- screen time
- clinical trials (epidemiology)
- data interpretation (statistical)
- (bias) epidemiology
The childhood obesity epidemic is close to 2 decades old. Millions if not billions of dollars and participant and researcher hours have been spent seeking effective preventive and treatment interventions. Why, then, are childhood obesity rates yet to fall?
We suggest that 1 reason may be an ingrained overoptimism as to how much population change can really be achieved via educational and motivational means. The evidence suggests that these broad approaches have, at most, small benefits, yet they continue to dominate research and policy. This overoptimism is potentially blocking the nimble responses required for a continuously learning health care system.1
Smith et al’s2 excellent trial in this issue of Pediatrics perhaps provides some insights. The authors tested a multicomponent, school-based intervention to prevent unhealthy weight gain in male adolescents recruited from disadvantaged areas of New South Wales, Australia. The trial aimed to increase physical activity, reduce screen time, and reduce sugar-sweetened beverage consumption via physical activity training, teacher-led activity programs (up to 33 hours in total for students), and a mobile phone app/Web site and pedometer to assist students in physical activity monitoring and goal setting.
As well as being ambitious and innovative, the trial was well designed, executed, and analyzed, which gives confidence in its findings. Unfortunately, like many childhood obesity trials with true controls,3 this was a negative study in terms of its primary outcomes (BMI, waist circumference) and most of the objective secondary outcomes. Also, like most multicomponent trials, a smattering of secondary outcomes did improve, but these were mainly in the subjective ratings that are so open to social desirability (sugar-sweetened beverages, daily screen time), and the latter was not reflected in the null findings for measured mean activity counts per day. Two of the 8 objective measures (resistance training competency, push-ups) did seem to show benefit, although intervention children already markedly outperformed the controls on the latter at baseline. Given the overall null findings, it therefore came as a surprise when the main conclusion was that “a multicomponent school-based intervention…can improve muscular fitness, movement skills, and key weight-related behaviors among low-income adolescent boys.”
Low reach must inevitably further dampen any possible population impacts. In this trial, 70% (14 of 20) of invited schools participated. Within these schools, 42% of eligible children (361 of 850) took part. Of the intervention children, 80% actually received the intervention. A back-of-the-envelope calculation suggests that, at most, the program could reach only 20% (0.7 × 0.4 × 0.8) of the eligible population. The intervention reach number is further eroded when one considers that participants only received ∼50% to 60% of the prescribed components. Given that rigorous efficacy trials probably represent the best possible uptake, the true reach when scaled to the population is likely to be even lower. It does not take a health economic analysis to see that the costs must surely outweigh any population benefits, even if the program is further optimized with the range of improvements that the authors suggest.
A third issue is that of presenting subgroup analyses that were not specified in the protocol a priori and were restricted to only those with high BMI (rather than the preferred interaction analyses). Participants selected on the basis of extreme values are as likely to regress to the mean in obesity as in any other chronic condition. When the numbers are small, as in their article, subgroup analysis is likely to throw up chance findings, and only those supporting differential resolution in favor of the trial are likely to be reported.4 Believing this is risky business.5
Smith et al are far from alone. In what they term “white hat bias,” Atkinson and Macdonald6 reported that negative results in obesity trials are usually ignored and secondarily positive analyses cited as the conclusions of the study. Does this matter, if the interventions are low risk and their components inherently healthful? We say it does. As authors ourselves of 3 negative childhood obesity trials,7–9 we have lived the pain of null findings. But every intervention carries an opportunity as well as a dollar cost. The practical opportunity cost is that schools and health departments will continue to implement ineffective interventions to the exclusion of interventions, whether for obesity or for other important conditions, that could produce more population health gain.10 The research opportunity cost is that, until we call it as it is, researchers may not make the bold paradigm shifts that the child obesity epidemic demands.
- Accepted June 27, 2014.
- Address correspondence to Melissa Wake, MD, Centre for Community Child Health, Royal Children’s Hospital, Flemington Rd, Parkville 3052, Australia. E-mail:
Dr Wake provided key ideas and critical input; Ms Lycett assisted with drafting the commentary and provided critical input; and both authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Opinions expressed in these commentaries are those of the author and not necessarily those of the American Academy of Pediatrics or its Committees.
FINANCIAL DISCLOSURE: Dr Wake is supported by Australian National Health and Medical Research Council Senior Research Fellowship 1046518. Ms Lycett was supported by a Murdoch Children’s Research Institute (MCRI) Postgraduate Health Scholarship. Research at the MCRI is supported by the Victorian Government’s Operational Infrastructure Support Program.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
COMPANION PAPER: A companion to this article can be found on page e723, online at www.pediatrics.org/cgi/doi/10.1542/peds.2014-1012.
- 1.↵Institute of Medicine. Best Care at Lower Cost: The Path to Continuously Learning Health Care in America. Washington, DC: The National Academies Press; 2013
- Smith JJ,
- Morgan PJ,
- Plotnikoff RC,
- et al
- Oude Luttikhuis H,
- Baur L,
- Jansen H,
- et al
- Sun X,
- Briel M,
- Busse JW,
- et al
- Oxman AD
- Wake M,
- Baur LA,
- Gerner B,
- et al
- Wake M,
- Lycett K,
- Clifford SA,
- et al
- Moyer VA,
- Butler M
- Copyright © 2014 by the American Academy of Pediatrics