Obesity is estimated to affect over 50 million Americans and its resulting morbidities remain significant problems for many individuals. Current treatments are only modestly effective and no treatment works for all individuals. Thus, it is critical to identify and evaluate new strategies for both efficacy and safety. Yet, several important questions about how to best design, interpret, and analyze randomized clinical trials for obesity treatments remain unanswered and in some cases completely unaddressed. Tens of millions of dollars are spent on obesity treatment studies annually by government and industry. Such studies could be better designed and more efficiently executed if they could draw on a well established body of empirical evidence regarding factors such as the effects of using run-in periods, expected drop-out rates, factors affecting drop-out rates, and the relation between population treatment effect and the expected degree of weight loss in a control condition. Information about all of these factors is critical for optimally designing studies, yet no rigorously established comprehensive body of knowledge about these factors exists. Thus, the purpose of the proposed research is to use meta-analytic methodology to develop this knowledge.
Three specific aims are proposed.
Aim 1. Comprehensively investigate of the effect of run-in periods on (a) subject drop-out rates (DORs);and (b) weight loss (WL).
Aim 2. Comprehensively characterize DORs and patterns in obesity RCTs and how such rates and patterns vary as a function of RCT and patient characteristics using meta-regression of published summary statistics AND analysis of multiple raw datasets from large RCTs.
Aim 3. Use a novel hierarchical meta-regression method to estimate the association between the true population (not sample) treatment effect and the true population (not sample) mean weight loss of the control group. This is important because it influences decisions about how vigorous the lifestyle modification interventions provided to control groups in RCTs of pharmaceuticals, surgery, and extra-lifestyle interventions should be. To accomplish these aims, we have assembled an experienced team of obesity researchers and statisticians. This research should give valuable answers and critical information for more evidence-based decisions about the design, analysis, and interpretation of obesity RCTs.

Public Health Relevance

The purpose of the research is to use meta-analytic methodology to develop this knowledge that would inform the design of randomized clinical trials for obesity treatment and prevention. It will entail comprehensively investigating the effect of run-in periods on subject drop-out rates (DORs) and weight loss;determining predictors of DORs, and estimating the association between the true population (not sample) treatment effect and the true population (not sample) mean weight loss of the control group. This research should give valuable answers and critical information for more evidence-based decisions about the design, analysis, and interpretation of obesity RCTs.

Agency
National Institute of Health (NIH)
Institute
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
Type
Research Project (R01)
Project #
5R01DK078826-03
Application #
8039196
Study Section
Kidney, Nutrition, Obesity and Diabetes (KNOD)
Program Officer
Everhart, James
Project Start
2009-03-01
Project End
2013-02-28
Budget Start
2011-03-01
Budget End
2013-02-28
Support Year
3
Fiscal Year
2011
Total Cost
$247,279
Indirect Cost
Name
University of Alabama Birmingham
Department
Biostatistics & Other Math Sci
Type
Schools of Public Health
DUNS #
063690705
City
Birmingham
State
AL
Country
United States
Zip Code
35294
Dawson, J A; Kaiser, K A; Affuso, O et al. (2016) Rigorous control conditions diminish treatment effects in weight loss-randomized controlled trials. Int J Obes (Lond) 40:895-8
Kaiser, Kathryn A; Affuso, Olivia; Desmond, Renee et al. (2014) Baseline participant characteristics and risk for dropout from ten obesity randomized controlled trials: a pooled analysis of individual level data. Front Nutr 1:
Affuso, O; Kaiser, K A; Carson, T L et al. (2014) Association of run-in periods with weight loss in obesity randomized controlled trials. Obes Rev 15:68-73
Kaiser, Kathryn A; Gadbury, Gary L (2013) Estimating the range of obesity treatment response variability in humans: methods and illustrations. Hum Hered 75:127-35
Kaiser, K A; Affuso, O; Beasley, T M et al. (2012) Getting carried away: a note showing baseline observation carried forward (BOCF) results can be calculated from published complete-cases results. Int J Obes (Lond) 36:886-9
Kaiser, K A; Cofield, S S; Fontaine, K R et al. (2012) Is funding source related to study reporting quality in obesity or nutrition randomized control trials in top-tier medical journals? Int J Obes (Lond) 36:977-81
Poulson, Robert S; Gadbury, Gary L; Allison, David B (2012) Treatment Heterogeneity and Individual Qualitative Interaction. Am Stat 66:16-24
Marrie, Ruth Ann; Cutter, Gary; Tyry, Tuula (2011) Substantial adverse association of visual and vascular comorbidities on visual disability in multiple sclerosis. Mult Scler 17:1464-71
Marrie, R A; Rudick, R; Horwitz, R et al. (2010) Vascular comorbidity is associated with more rapid disability progression in multiple sclerosis. Neurology 74:1041-7
Marrie, Ruth Ann; Cutter, Gary; Tyry, Tuula et al. (2009) A cross-sectional study of bone health in multiple sclerosis. Neurology 73:1394-8

Showing the most recent 10 out of 11 publications