In short
- Diet is the single largest modifiable cause of early death worldwide, and the risks that matter most are too much salt, sugar-sweetened drinks and processed meat and too little whole grain, fruit, vegetables, nuts and fibre.
- For weight, the balance of calories in and out decides the result and the named diet matters far less than whether you can keep it; people underestimate what they eat by close to half.
- Protein is the nutrient with the clearest target: around one point six grams per kilogram a day, spread across meals, if you train; less is fine if you do not.
- Ultra-processed food makes people eat about five hundred calories a day more without noticing, so the shape of the food matters as much as its label.
- Writing down what you eat is the best-evidenced single behaviour for changing it, which is what the Log button is for.
Fundamentals
The patterns the big studies agree on, and the two numbers to know.
Three things to hold onto: a pattern of food that the large studies keep finding protective, the arithmetic of energy that decides weight whatever the pattern is called, and the one nutrient with a number attached. Everything later in the book is a way of acting on those three.
The pattern that keeps coming up
Across countries and decades the same pattern protects against heart disease, diabetes and early death: mostly plants, whole grains, nuts, legumes, fish, olive oil and little processed meat, refined grain or sugary drink. The Global Burden of Disease analysis attributed around eleven million deaths a year to diet, with too much sodium and too little whole grain and fruit at the top. In a randomised trial, a Mediterranean pattern with extra olive oil or nuts cut major cardiovascular events by about thirty per cent. Eight hundred grams a day of fruit and vegetables, roughly ten portions, tracked with the lowest risk in the pooled cohorts, and each portion up to five brought a clear reduction.
Energy: the arithmetic that decides weight
Body weight follows the difference between energy eaten and energy spent. In the head-to-head trials, low-fat, low-carbohydrate and high-protein diets produced the same weight loss when calories matched, and the named diet programmes differed by a kilogram or two at most; adherence explained the results. The catch is that people are poor judges of intake: in a careful study, people who believed they could not lose weight were under-reporting what they ate by forty-seven per cent and over-reporting exercise by half.
Protein: the nutrient with a number
Protein has the clearest target of any nutrient. Pooled resistance-training trials show the muscle and strength benefit of protein rising up to about one point six grams per kilogram of body weight a day and flattening beyond it. About a fifth of that per meal, across four meals, uses it best, because a single meal can only use so much. Athletes and anyone in a calorie deficit sit at the upper end, around one point eight to two grams per kilogram, to protect muscle. The tab's protein tile is measured against a fifty-five-gram daily floor, well under those figures, so that it only speaks up when intake is clearly short.
Why it matters
What the food you eat does to the years ahead.
The blocks move outward in time: the diseases diet drives, the weight that drifts over decades, and the teeth, training and thirst that respond within weeks. The inpatient trial in the middle is the one to remember, because it shows the form of food working on its own.
Disease, one factor at a time
In the United States, close to half of deaths from heart disease, stroke and diabetes were attributable to ten dietary factors, led by excess sodium, too few nuts and seeds, too much processed meat and too little seafood, vegetables and fruit. Reducing saturated fat lowered cardiovascular events by seventeen per cent in the Cochrane pooled trials. Higher fibre intake, twenty-five to twenty-nine grams a day, was associated with fifteen to thirty per cent lower mortality and heart disease. Sugar-sweetened drinks raised the risk of diabetes and metabolic syndrome by about a quarter at one to two servings a day.
Weight over years, not weeks
Following over a hundred thousand adults for two decades, the foods most tied to weight gain were potato chips, potatoes, sugary drinks and processed and unprocessed red meat; yogurt, nuts, fruit, whole grains and vegetables tracked with less gain. In a tightly controlled inpatient trial, people offered an ultra-processed diet ate about five hundred calories a day more and gained nearly a kilogram in two weeks, then lost it on the unprocessed diet, with the same nutrients on offer. Free sugars added or removed from the diet moved body weight by close to a kilogram in the pooled trials.
Teeth, training and the rest
Diet reaches beyond weight. Free sugars are the pivotal cause of tooth decay, and the evidence supports keeping them under ten per cent of energy, ideally five. Protein timing and total intake decide how much of a training block turns into muscle. Water is the drink with no downside; most adults meet their needs from thirst, food and ordinary drinks, and the review evidence for extra intake is thin outside heat and exercise.
How to improve
The moves with the best evidence, in order of payoff.
Write it down first, because the record fixes the under-counting everything else depends on. Then change the form of the food, land the protein, and choose a pattern you can keep; each step is a smaller change with a smaller payoff than the one before it.
Write it down
Self-monitoring of food intake is the behaviour most consistently linked with weight change in the reviews, and the effect is largest when it is frequent. It works because it corrects the under-reporting everyone does, and because attention while eating itself reduces intake: people who ate while distracted ate more later, and people reminded of their last meal ate less. One line per meal in the composer is enough; the tab's consistency tile measures whether meals are being logged on most days.
Change the shape of the food
Swap ultra-processed for minimally processed versions of what you already eat before counting anything: the inpatient trial shows the five-hundred-calorie difference comes from the food's form, not its macronutrients. Smaller plates, packages and portions cut what people eat without them noticing in the Cochrane review. Trade sugary drinks for water or unsweetened alternatives, the single change with the clearest diabetes evidence, and add fruit, vegetables, nuts and whole grains rather than only removing things.
Hit the protein target, then relax
Put a protein source in each meal so that the day lands near one point six grams per kilogram if you train, spread across three or four meals. That is the only macronutrient target with trial-level evidence for most people; beyond it, the carbohydrate-versus-fat argument made no difference to weight in the two-year head-to-head trial, and carbohydrate quality, meaning fibre and whole grains, matters more than quantity.
Pick a pattern you can keep
The named diets lose about the same weight; the one you keep wins. Mediterranean-style eating has the strongest outcome trial behind it, and it is also the pattern people stay on. Time-restricted eating in an early window improved insulin sensitivity and blood pressure in a small crossover trial even without weight loss, so it is a reasonable structure for people who find it easy, but the evidence is early.
Common pitfalls
Where honest effort goes wrong.
The pitfalls are all versions of not seeing: calories that never reach the log, a plan too strict to survive a holiday, and portions that grow because the plate did. None is a failure of willpower, and each has a structural fix.
Counting what you cannot see
Untracked snacks, oils, sauces and drinks are where the under-reporting lives, and they are why people who are certain they eat little do not lose weight. Sugary drinks are the clearest example: liquid calories do not register as food and they raise diabetes risk on their own. Logging the small things is what makes the calorie tile trustworthy.
Perfect diet, no adherence
A strict plan that lasts three weeks loses to a loose one that lasts three years. In every comparison of diet types, adherence predicted the result and the type did not, and the foods tied to long-term weight change were ordinary ones eaten often. Choose changes that survive holidays and bad weeks.
Portion drift
Plates, glasses and packages set how much people eat before hunger has a say, and the effect was consistent across the pooled trials. Ultra-processed foods are eaten faster and in larger amounts than whole foods. Keeping the default small does more than willpower.
Myths
Claims the trials have tested and not supported.
Breakfast, meal frequency and carbohydrate have each carried a rule that the trials then tested. In every case the day's total and its quality turned out to matter, and the rule did not.
Breakfast is the key to weight loss
The randomised trials found that adding breakfast did not help weight loss and that skipping it did not lead to eating more later in the day; breakfast eaters simply ate more in total. Eat it if you like it. What matters is the day's total and its quality.
Small frequent meals boost metabolism
Meal frequency has no effect on body composition once an outlying study is removed from the pooled analysis. Spreading protein across meals helps muscle, but three meals do that as well as six. Choose the number of meals that keeps you comfortable and consistent.
Carbs make you fat
In the two-year trial that assigned people to diets ranging from thirty-five to sixty-five per cent carbohydrate, weight loss was the same. What differs is the type: whole grains and fibre are associated with lower mortality and heart disease, while free sugars and sugary drinks raise weight and diabetes risk. The carbohydrate question is a quality question.
Measuring progress
What the tiles measure and how to read them.
The stat card shows calories, protein and logged days because those are the three behaviours with evidence behind them. The second block is a caution: weigh the trend, since a day's number moves with water and salt.
Calories, protein and logged days
The calorie tile compares the day with the target you set in Goals; the strip reacts when the weekly average drifts more than a fifth from it either way, because chronic large surpluses and deficits both matter. The protein tile compares the daily average with a floor of fifty-five grams and the guidebook figure of one point six grams per kilogram for people who train. The consistency tile counts days with at least one logged meal, because logging is itself the behaviour with the best evidence.
Weigh the trend, not the day
Daily weight moves with water and salt; the weekly average is the signal. Waist measurement and how clothes fit track fat better than the scale during training. Over months, the foods that appear most often in your log predict the direction, as they did in the twenty-year cohort.
Tips you may see in the app
The line under the tab picks one of these for your own numbers. Each rests on the same standard of evidence.
- If you train, aim for about one point six grams of protein per kilo a day, over meals.Pooled trials show the muscle benefit of protein rising up to about one point six grams per kilogram of body weight a day and flattening beyond it; spread it across three or four meals.
- A protein source in every meal beats one large hit; a single meal can only use so much.About a fifth of the day's protein per meal, across three or four meals, uses it best; meal frequency beyond that made no difference to body composition in the pooled analysis.
- Ultra-processed food adds about five hundred calories a day unnoticed; swap the form.In a tightly controlled inpatient trial, people offered an ultra-processed diet ate about five hundred calories a day more and gained weight in two weeks, with the same nutrients on offer as the unprocessed diet.
- Sugary drinks are the clearest single risk in the diet studies; water is the easy win.Sugar-sweetened drinks raised the risk of diabetes and metabolic syndrome by about a quarter at one to two servings a day, and liquid calories do not register as food; swap them for water or unsweetened alternatives.
- Smaller plates and packages cut intake without hunger; set the default small.Across the pooled trials, smaller plates, glasses and packages reduced how much people ate without them noticing; keeping the default small does more than willpower.
- Named diets lose about the same weight; the one you can keep wins.In the head-to-head trials the diet programmes differed by a kilogram or two at most and adherence explained the results; a Mediterranean pattern has the strongest outcome trial behind it and is the one people stay on.
- Writing down what you eat is the best-evidenced way to change it; one line per meal.Self-monitoring of food intake is the behaviour most consistently linked with weight change in the reviews, most when it is frequent; it corrects the under-reporting everyone does.
- People underestimate what they eat by close to half; unlogged sauces and drinks hide it.In a careful study, people who believed they could not lose weight were under-reporting intake by forty-seven per cent and over-reporting exercise by half; the oils, sauces, snacks and drinks are where the gap lives.
Sources
Every study this guidebook rests on, in the order it is first cited.
- 1
GBD 2017 Diet Collaborators (Afshin A, Sur PJ, Fay KA, et al.) (2019). Health effects of dietary risks in 195 countries, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet.
- 2
Micha R, Peñalvo JL, Cudhea F, Imamura F, Rehm CD, Mozaffarian D (2017). Association between dietary factors and mortality from heart disease, stroke, and type 2 diabetes in the United States. JAMA.
- 3
Estruch R, Ros E, Salas-Salvadó J, et al. (2018). Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. New England Journal of Medicine.
- 4
Aune D, Giovannucci E, Boffetta P, et al. (2017). Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality: a systematic review and dose-response meta-analysis of prospective studies. International Journal of Epidemiology.
- 5
Wang X, Ouyang Y, Liu J, et al. (2014). Fruit and vegetable consumption and mortality from all causes, cardiovascular disease, and cancer: systematic review and dose-response meta-analysis of prospective cohort studies. BMJ.
- 6
Sacks FM, Bray GA, Carey VJ, et al. (2009). Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. New England Journal of Medicine.
- 7
Johnston BC, Kanters S, Bandayrel K, et al. (2014). Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis. JAMA.
- 8
Lichtman SW, Pisarska K, Berman ER, et al. (1992). Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. New England Journal of Medicine.
- 9
Hall KD, Ayuketah A, Brychta R, et al. (2019). Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism.
- 10
Mozaffarian D, Hao T, Rimm EB, Willett WC, Hu FB (2011). Changes in diet and lifestyle and long-term weight gain in women and men. New England Journal of Medicine.
- 11
Morton RW, Murphy KT, McKellar SR, et al. (2018). A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults. British Journal of Sports Medicine.
- 12
Schoenfeld BJ, Aragon AA (2018). How much protein can the body use in a single meal for muscle-building? Implications for daily protein distribution. Journal of the International Society of Sports Nutrition.
- 13
Phillips SM, Van Loon LJC (2011). Dietary protein for athletes: from requirements to optimum adaptation. Journal of Sports Sciences.
- 14
Hooper L, Martin N, Jimoh OF, Kirk C, Foster E, Abdelhamid AS (2020). Reduction in saturated fat intake for cardiovascular disease. Cochrane Database of Systematic Reviews.
- 15
Reynolds A, Mann J, Cummings J, Winter N, Mete E, Te Morenga L (2019). Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. The Lancet.
- 16
Malik VS, Popkin BM, Bray GA, Després JP, Willett WC, Hu FB (2010). Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes: a meta-analysis. Diabetes Care.
- 17
Te Morenga L, Mallard S, Mann J (2012). Dietary sugars and body weight: systematic review and meta-analyses of randomised controlled trials and cohort studies. BMJ.
- 18
Sheiham A, James WPT (2015). Diet and dental caries: the pivotal role of free sugars reemphasized. Journal of Dental Research.
- 19
Popkin BM, D'Anci KE, Rosenberg IH (2010). Water, hydration, and health. Nutrition Reviews.
- 20
Burke LE, Wang J, Sevick MA (2011). Self-monitoring in weight loss: a systematic review of the literature. Journal of the American Dietetic Association.
- 21
Robinson E, Aveyard P, Daley A, et al. (2013). Eating attentively: a systematic review and meta-analysis of the effect of food intake memory and awareness on eating. American Journal of Clinical Nutrition.
- 22
Harkin B, Webb TL, Chang BPI, Prestwich A, Conner M, Kellar I, Benn Y, Sheeran P (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin.
- 23
Hollands GJ, Shemilt I, Marteau TM, et al. (2015). Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Cochrane Database of Systematic Reviews.
- 24
Sutton EF, Beyl R, Early KS, Cefalu WT, Ravussin E, Peterson CM (2018). Early time-restricted feeding improves insulin sensitivity, blood pressure, and oxidative stress even without weight loss in men with prediabetes. Cell Metabolism.
- 25
Sievert K, Hussain SM, Page MJ, et al. (2019). Effect of breakfast on weight and energy intake: systematic review and meta-analysis of randomised controlled trials. BMJ.
- 26
Schoenfeld BJ, Aragon AA, Krieger JW (2015). Effects of meal frequency on weight loss and body composition: a meta-analysis. Nutrition Reviews.
This summarises research for healthy adults. It is not medical advice and does not replace a professional.