> ## Content Index
> Fetch the complete content index at: https://medical-archaeology.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# The sleep myth that ends in a broken hip
- URL: https://medical-archaeology.ghost.io/the-sleep-myth-that-ends-in-a-broken-hip/
- Published: 2026-08-10T15:21:07.000Z
- Updated: 2026-08-10T15:21:07.000Z
- Author: Amanda Jones

On June 4, 1938, Nathaniel Kleitman walked into Mammoth Cave, Kentucky, and didn't come back out for thirty-two days.

He brought a graduate student named Bruce Richardson, two lanterns, a table, and a bunk bed carried down from the hotel upstairs. Their room was a rock chamber roughly a hundred and forty feet underground. Fifty-four degrees, year-round. No daylight, no weather, no seasons. [A waiter from the Mammoth Cave Hotel came down with lunch and the mail.](https://www.nps.gov/articles/000/dreaming-underground-the-mammoth-cave-sleep-study.htm?ref=medical-archaeology.ghost.io)

The plan was to live on a 28-hour day. Nine hours asleep, ten working, nine more awake. [Six long days to the calendar week.](https://time.com/archive/6759335/science-cave-men/?ref=medical-archaeology.ghost.io) They took their temperatures every two hours and wrote everything down by lantern.

Richardson was in his twenties. Inside a week he'd more or less moved onto the new schedule — he slept when he was told to, and his temperature curve stretched to match.

Kleitman was forty-three. His body refused. His temperature cycle stayed near twenty-four hours no matter what the lanterns said.

And that difference is the entire reason you're reading this.

Because look at what actually happened down there. The older man's clock was harder to move. That's it. That's the finding. Nobody gave him less sleep. Nobody suggested he needed less. Both men were scheduled the same nine hours, and the only thing age changed was how hard it was to shift the timing of them.

Somewhere between that cave and your kitchen, the finding got rewritten.

Now it comes out of people's mouths as a fact. Older bodies need less sleep. You've heard it from a friend, maybe from a nurse, possibly from a doctor who was running forty minutes behind. And if you've been awake at three in the morning for six years, you may have quietly decided this is simply what your body does now.

It isn't.

Part 01 · Paper trail 

### How a complaint became a prescription — Matthew's part

In 1685, an English physician named [Thomas Willis](https://www.medlink.com/articles/restless-legs-syndrome?ref=medical-archaeology.ghost.io) published a description of patients who couldn't lie still at bedtime. Arms and legs jumping. Tendons contracting. He wrote that these people could no more sleep than if they'd been put in a place of torture.

> "…so great a Restlessness and Tossings of their Members…"  
> — Thomas Willis, 1685

He got it right the first time. Then medicine spent two and a half centuries getting it wrong.

In 1861 a physician named Theodor Wittmaack gave it a Latin name — *anxietas tibiarum*, anxiety of the shins — and filed it under hysteria. Which is a small thing on paper and an enormous thing in a consulting room. Once a symptom belongs to the nerves, nobody examines the legs.

Hysteria.

And there was always something on the shelf to calm a patient with.

It took until 1945 for a Swedish neurologist, Karl-Axel Ekbom, to write the thing up properly — eight cases, in *Acta Medica Scandinavica*, [under the name it still carries](https://www.nejm.org/doi/10.1056/NEJMe1313155?ref=medical-archaeology.ghost.io). Restless legs. Two hundred and sixty years after Willis, the complaint finally had a diagnosis.

But the sedatives had a head start, and they never lost it.

[Wallace Laboratories put Miltown on the American market in the mid-fifties.](https://www.washingtonpost.com/archive/politics/2005/10/01/inventor-of-valium-once-the-most-often-prescribed-drug-dies/2dc3ff01-d57d-4042-987b-299442d949ce/?ref=medical-archaeology.ghost.io) Librium came next. Then Valium, introduced in the United States late in 1963 — and [the most prescribed drug in the country from 1969 to 1982, with more than 2.3 billion doses sold in its peak year of 1978 alone](https://www.sciencedirect.com/science/article/pii/S104366182400255X?ref=medical-archaeology.ghost.io).

On September 10, 1979, a Senate health subcommittee chaired by Edward Kennedy heard a Navy addiction specialist testify that [he'd seen people become dependent on Valium in six weeks](https://digital.bentley.umich.edu/midaily/mdp.39015071754373/156/download%5Ftext?ref=medical-archaeology.ghost.io).

Thirteen years later, in December 1992, [the FDA approved zolpidem](https://www.accessdata.fda.gov/drugsatfda%5Fdocs/nda/pre96/019908%5Fs000%5Fambientoc.cfm?ref=medical-archaeology.ghost.io), and the shelf restocked.

So there's the paper trail, end to end. A specific, physical, perfectly describable complaint gets written down in 1685\. It gets reclassified as nervousness in 1861\. It gets its real name in 1945 — by which point there's a whole industry built on helping people who can't sleep stop mentioning it.

Nobody had to suppress the diagnosis. They simply never needed to reach for it.

Part 02 · Clinical picture 

### What the evidence actually says — Logan's part

Two things get welded together here, and once you pry them apart the picture changes completely.

The need doesn't fall. The ability does.

[The National Sleep Foundation's recommendation for adults 65 and older is seven to eight hours](https://www.sleephealthjournal.org/article/s2352-7218%2815%2900015-7/fulltext?ref=medical-archaeology.ghost.io) — effectively the same as for a forty-year-old. What shifts with age is how well you hold those hours: lighter stages, more awakenings, an internal clock that runs earlier. [The NSF states it plainly — the need for sleep doesn't decrease with age, the ability to sleep can.](https://www.sleephealthjournal.org/article/S2352-7218%2823%2900196-1/fulltext?ref=medical-archaeology.ghost.io)

And here's the part that surprises people. Most of that decline is finished by about sixty.

The largest analysis we have — [Ohayon and colleagues, 65 studies, 3,577 healthy people aged five to a hundred and two](https://academic.oup.com/sleep/article-abstract/27/7/1255/2696819?ref=medical-archaeology.ghost.io) — found that past sixty, sleep efficiency is essentially the only measure still significantly sliding. So if your sleep fell apart at seventy-one, the calendar isn't the explanation. Something else changed.

Daytime sleepiness is the tell. [It isn't a normal feature of ageing](https://www.mdpi.com/2076-3425/11/8/1003?ref=medical-archaeology.ghost.io). If you're fighting to stay awake at two in the afternoon, that's a finding, not a birthday.

Sleep apnea accounts for a lot of it. So do the legs. [Restless legs runs somewhere between three and ten percent of adults in North America and Western Europe, rises with age, and is badly under-recognised in primary care](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3056753/?ref=medical-archaeology.ghost.io) — partly because people don't link a leg sensation to a sleep problem. They report the insomnia. They never mention the legs.

> 15.8%of adults 65 and older use a sleep aid most days or every day

[The AASM rewrote its guideline in January 2025](https://jcsm.aasm.org/doi/abs/10.5664/jcsm.11390?ref=medical-archaeology.ghost.io), and the headline is iron. Everyone with clinically significant restless legs should have ferritin and transferrin saturation measured. Oral iron when ferritin sits at or below 75 µg/L, or transferrin saturation under twenty percent. Note that number: 75\. Plenty of labs will call a ferritin of 40 normal. The same guideline pushed dopamine agonists off standard first-line use — long-term they cause augmentation, where symptoms intensify and creep earlier into the day — and moved gabapentin enacarbil, gabapentin and pregabalin up in their place.

Now the sedatives. [The 2023 Beers Criteria say to avoid benzodiazepines and the Z-drugs — zolpidem, eszopiclone, zaleplon — in everyone 65 and over](https://health.uconn.edu/pharmacy/wp-content/uploads/sites/60/2023/11/2023-American-Geriatrics-Society-Beers-Criteria-1.pdf?ref=medical-archaeology.ghost.io), for cognitive impairment, delirium, falls and fractures. [A 2018 meta-analysis in *Age and Ageing* found Z-drug use carried roughly 63 percent higher odds of fracture, and across 160,502 people, about double the odds of injury with zolpidem.](https://academic.oup.com/ageing/article/47/2/201/4564456?ref=medical-archaeology.ghost.io) The falls signal in that same paper pointed the same direction but missed significance and the studies disagreed sharply with one another, so I'll call that one unresolved. The fracture and injury findings aren't.

[In April 2019 the FDA added a boxed warning to all three](https://www.fda.gov/drugs/drug-safety-and-availability/fda-adds-boxed-warning-risk-serious-injuries-caused-sleepwalking-certain-prescription-insomnia?ref=medical-archaeology.ghost.io), after 66 reported cases of sleepwalking, sleep-driving and worse — including deaths, at the lowest recommended dose, after a single tablet.

What the guidelines want instead: [the American College of Physicians recommends CBT-I as the initial treatment for chronic insomnia in every adult](https://www.acpjournals.org/doi/10.7326/M15-2175?ref=medical-archaeology.ghost.io). Strong recommendation.

One thing that isn't optional. If you're taking a benzodiazepine now, don't stop it on your own — abrupt withdrawal can cause seizures. That's a conversation with your prescriber and a written taper, not a decision you make on a Sunday night.

Start with the blood. Before you accept any explanation beginning with "at your age," ask for ferritin and transferrin saturation, and ask for the actual figures rather than the word *normal*. Seventy-five is the threshold the sleep guideline works to, and your lab's normal range starts a long way underneath it — which means you can be told you're fine while sitting squarely in the band where iron would help. Write both numbers down and keep them.

Then get precise about what's happening at night, because "I don't sleep well" is too blunt to act on. There's a real difference between not falling asleep, not staying asleep, and lying there with legs that won't settle until you get up and walk to the kitchen. That last one has a name and a treatment pathway attached to it. Say the words out loud at the appointment — the urge to move, worse in the evening, better once you're standing. Those are the words that get you assessed rather than reassured. And ask directly whether you've ever been screened for sleep apnea, because snoring plus afternoon sleepiness is a different road entirely.

Ask for CBT-I by name. Not sleep hygiene — that's one small component that gets handed out as a leaflet. Cognitive behavioural therapy for insomnia is a structured course, usually four to eight sessions, built around sleep restriction and stimulus control. It's what the ACP puts first for every adult, and it can be delivered in primary care or through a digital programme if there's no clinician nearby. Ask for the referral. The request travels further than the complaint does.

Our teacher was up at four most mornings. She graded at the kitchen table with the radio down low, and when we asked about it she said it was the best part of her day, and that people her age didn't need much anyway. She said it the way you say something you've heard often enough to stop testing.

Nobody ever ordered a blood test. Nobody ever asked about her legs.

Kleitman went a hundred and forty feet underground with a lantern, a thermometer and a sample size of two, and he had an excuse for guessing. The blood draw at your next appointment doesn't.

See you!

Amanda, Logan, Matthew

| [Click here to learn more](https://www.nps.gov/articles/000/dreaming-underground-the-mammoth-cave-sleep-study.htm?ref=medical-archaeology.ghost.io) |
| --------------------------------------------------------------------------------------------------------------------------------------------------- |