Prevention of obesity: a review of interventions

In 2004, the World Cancer Research Fund (WCRF) commissioned the Institute for health sciences and social care research at the University of Teeside, as one of its Collaborating Centres, to carry out a systematic review on the dietary and physical activity determinants of overweight and obesity. Observational studies with a follow-up period of more than 1 year, with outcomes in children aged above 5 years, were included. In addition, a systematic review of the mechanistic data (usually from laboratory experiments) that could potentially underpin the findings from the observational studies was conducted. A first draft of this work was presented to the WCRF in November 2005 and a final report produced in June 2006. Also in 2004, the National Institute for Clinical Excellence commissioned the Institute for health sciences and social care research at the University of Teeside, as one of its Collaborating Centres, to carry out a series of rapid critical reviews on the prevention of overweight and obesity. This work formed part of the supporting evidence for their obesity guidance (1). We conducted five reviews: interventions for the prevention of overweight and obesity in school children; interventions for the prevention of overweight and obesity in children aged 2–5 years and family-based interventions; interventions for the prevention of overweight and obesity in vulnerable groups; strategies around raising awareness of obesity; determinants of overweight and obesity. Inclusion criteria for each review varied, where the ‘best’ available evidence was used in preference. For reviews of interventions, ‘best’ means taken from randomized controlled trials. For the review of determinants, ‘best’ available evidence means taken from observational studies with a follow-up of more than a year. This review draws specifically on the evidence reviewed within these two pieces of work, and also our Cochrane review, Interventions for the Prevention of Obesity in Childhood, which only included controlled trials. There is a strong argument that, in evaluations of public health interventions, randomized controlled trials are never sufficient by themselves (2). Although randomized controlled trials may tell us about the efficacy of interventions, they do not necessarily inform us about effectiveness. Randomization, without further analyses for adequacy and plausibility, is never sufficient to support public health decision-making, regardless of the level of statistical significance observed. The Cochrane Health Promotion and Public Health Field is currently working on methodologies that address this very important issue. Table 1 shows those components of diet and physical activity that were deemed to be important determinants of overweight and obesity, as assessed from the observational and mechanistic evidence reviewed. Observational data from adults and children that have looked at the importance of total physical activity show a generally consistent finding that total physical activity decreases the risk of overweight and obesity. There is also evidence of a dose–response effect. The mechanistic data are strong and support the observational data. Of note, there is a trend for more recent studies (published after 2000) to more often report an inverse association between total physical activity and weight gain. The reasons for this are not clear, but could relate to generally larger sample sizes with great statistical power, better and more complete adjustment for confounding, the much greater emphasis on low physical activity as a cause of weight gain, or more precise measures of diet and physical activity. Observational data show a consistent relationship between breastfeeding and a decreased risk of overweight and obesity beyond 4 years of age, and there is some evidence of a biological mechanism (see contribution from Singhal). Observational data from adults and children that have looked at the importance of low energy-dense foods (e.g. wholegrain cereals and cereal products; foods rich in non-starch polysaccharides and dietary fibre) show conflicting results, but the mechanistic data are compelling. Observational data from adults and children that have looked at the importance of energy-dense foods show conflicting results, but the mechanistic data are strong. Many of the highly consumed processed and fast foods have a high energy density because of a high content of fat and refined carbohydrates. Several human clinical studies have shown that high energy-dense diets can undermine normal appetite regulation, a process that has been termed ‘passive overconsumption’. This inability to recognize the energy density of food and compensate appropriately to maintain energy balance can lead to unhealthy weight gain, and overweight and obesity. There is strong evidence from observational studies, and mechanistic evidence, that shows that an increased consumption of sugary drinks increases the risk of overweight and obesity. Evidence relating to the usefulness of interventions for the prevention of overweight and obesity in: school children; children aged 2–5 years and family-based interventions; vulnerable groups. is summarized as evidence statement in Tables 2–4. Evidence statements relating to weight outcomes, and dietary and physical activity outcomes, are presented separately. In collating the evidence base, we were mindful to include measures of environmental context (where these were presented in the primary papers), including both quantitative and qualitative information. In summarizing the information, some of this detail is lost, but where environmental context appeared to have a significant impact on the impact of interventions, this detail is included. No conflict of interest was declared.