Eating to Live to 100.

The same dinner, eaten later, is handled differently. As bedtime approaches, the body begins its shift toward rest — melatonin rising, insulin response softening — and meals eaten close to sleep are associated with higher overnight glucose and, in some studies, more fragmented sleep. Finishing the last bite around three hours before bed gives digestion a head start and lets cortisol hand over to melatonin roughly on schedule. Many of the long-lived communities eat their smallest meal in the late afternoon or early evening — and then simply stop.

Why it mattersEvening glucose handling and sleep quality are two levers you can move with a clock, not a diet.

Overnight, you lose fluid with every breath, and most people wake mildly behind on water. One or two glasses first thing is a gentle way to begin refilling before the day gets busy. If your blood pressure is normal, some clinicians suggest a small pinch of salt or an electrolyte mix, on the reasoning that a little sodium may help the water stay in the body's cells rather than pass straight through — though the evidence here is thin and largely extrapolated.

Important: skip added sodium entirely if you have elevated blood pressure or any kidney or heart condition — and ask your physician before adding salt to anything.

Why it mattersYou start every day mildly dehydrated; this is the least effortful fix in this guide.

The best-supported concern in the evidence isn't fat itself — it's what happens to oils under repeated or very high heat, where oxidation products accumulate; these have been linked, mainly in laboratory and some observational work, to vascular stress. A practical order of preference: whole-food fats first — nuts, seeds, avocado, olives — then extra-virgin olive oil used cold or over gentle heat. Favor unsaturated fats over saturated ones, and minimally processed oils with nothing added. What matters less than any single oil is how often the same oil meets a smoking pan.

Why it mattersThe dose of heat-damaged fat, not the presence of fat, is where the evidence points.

Ultra-processed foods are formulations of industrial ingredients — refined starches, oils, sweeteners, emulsifiers — built more for shelf life and craveability than nourishment. In the most careful test to date, a small NIH crossover trial, people offered ultra-processed meals ate roughly 500 calories a day more than the same people offered unprocessed meals matched for nutrients. Chocolate is a useful case study: cocoa itself is rich in flavanols, but most chocolate is a delivery vehicle for sugar — roughly 10–30% by weight even for dark bars, depending on cacao percentage, and around half for milk chocolate.

Why it mattersThe clearest trial we have suggests these foods make overeating nearly automatic.

Fruits and vegetables carry the vitamins, minerals, and fiber the rest of the modern plate quietly lacks — and the fiber matters as much as anything, feeding the gut bacteria that do real metabolic work. One heuristic worth keeping: if a fruit tastes suspiciously sweet, enjoy it the way you'd enjoy dessert. Some commercial varieties have been bred dramatically sweeter than their ancestors, and your taste buds notice before any label does. None of this is a reason to eat less fruit — only to notice which ones are candy in costume.

Why it mattersFiber and micronutrients are the two shortfalls most Western diets share.

"Breakfast is the most important meal" was always more slogan than science. The steadier claim is mechanical: calories eaten earlier give the body an entire day of movement, work, and warmth in which to use them, while the same calories at nine at night arrive as the metabolism is powering down. Trials of early time-restricted eating lean favorable — better morning insulin sensitivity and blood pressure in some studies — though the literature is still young. And if life forces you to skip a meal, most people appear to do better skipping dinner than breakfast.

Why it mattersThe same calories may land differently at 8 a.m. than at 9 p.m.

For years, moderate drinking wore a health halo. The largest recent analyses have dimmed it: the 2018 Lancet analysis of Global Burden of Disease 2016 data concluded that the level of drinking that minimizes health loss is zero, and in 2023 the WHO stated plainly that no level of alcohol is safe for health. Livers do differ — genetics vary the enzymes that clear alcohol — but for most people, habitual intake runs ahead of that capacity. Yes, some blue-zones communities drink wine socially. The longevity signal there, on the best reading, is the company — not the pour.

Why it mattersIf you drink, let it be for the table you share — not for your health.