Eating compared to skipping breakfast has no discernible benefit for obesity‐related anthropometrics: systematic review and meta‐analysis of randomized controlled trials.

Background Skipping versus eating breakfast has been consistently associated with higher weight in observational data, but the causal effects of eating versus skipping breakfast on obesity‐related anthropometric outcomes has not been systematically reviewed and meta‐analyzed. Objective To meta‐analyze randomized controlled trials investigating the effects of eating versus skipping breakfast on anthropometric measures related to weight, body size, and body composition. Methods This abstract extends our prior systematic review by updating the search and including meta‐analyses. Scopus, PsycInfo, CINAHL, Alt‐HealthWatch, and Proquest Global Thesis and Dissertation databases were searched for randomized controlled trials in humans that investigated the effects of skipping versus eating breakfast on obesity‐related outcomes. Studies were included that had interventions of 72 hours or longer, reported weight‐related outcomes measured objectively, included participants with normal weight or greater without diseases expected to influence weight, and investigated the effects of breakfast eating versus breakfast skipping in a way that could be isolated from other treatment components. Data were double coded. Outcome measurements (e.g., body weight and BMI) were synthesized separately using random effects meta‐analysis with inverse variance weighting. Analyses were stratified by baseline eating and skipping habits where possible. Results From 4732 search results we identified 8 articles that met all criteria for meta‐analysis. Of the 8 articles: 5 studies were conducted in the US, 3 in the UK; 7 in adults, 1 in adolescents (age 18+); n randomized per treatment arm ranged from 8 to 56, with a total of 381 participants; study length ranged from 2–16 weeks; 7 parallel arm RCTs, 1 cross‐over RCT; and 3 provided meals, 5 gave recommendations/meal plans. The meta‐analyzed outcomes are presented here as “outcome (number of studies): mean estimate [95% CI]”, with lower numbers favoring skipping breakfast. Body weight (7): −0.36 [−0.74, 0.03]. Fat mass (5): 0.13 [−0.62, 0.88]. Fat mass index (2): 0.00 [−0.26, 0.26]. Body fat percentage (4): 0.44 [−0.63, 1.50]. BMI (4): −0.17 [−0.35, −0.00]. Lean/fat free mass (4): −0.19 [−0.85, 0.47]. Waist circumference (3): −0.65 [−1.63, 0.32]. Waist to hip ratio (3): −0.01 [−0.02, 0.01]. Sagittal abdominal diameter (2): −0.17 [−0.62, 0.27]. No systematic differences were noted between habitual breakfast eaters and breakfast skippers, though these comparisons are limited by few study arms in each category. One study could not be included as reported, and additional data are being sought from authors of several papers to refine the estimates. However, the available published data from these studies indicate that changes to the meta‐analytic estimates are expected to be small. Conclusions Current randomized evidence does not support the general recommendation to eat and not skip breakfast for obesity‐related anthropometric outcomes. This systematic review and meta‐analysis was registered at http://www.crd.york.ac.uk/PROSPERO/ as CRD42016033290. Support or Funding Information AWB and DBA are funded in part by P30DK056336.