# Macro split for weight loss: what the trials show Canonical URL: https://www.hemed.com/blog/macro-split-for-weight-loss Plain-text URL: https://www.hemed.com/blog/macro-split-for-weight-loss/llms.txt Published: 2026-10-29 Modified: 2026-10-29 Author: The HeMed clinical team Topic: Living well Topic URL: https://www.hemed.com/blog/topic/living-well Tags: Macros, Keeping weight off Reading time: 16 min Publisher: HeMed (https://www.hemed.com/) ## Summary With calories matched, the macro split made little difference to weight loss at one and two years in the largest trials. Protein is the exception. ## Article With calories matched, the macro split makes little difference to weight loss. In a two-year trial of 811 adults, four reduced-calorie diets produced the same weight loss. They ranged from 35% to 65% carbohydrate and 20% to 40% fat. A 12-month trial of a healthy low-fat diet against a healthy low-carbohydrate diet found a 0.7 kg difference. That gap was not significant, meaning it was small enough to be chance. Protein is the exception. In pooled short trials, people on higher-protein diets lost slightly more fat and kept more lean mass at the same calories. Lean mass is everything in the body that is not fat, mainly muscle, bone and organs. Macros, short for macronutrients, are protein, carbohydrate and fat: the three nutrients that supply energy. This article sets out what the trials found, why protein stands out, and how to set a split you can keep. The figures are averages from groups of adults, not a prediction for any one man. ### What did the two large trials find? Two large trials found no meaningful difference between splits once calories fell. One ran for two years with four splits. The other ran for one year with two. Both were randomised controlled trials, where people are assigned to diets by chance so the groups start out alike. #### The two-year trial of four splits The POUNDS LOST trial randomised 811 overweight adults to one of four diets for two years. Men made up 36% of the group, and the average age was 51. Every diet cut 750 kcal a day from each person's estimated needs. Kcal is kilocalories, the calories shown on food labels. Diet target | Fat | Protein | Carbohydrate Low fat, average protein | 20% | 15% | 65% Low fat, high protein | 20% | 25% | 55% High fat, average protein | 40% | 15% | 45% High fat, high protein | 40% | 25% | 35% Percentages are shares of daily energy. At six months every group had lost about 6 kg, or 7% of starting weight. At two years the figures were 3.0 kg against 3.6 kg for 15% against 25% protein. Both fat groups were at 3.3 kg. For 65% against 35% carbohydrate the figures were 2.9 kg against 3.4 kg. None of those differences was statistically significant, meaning each gap was small enough to be chance. Hunger, fullness, cravings and satisfaction with the diet were also similar across the four groups. The authors concluded that reduced-calorie diets produce clinically meaningful weight loss regardless of which macronutrients they emphasise. #### What happened to blood fats in the two-year trial? The low-fat diets lowered LDL cholesterol more than the high-fat diets. The lowest-carbohydrate diet raised HDL cholesterol more than the highest-carbohydrate diet. All four diets lowered triglycerides by 12 to 17%. #### Low fat against low carbohydrate for a year The DIETFITS trial randomised 609 adults aged 18 to 50 to a healthy low-fat or a healthy low-carbohydrate diet for 12 months. About 43% were men. Nobody was given a calorie target. Both groups were told to eat plenty of vegetables and to cut added sugars, refined flours and trans fats. By 12 months the low-fat group was eating 48% carbohydrate and 29% fat. The low-carbohydrate group was eating 30% carbohydrate and 45% fat. Weight change was 5.3 kg lost against 6.0 kg lost. The 0.7 kg gap had a 95% confidence interval of 0.2 kg in favour of low fat to 1.6 kg in favour of low carbohydrate. A 95% confidence interval is the range within which the true average difference most likely sits. Because this one crossed zero, the trial could not say either diet was better. Individual results within each group ranged across about 40 kg, from roughly 30 kg lost to 10 kg gained. Both diets lowered blood pressure, insulin and glucose. LDL cholesterol favoured low fat; HDL cholesterol and triglycerides favoured low carbohydrate. ### What do the pooled analyses show? Pooled analyses agree: differences between splits are small at six months and smaller at a year. A meta-analysis pools the results of many trials into one estimate. One such analysis covered 48 trials with 7,286 adults. It found that low-carbohydrate and low-fat diets each produced about 7.3 kg of weight loss at 12 months compared with no diet. At six months that analysis put low-carbohydrate diets at 8.73 kg and low-fat diets at 7.99 kg against no diet. Its authors wrote that weight loss differences between named diets were small. They supported recommending any diet a person will stick to. A larger analysis covered 121 trials and 21,942 adults. At six months, low-carbohydrate and low-fat patterns produced 4.63 kg and 4.40 kg more weight loss than a usual diet. By 12 months the effects on weight had largely disappeared, at about 3 kg against a usual diet. The authors called the differences between diets trivial to small. They also noted that the trials rarely reported adherence, which may have been low by 12 months. ### Why does protein stand out? Because protein is the one macronutrient with a measurable, if modest, effect at matched calories. A meta-analysis of 24 trials with 1,063 adults compared higher-protein with standard-protein diets at the same calories and the same fat. Over about 12 weeks on average, the higher-protein groups lost 0.79 kg more weight and 0.87 kg more fat mass. They kept 0.43 kg more fat-free mass, the trial term for lean mass. The authors called the benefits modest. A 2015 review of the evidence pointed to 1.2 to 1.6 g of protein per kilogram of body weight a day during weight loss. It also pointed to at least about 25 to 30 g at each meal. That range comes from trials, not from UK guidance. The UK reference nutrient intake is 0.75 g per kilogram, which is 55.5 g a day for a man aged 19 to 64. #### What protein does not do Protein does not switch off appetite. The same review found a modest satiety effect, with greater fullness after higher-protein meals. Satiety is the feeling of fullness after eating. Even so, the review did not find evidence of less eating at the next meal. Protein does not boost metabolism by much. Higher-protein diets held resting energy expenditure (the energy your body uses at rest) 595.5 kJ a day higher, about 140 kcal. The authors called that difference modest. Protein does have the highest thermic effect of the three macronutrients. But the thermic effect of food overall is only about 10% of energy intake. The thermic effect is the energy used to digest and process a meal. For a calculator that works the trial range out from your body weight, see how much protein you need to lose weight. ### Does cutting carbohydrate burn more fat at the same calories? No. Researchers pooled 32 controlled feeding studies, where every meal was provided. Swapping carbohydrate for fat at equal calories and equal protein did not favour the low-carbohydrate diets. Energy expenditure was 26 kcal a day higher on the lower-fat diets, and fat loss was 16 g a day greater. The authors called those differences so small as to be physiologically meaningless. They wrote that for body fat and energy use, a calorie is a calorie for all practical purposes. #### What about the insulin theory? The carbohydrate-insulin model proposes that carbohydrate raises insulin, which pushes energy into fat stores and slows metabolism. Its supporters have predicted a metabolic advantage of as much as 400 to 600 kcal a day for very low carbohydrate diets. The feeding studies ran in the opposite direction, and the review authors wrote that the evidence does not support such an advantage. DIETFITS tested a related idea. People with higher insulin secretion at the start did not do better on the low-carbohydrate diet. A three-gene pattern did not predict the better diet either. The model is still debated, so treat it as contested rather than settled. Hall and Guo add that low-carbohydrate diets often raise protein intake, and that ketones may reduce appetite. Those effects may explain their short-term results in free-living trials. ### What about ketogenic diets? At a year or more, the advantage is under a kilogram and uncertain. A ketogenic diet limits carbohydrate to 50 g a day or less, so the body runs more on fat and on ketones made from fat. A meta-analysis of 13 trials compared a ketogenic diet with an energy-restricted low-fat diet over 12 months or longer. The ketogenic groups lost 0.91 kg more, with a 95% confidence interval of 0.17 to 1.65 kg. In a sensitivity analysis the difference fell to 0.73 kg and was no longer significant. A sensitivity analysis repeats the calculation under different assumptions to test how stable the result is. LDL cholesterol rose more on the ketogenic diets, by 0.12 mmol/L. HDL cholesterol rose 0.09 mmol/L more and triglycerides fell more. LDL is the cholesterol linked to heart disease and HDL is the kind often called good cholesterol. The unit mmol/L is the one UK blood tests use. Triglycerides are a type of fat carried in the blood. The authors also noted low adherence to the ketogenic diet. They added that carbohydrate intake in most studies had crept above the protocol by the end. ### Does the quality of carbohydrate and fat matter? Yes, for health and for appetite, even though the ratio matters little for the scales. Both large trials that found no difference between splits required a healthy pattern. POUNDS LOST required at least 20 g of fibre a day and capped saturated fat at 8% of energy. DIETFITS told both groups to minimise added sugars and refined flours. #### Carbohydrate: free sugars and fibre The Scientific Advisory Committee on Nutrition (SACN) advises that free sugars should not exceed 5% of total energy. Free sugars are sugars added to food or drink, plus those in honey, syrups and fruit juice. SACN also advises that adults should eat 30 g of fibre a day. SACN found that higher sugar intakes push up total energy intake. It also found that cutting free sugars can be part of eating less for people who are overweight. SACN kept the reference value for total carbohydrate at about 50% of energy. At the levels recommended in the UK diet, it found no association between total carbohydrate intake and cardiovascular disease, type 2 diabetes or colorectal cancer. The NHS adds that starchy foods carry fewer than half the calories of fat, gram for gram, and that wholegrain versions help you feel full. #### Fat: the saturated share UK guidance keeps total fat at no more than 35% of food energy and saturated fat at no more than 11%. SACN expresses the saturated figure as no more than about 10% of total energy, which includes alcohol. For a man aged 19 to 64, on the PHE basis, that is less than 31 g of saturated fat a day at 2,500 kcal. SACN recommends replacing saturated fat with unsaturated fat rather than with carbohydrate. In cohort studies, which follow groups of people over years, swapping saturated fat for carbohydrate went with more coronary events such as heart attacks. Fat carries 9 kcal per gram against 4 kcal for carbohydrate and protein. Even so, lower-fat foods are not always lower in calories, because the fat is sometimes replaced with sugar. ### Why does sticking to the plan matter more than the ratio? Because the trials saw weight loss track attendance and adherence, not the split. In POUNDS LOST, every counselling session attended went with 0.2 kg more weight lost at two years on every diet. People who attended two thirds of the sessions lost about 9 kg. On average, people did not hold their assigned split. Despite intensive counselling, reported intakes never reached the targets and drifted back towards habit over two years. The authors read the results as pointing to behaviour rather than macronutrient metabolism. In DIETFITS, both groups cut about 500 to 600 kcal a day without being given a calorie target. They were told which foods to cut, not how many calories to eat. No trial randomised people to adhere or not, so this is what the trials observed, not a proven cause. ### What split should a man losing weight use? Set protein first by body weight, keep fat within the UK ceiling, and let carbohydrate fill the rest from higher-fibre foods. That is the split the evidence supports, and it leaves the ratio to your preference. - Protein: 1.2 to 1.6 g per kilogram of body weight a day, the range used in weight-loss trials. - Fat: no more than 35% of food energy, with saturated fat no more than 11% and the rest unsaturated. - Carbohydrate: the remaining calories, with free sugars under 5% and 30 g of fibre a day. Keeping muscle also takes strength training, which UK guidance recommends on at least two days a week. #### A worked example Take an 85 kg man eating 2,000 kcal a day. His protein range is 102 to 136 g: 85 multiplied by 1.2 and by 1.6. At 4 kcal per gram that is about 20% to 27% of his energy. Fat at the 35% ceiling is 700 kcal, or about 78 g at 9 kcal per gram. Carbohydrate takes what is left: about 190 to 220 g a day. The arithmetic is ours, applied to the cited figures. If he prefers less fat, carbohydrate rises. If he prefers fewer carbohydrates, fat can take their place up to the ceiling. Either way the trials found a similar result, as long as the calories and the protein held. Macro calculator Enter your age, height, weight and activity level for an estimated daily energy figure and the macros that go with it. Work out your macros. A split only works if it fits your meals, your budget and your tastes. The authors of the two-year trial made the same point: diets tailored to personal and cultural preferences may have the best chance of lasting. ### What does Diabetes UK say about low carbohydrate in type 2 diabetes? A low-carbohydrate diet can be a short-term option, but not a proven long-term winner. Diabetes UK, working with SACN, recommends lower-carbohydrate diets as a short-term effective option for adults with type 2 diabetes who have obesity or overweight. The statement covers diets providing 50 to 130 g of carbohydrate a day. The same statement says there is no consistent evidence that a low-carbohydrate diet is more effective than other approaches in the long term. It should not be seen as the diet for everyone. Diabetes UK does not recommend low-carbohydrate diets for people with type 1 diabetes. This applies to people with type 2 diabetes. If that is you, talk to your diabetes team before changing your carbohydrate intake. Medicines may need adjusting, and your team may want you to check your blood sugar more often. ### Frequently asked questions **Is 40/30/30 the best macro split for weight loss?** No trial has shown it to be the best. The two-year trial found no effect of protein between 15% and 25%, fat between 20% and 40%, or carbohydrate between 35% and 65%. Pooled analyses of named diets found differences that were small at six months and negligible at a year. A 40/30/30 split is a reasonable pattern if you like it. It is not better than the others at the same calories. **Do men need a different macro split from women?** Neither large trial reported a split that worked better for men. Men were 36% of the two-year trial and about 43% of the one-year trial. The results are given for everyone together. What differs is the amount. A protein target set per kilogram scales with body weight. Government figures put a man's energy needs at about 2,500 kcal against 2,000 for a woman. **Should I count macros or calories?** The trials cut calories, and the split within them made little difference. If you track one thing, make it protein. It is the macronutrient with evidence for keeping lean mass and for fullness. Counting macros covers calories anyway, because each gram of protein, carbohydrate and fat has a known energy value. A protein target per meal is one number to keep rather than three. **What happens in the first months on a low-carbohydrate diet?** Slightly more weight loss, on average. At six months the pooled analyses put low-carbohydrate patterns 0.2 to 0.7 kg ahead of low-fat patterns. Hall and Guo suggest two reasons: these diets often raise protein intake, and ketones may blunt appetite. By 12 months the difference has shrunk to a kilogram or less in the pooled trials. Adherence to very low carbohydrate intakes was low. Early speed is not the same as a better year. **What does a higher-protein split do for muscle?** A higher-protein split helps. In 24 pooled trials, higher-protein groups kept 0.43 kg more fat-free mass than standard-protein groups at the same calories. The authors called the benefit modest. Protein supports muscle. It does not replace strength training, which UK guidance recommends on at least two days a week. **Does the split change my cholesterol?** Yes, in a broadly consistent pattern. In both large trials, LDL cholesterol changes favoured the lower-fat diets, and HDL cholesterol rose more on the lowest-carbohydrate diets. In DIETFITS, triglycerides fell more on the low-carbohydrate diet, and blood pressure, insulin and glucose improved on both diets. In POUNDS LOST, all four diets lowered triglycerides similarly, by 12 to 17%. Saturated fat raises LDL, and UK guidance is to keep it to about 10% to 11% of energy and replace it with unsaturated fat. HeMed provides clinician-led weight management. This page is general health information, not medical advice. If you are worried about symptoms, speak to your GP, or use NHS 111. Next step: see what the HeMed blood test measures. Sources (16) - Sacks FM and colleagues. "Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates" (the POUNDS LOST trial). New England Journal of Medicine 2009;360:859 to 873. Full text of the author manuscript. pmc.ncbi.nlm.nih.gov/articles/PMC2763382. Accessed 28 October 2026. - Gardner CD and colleagues. "Effect of low-fat vs low-carbohydrate diet on 12-month weight loss in overweight adults and the association with genotype pattern or insulin secretion: the DIETFITS randomized clinical trial". JAMA 2018;319(7):667 to 679. Full text. pmc.ncbi.nlm.nih.gov/articles/PMC5839290. Accessed 28 October 2026. - Wycherley TP and colleagues. "Effects of energy-restricted high-protein, low-fat compared with standard-protein, low-fat diets: a meta-analysis of randomized controlled trials". American Journal of Clinical Nutrition 2012;96(6):1281 to 1298. Abstract. doi.org/10.3945/ajcn.112.044321. Accessed 28 October 2026. - Johnston BC and colleagues. "Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis". JAMA 2014;312(9):923 to 933. Abstract. doi.org/10.1001/jama.2014.10397. Accessed 28 October 2026. - Ge L and colleagues. "Comparison of dietary macronutrient patterns of 14 popular named dietary programmes for weight and cardiovascular risk factor reduction in adults: systematic review and network meta-analysis of randomised trials". BMJ 2020;369:m696, with the correction BMJ 2020;370:m3095 (the six-month low-fat figure of 4.40 kg). Full text. pmc.ncbi.nlm.nih.gov/articles/PMC7190064. Accessed 28 October 2026. - Leidy HJ and colleagues. "The role of protein in weight loss and maintenance". American Journal of Clinical Nutrition 2015;101(6):1320S to 1329S. Abstract. doi.org/10.3945/ajcn.114.084038. Accessed 28 October 2026. - Scientific Advisory Committee on Nutrition (SACN), "Statement on nutrition and older adults living in the community", 20 January 2021, paragraphs 3.2 and 4.22 to 4.24 (the 0.75 g per kg reference nutrient intake). gov.uk/government/publications/sacn-statement-on-nutrition-and-older-adults. Accessed 28 October 2026. - Public Health England, "Government Dietary Recommendations: energy and nutrients for males and females aged 1 to 18 years and 19+ years", 2016, Table 2 (page 7) and footnotes. assets.publishing.service.gov.uk, government_dietary_recommendations.pdf. Accessed 28 October 2026. - Hall KD, Guo J. "Obesity energetics: body weight regulation and the effects of diet composition". Gastroenterology 2017;152(7):1718 to 1727. Full text of the author manuscript. pmc.ncbi.nlm.nih.gov/articles/PMC5568065. Accessed 28 October 2026. - Bueno NB and colleagues. "Very-low-carbohydrate ketogenic diet v. low-fat diet for long-term weight loss: a meta-analysis of randomised controlled trials". British Journal of Nutrition 2013;110(7):1178 to 1187. Full text. doi.org/10.1017/S0007114513000548. Accessed 28 October 2026. - Scientific Advisory Committee on Nutrition (SACN), "Carbohydrates and Health", 2015, Foreword, Summary paragraphs S.22 and S.27, and paragraphs 11.4, 11.16 and 12.24. assets.publishing.service.gov.uk, SACN_Carbohydrates_and_Health.pdf. Accessed 28 October 2026. - NHS, "Starchy foods and carbohydrates", page last reviewed 15 March 2023. nhs.uk/live-well/eat-well/food-types/starchy-foods-and-carbohydrates. Accessed 28 October 2026. - Scientific Advisory Committee on Nutrition (SACN), "Saturated fats and health", July 2019, Summary paragraphs S.3, S.20 and S.24, and chapter 1 footnote 3. assets.publishing.service.gov.uk, SACN_report_on_saturated_fat_and_health.pdf. Accessed 28 October 2026. - NHS, "Fat: the facts", page last reviewed 14 April 2023. nhs.uk/live-well/eat-well/food-types/different-fats-nutrition. Accessed 28 October 2026. - NHS, "Physical activity guidelines for adults aged 19 to 64", page last reviewed 22 May 2024. nhs.uk/live-well/exercise/physical-activity-guidelines-for-adults-aged-19-to-64. Accessed 28 October 2026. - Diabetes UK, "Low carb diets for people with diabetes" (position statement for professionals, May 2021) and "Low-carb diet and meal plan". diabetes.org.uk, low-carb-diets-for-people-with-diabetes and diabetes.org.uk/living-with-diabetes/eating/meal-plans/low-carb. Accessed 28 October 2026. Numbered markers in the page text link to these entries. Statistics describe populations, not any one person’s results. Published 29 October 2026. --- This is the plain-text version of the article at https://www.hemed.com/blog/macro-split-for-weight-loss, generated from the page on 2026-10-29. It is for information only and is not medical advice. Whether any treatment is right for a person is a clinical decision. Statistics describe populations, not any one person's results.