Intermittent fasting (IF) has become popular in recent years due to its simplicity and potential to facilitate weight loss. The most studied forms of IF are:
Alternate-day fasting (ADF): Alternating “fast days” (less than 500 kcal) with “feast days” (ad libitum intake).
5:2 diet: A modified version of ADF, which involves two fast days (500–1,000 kcal per day) and five feast days per week.
Time-restricted eating (TRE): Individuals confine their eating window to a specified number of hours per day (typically 4–8 hours) and fast for the remainder of the day.
This narrative review summarized the body of literature assessing the effectiveness of IF for reducing body weight and decreasing cardiometabolic disease risk. The authors also assessed the safety of IF and provided practical recommendations and directions for future research.
The authors included randomized controlled trials that were conducted in adults, lasted at least 5 weeks, and included either a control group (no change in diet or physical activity) or a daily calorie restriction group as a comparator.
They ultimately reviewed 6 studies that assessed ADF, 7 that assessed the 5:2 diet, and 9 that assessed TRE (two TRE studies used a 4-hour eating window; two used a 6-hour eating window; five used an 8-hour window). Most of the ADF studies were 8–12 weeks in duration (one was 52 weeks long), most of the 5:2 studies were 24–52 weeks in duration (one was 8 weeks; one was 12 weeks), and the TRE studies were 5–12 weeks in duration.
All 4 studies comparing ADF to a control group reported that ADF resulted in greater weight loss. All 3 comparing it to a 1,500-calorie restriction group reported similar effects on body weight. ADF reduced body weight by 4–8% from baseline in all included studies.
Both studies comparing 5:2 to a control group reported that 5:2 resulted in greater weight loss. All 7 studies comparing it to a 1,500–1,600-calorie restriction diet reported similar reductions in body weight. 5:2 reduced body weight by 4–7% from baseline in all of the included studies.
None of the 9 studies assessing TRE reported a greater loss in body weight than the control group, and only 3 reported a reduction in body weight from baseline, ranging from 3–4% (one used a 6-hour feeding window and two used an 8-hour window). One study using an 8-hour feeding window reported a within-group weight gain of 2% (the participants in that study were women with a BMI of less than 25 who performed resistance training). None of the included studies compared TRE to a calorie-restricted diet.
For ADF and 5:2, weight loss at 52 weeks was similar to that at 8–12 weeks, suggesting that weight loss efficiency peaks at 12 weeks. Additionally, weight loss was similar in participants both with and without diabetes.
Most of the included studies reported a ratio of fat mass loss to lean mass loss similar to that of calorie restriction (75% of weight lost as fat, 25% lost as lean mass). However, in 2 TRE studies using an 8-hour feeding window in participants with a BMI of less than 25, the participants experienced muscle mass maintenance or accretion and fat mass loss compared to a control group with a similar energy intake.
The participants reduced their overall energy intake by approximately 10–30% relative to their baseline. The researchers also noted that participants following ADF or 5:2 only consumed an additional 10–15% above their energy needs on feast days. The authors noted that, contrary to concerns expressed by some clinicians that people might overeat during their TRE eating window, most participants experienced an unplanned reduction in energy intake.
Eight studies assessed changes in nutrient and beverage intake from baseline, all of which found no changes in the intakes of dietary fat, carbohydrate, protein, sugar, saturated fat, monounsaturated fat, polyunsaturated fat, cholesterol, sodium, coffee, tea, diet sodas, energy drinks, or fiber, even though 5 of these studies provided dietary counseling to increase fruit, vegetable, and whole-grain intake.
Of the 4 studies assessing ADF and blood pressure, one reported a reduction in systolic and diastolic blood pressure from baseline. The reduction in systolic blood pressure was greater than that observed in a calorie restriction group.
Of the 5 studies assessing the 5:2 diet and blood pressure, 2 reported reductions in both systolic and diastolic blood pressure from baseline. The reductions were similar to those observed in a calorie restriction group in both studies.
Of the 6 studies assessing TRE and blood pressure, one reported reductions in systolic and diastolic blood pressure from baseline, and another reported a reduction in systolic blood pressure from baseline. None of the studies reported a difference compared to a control group.
The authors noted the studies reporting decreases in blood pressure involved participants with elevated blood pressure at baseline. They also noted that greater degrees of weight loss didn’t produce greater reductions in blood pressure, suggesting that reductions in blood pressure could be due to IF itself rather than weight loss.
All 6 ADF studies assessed blood lipids. One reported a reduction in LDL from baseline, which was similar to the reduction observed in a calorie-restriction group. One reported a reduction in HDL from baseline, and another reported an increase from baseline, which was greater than that observed in a control group. Three of the 6 studies reported reductions in triglycerides from baseline. In one of these studies, the reduction was greater than that observed in a calorie-restricted group.
All 7 studies assessing the 5:2 diet assessed blood lipids. One reported reductions in LDL and triglycerides from baseline, another reported reductions in HDL and triglycerides from baseline, and another reported reductions in LDL, HDL, and triglycerides. In all 3 studies, the changes in blood lipids were similar to a calorie-restricted control group.
Seven TRE studies assessed blood lipids. One 6-hour TRE study reported an increase in triglycerides from baseline, and one 8-hour TRE study reported a reduction in triglycerides from baseline. In both studies, the changes did not differ from a control group.
Five ADF studies assessed fasting glucose and fasting insulin. One reported an increase in fasting glucose from baseline, one reported decreases in fasting glucose and fasting insulin, and one reported a reduction in fasting insulin. Four studies assessed insulin resistance, one of which reported a reduction from baseline. Two studies assessed HbA1C, which did not report an effect of ADF.
Six 5:2 studies assessed fasting glucose, 2 of which reported reductions from baseline. Four 5:2 studies assessed fasting insulin, 3 of which reported reductions from baseline. Three 5:2 studies assessed insulin resistance, 2 of which reported reductions from baseline and compared to a calorie restriction group. Three studies assessed HbA1C, 2 of which reported reductions from baseline.
Eight TRE studies assessed fasting glucose, one of which reported an increase from baseline. Eight TRE studies assessed fasting insulin, 2 of which reported reductions from baseline. Six TRE studies assessed insulin resistance, 2 of which reported reductions from baseline. Three TRE studies assessed HbA1C, none of which reported a change from baseline.
The authors noted that the majority of the studies reporting no effects of IF on glycemic factors involved participants who did not have diabetes at baseline. This may explain why many of the studies found no effect of IF on glycemic factors.
Two ADF studies, one 5:2 study, and 4 TRE studies indicated that IF did not affect circulating inflammatory markers.
ADF and TRE did not increase the frequency of constipation, diarrhea, nausea, dry mouth, bad breath, irritability, fatigue, or dizziness. Headaches due to dehydration were occasionally reported.
Two trials (one ADF and one TRE study) indicated that IF didn’t increase the rates of binge eating, purging, depression, or fear of becoming overweight. However, as participants with an eating disorder history were excluded from these trials, it remains unknown whether IF is safe in individuals with a diagnosed eating disorder or those at risk of developing an eating disorder.
One study found that 8 weeks of 8-hour TRE decreased free and total testosterone concentrations in men. In contrast, a 24-week 5:2 study found that IF did not affect androstenedione, testosterone, dehydroepiandrosterone sulfate, sex hormone-binding globulin, or prolactin in premenopausal women. Two studies (one 8-hour TRE study and one ADF study) reported that circulating thyroid hormone levels remained unchanged after IF.
Finally, the authors reported that IF either had no effect on resting metabolic rate when weight was maintained or resulted in minor reductions (100–200 kcal) when weight was reduced by 5–7% from baseline.
The authors indicated that the following populations should not engage in intermittent fasting:
Children less than 12 years old
Adolescents of normal weight
Women who are pregnant or lactating
People with a history of an eating disorder
People with a BMI of less than 18.5 kg/m2
People over 70 years old
The authors noted that healthy adults with a BMI of at least 25 and who do not have type 1 diabetes, type 2 diabetes, or other comorbidities can safely engage in IF without medical supervision. They noted that IF might be effective for weight management in adolescents with obesity, but also that IF in this group should only be undertaken when weight loss is clinically indicated and under medical supervision.
The authors noted an adjustment period of 1–2 weeks, during which headaches are common. They noted that participants should be encouraged to consume plenty of fruits, vegetables, and whole grains to increase their fiber and micronutrient intake and consume at least 50 grams of protein on fast days (for ADF and 5:2) to mitigate hunger and prevent excessive lean mass loss. They advised against alcohol on fast days, as energy intake is greatly limited. Behavioral change strategies should be encouraged as an adjunct therapy to fasting to promote weight management.
Considerations for diabetes mellitus
Preliminary findings suggest that intermittent fasting might be safe for people with type 1 or 2 diabetes. However, these individuals should be in weekly contact with their physician to monitor for adverse effects and should test their fasting glucose levels throughout the day. People with type 1 diabetes who are on ADF or 5:2 might be advised to reduce their basal insulin dose by 50% on fast days and 10% on non-fast days. People with type 2 diabetes and an HbA1C of less than 7% might be advised to discontinue sulfonylureas and insulin while undergoing ADF or the 5:2 diet. An endocrinologist should work with each person individually, and medications should not be adjusted without consultation with a medical professional.
The authors noted the following directions for future research to address gaps in the IF literature:
Long-term (more than 1 year) studies to assess the feasibility and long-term efficacy of ADF, 5:2, and TRE.
Studies that assess the effects of IF in people with medical conditions such as type 1 diabetes, type 2 diabetes, polycystic ovary syndrome, and thyroid disorders.
Studies comparing ADF, 5:2, and TRE to each other to determine whether one regimen is superior to the others.
Studies that are able to separate the effects of IF from the effects of weight loss to further elucidate the mechanisms that underlie metabolic improvements observed with IF.
Limitations of the current evidence
The authors noted that the majority of the included studies had small sample sizes (10–150 participants), were conducted in the U.S., and were at high risk of bias primarily due to missing participant outcome data.
The big picture
The findings of the current review regarding body weight and body composition are consistent with recent meta-analyses (one of which is covered in this previous review), suggesting that IF and continuous energy restriction result in similar reductions in body weight and fat mass.
Here are some additional key takeaways from previous Examine research breakdowns on this topic:
Individuals who combine 8-hour TRE with resistance training shouldn’t have any issues gaining strength and muscle mass. When paired with a caloric deficit, 8-hour TRE might preserve lean mass, but those findings aren’t certain.
8-hour TRE is well-tolerated by individuals with obesity, who have relatively high adherence rates.
Individuals at risk for type 2 diabetes can engage in 8-hour TRE safely, and simply engaging in TRE with no other dietary restrictions may provide some metabolic benefits.
One study has investigated the effect of 8-hour TRE on the gut microbiome and didn’t find any noteworthy changes.
Limited evidence suggests that 8-hour TRE does not affect sleep.
A recent Study Summary also covered a 12-week randomized controlled trial comparing 5:2 IF with a 70% energy restriction on fast days to a continuous 20% calorie restriction group. Both groups experienced similar increases in lean mass, thigh muscle size, and reductions in body weight and body fat. However, the continuous energy restriction group demonstrated greater increases in muscle surface area than the IF group.
There are 8 more summaries in the Editor's Picks category for June 2022 including ...
- Beans, they’re good for your heart. The more you eat, the more … fasting blood glucose decreases?
- Can vitamin D prevent COVID-19 infection?
- Psilocybin, the brain, and depression
Become an Examine Member to view the latest study summaries across 25 categories.