. Lifestyle improvement through food and exercise (LIFE): randomized trial of a self-management intervention for underserved African Americans with type 2 diabetes. Diabetes. 2016; 65(suppl1): 282-OR.