24-Hour Fasting: What the Evidence Shows

What Is a 24-Hour Fast, and How Is It Different From 5:2 or Alternate-Day Fasting?

A 24-hour fast, sometimes called Eat-Stop-Eat, is a complete, zero-calorie fast lasting a full day, done once or twice a week. It is not the 5:2 diet, which allows a reduced-calorie meal (commonly about 500 to 600 kcal) on fast days, and it is not alternate-day fasting, which repeats every other day rather than once or twice weekly.

These three schedules are tested in separate trials and should not be read as interchangeable. See the 5:2 diet for the reduced-calorie version and alternate-day fasting for the every-other-day version; this page covers only the complete, once- or twice-weekly schedule and its own dedicated trial evidence. "Eat-Stop-Eat" is a popular-book term (Brad Pilon's book of that name), not a named protocol tested in any trial cited here; it is used on this page only as a reader-recognition aid, never as a label for a specific study's finding.

Does a 24-Hour Fast, Done Weekly, Improve Your Health Markers?

Before looking at what this schedule's trial evidence shows: nearly all of the dedicated research on this exact protocol comes from one trial cohort, run by one research group. The WONDERFUL trial (Bartholomew et al., 2021) [1] and four secondary analyses of the same trial cohort, each using the 67 to 71 participants who completed it and had the relevant measurements (citations 2 through 5 below), share overlapping authorship, with Benjamin D. Horne an author on every one, and funding and advisory ties disclosed paper by paper in Authors and Funding below. The one citation on this page from outside that trial, a Brigham Young University study of how a pre-fast meal affects a 24-hour fast (citation 7), is not fully separate from it either: its senior author, Bruce W. Bailey, is also a co-author of the WONDERFUL trial paper. This does not disqualify the findings; it means they describe one well-documented research program's single cohort, not a broad literature, and should be read that way.

In the WONDERFUL trial itself, 103 adults aged 21 to 70, about two-thirds women, with modestly elevated LDL cholesterol and at least one metabolic-syndrome feature, not taking statins or diabetes medication, were randomized to a 24-hour water-only fast twice weekly for 4 weeks, then once weekly for 22 weeks, or to ad-libitum eating; 71 completed the 26-week trial. The fasting schedule produced no significant change in LDL cholesterol, the trial's pre-specified primary outcome, but did significantly improve insulin resistance (HOMA-IR) and a composite metabolic-syndrome score compared with controls. Weight (-1.7 vs. 0.2 kg, P=0.06), a general cognitive score, and brain-derived neurotrophic factor (BDNF), a protein involved in nerve-cell growth and survival, were not significantly different between groups (Bartholomew et al., 2021) [1].

A secondary analysis of the same trial, in the 68 participants with growth-hormone measurements available, found that people with lower baseline growth hormone had a significantly larger 26-week HOMA-IR improvement on the fasting schedule than people with higher baseline growth hormone (HOMA-IR change -1.04 vs. 0.60 with lower baseline growth hormone, p=0.003; -0.69 vs. -0.42 with higher baseline growth hormone, p=0.26; p-interaction=0.004). In this same analysis, weight loss was not significantly different between the fasting and control groups (-1.74 kg vs. 0.21 kg, p=0.08), and weight loss did not correlate with the insulin or glucose changes the analysis measured (Horne et al., 2024) [2].

A separate post hoc subgroup analysis of the same 71 completers found fasting lowered insulin more in people at or above the group's median red cell distribution width (RDW-SD of 42.6 fL or higher) than below it (insulin change -3.45 vs. 0.48 mIU/L at or above the median, -1.99 vs. -1.08 below it; p-interaction=0.039), with a similar but weaker pattern for glucose, HOMA-IR, and the metabolic-syndrome score; RDW-SD itself did not change significantly with fasting (p=0.34). This is an exploratory subgroup finding, not a general result, and the paper's authors report that funders had no role in the study's design, analysis, or the decision to publish (Horne et al., 2021) [4].

Does a 24-Hour Fast Affect Long-Term Disease or Mortality Risk?

A secondary analysis of the WONDERFUL trial found this schedule increased a 1-year mortality-risk score while significantly decreasing one 10-year chronic-disease-risk score, with a non-significant trend toward a lower second one, over 26 weeks. The trial's own authors read the short-term increase as a physiological stress response that may drive the longer-term benefit, and also as a possible short-term safety concern; it is a risk-score change, not a measured mortality outcome, and both directions are stated here together, not separately.

In this analysis of the same 71 completers, the fasting group's 1-year Intermountain Mortality Risk Score (IMRS) rose relative to controls (0.78 vs. -0.61, p=0.010), while the 10-year Intermountain Chronic Disease score (ICHRON) fell significantly more in the fasting group (-0.92 vs. 0.58, p=0.035), with a non-significant trend toward a lower Pooled Cohort Risk Equations score as well (p=0.054); HOMA-IR fell more in participants with a higher baseline IMRS. This analysis describes disease-free adults: the WONDERFUL trial excluded people with existing chronic disease, and its own discussion names "advanced comorbidities, frailty, or uncontrolled metabolic disease" as groups for whom the stress of fasting "may be too great." The paper's own authors interpret the short-term IMRS increase as evidence that fasting "may elevate short-term mortality risk as a central trigger for myriad physiological responses that elicit long-term health improvements," adding that it "may also reveal short-term fasting-induced safety concerns" (Horne et al., 2023) [3]. HOMA-IR is a calculated index of insulin resistance, and IMRS, PCRE, and ICHRON are risk scores calculated from blood tests and clinical data; none is a diagnosis, and this finding describes a change in a risk score, not a measured case of disease or death in either group. Anyone with cardiovascular disease, or taking a medication that affects heart rate or rhythm, should talk to a clinician before trying this schedule, given this finding; see Who Should Not Try a 24-Hour Fast Without a Clinician below.

A Possible Biological Mechanism (and Its Limits)

One secondary analysis points to a possible biological mechanism behind this schedule's metabolic effects, and the same research group's principal investigator has also written a commentary on the wider question of weight-loss-independent benefits; both carry important limits.

In a secondary analysis of 67 completers with galectin-3 results available, the fasting schedule significantly increased galectin-3, a protein involved in inflammation and tissue remodeling, compared with controls (median change 0.793 vs. -0.332 ng/mL, p=0.021); the increase correlated inversely with the improvement in HOMA-IR (r=-0.288, p=0.018), with a similar but non-significant pattern for the metabolic-syndrome score (p=0.052); the other heart-failure biomarkers the analysis measured did not change with fasting. The paper's own authors label their explanation speculative: they suggest the galectin-3 rise may be "an evolutionary adaptive survival response." This is the authors' own interpretation, not an established mechanism, and should be read that way (Horne et al., 2022) [5].

In a short News & Views commentary in Nature Metabolism, not a new trial or a systematic review, the WONDERFUL trial's principal investigator discusses a separate 7-day water-only fasting study (Pietzner et al., 2024, cited in full on water fasting) under the title "The weight-loss-independent benefits of fasting", and cites the WONDERFUL trial among its references (Horne, 2024) [6]. It is the same research group's own framing, offered as a commentary on another group's study, and is cited here as that, not as confirmation from outside the group; its full text is behind a paywall and only its summary was read for this page.

What Happens If You Only Fast Once (Not on a Repeating Schedule)?

A single 24-hour water-only fast, done once rather than on a repeating schedule, is a different question with its own trial evidence. In a randomized crossover trial of 30 apparently healthy volunteers, a single fast significantly raised growth hormone and some lipid markers and lowered triglycerides and weight compared with a day of normal eating, and in the 16 participants who fasted first most of these changes were back to baseline the following day except weight (Horne et al., 2013, cited in full on water fasting).

This single-fast evidence is fully covered on a single 24-hour water-only fast under "24 and 36 Hour Fasts" and is not re-explained here. A secondary analysis of that same trial, examining the growth-hormone mechanism (Horne et al., 2025), is fully covered on prolonged fasting. That trial comes from the same Horne-led research group as the sections above, but it is a different trial testing a different question: a one-time fast rather than a repeating weekly schedule.

What to Expect When You Start and End the Fast

In a randomized crossover trial in 24 older, sedentary adults with overweight or obesity, run at Brigham Young University by a group that shares a senior author with the WONDERFUL trial paper, a lower-carbohydrate, higher-fat pre-fast shake brought participants into nutritional ketosis (a blood beta-hydroxybutyrate level at or above 0.5 mmol/L) at about 12 hours into a 24-hour fast on average; a higher-carbohydrate, lower-fat shake did not reach that threshold during the fast on average. One hour after the shake, before the fast had begun, insulin was lower and glucagon, GLP-1, and GIP were higher after the lower-carbohydrate shake, and glucose ran higher after the higher-carbohydrate shake for roughly the first two hours (Gipson et al., 2025) [7]. This is a measured timing and hormone-kinetics finding about how a pre-fast meal affects the fast that follows, not a claim that reaching ketosis faster is itself a health benefit, and this page does not set a specific macronutrient target.

How you end a 24-hour fast matters: start with something modest rather than a large meal, and avoid pairing a big carbohydrate load with a long gap since your last meal. At exactly 24 hours, this sits at the boundary of this site's guidance for shorter and longer fasts, so follow both rather than picking one: see how to break a fast safely for the full detail on breaking a 16-to-24-hour fast and a 24-plus-hour fast.

If you are fasting for a full day, see electrolyte guidance for a daily fast rather than guessing at amounts; this page does not repeat specific dosing.

Who Should Not Try a 24-Hour Fast Without a Clinician?

Talk to a clinician before trying a 24-hour fast if you are pregnant, under 18, have a history of an eating disorder, manage diabetes or blood pressure with medication, or manage any chronic condition that requires food or medication on a consistent schedule.

Anyone taking insulin or a sulfonylurea should talk to a clinician first: a full day without food raises the risk of low blood sugar specifically on these medications. No trial cited on this page measured hypoglycemic events directly, so this is a medication caution, not a trial finding. Anyone with cardiovascular disease, or taking a medication that affects heart rate or rhythm, should also talk to a clinician before trying this schedule, given the mortality-risk-score finding described above; that finding is a reason for clinician guidance, not evidence that the schedule is unsafe for everyone.

See choosing a schedule for help deciding between this and other fasting patterns.

Safety and Cautions

This is general information, not medical advice. The research on this exact schedule comes mostly from one trial cohort studied by one research group, as described above, and even its own positive findings are limited to specific markers, not a broad set of health outcomes. Stop and seek care for any severe or concerning symptom during a fast.

Authors and Funding of the Cited Studies

• Bartholomew et al., 2021 [1]: Funding: the Intermountain Research and Medical Foundation (grant 759; Dell Loy Hansen Heart Foundation philanthropy). Disclosure: "B.D.H. is PI of other intermittent fasting grants from the Intermountain Research and Medical Foundation and a member of the scientific advisory board of Sylfph."
• Horne et al., 2024 [2]: Disclosure: authors declare no conflicts directly related to this study; outside this work, B.D.H. is a member of the advisory board of Unleash Health, previously consulted for Pfizer on risk scores (funds paid to Intermountain), and is site PI of a Task Force for Global Health grant; K.U.K. and B.D.H. are site PIs of PCORI and NIH RECOVER grants. Funding: the Intermountain Research and Medical Foundation (grant 759) and a supplemental grant through the philanthropy of the Dell Loy Hansen Heart Foundation; the paper states the funders had no role in design, analysis, writing, or the decision to publish.
• Horne et al., 2023 [3]: Disclosure: B.D.H., H.T.M., and J.L.A. are inventors of risk scores (including IMRS; not ICHRON) licensed by Intermountain to Alluceo and CareCentra; B.D.H. is an advisory-board member of Opsis Health and Lab Me Analytics, a consultant to Pfizer on risk scores (funds paid to Intermountain), PI of a risk-prediction grant from AstraZeneca, and site PI of a Task Force for Global Health grant; K.U.K. and B.D.H. are site PIs of PCORI and NIH RECOVER grants. Funding: two grants from the Intermountain Research and Medical Foundation (grant 759 and a supplemental grant through the philanthropy of the Dell Loy Hansen Heart Foundation); the paper states the funders had no role in design, analysis, writing, or the decision to publish.
• Horne et al., 2021 [4]: Disclosure: B.D.H., H.T.M., and J.L.A. are inventors of RDW-based clinical decision tools licensed to CareCentra and Alluceo; B.D.H. is PI of grants (Intermountain Research and Medical Foundation, CareCentra, GlaxoSmithKline, AstraZeneca) involving those tools and of other intermittent-fasting grants from the Intermountain Research and Medical Foundation; B.D.H. is on the scientific advisory boards of LabMe and Opsis Health; K.U.K. is PI of, and B.D.H. co-investigator on, a PCORI grant. Funding: the Intermountain Research and Medical Foundation (grant 759) and a supplemental grant through the philanthropy of the Dell Loy Hansen Heart Foundation. The paper states funders had no role in study design, analysis, or the decision to publish.
• Horne et al., 2022 [5]: Disclosure: B.D.H. is PI of other intermittent-fasting grants from the Intermountain Research and Medical Foundation and a scientific advisory board member of Opsis Health and Lab Me Analytics.
• Horne, 2024 [6]: Single-author News & Views commentary, not a trial or systematic review. Disclosure: "The author declares no competing interests." Funding: not stated in the available record.
• Gipson et al., 2025 [7]: Disclosure: "The authors declare no competing interests." Funding: a gerontology research grant from Brigham Young University and the National Center for Advancing Translational Sciences of the National Institutes of Health (award TL1TR002368).
• Disclosures are listed where the full text of the paper was available to us; a study not listed here is not a statement that its authors have no ties.