Intermittent fasting (IF) and Paleolithic (Paleo) diets produce weight loss in controlled trials, but minimal evidence exists regarding long-term efficacy under free-living conditions without intense dietetic support.
This exploratory, observational analysis examined adherence, dietary intake, weight loss, and metabolic outcomes in overweight adults who could choose to follow Mediterranean, IF, or Paleo diets, and standard exercise or high-intensity interval training (HIIT) programs, as part of a 12-mo randomized controlled trial investigating how different monitoring strategies influenced weight loss (control, daily self-weighing, hunger training, diet/exercise app, brief support).
A total of 250 overweight [BMI (in kg/m2) ≥27] healthy adults attended an individualized dietary education session (30 min) relevant to their self-selected diet. Dietary intake (3-d weighed diet records), weight, body composition, blood pressure, physical activity (0, 6, and 12 mo), and blood indexes (0 and 12 mo) were assessed. Mean (95% CI) changes from baseline were estimated using regression models. No correction was made for multiple tests.
Although 54.4% chose IF, 27.2% Mediterranean, and 18.4% Paleo diets originally, only 54% (IF), 57% (Mediterranean), and 35% (Paleo) participants were still following their chosen diet at 12 mo (self-reported). At 12 mo, weight loss was -4.0 kg (95% CI: -5.1, -2.8 kg) in IF, -2.8 kg (-4.4, -1.2 kg) in Mediterranean, and -1.8 kg (-4.0, 0.5 kg) in Paleo participants. Sensitivity analyses showed that, due to substantial dropout, these may be overestimated by ≤1.2 kg, whereas diet adherence increased mean weight loss by 1.1, 1.8, and 0.3 kg, respectively. Reduced systolic blood pressure was observed with IF (-4.9 mm Hg; -7.2, -2.6 mm Hg) and Mediterranean (-5.9 mm Hg; -9.0, -2.7 mm Hg) diets, and reduced glycated hemoglobin with the Mediterranean diet (-0.8 mmol/mol; -1.2, -0.4 mmol/mol). However, the between-group differences in most outcomes were not significant and these comparisons may be confounded due to the nonrandomized design.
Small differences in metabolic outcomes were apparent in participants following self-selected diets without intensive ongoing dietary support, even though dietary adherence declined rapidly. However, results should be interpreted with caution given the exploratory nature of analyses. This trial was registered with the Australian New Zealand Clinical Trials Registry as ACTRN12615000010594 at https://www.anzctr.org.au.