Scant Data on Dietary Advice in Diabetes, but Exercise Helps

MIDDLESBROUGH, England, July 18 -- High-quality data are lacking to assess dietary advice for patients newly diagnosed with diabetes, but evidence does show that exercise improves glycemic control.
Although initial dietary management is a cornerstone of type 2 diabetes treatment, a formal systematic review of its efficacy and method of delivery was not possible, said Lucie Nield, a researcher at the University of Teesside here, in an online July 18 Cochrane review.
The findings from a search of the Cochrane Library, Medline, Embase, Cinahl, AMED, bibliographies, and relevant experts, produced 36 articles that reported work from 18 different trials, including 1,467 patients with type 2 diabetes.
Only a minority of the trials examined hard clinical endpoints, such as death, development of macrovascular or microvascular diabetic complications, or quality of life, the Cochrane reviewers said. Those that did, they noted, offered no details, and most of the articles centered on reporting and discussing the participants' weight and blood-glucose control, factors easier to measure.
Meta-analyses could not be carried out for various forms of dietary advice, such as suggestions for very-low-calorie diets versus merely low-calorie diets, because data were insufficient, the researchers reported.
Nor was there sufficient data in trials to assess dietary advice plus behavioral changes, they added.
On the other hand, studies that included exercise did appear to yield useful guidance. In seven studies comparing dietary advice with dietary advice plus exercise, the exercise patients at six months achieved a statistically significant decrease (pooled weighted mean difference) in glycated hemoglobin of 0.9% (confidence intervals of 0.4 to 1.3).
At 12 months, the exercise patients recorded a statistically significant mean decrease in glycated hemoglobin of 1.0% (CI 0.4 to 1.5), the researchers reported.
The studies that examined dietary advice plus physical activity do suggest a benefit from increasing physical activity levels alongside a reduced energy diet, the researchers said.
Exercise programs, for example, included 20 minutes of low- impact aerobic exercise, joining an aerobic exercise group or a structured exercise program, or walking three miles a day three times a week plus other additional exercise.
There is a need for far more research into effects of dietary change (with and without the addition of physical activity) on micro- and macrovascular diabetic complications, weight, and glycemic control, the researchers said. Many of the outcomes they initially wanted to study in this review were not investigated in the included studies.
Work carried out in the future should take care to record and publish mortality, change (or delay) in initiation of anti-diabetic medication, and also quality of life, as these are outcomes of importance to people with type 2 diabetes. It would also be desirable if measures of compliance were reported in published works, they advised.
Even the exercise data were flawed with a risk of bias, the researchers said. There is a need for well-designed studies that examine a range of interventions, they added, noting that a promising study is currently underway.
That trial, a long-term study, known as "look AHEAD" (Action for Health in Diabetes), sponsored by the National Institutes of Health, began in 2002 and aims to examine the long-term health effects of weight loss in men and women, ages 45 to 75, who are overweight and have type 2 diabetes.
It is planned that participants will be assessed up to 11 and a half years after enrolling in the program, which should reveal the long-term health effects of the trial, the researchers said.
The reviewers reported no financial conflicts. The review was funded by a dissertation grant from the National Institute on Aging of the NIH. Primary source: Cochrane LibrarySource reference: Nield L, et al "Dietary advice for treatment of type 2 diabetes mellitus in adults (review)" The Cochrane Library 2007; Issue 3.Art.No.:CD004087: doi:10.1002/14651858.CD004097.pub4.