In 2018, intermittent fasting was the diet of the moment. Podcasts breathlessly announced metabolic benefits. Influencers posted 16:8 schedules with the certainty of converts. The framing implied that when you ate was as important as, or more important than, what or how much.
Eight years later, the evidence has settled. The picture that’s emerged is less exciting than the original pitch and more useful than the backlash.
Here is roughly where things stand.
What the evidence consistently shows
Three findings have replicated reliably across the major trials.
Intermittent fasting causes weight loss when it produces a calorie deficit. This is the central finding from Varady and colleagues’ work, the TREAT trial, the Trepanowski alternate-day-fasting trial, and the more recent matched-calorie comparisons. IF works. The mechanism, as best as the evidence supports, is that compressing eating into a smaller window or skipping days tends to reduce total intake, even without explicit calorie counting. The fasting itself isn’t separately doing magic; it’s making a deficit easier to maintain for some people.
Head-to-head against matched-calorie continuous restriction, IF performs about the same. This is the finding that punctured the original hype. When researchers control for total calorie intake — feeding two groups the same number of calories per week, one continuously and one in an IF pattern — the weight loss outcomes converge. The Liu et al. trial in 2022 was particularly clean on this point: 16:8 time-restricted eating combined with calorie restriction produced essentially the same weight loss as calorie restriction alone over 12 months.
Adherence is the differentiator. This is where IF still earns its place. For some people, “stop eating after 7 PM” is an easier rule to follow than “track and limit your calories every day.” For others, the opposite is true. The trial-level data shows comparable outcomes; the patient-level data shows wide variation in who finds which approach sustainable.
Where the evidence is mixed
A few claims that are not as settled as the popular discourse implies.
Metabolic health benefits independent of weight loss. The early enthusiasm for autophagy, insulin sensitivity improvements, and cardiometabolic gains beyond what weight loss alone would explain has not fully held up. There are studies showing modest improvements; there are well-conducted studies showing the effects largely disappear when you control for weight change. The honest summary: most of the metabolic benefits you’ll see from IF are mediated by the weight loss it produces. Independent effects, if they exist, are small.
The specific eating window. 16:8, 18:6, 20:4 — the protocols are quite different in feel and in adherence, but the outcome differences across windows are smaller than the protocol differences would imply. There isn’t strong evidence that 18:6 is meaningfully better than 16:8 for typical goals.
Effects on muscle mass during weight loss. The emerging concern, paralleling the GLP-1 conversation, is that IF protocols may make it harder to hit per-meal protein thresholds (25–40 g of complete protein per meal, 3–5 meals per day) that the muscle-protein-synthesis literature supports. A 16:8 window with two meals can work for protein distribution; a 20:4 window with one large meal probably can’t. People doing IF while also trying to maintain or build muscle should pay attention to per-meal protein math.
Where the evidence is clearer than the discourse implies
A few things that have settled but still get debated:
Breakfast is not magically necessary. The “breakfast is the most important meal” framing was always more cultural than scientific. Skipping breakfast as part of an IF protocol does not produce the negative metabolic outcomes the older literature implied — most of those studies were confounded by lifestyle correlates of breakfast skipping (irregular schedules, lower socioeconomic status, less overall dietary structure) rather than the meal omission itself.
Eating late at night is, however, slightly worse. The evidence for circadian effects on glucose handling is more consistent than the evidence for “always eat breakfast.” Late-night eating, especially of carbohydrate-heavy meals, produces somewhat worse postprandial glucose responses than the same meals eaten earlier in the day. For IF protocols, this argues for eating windows that finish earlier (e.g., 10 AM – 6 PM) rather than later (e.g., 2 PM – 10 PM), all else equal.
Hydration during the fasting window matters. Water, plain coffee, plain tea — fine. The “is black coffee allowed” debates were a distraction; the practical effect on outcomes is negligible.
Who should consider it (and who shouldn’t)
A reasonable summary of the patient-selection literature.
Likely to benefit: people who don’t eat breakfast naturally, people whose social and work schedules support a compressed eating window, people who find calorie counting psychologically taxing and prefer time-based rules, and people whose problematic eating tends to be late-evening snacking.
Likely to do equally well or better with continuous calorie restriction: people with consistent breakfast habits, athletes during heavy training blocks, people building or maintaining muscle who need 4+ feeding windows for protein distribution, and people whose food-and-mood relationship benefits from regular, predictable meals throughout the day.
Should probably avoid IF: people with active or recent disordered eating, people on medications that require food, pregnant or breastfeeding people, and people with diabetes who are not under close clinical supervision.
The cleanest synthesis in 2026
After the hype cycle, intermittent fasting is best understood as one of several strategies for managing calorie intake rather than as a separate metabolic intervention. It works for some people because it makes adherence easier. It doesn’t work better than calorie counting on average. It doesn’t unlock metabolic benefits beyond what weight loss alone provides. The specific window matters less than people think; the consistent calorie deficit matters more.
That’s a less exciting story than the 2018 version. It’s also a more useful one, because it tells you the right question to ask: not “is IF a magic intervention?” — it isn’t — but “is IF the calorie-management strategy that fits my life?” That question has different answers for different people, and the answer is the actual point.
