Identifying causal relations is fundamental to a science of intervention and prevention. Obesity is a major problem for which much progress in understanding, treatment, and prevention remains to be made. Behavior is a vital contributor to variations in energy balance and body composition, the final common pathways of obesity. Social, environmental and physiological factors are also key influences on behaviors which affect energy balance. Evidence for causation of these hypothesized factors exists on a continuum from weakest to strongest. Yet, most obesity research does not consider the evidence continuum between ordinary association tests (OATs) (observational, non-intervention studies among unrelated individuals), which do not offer strong evidence of causal effects, and randomized controlled trials (RCTs), which do offer strong evidence, but cannot be done in all circumstances. In contrast, there are techniques that lie intermediary between OATs and RCTs, including but not limited to quasi-experimental studies and natural experiments. Such designs are increasingly used, especially in the disciplines of economics and genetics, but are used by obesity researchers less often than seems warranted. Our ability to draw causal inferences in obesity research could be strengthened by using such approaches. In-depth understanding and appropriate use of the full continuum of these methods requires input from disciplines including statistics, economics, psychology, epidemiology, mathematics, philosophy, and behavioral or statistical genetics. Applying these techniques, however, does not involve routine ?cookbook? approaches but requires understanding of underlying principles, so the investigator can tailor approaches to specific and varying situations. Yet, other than our annual 5-day short course, no resource provides such training, particularly for behavioral and social science researchers. Our short course on methods for causal inference in obesity research, which features some of the world?s finest scientists, has been consistently evaluated by attendees as essential for their research and teaching. The course provides rigorous exposure to the key fundamental principles underlying a broad array of techniques and experience in application through guided discussion using real examples. The course is dynamic in that we refine and modify its content and delivery methods based on feedback from stakeholders. In this renewal application, we propose to continue to offer this course, alternating its location between Indiana University and the University of Alabama at Birmingham. Given the prevalence of obesity and its related health problems, training behavioral and social science researchers to better assess causal effects is more important than ever.

Public Health Relevance

Obesity affects about 40% of the US adult population generating an urgent need for novel interdisciplinary approaches to identify factors causing (not merely correlating with) obesity and strategies which cause reductions in obesity. Many research techniques exist to help scientists make more informed conclusions about causal effects, but many are unfamiliar with these techniques. We propose a course to train behavioral and social science researchers in understanding and using such techniques, so they will be better positioned to help identify ways of reducing the burden of obesity.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Education Projects (R25)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Pratt, Charlotte
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Indiana University Bloomington
Schools of Public Health
United States
Zip Code
Dickinson, Stephanie L; Brown, Andrew W; Mehta, Tapan et al. (2018) Incorrect analyses were used in ""Different enteral nutrition formulas have no effect on glucose homeostasis but on diet-induced thermogenesis in critically ill medical patients: a randomized controlled trial"" and corrected analyses are requested. Eur J Clin Nutr :
McComb, Bryan; Frazier-Wood, Alexis C; Dawson, John et al. (2018) Drawing conclusions from within-group comparisons and selected subsets of data leads to unsubstantiated conclusions: Letter regarding Malakellis et al. Aust N Z J Public Health 42:214
Hannon, Bridget A; Thomas, Diana M; Siu, Cynthia et al. (2018) The claim that effectiveness has been demonstrated in the Parenting, Eating and Activity for Child Health (PEACH) childhood obesity intervention is unsubstantiated by the data. Br J Nutr 120:958-959
Brown, Andrew W; Kaiser, Kathryn A; Allison, David B (2018) Issues with data and analyses: Errors, underlying themes, and potential solutions. Proc Natl Acad Sci U S A 115:2563-2570
Ejima, K; Pavela, G; Li, P et al. (2018) Generalized lambda distribution for flexibly testing differences beyond the mean in the distribution of a dependent variable such as body mass index. Int J Obes (Lond) 42:930-933
Allison, David B (2018) The Conclusions Are Unsupported by the Data, Are Based on Invalid Analyses, Are Incorrect, and Should be Corrected: Letter Regarding ""Sleep Quality and Body Composition Variations in Obese Male Adults after 14 weeks of Yoga Intervention: A Randomized Con Int J Yoga 11:83-84
Dawson, J A; Brown, A W; Allison, D B (2018) The stated conclusions are contradicted by the data, based on inappropriate statistics, and should be corrected: comment on 'intervention for childhood obesity based on parents only or parents and child compared with follow-up alone'. Pediatr Obes 13:656-657
Speakman, J R; Loos, R J F; O'Rahilly, S et al. (2018) GWAS for BMI: a treasure trove of fundamental insights into the genetic basis of obesity. Int J Obes (Lond) 42:1524-1531
Smith Jr, D L; Thomas, D M; Siu, C O et al. (2018) Regression to the mean, apparent data errors and biologically extraordinary results: letter regarding 'changes in telomere length 3-5 years after gastric bypass surgery'. Int J Obes (Lond) 42:949-950
George, Brandon J; Li, Peng; Lieberman, Harris R et al. (2018) Randomization to randomization probability: Estimating treatment effects under actual conditions of use. Psychol Methods 23:337-350

Showing the most recent 10 out of 32 publications