The first time I woke at four forty-seven in the morning and could not fall back asleep, I was fifty-three years old, newly moved to Madison, and fairly certain something was wrong.
Not wrong in a frightening way. I was not in pain. I was not anxious. I was simply awake, in a way that felt categorical rather than temporary — the way you know you are not going back to sleep, not with willpower or chamomile tea or any further lying there. I had always been a good sleeper. Seven hours, consistent, functional. Now I was lying in the dark at not-quite-five in the morning, listening to a radiator knock, completely and uselessly awake.
This went on. Not every night, but most nights. Four-fifty, four-thirty-eight, occasionally a late morning of five-fifteen. I tried going to bed later, which made me tired the next day and did not move the wake time. I tried a white noise machine. I tried, once, half a glass of wine before bed, which is a thing I will come back to. None of it changed the fundamental fact: my sleep had done something, and I did not understand what.
What I did, eventually, was read the literature. Fourteen years as a nurse gave me the vocabulary and a useful distrust of shortcuts. What I found was not alarming. It was, in fact, the most useful thing anyone could have told me that first morning, and it is what I wish had been in every piece of writing about sleep and aging I had ever encountered.
Here is what actually happens.
The architecture changes
Sleep is not a uniform state. It runs in cycles of roughly ninety minutes each, and within those cycles you move through distinct stages: light sleep, deeper sleep, and what researchers call slow-wave sleep, sometimes called N3 or deep sleep. Slow-wave sleep is the deepest restorative phase. It is when your body does its most significant repair work, when the brain consolidates memory and clears metabolic waste. It is the kind of sleep that makes you feel, in the morning, as though you actually slept.
After about sixty, the proportion of slow-wave sleep in each cycle decreases. This is not a malfunction. It is a physiological change that occurs consistently across the aging population, documented across decades of research. You still cycle through sleep stages. You simply spend less time in the deepest stage, and your overall sleep becomes lighter and more fragmented. You wake more easily. Noise that would not have disturbed you at forty does now. A full bladder that you would have slept through at forty-five surfaces you at three in the morning.
This is the mechanism behind the phrase “I’m just a light sleeper now” that you hear from almost every person in their sixties and seventies. It is accurate. It is also not a sign of illness.
The clock shifts forward
The second change is circadian. Your body runs on a roughly twenty-four-hour internal clock that regulates when you feel sleepy and when you feel alert. This clock is influenced by light, by temperature, by when you eat and exercise, and by a cluster of neurons in the brain called the suprachiasmatic nucleus, which processes light information from the eyes.
After sixty, this clock tends to shift earlier. The technical term is circadian phase advance. What it means practically is that your body begins producing melatonin, the hormone that signals darkness and prepares you for sleep, earlier in the evening. You get drowsy at nine instead of eleven. And because your sleep onset has moved earlier, your wake time moves earlier too. Four-thirty. Five o’clock. Five-fifteen. Not because you have slept badly, but because, biologically, you are done. Your brain considers sunrise somewhere around that hour and begins signaling alertness accordingly.
This is why going to bed at your old time, hoping to sleep until your old time, frequently does not work. The math no longer adds up the way it did at forty-five. Staying up until midnight because you want to sleep until seven does not force the issue. It mostly produces a tired person who is still awake at five.
What this does and doesn’t mean
Let me be precise about this distinction, because it matters.
What I have described above is normal. Not normal as in ideal, not normal as in pleasant, but normal as in consistent with healthy aging, documented across populations, not associated with elevated risk of dementia or cardiovascular disease or shortened life span in and of itself. The sleep architecture of a healthy seventy-year-old is measurably different from the sleep architecture of a healthy thirty-year-old. That difference is not a disease.
What is a disease is chronic insomnia: the persistent inability to fall asleep or stay asleep paired with significant daytime dysfunction. If you are lying awake for long periods most nights, if your daytime functioning is genuinely compromised, if you are exhausted rather than just early-rising, that warrants a conversation with a physician. There is also sleep apnea, which is common in older adults, frequently undiagnosed, and worth ruling out if you snore, if you wake with headaches, or if a bed partner has noticed that you stop breathing during the night. These things are treatable. They are not the same as the normal changes I have been describing, and conflating them is how people end up either dismissing something real or catastrophizing something ordinary.
The four forty-seven problem I was having – lying awake, fully alert, unable to return to sleep – was not a disease. It was a circadian phase advance. My clock had moved earlier and my body was done with sleep before the clock said morning. Understanding this did not make me fall back asleep. But it changed what I did about it.
What the evidence actually supports
The research on sleep in older adults is extensive enough that we have a short list of things that reliably help. Here it is without embellishment.
Morning light exposure. Getting bright light in the first hour after waking anchors your circadian clock. This is not metaphor. Light is the primary signal your suprachiasmatic nucleus uses to set the clock each day. Twenty to thirty minutes of actual outdoor daylight in the morning, not through a window and not from a screen, is the single most effective circadian intervention the research supports. For people dealing with phase advance, doing this deliberately helps stabilize the clock. It will not reverse the shift entirely, but it can moderate it. I run at six in the morning, partly for exercise and partly for this reason, though I would not have articulated it that way when I started.
A consistent sleep schedule, including weekends. This sounds like something printed on a poster in a pediatrician’s office. It works. Your circadian clock learns from when you consistently sleep and wake. When you sleep in on Saturday, you shift the clock slightly later and spend the rest of the week recovering. Older adults are not exempt from this effect. Consistency is what the clock needs. Give it that.
Alcohol before bed, reduced or eliminated. This surprised me more than I expected when I looked at the data. Alcohol at low doses is sedating. It helps people fall asleep faster. It also fragments the second half of sleep significantly, suppressing REM and producing more awakenings in the hours after it is metabolized. If you are already dealing with lighter, more fragmented sleep, alcohol amplifies the problem while the first part of the night feels like it helped. My one-glass-before-bed experiment was the night I woke at three and did not sleep again until nearly six. The data, afterward, made sense.
Regular exercise. Specifically: consistent aerobic activity, even moderate, improves sleep quality in older adults in ways that are measurable and replicated across studies. This does not require a gym. Walking is medicine. I have said this before and will continue to say it because it remains true and people keep expecting a more complicated answer. Thirty minutes most days. The effect is not immediate; it accumulates over weeks. But it is real.
What does not work
Melatonin supplements, taken at the doses sold over the counter, are almost certainly not doing what most people think they are doing.
A pharmacologically relevant dose of melatonin is between 0.3 and 0.5 milligrams. Most commercial supplements sell three, five, or ten milligrams, which is six to thirty times the therapeutic dose. The research on high-dose melatonin in older adults is not strong. At low doses, melatonin may help with circadian shifting in specific circumstances, and there is some evidence for this. But the product on the shelf at your pharmacy, taken at the time most people take it and in the dose it comes in, has not demonstrated meaningful benefit for the sleep architecture changes I have described. Your body produces melatonin on its own. After sixty, that production shifts earlier. Adding large quantities of exogenous melatonin on top of it has not been shown to reliably move the clock back to where it was at forty.
Sedative-hypnotics are a more serious concern. These are the medications commonly called sleep aids: older benzodiazepines and the newer z-drugs, including zolpidem, marketed under various brand names as a prescription and in lower doses over the counter. In older adults, these medications are associated with increased fall risk, next-day cognitive impairment, physical dependence, and in research published over the past decade, association with dementia risk over long-term use, though the causality there is still being studied. They are not appropriate first-line treatment for the sleep changes I have been describing. If you are taking one of these regularly and have been for some time, that is worth raising at your next appointment specifically. Not to alarm you. To make sure it is still the right decision given your current circumstances.
The intervention with the strongest evidence for genuine chronic insomnia is cognitive behavioral therapy for insomnia, abbreviated CBT-I. It is not fast, it requires working with someone trained in it, and it is not the right tool for simple phase advance. But for people with actual chronic insomnia who have not been offered this, asking about it specifically is worthwhile. It outperforms medication for most people over the long term and does not carry the fall risk.
What I eventually did, at fifty-three, was stop fighting the earlier wake time and start adjusting to it. I moved my bedtime earlier to match. I started running at six instead of seven. I stopped looking at my phone in the hour before bed, less because of blue light theory and more because I did not want to fill my brain with information at the moment I was trying to quiet it.
I still wake some mornings before five. I do not lie there trying to force sleep that is not coming. I read. I am usually at my desk by seven-thirty, which is earlier than I would have predicted at forty-five, and which, with time, I have come to find genuinely useful. The mornings are good for writing.
This is not a story about conquering something. My sleep changed and I adjusted. What I want you to take from it is the mechanism, because understanding what your body is actually doing changes the quality of your response to it. You are not broken. Your clock has moved, and your deep sleep has decreased, and these are things that happen to people who are healthy and aging. Work with what is, not against it.
Questions to raise with your doctor if you are concerned about sleep:
- Could my sleep difficulties be related to sleep apnea, and should I be evaluated for it?
- Are any of my current medications known to affect sleep quality or architecture?
- If I have been using a sleep aid regularly, is this still the right approach, and have we discussed alternatives?
Carol Gifford spent fourteen years as a registered nurse, including six in hospital case management. She writes about health and medicine for people who want accurate information without the alarm or the sales pitch. She lives in Madison, Wisconsin.
