AI-generated research brief — verify at source
Intermittent Fasting Outperforms Ad Libitum Diet and Matches Continuous Energy Restriction for Glycemic Control in T2DM
In a meta-analysis of 14 randomized controlled trials involving 899 patients with type 2 diabetes or prediabetes, intermittent fasting produced meaningfully greater reductions in HbA1c compared to an unrestricted diet (moderate effect size: SMD −0.64, 95% CI −1.04 to −0.24), while achieving outcomes statistically comparable to continuous energy restriction diets across all metabolic markers. These findings position intermittent fasting as a viable dietary strategy for glycemic and weight management in this patient population.
What Was Studied
This study examined whether intermittent fasting (IF) could effectively reduce key metabolic indicators — specifically glycated hemoglobin (HbA1c), fasting blood glucose (FBG), body weight (BW), and body mass index (BMI) — in adults diagnosed with type 2 diabetes mellitus (T2DM) or prediabetes. A secondary objective was to determine whether IF performed differently from both unrestricted eating and structured caloric restriction approaches.
How It Was Studied
This was a systematic review and meta-analysis drawing on searches across five major databases — PubMed, Cochrane Library, Embase, Scopus, and Web of Science — for randomized controlled trials published before September 30, 2025. Fourteen eligible studies comprising 16 trial arms and 899 patients were included. The evidence was organized into two comparison streams: four studies (five arms, 316 patients) compared IF against an ad libitum (unrestricted) diet, and ten studies (eleven arms, 583 patients) compared IF against continuous energy restriction diets (CERD). Statistical analyses were conducted using RevMan 5.3 and Stata 17.0, with effect sizes reported as standardized mean differences (SMD, Hedges’s g).
What Was Observed
- HbA1c — IF vs. ad libitum: Intermittent fasting was associated with a moderate and statistically significant improvement in long-term glycemic control relative to unrestricted eating (SMD −0.64, 95% CI −1.04 to −0.24, p < 0.05). This represents the largest effect size observed across all metabolic outcomes in this comparison.
- Fasting blood glucose and body weight — IF vs. ad libitum: IF also produced significantly lower FBG (SMD −0.39, 95% CI −0.64 to −0.15, p < 0.05) and lower body weight (SMD −0.30, 95% CI −0.55 to −0.05, p < 0.05) compared to an unrestricted diet. BMI showed a borderline significant reduction (SMD −0.26, 95% CI −0.51 to 0.00, p = 0.05).
- IF vs. continuous energy restriction (CERD): Across all four outcomes — HbA1c, FBG, body weight, and BMI — intermittent fasting produced effects that were statistically indistinguishable from those of CERD, suggesting the two approaches are broadly equivalent in clinical impact for this population.
- Dropout rates: Adherence patterns differed across dietary arms. Dropout rates were 21.5% for IF, 26.2% for CERD, and 15.9% for the ad libitum group. CERD showed a significantly higher dropout rate than ad libitum, while no statistically significant differences in dropout were detected between IF and either comparator.
Why This Matters
For clinicians and researchers focused on dietary interventions in T2DM and prediabetes, these results suggest that intermittent fasting can achieve clinically meaningful improvements in glycemic control and body composition without requiring continuous caloric counting — and that it performs on par with the more demanding continuous energy restriction approach. The numerically lower dropout rate of IF compared to CERD, though not statistically significant, may have practical relevance for long-term dietary adherence in this population. These data support further investigation into IF as a structured alternative within the existing landscape of diabetes dietary management.
How to Read This Result
While directionally positive, these findings should be interpreted with caution given the modest total sample size (N = 899 across 14 studies), the use of standardized mean differences that limit direct clinical translation, and the medium overall quality signal — all of which constrain the generalizability and precision of the pooled estimates.
Limitations
The abstract does not explicitly report study limitations.