Alternate-day fasting asks a more demanding question than moving dinner earlier: what happens when low-intake days alternate with ordinary eating days? Human trials suggest it can reduce weight, but its advantage over daily calorie restriction is modest. For anyone considering it in midlife, the useful distinction is between a change on the scales and evidence of better long-term health.
First, define the intervention
Alternate-day fasting is not one uniform treatment. Some studies use days without calories; others allow a small amount of food. That matters when interpreting a headline about “fasting”. A result from a modified regimen cannot automatically be transferred to a water-only fast. The quantity of food, the comparison diet, and the support provided are part of the intervention.
In John Trepanowski and colleagues’ 2017 randomised trial in JAMA Internal Medicine, fasting days allowed 25% of participants’ energy needs, alternating with days prescribed at 125%. The daily-restriction group was prescribed 75% every day. Both approaches therefore aimed at the same average energy reduction across two days. This was a comparison of schedules as much as a comparison of diets.
That design helps clarify the central question. If two plans intend to reduce average intake similarly, does alternating the restriction produce a clinically useful extra benefit? It also prevents a misleading comparison between a carefully supported fasting programme and an unchanged diet. Those comparisons answer different questions, and readers deserve to know which one is being asked.
The weight-loss advantage needs a scale
A 2025 systematic review and network meta-analysis in The BMJ, led by Zhila Semnani-Azad, brought together 99 randomised trials involving 6,582 adults. Alternate-day fasting produced an estimated 1.29 kg more weight loss than continuous energy restriction. The confidence interval ranged from 0.59 to 1.99 kg more, and the evidence was rated moderate certainty.
This is an average difference between approaches, not the total weight someone lost. The authors’ threshold for a minimally important weight difference was 2 kg; the estimate fell below it. That does not make the finding meaningless, but it puts “better” into proportion. Longer studies, lasting at least 24 weeks, did not establish the same advantage over continuous restriction.
The review’s median follow-up was 12 weeks. It mainly informs a discussion about relatively short-term weight management. My reading is that the evidence can support considering different approaches with appropriate clinical help; it cannot justify presenting the most restrictive schedule as the default. A statistically detectable difference still needs to be weighed against what the intervention asks of the person following it.
A year-long trial adds the question of adherence
The 2017 JAMA Internal Medicine trial enrolled 100 adults with obesity. At 12 months, weight loss relative to the control group was about 6.0% with alternate-day fasting and 5.3% with daily restriction. It did not establish superior weight loss with the alternate-day approach. Dropout was 38% in that group, compared with 29% with daily restriction.
Participants also ate more than prescribed on fasting days and less than prescribed on the higher-intake days. The schedule people actually followed therefore moved closer to a less sharply alternating pattern. This is a useful observation, not a moral judgement about compliance. A dietary strategy has to be assessed as people can live with it, alongside its intended design.
A single trial does not settle what everyone will prefer. Nor does the dropout difference prove that the schedule alone caused people to leave. It does, however, make persistence a legitimate question for a consultation. If the proposed benefit depends on months of repetition, asking whether the arrangement is workable is part of evaluating the treatment rather than an afterthought.
Short trials cannot establish a longevity benefit
The 2025 BMJ review evaluates weight and cardiometabolic risk factors. Those measurements can inform a discussion of dietary options, but they do not establish that alternate-day fasting extends human life. A positive finding needs to remain attached to the outcome and follow-up actually studied.
What we have is evidence about intermediate outcomes. What a longevity claim would require is evidence about later illness, function, or survival. A difference on the scales cannot supply that missing link merely because the intervention sounds biologically interesting. The distinction matters whenever short-term results are used to encourage a lifelong habit.
Mayo Clinic’s discussion of the long-term evidence makes the same broad caution: benefits and risks over extended periods remain uncertain. No number of days without food should be read as a validated measure of slower ageing.
Safety can change the decision entirely
For someone with diabetes, the medication question comes before any comparison of weight-loss averages. NIDDK’s clinical discussion of fasting with diabetes identifies low blood glucose as a principal risk. Reducing treatment without an appropriate plan can also cause high glucose and, in some circumstances, diabetic ketoacidosis.
The same guidance explains why an every-other-day pattern can make medication adjustment more complicated than a consistent daily schedule. Insulin and sulfonylureas deserve particular attention. This is not a reason to improvise a dose reduction, or to assume that omitting a tablet makes fasting safe. Assessment and any changes belong with the treating team.
Dehydration is another concern, especially when fluids are restricted or medicines increase fluid loss. NIDDK identifies additional risks in type 1 diabetes. A discussion of fasting should therefore include the exact form of the fast, existing conditions, and every medicine taken. “Intermittent fasting” is too broad a label to function as a safety assessment.
NHS-hosted guidance on intermittent fasting advises that pregnant or breastfeeding people and those with a history of eating disorders should not attempt it. These are substantive exclusions, not a small-print qualification to a recommendation. A dietitian or clinician can help discuss alternatives without making fasting the goal.
Discomfort is part of the outcome
Mayo Clinic dietitian Romi Londre’s explanation lists possible effects including hunger, fatigue, insomnia, irritability, reduced concentration, nausea, constipation, and headaches. A schedule that produces an acceptable average result in a study may still be a poor fit for someone’s working day or wellbeing.
Those effects should not be relabelled as proof that a beneficial process has begun. The clinical question is whether the eating pattern is appropriate and tolerable, not how much discomfort someone can endure. Londre advises discussing fasting with the primary healthcare team beforehand, including concerns about other conditions and overall nutrition.
For the reader, this changes the standard of success. A useful conversation can include symptoms, daily function, and whether adequate meals remain practical, alongside any intended weight change. There is no requirement to choose the schedule with the most striking name. The modest comparative evidence leaves room for an approach that is less disruptive.
What this means in practice
- Check what a fasting claim actually describes: zero-calorie days and modified low-intake days are different interventions.
- When reading a result, identify the comparison group and follow-up period. Extra weight loss compared with another diet is different from total weight lost.
- If fasting is under consideration, bring a complete medication list to a clinician or pharmacist; do not change doses independently.
- Discuss alternatives with a dietitian if restriction affects concentration, mood, sleep, or the ability to eat adequately. Symptoms are relevant to the decision.
- If pregnant, breastfeeding, or living with a history of an eating disorder, avoid this approach and seek advice suited to your circumstances.
What we don’t know
We do not yet have a reliable basis for saying that alternate-day fasting extends human life or that its short-term weight advantage persists for years. A larger review improves the precision of some comparisons, but it cannot turn short follow-up into long-term evidence. The pooled findings also cannot resolve the individual safety questions facing an older person with several conditions.
There is also a gap between finding a tolerable schedule in a supported trial and maintaining it outside that setting. The available findings leave individual suitability unresolved. That uncertainty argues for careful discussion of goals and risks, rather than choosing a regimen because it sounds more intensive.
Alternate-day fasting belongs in an evidence-based discussion of dietary options. Its value depends on the benefit that can reasonably be expected, the burden of following it, and the person’s clinical circumstances. The calendar alone cannot answer those questions.
Photo: Alexey Demidov on Pexels.
