Hi {{first name | there}},

You are probably thinking about body recomposition, even if you have never used the phrase.

Lose fat. Keep the muscle. Grow muscle if you can. It is a harder project than simply "losing weight," as we've learned from the GLP-1 revolution and the resulting frailty across those who aren't paying attention to muscle.

The advice around body recomposition is noisy. What gets the most hype online is unfortunately the weakest lever, and the two things with the best evidence behind them are so free and boring enough that nobody wants to talk about them.

Until now.

Glucose rises after you eat. Insulin is what your pancreas sends to put it away.

A meal that arrives fast spikes both, then drops them, often overshooting on the way down. That is the 3pm crash you have probably felt after a carb-heavy lunch.

But you could have the same exact meal, eaten differently, that won't result in a crash. It'll give you a curve that is wider and lower, plus level energy.

This concept will repeat in various ways throughout our issue below:

Your muscles store fuel as glycogen (storage form of sugar, glucose), and you’ve probably heard that you should top off your tank with a banana or something else filled with fast carbs as a kind of pre-workout.

A team in Maastricht went and looked, using magnetic resonance imaging to watch glycogen inside muscle and liver directly.

They fed trained cyclists a large carbohydrate breakfast and measured for three hours. Muscle glycogen levels did not change. Liver glycogen content did not move either.

Then the harder version. They emptied the stores with two hours of brutal cycling, then fed people an enormous carbohydrate load over twelve hours. The liver refilled fast and blew past its starting point within six. Muscle was still about a third empty at hour twelve.

Muscle fuel refills over roughly a day. Not in an hour, which practically means your banana snack pre-gym won’t do anything.

Your banana is not loading your legs. It is putting sugar in your blood, settling your stomach, and keeping you from thinking about lunch mid-set. All real benefits, maybe making you feel subjectively better, but not really powering your workout like you’ve been sold.

Instead: for anything under about 90 minutes, you are running on fuel you banked yesterday. Eat beforehand if you like it, but it’s not a reason to scramble for a snack.

I believed this for years, and there is some truth to the theory.

There is a real hierarchy in what your body burns. Alcohol goes first, because you cannot store it and its breakdown products are toxic, so clearing it jumps the queue. Carbohydrate is “burned” next. Fat waits till those are done. This is why a few drinks cost you more than their calorie count suggests: ethanol cuts 24-hour fat oxidation by roughly a third. Obvious tip: if you want to lose fat, cut out alcohol as much as you can.

Train fasted and you will burn a higher proportion of fat during the session. This is actually true.

But then we have a problem. Because it turns out that burning more fat during a workout does not make you lose more fat.

Researchers randomized 20 women to fasted or fed aerobic exercise on a calorie-restricted diet for four weeks. Both groups lost fat. No significant difference between them on any measure. A meta-analysis pooling the available trials found the same nothing. These are small studies, and I would love a big one, but the findings make sense.

Your body settles the books over 24 hours, not 45 minutes. Burn more fat at 7am and you burn less later. What you actually lose comes down to how much you eat across days, not which fuel you happened to be burning at 7am.

Same problem as the "fat burning zone" sticker on the treadmill. A slow walk burns a higher percentage of fat than a hard interval. The high intensity interval workout still burns more fat overall for your body recomposition goals, because it burns more of everything.

Instead: train fasted if it suits your morning. I do, if only to get myself out the door and to the gym faster. Just do it for convenience, not because you think it unlocks something. And if holding onto muscle is the whole point, there is a reasonable case against going hard fasted, since you are asking your body to find amino acids somewhere.

This is an interesting timing one.

The standard pattern: light breakfast, sandwich at lunch, most of the day's protein at dinner. Add it up and the number can be perfectly respectable.

Researchers tested that against the same protein spread evenly. Not more protein. The same amount protein, spaced differently throughout the day.

One group ate about 30 grams at each meal. The other ate 10 grams, then 16, then 63 at dinner. Over a day, the evenly spread group built 25% more muscle, and it held up after a week on each pattern.

Muscle treats each meal like a switch, not a dial. Under roughly 25 to 30 grams, weak signal. Far above it, no bonus.

A 63-gram dinner cannot bank the extra for tomorrow. You just threw away a chance to flip the switch a second time.

This matters most when you are dieting. Eating less than you burn means your body spends the day shopping for something to break down, and protein arriving every few hours is what tells it to leave your muscle alone.

That finding is a lab measurement, though. What you want to know is whether it builds muscle you can actually see.

Another trial examined this question carefully. Twenty-six young men lifted three times a week for twelve weeks on identical daily protein, with one group eating more of it at breakfast and the other saving it for dinner. The breakfast group gained 2.5 kg of lean tissue against 1.8 kg. A real gap, and a big one, but it landed just outside statistical significance at P = 0.06.

So: promising, not settled. How much protein you eat in a day is still the most important (you need a solid baseline), and when studies account for that, the timing effect mostly fades. Spreading it out is a good bet with one encouraging trial behind it. It is not a rule.

What is not in question is the amount. Young men ate at a roughly 40% deficit for four weeks while training six days a week. One group ate moderate protein, the other about double. The high-protein group gained 1.2 kg of lean mass and lost 4.8 kg of fat. The moderate group held lean flat and lost 3.5 kg.

Gaining muscle in a 40% deficit sounds ridiculously difficult, but this group got it done.

Caveats: young men, supervised, six days a week, and the high intake was about 2.4 grams per kilogram, which for a 150-pound person runs near 165 grams a day. That is pretty difficult for many folks unless you meal plan and track.

What to do? Still hit your total first, because that is the part with the strongest evidence behind it. Then move some of it to breakfast, which costs nothing and might be worth another half kilo of muscle over a few months.

Side note: It’s really beyond me why breakfast foods tend to be so carb-centric, when really we should emphasize protein!

You can change what a carbohydrate does to you without changing the carbohydrate.

Eat it last. Researchers gave people with prediabetes the identical meal three days running and changed only the sequence. Starch last, after protein and vegetables, cut the blood sugar spike by more than 40%, and the total rise over three hours was about 39% lower.

Protein and fiber first slow how fast your stomach empties, so the starch trickles in instead of arriving all at once. In practice: salad before pasta, vegetables before rice, bread last.

Two caveats: That study was in prediabetes, where there is more room to improve, so a healthy metabolism will see a smaller effect. And when researchers ran food order as a 16-week behavioral intervention in 45 adults with prediabetes, the food-order group lost about 3.6 pounds, but the change was not significantly different from the control group's.

So eating your starch last is proven to flatten blood sugar and is not proven to make you leaner. I still do it, because it is free and steadier blood sugar is worth having on its own.

Eat more fiber, full stop. This is the part of the carbohydrate story with the best outcome evidence, and it has nothing to do with timing.

Researchers randomized 240 adults with metabolic syndrome to either the American Heart Association's 13-component diet or a single instruction: eat 30 grams of fiber a day. That was the whole intervention. At twelve months the fiber group had lost 2.1 kg and the AHA group 2.7 kg, a gap of 0.6 kg that was not statistically significant.

The trial was built to find a difference, but it shows that one simple instruction produced real weight loss over a year, in people who had struggled with complicated diets.

If you change one thing about how you eat, make it this one.

Cook it, cool it. Chill cooked starch and some of it recrystallizes into resistant starch, which your small intestine cannot digest. It travels to your colon and feeds bacteria instead of spiking you. Rice cooled for 24 hours held roughly two and a half times the resistant starch of fresh, and reheating did not undo it. The cooled-then-reheated rice still produced a lower glucose response than fresh.

Cook your rice and potatoes the day before. That is the entire trick.

Buy the chemistry. Some bakers build with resistant wheat starch directly. ROYO is the one I point people toward, founded by a dietitian, using resistant wheat starch and psyllium to land near 2 grams of net carbs a slice. If you miss bread, it beats most of the keto aisle.

Add psyllium, slowly, and for the right reason. Psyllium is a soluble fiber that gels in your gut and slows glucose absorption. People quit because they start at 15 grams, get miserably gassy, and decide fiber is not for them. Start at a quarter to a half teaspoon, sit there a week, then creep up, always with a full glass of water. Most studies use 7 to 15 grams a day, and nobody warns you about the weeks it takes to get there. I am still climbing that ladder myself, which I will get to below.

One thing psyllium is not: a weight-loss supplement. A meta-analysis of 22 randomized trials found no significant effect on body weight, BMI, or waist circumference. It has good evidence for glucose and cholesterol, and that is the reason to take it.

A scoop of psyllium is a blood sugar tool that happens to be made of fiber. Fiber from food does seem to help a bit with weight loss.

And sometimes, a prescription. Acarbose is an old diabetes drug, about $20, that blocks the enzymes breaking starch into glucose. The undigested starch goes to your colon instead of your bloodstream. I prescribe it to help with metabolic health and wellbeing, plus the potential benefits to the gut and beyond.

Acarbose is also one of the few compounds to extend lifespan in the NIA's Interventions Testing Program: 22% in male mice, 5% in females. Side note: people see that gap and conclude it does not work for women. I disagree. Female mice do not go through menopause the way we do; they keep estrogen protection we lose. My hunch is that after menopause, women look more like those male mice than like the female ones.

It is a prescription, not a supplement, and the side effect is exactly what the mechanism predicts 😅. Send starch to your colon and the bacteria there will throw a party. Eat a lot of carbohydrate on it and you will find out.

Everything so far flattens the curve, which is what we typically want, but not all the time…

Fast glucose is what you want when:

Slow is what you want when:

Building muscle gradually needs a growth signal, not a flood. A steady stream does the same job without the cost.

Researchers fed people identical food, matched for nutrients, activity, sleep, and light exposure, and moved only the timing later in the day.

Eating late made people hungrier, shifted their appetite hormones in the same direction, lowered how many calories they burned while awake, dropped their body temperature, and changed how their fat tissue handled fat.

Your body also handles sugar better in the morning and worse as the day goes on, so the same meal spikes you harder at 9pm than at 9am.

Calories do not turn into something else after dark. It works through hunger and how much you burn, which is a duller explanation and a more useful one, because it explains why eating late mostly makes tomorrow harder.

Instead: move dinner earlier. Because it makes tomorrow easier.

Walk after you eat.

Pooling the trials, two to five minutes of easy walking cut blood sugar after a meal by about 17% compared with sitting. Standing managed 9.5%, and did almost nothing for insulin.

Two minutes. To the corner and back.

Nutrition advice often over-generalizes study findings. “This improved a number in a study” does not always mean, “This made people leaner or stronger.”

So I checked. Here is the whole list of study findings and whether you’ll notice the effects of specific practices in your body recomposition.

Protein, lifting, and fiber are the gold standard. Everything below is either a good mechanism with thin evidence behind it, or good for some things but not necessarily others (good for metabolic health, not for losing fat).

Flatter blood sugar is worth having for its own sake, and several of these cost nothing. But none of it pays off until the basics are handled, and most people are polishing the last five percent before the first 90% is done.

So here is the order I would fix things in.

The reason most eating plans collapse is that they hand you a single target. One number turns every restaurant dinner into a failure, and failure is what makes people quit in week three.

Give yourself three numbers instead - for calories, fiber, and protein especially.

A floor. What you do not go under, even on a bad day. This is the one that keeps you safe.

An ordinary day. What a normal Tuesday looks like.

A flex day. Room for the dinner out, the birthday, the work trip. Planned is the whole point. A flex day you chose is not the same as a day that got away from you.

Your body responds to the overally week, not the day. Two big days inside five normal ones is a different thing from two big days that ambushed you, even when the calories match exactly.

When something has to give, this is the order I use. Each step only matters once the one above it is handled.

1. Manage the week, not the day. One big dinner is fine. Three in a row is a pattern.

2. Hit a protein floor every single day. A reasonable daily target is about 0.7 grams of protein per pound of bodyweight — roughly 110 grams for someone at 150 pounds. The floor is around 0.55 g per pound; past about 1 gram per pound you stop getting much back (although it’s useful for people trying to recomp since it keeps you full). Split it across three or four meals so each one clears the 30-gram threshold, and put one of them at breakfast.

3. Get fiber up, and aim for 30 grams. From food. The evidence above is about what you eat, not what you scoop.

4. Do not keep under-eating fat. It drifts down while you are busy counting protein and calories, and you will feel it in your hunger and your hormones long before it shows up anywhere else. FWIW, this is not practically a danger for many of us living in modern society, since fat is hiding everywhere, but it’s worth a mention if you’re being particularly restrictive.

5. Let carbs take whatever is left, calorie-wise. This is the part you get to move around, once the four above are handled. It is also where every timing trick in this issue lives, which tells you roughly what they are worth.

One warning from lived experience: a low-fiber day is not a reason to slam 15 grams of psyllium before bed. That is how people end up cramping, and it is worse if you are on something like acarbose that is already sending starch to your colon. Rebuild the baseline gradually, with fluid.

Your own numbers depend on your size, your training, your sex, and where you are starting from.

So I built you the numbers. There is a body recomposition calculator on my site now that gives you all three, a floor, an ordinary day, and a flex day, across calories, protein, fiber, and fat. It also shows you my own targets, so you can see what this looks like on a real person rather than a formula.

Mine, for the curious: I hold a protein floor in the low 120s and aim for the mid 130s. Fiber lives around 25 to 30 grams. Fat I keep from drifting under about 50. Carbohydrate takes whatever is left, which on a training day is a lot more than people expect from a longevity doctor.

Protein at breakfast, which I resisted for years because I did not want to think about food before coffee. Now I sneak it in with ultrafiltered milk in my latte (it has higher protein, lower carbs). Aim for 20-30 grams. Eggs, Greek yogurt, or whatever protein survived dinner.

I eat in order without making it a production. Vegetables, protein, then the starch.

I eat resistant starch keto products. This is relatively out of the box, but something I’ve found works with my lifestyle.

Psyllium, one teaspoon a day. I am at the beginning of this and will not pretend otherwise. Some days a second teaspoon, and that is my current ceiling. Two teaspoons is still under what the studies use, but going slowly is the only way to stay comfortable, and rushing it is how people quit.

Acarbose, which I prescribe for myself. The least intuitive thing on this list and the one I get the most questions about. You’ll need to work with a clinician since it is prescription.

I walk after lunch, dinner, and as much as I can in general. I work at a desk most of the day, so my walking pad is a lifesaver.

I am borrowing the direction of the study findings, knowing there are always limitations (small studies, male subjects, etc). Again, at the end of the day, it’s about enough protein, enough lifting, and enough fiber.

Nutrition advice is full of unfounded claims. What actually works is spread thin and boring. Protein at breakfast. Starch last. Yesterday's rice. Dinner at seven. A walk around the block.

Nobody can sell you that, which is most of the reason you have not heard about it much!

Keep learning,

Hillary Lin, MD

What Human Longevity Found After Screening 10,000 "Healthy" People

Wei-Wu He and the Human Longevity team has run whole-body MRI, deep genomics, and advanced cardiac imaging on thousands of people who walked in certain they were fine. Watch it here, and subscribe if you have not already!

🏋️ Half of muscle aging looks optional. Trained older adults were simply missing 50% of the energy-metabolism gene changes their sedentary peers carried.

💪 Grip strength outpredicted how much people actually moved. In 5,472 women aged 63 to 99, strength tracked survival even after adjusting for accelerometer-measured activity. Lift those weights!

😴 Nine hours of sleep looked as bad as five for dementia. Across 38,816 UK Biobank brains, the Alzheimer's-atrophy curve was a U, not a slope.

🩺 Five proteins predicted fatty liver coming 16 years out. Validated in over 50,000 people across four cohorts in the UK and China.

⏱️ Someone finally tested meal timing properly. Four arms, twelve weeks, assessor-blinded, which almost never happens in this literature.

If you want physician-guided biological age testing, start on the testing page. It includes epigenetic age testing with consultation, plus organ-specific proteomic and SystemAge assessments when they fit the clinical question.

Have a question or topic you want covered? Reply to this email.

Was this forwarded to you? Subscribe here to get The Longevity Letter in your inbox.

Science of Skin Summit, September 17–20, Austin, TX. Speaking on AI in dermatology and skin and hair as windows into biological age.

MVMNT Longevity Summit, September 22–23, Coronado, CA. Leading a session on turning longevity data into clinical action.

Livelong Women's Health Summit, September 25–26, Hilton Midtown, New York City. Joining the expert faculty on AI-driven personalized medicine.

Science of Skin Longevity Summit, February 19–21, 2027, Scottsdale, AZ. Speaking on AI in precision diagnostics and joining the peptides panel.

A few reader-supported partners:

If you want physician-guided biological age testing, start on the testing page. It includes epigenetic age testing with consultation, plus organ-specific proteomic and SystemAge assessments when they fit the clinical question.

Have a question or topic you want covered? Reply to this email.

Was this forwarded to you? Subscribe here to get The Longevity Letter in your inbox.