Eating for Better Sleep & Foods that Improve Metabolic Health | Dr. Marie-Pierre St-Onge

KEYPOINTS

• 01

  Sleep and diet affect each other in both directions.

  St-Onge came to sleep research from obesity research, where population studies kept linking short sleep to higher body mass index (BMI) and more weight gain over time. Her lab tests causality both ways: how sleep loss changes eating, and how what you eat changes that night's sleep.

• 02

  Short sleep raises appetite through different hormones in men and women.

  In her inpatient study, short sleep increased ghrelin, the hormone that drives hunger, in men. In women it lowered GLP-1, a fullness hormone. Either way, participants ate about 300 more calories a day when they could choose their own food, and brain imaging showed reward centers responding more strongly to food after sleep restriction.

• 03

  Even mild, sustained sleep loss harms blood sugar control and blood pressure.

  In a free-living study of women, cutting sleep by about an hour and a half (from roughly 7.5 hours to around 6, close to a full sleep cycle) for 6 weeks increased insulin resistance and raised blood pressure. Insulin sensitivity fell more in postmenopausal than premenopausal women.

• 04

  Sleep loss burns a few extra calories, nowhere near enough to cover the extra eating.

  In a small metabolic-chamber study of women at Columbia, energy expenditure rose about 5%, roughly 90 calories, under sleep restriction, against the ~300 extra calories eaten. She also thinks sleep-deprived people become more sedentary, with less spontaneous daytime activity. Eating may also be a subconscious way to get a jolt of energy when tired.

• 05

  What you eat during the day changes deep sleep that night.

  Using nightly sleep lab recordings (polysomnography), her lab found that more fiber was associated with more deep (slow-wave) sleep, more saturated fat with less deep sleep, and more refined carbohydrates and simple sugars with more arousals during the night. When participants switched from a controlled diet to choosing their own food, they ate about 450 more calories and 33% more saturated fat. They took over 70% longer to fall asleep and had about 20% less deep sleep.

• 06

  Mediterranean and DASH-style diets are linked to less insomnia.

  With Susan Redline in the MESA cohort, diets closer to Mediterranean were associated with a better chance of adequate sleep and fewer insomnia symptoms. In Women's Health Initiative data, women whose diets aligned with Mediterranean or DASH (Dietary Approaches to Stop Hypertension: fruits, vegetables, nuts and more) were less likely to have insomnia three years later. These are observational links, not proof of cause.

• 07

  Eating later in the day reduces fat burning, even with identical food.

  In a metabolic-chamber study, participants ate the same foods in the same amounts with a 10-hour eating window, starting either 1 hour or 5 hours after waking. Meals eaten later led to less fat oxidation. Huberman summarizes the takeaway as shifting meals into roughly the first two-thirds of the day, which may help sleep and appetite too.

• 08

  Go to bed in line with your own clock; shift workers fight their biology.

  St-Onge knows of no data behind 'every hour before midnight is worth two.' But going to bed far later than your usual time puts you out of sync with your circadian system and makes good sleep harder. Shift workers try to sleep during the day when their biology isn't set up for it. On naps, she says the data are thin, but avoid napping close to bedtime so you keep your sleep pressure.

• 09

  Small dietary tweaks can modestly raise calorie burn, and they may add up.

  Medium-chain triglycerides (MCTs) go straight to the liver and are burned more readily than long-chain fats, raising the thermic effect of a meal in men and women. Over time her studies found better body composition and greater weight loss with MCT oil than with olive oil. Ginger raised calorie burn after a meal, possibly via the capsaicin receptor. She thinks such effects could be additive since they act through different mechanisms, though that hasn't been tested.