This article is more than 5 years old. Information may no longer be current.

Anti-obesity studies lack strict standards

Natural experiments in obesity prevention and control do not contain adequate methodological and analytic standards and have a high risk for bias, according to a review and position paper published in Annals of Internal Medicine.

“The rapid increase in overweight and obesity poses a significant and worsening public health threat,” Karen M. Emmons, PhD, from Harvard T.H. Chan School of Public Health, and colleagues wrote in the position paper. “About 70% of U.S. adults are overweight or obese... The complexity of obesity prevention and control requires addressing its multilevel and interrelated drivers. Evidence on effectiveness is needed to guide interventions to reduce overweight and obesity.”

Wendy L. Bennett, MD, MPH, from Johns Hopkins School of Medicine, and colleagues examined the population characteristics, data sources, measures, analytical methods and risks for bias of obesity prevention programs and policies using data from 294 eligible studies. Of the reviewed studies, 156 were natural experiments, 118 were experimental studies and 20 had an unclear design.

Data indicated that there was a lack of strict standards for implementation and measurement of diet and physical activity among the studies of anti-obesity experiments. Most studies demonstrated a high risk of bias in the study design and outcomes and poor handling of withdrawals and dropouts.

“Our findings reinforce the need for methodological and analytic advances, including reporting standards that would strengthen the evaluation of efforts to improve obesity prevention and control,” Bennett and colleagues concluded.

In the position paper, Emmons and colleagues offered several recommendations based on the findings of Bennett and colleagues to help close the identified research gaps for studies on obesity prevention programs.

Emmons and colleagues recommended that researchers maximize the use of existing surveillance and research databases and use standardized and validated measures of obesity-related outcomes and exposures. They also suggested that researchers systematically measure co-benefits and unintended consequences, as well as expand the use of validated technologies for measurement.

Guidance, documentation and communication about methods should be improved within the study design, according to Emmons and colleagues. They also suggested that bias be minimized.

Additionally, Emmons and colleagues recommended that funders enhance the rigor of natural experiments in obesity.

“The obesity epidemic is likely to worsen for decades,” Emmons and colleagues concluded. “To combat this significant public health threat, researchers should continue to take advantage of natural experiments.” – by Alaina Tedesco

Disclosure: Bennett and Emmons report no relevant financial disclosures. Please see review and position paper for all other authors’ relevant financial disclosures.