Collins CE, Warren J, Neve M, McCoy P, Stokes BJ. "Measuring Effectiveness of Dietetic Interventions in Child Obesity." Arch Pediatr Adolesc Med, Sep 2006. 160:906-922.
Collins and colleagues published a study in the September 2006 issue of the Archives of Pediatric and Adolescent Medicine on the lack of good studies on the effectiveness of dietary interventions in childhood obesity. The authors had originally intended to determine which dietary interventions were most efficacious by doing a meta-analysis on all previously published randomized controlled trials on obese pediatric (<18 yrs) patients that included a dietary intervention arm either alone or in combination with lifestyle modifications and/or psychological therapies. However, the 37 trials they found that met these criteria were so poorly described, inadequately designed, and hetereogenous in nature that the authors were only able to make two very qualified meta-analyses: that pediatric obesity studies including dietary interventions do seem to achieve relative weight loss but that the effect seems to diminish over time.
The authors cited a number of problems with the existing studies on dietary intervention. For one thing, very few studies (n=7) had dietary intervention as the sole intervention; those that had multiple interventions were often too diverse to draw overall conclusions. Many studies did not adequately describe the details of the dietary intervention taken. In addition, the outcome variables measured varied widely from trial to trial, also rendering meta-analyses difficult. Changes in food intake were often not reported and it was unclear whether this was because the data were not measured or if they were not used because of poor quality. Lengths of studies and follow-up periods tended to be very short or even nonexistent (nine studies had no follow-up period). Not only were the trials usually underpowered, but many had overly broad subgroup age ranges that included children at very different developmental stages (i.e. 3-12 years and 12-18 years).
Clearly, this is a deplorable state of affairs. On the one hand, we're surprised that more work hasn't been done in this field considering the burgeoning problem of childhood obesity; on the other hand, it makes sense that studies on dietary intervention would be underfunded and undersized -- this is not a very lucrative area and there is no financial incentive for doing these studies. We applaud the authors for pointing out the flaws in existing research; hopefully as obesity becomes more of a public health issue and more funding becomes available for research in combating obesity, future investigators will take their criticisms into account and avoid the pitfalls in study design that they have pointed out.