select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
A review published in the Cochrane Database of Systematic Reviews looks at intermittent fasting for adults with obesity.
Dr Baptiste Leurent, Associate Professor in Medical Statistics, UCL Department of Statistical Science, UCL, said:
“This study is a systematic review evaluating intermittent fasting in overweight or obese adults. It combines results from 22 randomised controlled trials and looks at the effect of intermittent fasting on weight loss, quality of life and adverse events.
“It appears a very well conducted review, with a robust methodology and appropriate interpretation from the authors. Cochrane reviews such as this are considered the gold standard in terms of reviewing the available evidence on health interventions.
“This review found absolutely no benefit of intermittent fasting on weight loss, when compared to standard dietary advice.
“When compared to not doing anything, however, it did identify a small benefit, with a 2 to 5% average reduction in weight 6 to 12 months later. This was statistically significant but remains small.
“Although the individual trials were small and of limited quality, taken together they provide a clear indication that intermittent fasting offers little benefit.
“This is yet another example of a misalignment between public perception and the scientific evidence.”
Prof Keith Frayn, Emeritus Professor of Human Metabolism, University of Oxford, said:
“This is a very authoritative review that considers all relevant published work comparing intermittent fasting with other means of weight loss. And yet the authors fail to find evidence that the various forms of intermittent fasting are superior to ‘regular dietary advice’, albeit conceding that many of the individual studies are of poor quality. Intermittent fasting has been widely promoted as a means of achieving weight loss, but often on the basis of claims that it has special effects on metabolism beyond simple restriction of calories. This study shows that such claims can have little relevance. The message is that, for people with overweight or obesity, calorie intake must be reduced to produce weight loss and there are no ‘quick fixes’. Nevertheless, anecdotally many people find intermittent fasting a tolerable way to avoid weight gain, but this is not an issue investigated in the present review.”
Prof Paul Garner, Emeritus Professor, Evidence Synthesis in Global Health, Liverpool School of Tropical Medicine, said:
“Many public figures have promoted Intermittent fasting as a practical way to reduce calorie intake and thus lose weight. This analysis examines trials that evaluated what happens when doctors in medical outpatient clinics recommend this strategy over regular dietary advice or no advice. None of the studies examine people who decide themselves to follow these diets, and who are therefore motivated to embark on intermittent fasting themselves. The authors conclude across the 22 studies conducted in hospital medical clinics that intermittent fasting has little or no effect on weight loss, and use a standard tool to evaluate evidence. The authors are correct in saying that we don’t really know yet – it needs more research, and their analysis doesn’t evaluate whether the strategy works for highly motivated people that decide to adopt this strategy of weight loss themselves, rather than have it foisted on them in medical outpatients.”
Dr Adam Collins, Associate Professor of Nutrition, University of Surrey, said:
“This robust review and meta-analysis of intermittent fasting (IF) focuses specifically on weight loss and quality-of-life benefits. The headline finding that intermittent fasting is no better (or worse) than standard dietary advice or doing nothing is attention-grabbing, but should, I feel, be interpreted with a little caution.
“Firstly, the authors describe no difference in comparison with “regular dietary advice”, which is a little misleading. In the studies included, “regular dietary advice” translates as continuous energy restriction (CER), for example, prescribing a daily energy deficit, as is common practice in weight management. Intermittent fasting regimens typically create an energy (calorie) deficit analogous to that seen in CER, whether through intermittent energy restriction (e.g., 5:2 protocols) or time-restricted eating (TRE). Furthermore, many of these studies are designed to examine the effect of IF on other outcomes (e.g. changes in metabolic variables or meal handling) and therefore incorporate CER (with a similar overall calorie deficit) as a comparator group by design.
“What is more intriguing is their conclusion that IF shows no greater weight loss than “doing nothing”. This is, again, a little misleading. The resultant box plot (Figure 6) does imply that weight loss is more favourable in the IF group, as you would expect. This appears to be at odds with the statement of results in the text. Nevertheless, even if accepting their claim that there was no clear difference in weight-loss outcomes, there are two possible explanations.
“First, the handful of studies included may not be representative of typical IF regimens. For example, one involves participants undertaking TRE just twice per week (compared with every day, as is standard), whilst another asks participants to fast just once per week (instead of the more common twice per week or every other day). Hence, these interventions will result in a smaller overall calorie deficit and less weight loss.
“The second explanation is one that is common in most dietary intervention studies, namely the Hawthorne effect – the phenomenon where participants change their behaviour simply because they know they are being observed or are part of a study. In other words, people are unlikely to be “doing nothing”. The sheer act of filling in food diaries or being weighed can lead people to change their eating behaviour. It is very common for people in control groups to lose some weight or become more active, just from the awareness that they are “in a study” and might be monitored or asked about their habits. Those on the waiting list may also change their behaviour in anticipation or preparation for undertaking the intervention.
“In conclusion, whilst this review is of interest and will generate clicks, it should not distract from the consensus from other studies in the scientific literature that IF regimens are an effective tool for weight loss. They can also be practically advantageous for many people. In addition, several studies (including our own) have suggested IF regimens may offer other metabolic benefits that are independent of weight loss, which is where much of the research on these regimens is now focused. This makes these approaches more universally useful, not just for those who are overweight or living with obesity – for example, in the management of metabolic disease, and the maintenance of weight once it is lost (e.g. post-GLP-1).”
Comment from our friends at the German SMC:
Prof Leonie Heilbronn, PhD, Group Leader, Obesity and Metabolism, School of Medicine, College of Health, Adelaide University:
“There have been dozens of meta-analysis on this topic and they have had a similar results. IF (intermittent fasting) produces similar weight losses to other moderate CR (calorie restriction) type weight loss interventions – typically around 7% after 6-12 months in both groups. This weight loss average is very common over 6-12 months. More can be achieved by very low calorie diet (10-15%), pharmacotherapy (15-20%) and surgery (>20%).
“Comment: This review combines many forms of intermittent fasting, which is a mistake. There are many types of ‘IF’ and they are not equivalent for weight loss and should not be mashed together.
“TRE (time-restricted eating) is a milder intervention than IF – it implements a 2-3 hour time restriction each day, resulting in ~300kcal/d restriction and 2-3 kg of weight loss over 6 -12 months (unless added on top of other diets). TRE trials typically compare against no intervention control or changing diet quality. There is some evidence that health is improved despite modest weight loss by TRE due to better alignment of circadian rhythms (but jury is out – more trials required).
“IF, or ADF (alternate-day fasting), is a 75-100% energy restriction for 2 or 3 days per week and regular eating practice on other days. Most trials take it against moderate 20-30% CR over 6-12 months, some incl. a third arm ‘standard care or control’. Most of the trials taking CR against IF have around 5-7% weight loss in both groups and similar improvements in body weight and in diabetes and CVD risk factors.
“Claim 1 : not really sure what you mean here, but in their first claim they show that IF produces similar outcomes to weight loss interventions with 21 studies included.
“Claim 2: Claim 2 seems to claim that IF has similar outcome to doing nothing. To me that does not make sense in light of claim 2 – that it is as good as weight loss interventions. Probably as there are only 6 studies involved in claim 2.
“IF produces statistically significant weight loss. Across the six studies included in claim 2, intermittent fasting resulted in about 3.4% greater weight loss than ‘control’. That’s arguably clinically meaningful on its own – for example, NICE guidelines consider 3% meaningful, but US guidelines do use 5% as a benchmark. But many of the control participants in those study still lost about 2%, so the actual weight loss with IF is closer to 5%.
“Very few studies were included. Claim 2 draws on only six studies compared with 21 in claim 1. Of these six, two used time-restricted eating, which generally produces smaller weight loss, and one involved fasting just one day per week, which would also be expected to have less impact. That leaves only three intermittent fasting trials. In one of these, only the 18-month follow-up was reported, they left out the primary 6-month time point when participants were undertaking the diet. In that study, the differences between IF and control was around 5% weight loss versus ‘control’.
“Type of control matters. Some of the control groups were not completely inactive or waitlist – they received some information and guidance for weight loss. This reduces the apparent difference between IF and control.
“Completely agree, they have heterogeneous results because they are lumping together diets that should not be put together. Which will result in heterogeneity… some also include those with diabetes who are typically refractory to weight loss (even with semaglutide).
“You can either look at the results of individual trials or combine like in a meta-analysis. Intermittent fasting is a proven alternative to continuous calorie restriction over 6–12 months. Both IF/ADF and calorie restriction typically produce around 7% weight loss when dietary support is provided. Time-restricted eating is a more modest approach, usually resulting in about 3 kg of weight loss – unless the eating window is further shortened or combined with other dietary strategies.
“Many people find it very difficult to keep weight off beyond 12 months, no matter what approach they use. This is because the body has biological mechanisms that encourage weight regain after weight loss. For example, levels of satiety peptides and other hormones that make you feel full drop, leptin also drops decreasing your energy expenditure, and hunger hormones like ghrelin increase. As a result, weight often returns after a year, whether someone followed a calorie-restricted diet or intermittent fasting. Exercise and weight loss medications can help maintain the loss by counteracting these biological signals.”
Conflicts of interest: No conflicts.
‘Intermittent fasting for adults with overweight or obesity’ by first author et al. was published in the Cochrane Database of Systematic Reviews at 01:00 UK time on Monday 16 February 2026.
DOI: 10.1002/14651858.CD015610.pub2
Declared interests
Dr Baptiste Leurent: “No conflict of interest.”
Prof Keith Frayn: “I have no conflict of interest to declare other than that I am an author of books on metabolism and nutrition.”
Prof Paul Garner: “No conflicts of interest.”
Dr Adam Collins: “I declare no conflict of interest.”