Skip to main content
ARS Home » Northeast Area » Boston, Massachusetts » Jean Mayer Human Nutrition Research Center On Aging » Research » Publications at this Location » Publication #340137
Title: Comparison among criteria to define successful weight-loss maintainers and regainers in the Action for Health in Diabetes (Look AHEAD) and Diabetes Prevention Program trials

Author
item BERGER, SAMANTHA - Tufts University
item HUGGINS, GORDON - Tufts University
item MCCAFFERY, JEANNE - Brown University
item LICHTENSTEIN, ALICE - Jean Mayer Human Nutrition Research Center On Aging At Tufts University

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/12/2017
Publication Date: 10/18/2017
Citation: Berger, S., Huggins, G.S., McCaffery, J.M., Lichtenstein, A.H. 2017. Comparison among criteria to define successful weight-loss maintainers and regainers in the Action for Health in Diabetes (Look AHEAD) and Diabetes Prevention Program trials. American Journal of Clinical Nutrition. 106(6):1337-1346. https://doi.org/10.3945/ajcn.117.157446.
DOI: https://doi.org/10.3945/ajcn.117.157446

Interpretive Summary: Weight loss decreases cardiovascular disease and type 2 diabetes risk, but risk does not always stay lowered when weight loss is not maintained. Weight loss interventions can result in successful weight loss, but rates of weight regain are high. There is currently no standard definition for successful weight loss maintenance. The objective of this study was to compare published categorization criteria used to define individuals who lost weight and then either successfully maintained the weight loss (maintainers) or regained the lost weight (regainers.) Data from two weight loss intervention trials, Action for Health in Diabetes (Look AHEAD) and Diabetes Prevention Program (DPP,) were used to compare the published categorization criteria using weight data 48 months post intervention. Eight categorization criteria were identified. Regainers had significantly higher body weight at month 48 than maintainers using all categorization criteria. Criteria were classified as strict (no regain,) moderate (partial regain,) or liberal (all regain.) Agreement between criteria was only among the strict or moderate. Redundancy among criteria was low; most comparisons did not reach agreement. While all criteria discriminated maintainers and regainers based on weight at 48 months post-intervention, the criteria showed little agreement. A standard definition of weight loss maintenance is needed for cross-study comparisons to facilitate the identification of behavioral characteristics associated with successful weight loss maintenance.

Technical Abstract: Objective: Compare published categorization criteria that define people who successfully maintain weight loss (maintainers) and those who regain lost weight (regainers). Methods: Publically available data from Look AHEAD (n=1791) and DPP (n=631) who lost >/=3% initial body weight in response to lifestyle weight loss intervention and had 48-month follow-up data. Criteria that defined maintainers and regainers used previously were compared with respect to month 48 weight, number of participants, and redundancy among the criteria via agreement statistics. Results: Eight categorization criteria were identified. For all criteria regainers had significantly higher weight at month 48 compared with maintainers (p<0.0001). We found strict (no regain), moderate (partial regain) and liberal (all regain) criteria. Moderate criteria generated a 50/50 split in maintainers/regainers, extreme a 20/80 split, and two in-between 34/66 split. Any concordance across criteria was among strict or moderate, but redundancy was low, with no comparisons reaching nearly perfect agreement (kappa<0.80) in Look AHEAD, and two agreeing in DPP (kappa=0.86 and kappa=0.82). Conclusions: While all criteria discriminated based on weight, the criteria showed little redundancy in defining maintainers and regainers, which limits cross-study comparisons. The findings in Look AHEAD were highly reproducible when applied in DPP.