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

We asked what people ate that day and how they slept that night, and saw more fiber linked to deeper sleep, more saturated fat tied to less deep sleep, and more refined carbs and sugars driving more night-time arousals.

Welcome to the Huberman Lab Podcast; today we explore the two-way street between sleep and nutrition with Dr. Marie-Pierre St-Onge, and we’ll get into practical steps to improve sleep, appetite regulation, metabolism, and overall health.

Let’s start with how losing small chunks of sleep shifts hunger and food choices, then circle back to how eating patterns reshape sleep.

Large cohorts link short sleep with higher BMI and weight gain over time, but lab studies show causality: after five nights of severe sleep restriction under controlled feeding, men had higher ghrelin, women had lower GLP-1, food reward responses rose, and when allowed, people ate roughly 300 extra calories.

So men’s hunger signal goes up, while women’s braking system goes down—was that expected?

It surprised us; earlier work was mostly in men, so pooling sexes hid the split until we analyzed them separately.

Sleep restriction was intense and supervised with no naps, and fatigue nudged choices toward palatable foods; across studies, people overeat about 250 to 400 calories, and just two weeks at around five hours a night led to about half a kilo gained.

Is that simply stress physiology?

In-lab, cortisol rhythms didn’t change, and with identical meals we saw no glucose, insulin, or lipid shifts, suggesting real-world metabolic harm comes from the combo of short sleep plus worse diet and activity patterns.

In a six-week free-living cut of about ninety minutes a night down to six hours, insulin resistance rose, insulin sensitivity dropped, and blood pressure climbed, especially in postmenopausal women.

Energy expenditure is slightly higher when you stay awake—about five percent or roughly ninety calories—but intake usually rises more, so the balance still tilts toward weight gain.

A simple takeaway is to recognize sleep-loss cravings in the moment, pause, and decide if you truly need the extra energy or just a short-term mood lift.

Now the other direction—how does diet shape sleep, including timing?

Across cohorts, Mediterranean- and DASH-style eating are linked to better sleep and fewer insomnia symptoms, and in the Women’s Health Initiative, healthier dietary patterns predicted moving from insomnia toward better sleep three years later.

In our inpatient study, self-selected diets were higher in calories, saturated fat, and sugar, and led to longer time to fall asleep and less deep sleep; more fiber tracked with more slow-wave sleep, while refined carbs increased arousals.

I finish eating about three hours before bed and feel better with earlier meals, likely from the thermic effect and overall metabolic alignment.

Going to bed out of sync with your circadian preference feels worse than missing an hour or two before midnight, which is why shift work is so hard.

Short, early naps can help, but avoid late naps that drain sleep pressure; if you need naps despite adequate time in bed, investigate your night-time sleep quality.

A recent analysis suggests a sweet spot of roughly six and a half to about eight hours for healthy aging, with women tending slightly higher, and both too little and too much sleep looking unfavorable.

Women report more insomnia and may show stronger blood pressure and metabolic impacts at lower apnea thresholds; a good clinical opener is simply, “How’s your sleep?” to surface more than hours alone.

Loud snoring and gasps suggest sleep apnea—home testing is common, CPAP needs the right pressure, and weight loss can lessen severity.

Kefir didn’t change cholesterol synthesis in our trial, yet fermented foods may still benefit gut health depending on outcomes; a coffee-derived oligosaccharide drink reduced fat mass in men only, and ginger in hot water modestly raised post-meal energy burn.

Delay of the first meal by about four hours reduced fat oxidation later in the day, so front-load most calories into the first two-thirds of your waking hours.

Replacing other fats with one to two tablespoons of purified MCT oil can slightly boost thermogenesis and trim fat mass, with brief GI adjustment; the cholesterol-lowering “portfolio” diet—soy protein, nuts, plant sterols, and soluble fiber with healthy fats—can rival statins in trials.

Lead with high-volume plants, add nuts and moderate animal portions for staying power, avoid refined starch–fat combos and “white” grains, and note that U.S. portions and added sugars tend to run high.

In a crossover study, corn-oil-fried chips improved lipid markers versus low-fat or high–saturated-fat snacks; choose oils by smoke point and purpose, keep portions sensible, and cook at home when you can.

When you read a nutrition label, focus on what matters for you. Salt might be an issue if you’re sensitive, but active people may need more, so choose what to track or the label becomes noise.

You mentioned industry funding. What’s the difference between a company paying for a study and a real conflict of interest, and how do you keep pressure off the results?

We design the question, collect and analyze the data, and our contracts secure the right to publish. I’ve taken industry-funded null results to several journals and seen rejections for being unexciting, not due to sponsor pressure, which points to publication bias; misconduct also shows up in government-funded work, so integrity matters more than who pays.

Given tight federal budgets, are you working with companies mainly to keep the research moving, and would you prefer government grants if they were available?

Yes, government grants are the target and usually support nutrition work better, but they’re scarce. Industry can be essential when a specific food or product is required and access would otherwise be impossible.

There’s a push to publish more null results and more interest in nutrition, yet people suspect bias in food studies and pharma hides negatives. How do you see that landscape?

Skepticism toward food research is strong even though drug companies often study their own products, and many projects that do not show benefit simply never advance. What we need are transparent nulls and a culture that values them.

Inside the lab, how do you keep results honest and avoid wishful thinking?

We teach trainees to check statistics and effect sizes, not just numbers that look different. The best habit is healthy doubt, where you challenge a finding until it survives scrutiny.

What are people asking you about now, and do you personally use supplements or stick to food?

Fads pop up all the time, peptides included, but I prefer to meet needs with whole foods because it’s more satisfying and comprehensive. Some people do require supplements, yet food first is a sound goal.

Doctors are recommending psyllium and other fiber supplements more often, partly because people avoid produce. What’s your take?

Fiber helps, but fruits and vegetables deliver fiber plus polyphenols and other compounds that nourish the gut and may amplify fiber’s effects. When possible, choose the whole package, not just the ingredient.

My big takeaway is to align eating with sleep to build momentum. Is that the cycle you aim to create?

Yes, flip the vicious cycle into a healthy one. Good sleep supports better choices, those choices improve sleep, and the loop keeps moving you toward better health.

Thank you for joining me for this conversation with Dr. Marie Pierre Saint-Onge. You can find links to her lab and her book, Eat Better, Sleep Better, in the show notes. If you’re learning from this podcast, please subscribe on YouTube, follow on Spotify and Apple, and consider leaving a review and comments on those platforms and on YouTube. Supporting our sponsors helps keep the show going. If you have questions or topic requests, post them in the YouTube comments, which I read. My first book, Protocols: An Operating Manual for the Human Body, is available for preorder at protocolsbook.com. I’m Huberman Lab on Instagram, X, Threads, Facebook, and LinkedIn, where I share science and tools that often extend beyond the podcast. You can also join our zero-cost Neural Network newsletter at hubermanlab.com for concise summaries and one to three page protocol PDFs on sleep, dopamine, deliberate cold exposure, and a foundational fitness program. We do not share your email. Thanks again for your interest in science and health.