The Diabetes Prevention Program (DPP) published in 2002 remains a landmark paper. It showed that diet and lifestyle interventions alone can induce sufficient weight loss to prevent diabetes in people with impaired fasting glucose or impaired glucose tolerance – a 58% reduction in cumulative diabetes incidence at 4 years – after an initial weight loss of around 7 kg after the 6-month core treatment, dwindling to less than 4 kg at 4 years. The remarkable conclusion was that lifestyle intervention knocks metformin into a cocked hat in diabetes prevention, showing its importance in any glucose-lowering strategy.
The DPP team have produced numerous papers, but the prospect of a new publication has the feel of one’s favourite old rock band reuniting one more time for another comeback album. One such comeback was a paper last year that looked at the predictors of modest weight loss – which individual should we best dedicate our time and resources to for the best bang for our buck? The answer: someone of White origin, older age, older age when first over-weight and fewer past self-implemented weight loss attempts. These were the independent, non-modifiable pre-treatment predictors for achieving 7% weight loss (Delhanty et al, 2013).
So, to complete the circle, the next question was, which weight loss characteristics are the best predictors for the prevention of diabetes? The resulting paper, summarised alongside, was limited to analysing data from those individuals receiving lifestyle intervention for at least 2 years, in order that subsequent weight fluctuations after the 6-month “core curriculum” could be monitored. Greater weight loss in the first 6 months and from 6 months to 2 years were both predictive of reduced diabetes incidence; although overall weight loss from baseline to 2 years was the strongest single predictor of lower diabetes incidence. Absolutely no surprise there – weight loss reduces risk of diabetes and prolonged weight maintenance is more effective than brief weight loss.
However, the crunch is that weight cycling was predictive of an increased incidence of diabetes. Conventional wisdom has always decreed that “yo-yo” dieting is probably a mistake, but the proof has been difficult to provide – studies have been unconvincing and methodologically flawed (Atkinson et al, 1994). But now it’s official; yo-yo dieting is dangerous. A random search of the internet revealed that 54% of us are currently dieting, and on average we try four different diet plans per year. This paper confirms what bad news this is; cumulative weight gain, regardless of the weight loss episodes in-between, is harmful and increases diabetes risk. Diabetes prevention programmes, however, seem to be best targeted on white, more elderly, weight loss novices who should be persuaded to lose weight gradually and maintain it. Woe betide the rest!
To view the summaries of each paper, please download the PDF of this article.
