Browse the evidence

Higher dietary glycemic index (GI) and glycemic load (GL) are robustly associated with an increased risk of incident type 2 diabetes (T2D) in adults initially in good health.

To lower your risk of type 2 diabetes, prioritize foods with a lower glycemic index and load. This means choosing carbohydrates that digest more slowly, rather than just reducing the total amount of carbs. The evidence suggests that the quality of carbohydrates, specifically their impact on blood sugar rise, is a significant predictor of diabetes risk in healthy adults.

Obese individuals exhibit increased activation in reward-related brain areas (insula, orbitofrontal cortex) in response to visual high-calorie food cues compared to healthy-weight individuals, particularly when satiated.

If you are obese, your brain may react more strongly to seeing high-calorie foods than a healthy-weight person's brain, even when you are full. This is a biological difference in reward processing, not a lack of willpower. To manage this, focus on reducing exposure to visual food cues (e.g., keeping junk food out of sight) rather than relying solely on willpower during moments of temptation.

Triple hormone receptor agonist retatrutide (1-12 mg weekly) significantly reduces liver fat in patients with metabolic dysfunction-associated steatotic liver disease (MASLD), achieving near-maximal reduction (~80%) at 24 weeks with high doses (8-12 mg).

For patients with MASLD, high-dose retatrutide (8-12 mg weekly) offers a highly effective treatment to significantly reduce liver fat, often normalizing it within 24 weeks. This hormonal intervention addresses the root metabolic drivers of liver fat accumulation more potently than current GLP-1 mono-agonists.

Higher circulating or adipose tissue levels of the dairy fat biomarkers pentadecanoic acid (15:0), heptadecanoic acid (17:0), and trans-palmitoleic acid (t16:1n-7) are associated with a significantly lower incidence of type 2 diabetes.

High levels of specific fatty acids (15:0, 17:0, and t16:1n-7) in your blood or fat tissue, which serve as biomarkers for dairy fat consumption, are linked to a lower risk of developing type 2 diabetes. This suggests that moderate consumption of dairy products may be metabolically beneficial, contrary to the general advice to avoid all saturated fats.

Obese, insulin-resistant individuals exhibit a defective (reduced) thermic effect of food, specifically for glucose, which may contribute to weight gain or impaired weight loss, though it cannot explain the majority of excess body weight.

If you are obese and insulin-resistant, your body may burn slightly fewer calories from food (50-150 kcal/day) than a lean person. While this makes weight loss slightly harder, it is not the main reason for obesity. Focus on sustainable caloric deficits and improving insulin sensitivity through exercise and diet quality rather than blaming a 'broken metabolism'.

Using a ±1 standard deviation (SD) cut-off to screen out inaccurate dietary energy intake reports (comparing reported intake to predicted total energy expenditure) is preferable to using a ±2 SD cut-off or no cut-offs when analyzing relationships between diet and health outcomes like BMI.

If you are analyzing dietary data to understand health outcomes, do not just throw out the most extreme outliers. Use a stricter screening method (like comparing reported intake to predicted energy needs using a ±1 SD threshold) to remove inaccurate reporters. This reveals the true relationship between diet and body weight, which is often hidden if you rely on self-reported 'usual' intake or loose exclusion criteria.

Daily butter consumption (approx. 14g) is associated with a small but statistically significant increase in all-cause mortality, while showing no significant association with cardiovascular disease or stroke, and a modest inverse association with type 2 diabetes incidence.

If you eat about one tablespoon of butter a day, current evidence suggests it will not increase your risk of heart disease or stroke, and may even be linked to a slightly lower risk of type 2 diabetes. However, it is associated with a very small (1%) increase in overall mortality risk. You do not need to eliminate butter, but you also shouldn't rely on it as a health food; prioritize overall dietary patterns over single ingredients.

Novel plant-based meat alternatives should be treated as sensory meat alternatives but not as true nutritional replacements for animal meat due to differences in nutrient bioavailability and complex food matrix interactions.

If you are swapping animal meat for plant-based alternatives to improve health, recognize that you are gaining sensory satisfaction but may lose specific nutrient benefits. To maintain nutritional status, ensure your overall diet includes diverse plant sources to compensate for lower bioavailability of nutrients like B12, Iron, and Zinc, or consider supplementation if you are not consuming fortified products.

Plant-based meat alternatives generally provide lower bioavailability of key micronutrients (B12, Iron, Zinc) compared to animal meat, even when fortified, due to food matrix effects and anti-nutrients.

When eating plant-based meats, be aware that you might not absorb as much B12, Iron, or Zinc as you would from animal meat, even if the label says they are equal. To mitigate this, pair these foods with Vitamin C (to help iron absorption) and consider getting your B12 and Zinc status checked if you rely heavily on these products.

Anthropometric prediction equations using age, race, height, weight, and waist circumference provide a valid, high-accuracy estimate of lean body mass and fat mass, significantly outperforming BMI in predicting obesity-related biomarkers.

Stop relying solely on BMI to judge your health or body composition. You can get a highly accurate estimate of your lean and fat mass using simple measurements: height, weight, waist circumference, age, and race. This method is as effective as expensive scans for predicting metabolic health risks like cholesterol and blood sugar, making it a practical tool for anyone.

High protein intakes (up to 3.0 g/kg/day) do not negatively impact renal function in healthy individuals and may positively influence bone health by increasing IGF-1 and peak bone mass.

If you have healthy kidneys, you don't need to worry about high protein intakes harming your kidneys or bones. In fact, higher protein may help build stronger bones.

For resistance-trained individuals, training to failure may provide a significant advantage for muscle hypertrophy compared to non-failure training.

If you are an experienced lifter who has plateaued, incorporating sets to failure may help break through stagnation and stimulate additional hypertrophy. For beginners or untrained individuals, stopping short of failure is sufficient and likely more sustainable.

Current global agricultural production overproduces grains, fats, and sugars while underproducing fruits, vegetables, and protein, creating a nutritional imbalance that contributes to obesity and diabetes.

Stop assuming that having enough food means you are eating well. Current global diets are skewed heavily toward grains, fats, and sugars, which drives obesity and diabetes. To fix this, prioritize increasing the intake of fruits, vegetables, and plant-based proteins, and reduce reliance on energy-dense processed foods.

A complete shift to plant-based protein (vegetarian diet) reduces total agricultural land use and GHG emissions compared to the HHEP diet, despite requiring more arable land for crops.

Shifting to plant-based protein sources (legumes, seeds, nuts) reduces GHG emissions and pasture land use compared to animal-based proteins. While it may require more arable land for crops, the overall environmental footprint is lower.

Intramuscular factors, specifically changes in Androgen Receptor (AR) protein content and p70S6K phosphorylation, are significant predictors of muscle hypertrophy, explaining more variance than systemic hormones.

Muscle growth is driven by local signals within the muscle, such as Androgen Receptor content and p70S6K phosphorylation, rather than systemic hormone levels. Consistent resistance training upregulates these local factors, which are the true drivers of hypertrophy.

The acute post-exercise increase in Interleukin-6 (IL-6) is correlated with muscle hypertrophy, but it does not add predictive value beyond local muscle factors like AR and p70S6K.

The inflammatory response to exercise, specifically IL-6, is correlated with muscle growth. However, this is likely a byproduct of local muscle adaptation rather than the primary driver. Focus on training consistency rather than trying to manipulate inflammation.

Food pricing strategies, specifically taxes on unhealthy foods and subsidies for healthy foods, effectively alter consumption, with stronger effects on lower-income populations.

To improve diet at scale, use financial incentives. Subsidize healthy options (like fruits/veg) to make them cheaper, and tax unhealthy options (like sugary drinks) to make them more expensive. This works best when targeted at lower-income groups who are more price-sensitive.

School procurement policies that restrict unhealthy competitive foods and beverages are effective in reducing student intake and obesity prevalence.

Schools can significantly reduce childhood obesity by banning sugary drinks and limiting high-fat/sugar snacks in cafeterias and vending machines. Government-mandated policies are more effective than voluntary ones.

Total energy expenditure (TEE) and physical activity level (PAL) decline progressively throughout adult life (from age 20 to 96) in both normal weight and overweight individuals, primarily driven by a decrease in physical activity rather than just basal metabolic rate changes.

As you age, your body burns fewer calories, but this isn't just because your organs slow down. A large part of the drop comes from moving less (lower Physical Activity Level). To maintain your weight, you need to account for this gradual decline in energy expenditure (about 150 kcal/day less per decade for normal-weight adults) by adjusting your food intake or increasing your physical activity levels.

Genetic factors account for specific portions of BMI variability through three measurable metabolic traits: metabolic rate (12%), respiratory quotient/fat oxidation (5%), and spontaneous physical activity (10%), leaving 40% attributable to food intake regulation.

Your body burns calories differently based on genetics. Some people burn less at rest (metabolic rate), oxidize less fat (high RQ), or move less unconsciously (low NEAT). Understanding this helps explain why some people gain weight easily despite similar calorie intake to others.

A theory-based group weight loss intervention significantly increases dietary fiber intake, which mediates weight loss, but fails to significantly increase objectively measured physical activity despite improvements in self-efficacy and social support for activity.

This structured group program helps people increase fiber intake, which leads to weight loss. However, it does not significantly increase physical activity levels, even though participants feel more confident and supported in doing so. For weight loss, prioritize increasing fiber intake through this structured approach rather than relying solely on increased physical activity.

Improvements in self-efficacy and understanding of the behavior change process mediate successful dietary change through engagement in coping planning and self-monitoring, but these same mediators do not translate to increased physical activity.

To change diet, focus on building self-efficacy and understanding the process, then use coping planning and self-monitoring. These steps may not work for increasing physical activity, suggesting different strategies are needed for exercise.

Dietary interventions produce observable changes in gut microbiome composition within days, whereas microbiome-mediated effects on host phenotypes or biomarkers require interventions lasting weeks to months.

If you change your diet, your gut bacteria will likely adapt within a few days. However, do not expect to feel health benefits or see changes in blood markers immediately. You need to maintain the dietary change for several weeks or months to see the downstream health effects driven by those microbial changes.

Microbiome responses to dietary interventions are highly personalized, meaning the same diet can produce different microbial and metabolic outcomes in different individuals based on their baseline microbiome composition.

Standard dietary advice may not work for everyone because your gut bacteria are unique. If a specific diet (like high fiber) doesn't improve your health markers, it might be due to your baseline microbiome. Future personalized nutrition may involve testing your microbiome to predict which foods will work for you.