Intermittent fasting: honest review of the research

Intermittent fasting can help some people lose weight and improve metabolic markers, but most of the benefit appears to come from eating less overall rather than a magic fasting switch. The evidence is strongest for modest weight loss in people who can stick to the pattern, and weaker for claims about longevity, detox, hormones or "metabolic reset". The honest question is whether fasting helps you eat well consistently without creating risk.

What counts as intermittent fasting?

Intermittent fasting is an umbrella term. Time-restricted eating limits food to a daily window, such as 10am to 6pm. The 5:2 approach uses 2 low-calorie days each week and 5 more usual days. Alternate-day fasting alternates very low-calorie or fasting days with eating days. Longer fasts, such as 24 to 72 hours, are a separate risk category and should not be treated as a casual wellness challenge.

This matters because research on one pattern cannot be applied to all fasting. A 10-hour eating window is not the same as 36-hour alternate-day fasting. Early time-restricted eating may affect glucose and appetite differently from skipping breakfast and eating late at night. A plan that is easy for one person may worsen binge eating, migraines, reflux, sleep or training in another.

Do not confuse fasting with diet quality. A short eating window filled with ultra-processed food is still a poor pattern. A normal eating schedule with enough protein, fibre, fruit, vegetables, unsaturated fats and fewer snacks may outperform fasting if it is easier to sustain.

What the weight-loss evidence says

The most consistent finding is modest weight loss when fasting helps people reduce energy intake. The 2025 BMJ network meta-analysis included randomised clinical trials of intermittent fasting strategies and concluded that alternate-day fasting, time-restricted eating and whole-day fasting reduced weight compared with unrestricted eating, while differences compared with continuous energy restriction were usually small or uncertain.¹

The 2022 New England Journal of Medicine trial is a useful reality check. Adults with obesity were assigned to calorie restriction with or without an 8-hour eating window. After 12 months, time-restricted eating did not produce significantly greater weight loss than calorie restriction alone.² In other words, the window was not a magic multiplier when calories were already controlled.

Alternate-day fasting has also been tested directly. A 2017 JAMA Internal Medicine randomised trial found alternate-day fasting was not superior to daily calorie restriction for weight loss, weight maintenance or cardiometabolic risk indicators after one year, and dropout was higher in the alternate-day fasting group.³

The practical conclusion is not "fasting does not work". It is more specific: fasting can work if it creates a calorie deficit you can maintain, but it is not reliably better than continuous restriction. If fasting makes you feel in control, eat less late at night and stop grazing, it may help. If it leads to rebound eating, poor training, irritability or obsession, it is the wrong tool.

Metabolic health, diabetes and cholesterol

Fasting can improve glucose, blood pressure, triglycerides or waist size in some studies, but these changes often track weight loss. If two diets produce the same weight loss, the metabolic difference is usually much smaller than marketing suggests. Claims about insulin "healing" after a few fasts are too simplistic.

There are plausible mechanisms. A 2019 New England Journal of Medicine review discussed metabolic switching, circadian biology and cellular stress responses as possible pathways.⁴ Plausible is not the same as proven clinical advantage for every person. Human outcomes depend on the whole pattern: food quality, energy intake, sleep, movement, medications and adherence.

For type 2 diabetes, fasting must be treated more carefully. Diabetes UK warns that fasting can increase the risk of hypoglycaemia for people taking insulin or certain tablets, and medication may need adjustment before fasting.⁶ Do not start intermittent fasting if you use insulin, sulfonylureas or other glucose-lowering drugs without clinical advice.

Lipids and blood pressure also need context. If fasting leads to weight loss, less alcohol and less late-night snacking, markers may improve. If it leads to large portions of saturated fat, low fibre and poor training recovery, markers may not. Use the stack builder to track supplements and medicines, the health library to understand metabolic conditions, insights to challenge fasting claims, and Start Here to record weight, waist, sleep and symptoms.

Safety and who should avoid fasting

Fasting is not automatically safe because it is "natural". Avoid self-directed fasting if you are pregnant or breastfeeding, under 18, underweight, frail, recovering from surgery or illness, or have a current or past eating disorder. NHS information describes eating disorders as serious mental health conditions involving unhealthy eating behaviours that can affect physical health.⁷ A fasting plan can easily become a socially approved form of restriction.

Get medical advice first if you have diabetes, kidney disease, gout, migraine, reflux, gallstones, adrenal or thyroid disease, severe mental health symptoms, a history of fainting, or if you take blood pressure medicine, diuretics, anticoagulants, lithium, insulin or glucose-lowering tablets. Fasting can change hydration, blood pressure, glucose, medicine timing and side effects.

Pregnancy is a separate issue. NHS pregnancy nutrition guidance focuses on a healthy, balanced diet and avoiding certain foods, not weight-loss fasting.⁸ If appetite, nausea or gestational diabetes makes eating difficult, get personalised maternity advice rather than following an online fasting protocol.

How to try it sensibly

If fasting is suitable for you, start gently. A 12-hour overnight fast, such as 8pm to 8am, is mostly a routine reset. If that is easy, try 14 hours before considering 16. Keep protein high enough, include fibre, drink fluids, and avoid turning the eating window into a processed-food sprint.

Keep the wider weight-loss basics in place. NHS healthy-weight guidance focuses on sustainable eating patterns, activity and realistic goals rather than extreme rules.⁵ Fasting should support those basics, not replace them with a timer-only diet.

Choose the behaviour you are trying to change. If the problem is evening snacking, set a kitchen closing time. If the problem is grazing, use three structured meals. If the problem is total calories, fasting may help, but so can smaller portions. If the problem is poor diet quality, fasting is a distraction.

Track the right outcomes for 4 weeks: weight trend, waist, hunger, energy, sleep, training, mood, binge urges, menstrual changes, reflux, headaches and social life. If fasting improves one metric while damaging several others, it is not working as a long-term strategy.

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.