So I had a patient walk in last month. Let's call him R. He's 43, runs half marathons, eats clean, takes about 15 different supplements every morning. The guy is disciplined. Seriously impressive routine.
We ran his biological age test. It came back at 48.
Five years older than his actual age. He was gutted. "Doc, I don't understand. I'm doing everything right."
First question I asked him. "How's your sleep?"
He laughed. "Sleep? I get about five hours. Sometimes less. I just don't need that much."
I hear this all the time. And I get it. In Singapore especially, sleeping less feels like a flex. You're grinding. You're hustling. You're "optimising" your time. But the research that's come out in the past few months has completely changed how we should be thinking about this. And honestly, it surprised even me.
Sleep doesn't just make you feel better. It predicts how long you'll live more strongly than your diet or your exercise routine.
Let me show you the data. Then I'll tell you what actually happened with R.
Sleep Beats Diet and Exercise. Yes, Really.
A team from Oregon Health and Science University looked at sleep data across more than 3,000 US counties. They used CDC survey data collected over six years, from 2019 to 2025, and compared counties where people weren't getting enough sleep (less than 7 hours) against life expectancy numbers.
Here's what they found.
By 2024, every single US state showed a statistically significant link between not sleeping enough and dying earlier. Every. Single. One.
Now look. I'm not saying exercise and nutrition don't matter. They absolutely do. But think about how most people approach their health. They've got a gym routine. They've got a supplement stack. They've got strong opinions about seed oils. But ask them about their sleep? They shrug. "It's fine."
It's usually not fine.
In Singapore, a 2024 YouGov survey found that only about 1 in 4 of us actually get 7 or more hours of sleep per night. But here's the ironic part: 80% said they wished they could sleep more. So we know it matters. We just don't treat it like it matters.
Your Brain Is Ageing Faster Than You Think
If the lifespan stuff doesn't get your attention, maybe this will.
Researchers at the Karolinska Institutet in Stockholm studied 27,500 adults. They used machine learning on over 1,000 brain MRI features to estimate something called "brain age." Basically, how old does your brain look compared to how old you actually are?
They scored everyone's sleep on five things: sleep timing (chronotype), whether they got 7 to 8 hours, insomnia symptoms, snoring, and daytime sleepiness. Simple 0 to 5 scale.
Poor Sleepers (score 0-1)
Brains looked 0.46 years older than their actual age on average.
That might not sound like much from a single measurement. But this is a snapshot. Imagine that accumulating over 10, 15, 20 years of bad sleep.
Healthy Sleepers (score 4-5)
Brain age matched or was younger than chronological age.
Good sleep doesn't just prevent damage. It actively protects your brain.
And here's the part that really got me. The researchers dug into why this happens. Turns out, about 10.4% of the connection between bad sleep and older-looking brains is explained by chronic inflammation. Specifically, markers like C-reactive protein and white blood cell counts that we check in our biological age panels.
Think of it this way. Every night of bad sleep is a small deposit into an inflammation bank account. You don't notice the balance growing. But it's compounding. By the time you feel the effects, the fog, the forgetfulness, the slower processing, years of damage have already built up.
This is exactly what we saw with R. His inflammatory markers were elevated. Not dramatically. Not "something is wrong" elevated. More like a slow burn that doesn't set off any alarms on a standard blood test. But on a biological age test, it shows up clearly.
It's Not Just About How Long You Sleep
OK so this next study is the one that really changed how I talk to my patients about sleep.
A team at Monash University tracked over 60,000 people using wrist accelerometers. That's 10 million hours of real sleep data. Not self-reported "I think I sleep 7 hours" data. Actual measured data.
They looked at two things. How long people slept. And how regular their sleep patterns were.
Read that again. Regularity mattered more than duration.
So what does that mean for you? It means going to bed at 11pm one night, 1am the next, then 10pm the night after is probably doing more damage than you think. Even if the total hours add up to "enough."
Your body runs on a clock. Hormones, immune function, DNA repair, glucose metabolism. All of it follows a rhythm. And every time you shift your sleep window around, you're basically giving your body jet lag.
I see this constantly with my patients here in Singapore. Especially the ones who say they "catch up" on weekends. Sleep 5 hours Monday to Friday, then 9 hours on Saturday. The data suggests a consistent 7 is better than that roller coaster.
The Magic Numbers
OK so you're probably thinking: just tell me what to aim for.
A massive study from the University of Sydney, published in January 2026, gives us the clearest answer yet. They looked at 59,078 people and measured the combined impact of sleep, exercise, and diet on both lifespan AND healthspan (meaning years lived without heart disease, cancer, diabetes, COPD, or dementia).
But here's the part I love. Because most people hear "9 extra years" and think, "Great, but I can't overhaul my entire life tomorrow."
You don't have to.
The same study calculated the minimum combined improvement needed for just one extra year of life. It was tiny. Five extra minutes of sleep per night. Two extra minutes of moderate activity. And half a serving more vegetables per day. That's it. One extra year. For changes you'd barely notice.
You don't need a perfect routine. You need a slightly better one. But sleep has to be part of the equation. Without it, the exercise and diet gains start losing their edge.
What to Actually Do About It
Here's what I tell my patients. Not the generic "sleep hygiene" stuff you've read a hundred times. Actual things that work.
Anchor your wake time, not your bedtime
Pick a wake time and stick to it every day. Yes, weekends too. Keep it within a 30 minute window. Your circadian clock is set by morning light and wake time. Lock that in and your body will start feeling sleepy at the right time in the evening on its own. This is the single best thing you can do for sleep regularity.
Get outside within 30 minutes of waking
Sunlight suppresses melatonin, raises cortisol (the good kind), and resets your internal clock. We're in Singapore. Morning light is free and abundant. Even 10 minutes helps. If you wake before dawn, a 10,000 lux light therapy box for 20 minutes does the same thing.
Set a caffeine curfew at 2pm
Caffeine has a half life of about 5 to 6 hours. But it varies by your genetics (something we can actually test with genomic profiling). If you're a slow metaboliser and many people are without knowing it, that 3pm kopi is still floating around your system at midnight. Just cut it off at 2pm. Simple.
Cool your bedroom to 24-25°C
Your core body temperature needs to drop by about 1 degree to fall asleep. In Singapore's humidity, that means aircon or a good fan is not a luxury. It's biology. 24 to 25 degrees Celsius is the sweet spot for us in the tropics (it's the equivalent of the 18 to 20 degrees recommended in temperate climates).
Use the 10-3-2-1-0 rule
My patients remember this one. 10 hours before bed: no more caffeine. 3 hours before: no more alcohol or big meals. 2 hours before: no more work. 1 hour before: no more screens. 0: the number of times you hit snooze in the morning.
Track it
An Oura ring, Whoop, or Apple Watch gives you data on your regularity, duration, and sleep stages. You don't need a sleep lab. You just need consistent tracking so you can spot trends. I review my patients' wearable data as part of their assessments and the patterns are often a wake up call. No pun intended.
What Happened With R.
So back to R. The 43 year old marathon runner with a biological age of 48.
We didn't change his exercise. We didn't touch his supplements. We didn't overhaul his diet. The only thing we focused on for the first three months was sleep.
He set a consistent wake time. Got morning light. Moved his last coffee to 1pm. Started tracking with a wearable. Nothing fancy.
Three months later, we re-tested. His biological age dropped to 44. Not all the way to 43 yet. But a four year improvement from one change. His inflammatory markers came down. His metabolic numbers improved. He said he felt sharper at work and recovered faster from his runs.
One change. Sleep.
If you're investing in your health but sleeping badly, you're building a house on a cracked foundation. Fix the foundation first.
Sleep is not a passive thing that happens when you run out of stuff to do. It's when your body repairs DNA, clears waste from your brain, consolidates memory, restores immune function, and regulates blood sugar. It is the most active recovery process you have. And it's almost entirely within your control.
Start tonight.
And if you want to know exactly how your sleep is affecting your biological age and organ health, a comprehensive longevity assessment can give you those answers with real data. Not guesswork.
- McAuliffe KE, Wary MR, Pleas GV, et al. Sleep insufficiency and life expectancy at the state-county level in the United States, 2019-2025. SLEEP Advances. 2025;6(4):zpaf090. doi:10.1093/sleepadvances/zpaf090
- Miao Y, Wang J, Li X, et al. Poor sleep health is associated with older brain age: the role of systemic inflammation. eBioMedicine. 2025;120:105941. doi:10.1016/j.ebiom.2025.105941
- Windred DP, Burns AC, Lane JM, et al. Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study. SLEEP. 2024;47(1):zsad253. doi:10.1093/sleep/zsad253
- Koemel NA, Biswas RK, Ahmadi MN, et al. Minimum combined sleep, physical activity, and nutrition variations associated with lifeSPAN and healthSPAN improvements: a population cohort study. eClinicalMedicine. 2026;92:103741. doi:10.1016/j.eclinm.2025.103741