David C. Goff Jr, MD, PhD, dean of the Colorado School of Public Health, Aurora, principal investigator for both studies stated that, "taken together, the new results suggest that "lifestyle interventions are very effective and can be delivered in an affordable manner in the community." "Many physicians believe there is little they can do to help their patients modify their health behaviors. However, well-run community-based programs can be effective and provide important health benefits," added Dr. Goff.
HELP PD is a randomized trial that attempted to translate the findings of the landmark Diabetes Prevention Program (DPP) to a less costly model, using an existing community health system. It was administered through a diabetes care center by registered dieticians, with group sessions led by community health workers who themselves had well-controlled type 2 diabetes (HbA1c < 7%) and healthy lifestyles.
The intervention was aimed at reducing caloric intake and increasing moderate-intensity aerobic activity to produce a 0.3-kg/week weight loss over the first 6 months for a total 5% to 7% reduction in body weight, after which the focus was on maintenance of weight loss. Group sessions were conducted weekly for the first 6 months and less frequently thereafter. Subjects also met with the dietician for 3 consultations.
In a comparison intervention used as a control, dubbed "enhanced usual care," individuals with prediabetes had 2 individual sessions with the dietician during the first 3 months and received a monthly newsletter about healthy lifestyle and community resources.
Of 301 overweight or obese patients with prediabetes (mean age, 58 years) randomly assigned to the lifestyle intervention or enhanced usual care, 261 completed 2 years of the study. No other study of DPP-based lifestyle interventions has extended beyond 1 year, the authors note.
In the outcomes analysis, published with lead author Jeffrey A. Katula, PhD, from Wake Forest University, Winston-Salem, North Carolina, at 2 years, the lifestyle-intervention group experienced significantly greater reductions compared with controls in weight (-4.19 kg), BMI (-1.4), waist circumference (-3.23 cm), fasting glucose (-4.35 mg/dL), and weight-loss percentage (-4.59%) (P < .001 for all). They also had greater decreases in insulin resistance (P = .006).
Of those in the lifestyle-intervention group, 46.5% maintained body weight 5% or more below baseline at 2 years compared with just 15% of the control group (P < .001). And 21.3% of the intervention group were able to keep body weight 10% or more below baseline, vs 5.3% of controls (P < .001). Both blood glucose and body weight were higher at 2 years than they had been at 1 year, but the differences between the 2 groups remained significant, the authors note.
At 2 years, 4 subjects in the lifestyle-intervention group and 11 in the usual-care group had developed type 2 diabetes, a nonsignificant trend. However, the authors point out, the -4.35-mg/dL difference in glucose reduction between the 2 groups is roughly equal to that seen in the DPP, which resulted in a 58% reduced incidence of type 2 diabetes. "Although the present study was not designed to detect differences in diabetes incidence, the glucose effect reported here appears to be large enough to result in significant reductions in diabetes incidence," they write.
In the second analysis, with lead author Michael S. Lawlor, PhD, from Wake Forest University, the direct medical and nonmedical costs incurred by the 2 HELP PD groups were assessed and compared with those of the DPP.
Over the 2 years, the total per capita direct medical costs of the interventions were $850 for the lifestyle-intervention group, compared with $142 for the usual-care group (in 2010 dollars). This $850 is significantly less than the $2631 spent during the first 2 years of the DPP, the authors note. The mean direct cost of identifying 1 person with prediabetes (301 of 746 screened) was $16.85. Direct medical costs outside of the study — including physician visits, hospitalizations, and prescriptions — were assessed via participant questionnaires. Here, expenses were significantly lower for the lifestyle-intervention group, $5177, vs $7454 for the usual-care patients. The greatest difference was for hospital days, $4778 for the lifestyle-intervention group vs $6994 for usual care.
Direct nonmedical costs, including the time and money spent on food and exercise, totaled $955 more per person for the lifestyle-intervention group compared with usual care ($13,836 vs $12,881).
"We believe the HELP PD model can help reach people who prefer a group-based approach that is more community or neighborhood oriented, rather than YMCA-based. The combination of these 2 models will help reach a larger proportion of the population that needs this intervention," Dr. Goff said.