In a recent study, Dr. Kathryn Kaiser, instructor in the office of energetics at the University of Alabama at Birmingham — working in conjunction with Dr. John Dawson, former UAB postdoctoral trainee in UAB’s department of biostatistics, section on statistical genetics, and current assistant professor in the department of nutritional sciences at Texas Tech University — noted that “[i]t has not been established whether control conditions with large weight losses (WLs) diminish expected treatment effects in WL or prevention of weight gain (PWG) randomized controlled trials (RCTs).” Therefore, they conducted a meta-analysis of 239 WL/PWG RCTs, all of which included a control group as well as at least one treatment group. Co-investigators were Dr. Olivia Affuso, associate professor in the department of epidemiology; Dr. Gary R. Cutter, professor in the department of biostatistics; and Dr. David B. Allison, director of the office of energetics and Nutrition Obesity Research Center (NORC).
Using a maximum likelihood meta-analysis framework to estimate the association between treatment effects and control group outcomes, results indicated that “an increase in control group WL of one kilogram corresponds with an expected shrinkage of the treatment effect by 0.309 kg [95 percent confidence interval (−0.480, −0.138), P=0.00081],” which was determined to be strong against violations of the model assumptions.
The researchers’ conclusions that control conditions with large WLs lessen expected treatment effects may prove to be useful to clinicians designing future WL/PWG studies, in order to be more confident in the design and targeted enrollment.
“Rigorous Control Conditions Diminish Treatment Effects in Weight Loss Randomized Controlled Trials” was published online in October in the International Journal of Obesity.
Journal article: http://www.nature.com/ijo/journal/vaop/naam/abs/ijo2015212a.html