Revisiting the impact of Health at Every Size® interventions on health and cardiometabolic related outcomes: An updated systematic review with meta-analysis.

Abstract

Aims: To (1) synthesise evidence from Health at Every Size® interventions on physical and psychological health in people with overweight and obesity and (2) report between-group differences within interventions evaluating the impact of Health at Every Size® interventions on health and health-related outcomes.

Methods: Six databases (Medline, Embase, Cochrane, PsychInfo, CINAHL, and Scopus) were searched from inception until November 2022. Included studies were conducted in adults with overweight or obesity, used Health at Every Size®-based interventions compared with control interventions and reported dietary, physical and/or psychological outcomes, including diet quality, anthropometry, or quality of life. Data on between-group differences were extracted. Risk of bias was assessed using ROB2. Random-effects meta-analyses were undertaken for outcomes with at least three studies reporting the same or comparable data.

Results: From 128 studies identified, 19 full-text articles (10 unique studies, 6 published since 2017), were included. Meta-analysis found a significant reduction for susceptibility to hunger in Health at Every Size® intervention groups relative to controls (p = 0.005), with no significant difference (p > 0.05) between Health at Every Size® interventions and control groups for anthropometric, psychological or cardiometabolic outcomes (total cholesterol, HDL cholesterol, triglycerides, systolic or diastolic blood pressure).

Conclusion: Health at Every Size® interventions had similar results compared with weight-based interventions on anthropometric outcomes and cardiometabolic outcomes. Health at Every Size® interventions had a significant benefit for reducing susceptibility to hunger. The decision to use a Health at Every Size®-based intervention should be personalised to individual needs. Further research in more diverse populations is required using standardised outcome measures to facilitate future meta-analyses.