Home Conditions Practical Health Sleep
📋 Practical Health Series · Article 09

Sleep

One of the most important things you do for your health — and one of the most commonly underestimated. What the evidence actually shows.

📖 9 min read 👨‍⚕️ Dr Paul · Retired NHS GP 📅 April 2026 🔬 Evidence-based
Dr Paul · Retired NHS GP · 20+ years NHS experience
Health education — not medical advice. Anything personally relevant is a conversation for you to have with your GP or healthcare professional.
1

What is healthy sleep?

Sleep is often described as a pause button — the hours lost in order to live the rest of the day. Biologically, it is almost the opposite. Sleep is one of the most active, tightly regulated processes the body performs. Brain tissue is cleared of waste products, memory is consolidated (moved from short-term into long-term storage), hormones that control appetite, blood pressure, and blood sugar are reset, and tissues repair themselves. Cut sleep short, and nearly every one of these systems runs less well the following day.

Healthy sleep has three components, not one. How much you sleep matters — but so does when you sleep, and how well you cycle through the stages of sleep while you are doing it.

How much sleep you need

The NHS recommends 7 to 9 hours per night for most adults. The American Academy of Sleep Medicine and Sleep Research Society agree: adults should sleep 7 or more hours regularly for optimal health.8 Needs change across life stages:

Newborns
14–17
hours / 24h
Toddlers
11–14
hours / 24h
Children
9–12
hours / night
Teens
8–10
hours / night
Adults
7–9
hours / night
Older adults
7–8
hours / night

Eight hours is often treated as the universal target, but it is really an average. What matters more is consistency and how you feel during the day. If you are persistently tired, needing caffeine to function, or falling asleep unintentionally when still, you are probably not getting enough good-quality sleep.

Sleep architecture — the four stages

A healthy night is a series of cycles, each lasting roughly 90 minutes. Most adults go through 4 to 6 cycles per night, moving through four different stages of sleep, each with distinct functions:

N1
~5% of night
Light sleep — the doorway
The transition from wakefulness to sleep. Easy to wake from. Brain activity starts to slow, muscles relax. You may feel a brief falling sensation (a hypnic jerk — harmless and very common).
N2
~45% of night
Light sleep — the bulk of the night
The majority of sleep time. Body temperature drops, heart rate slows. Important for memory consolidation (moving new information from short-term into long-term storage) and learning.
N3
~20% of night
Deep sleep — slow-wave sleep
The deepest, most restorative stage. Growth hormone is released, tissues repair, the immune system resets, and the brain clears metabolic waste products. Most deep sleep happens in the first half of the night.
REM
~25% of night
Dreaming sleep — rapid eye movement
The brain is almost as active as when awake, but the body is temporarily paralysed (REM atonia — a safety feature that stops you acting out dreams). Critical for emotional processing, learning, and creativity. Most REM sleep happens in the second half of the night — which is why cutting sleep short hits REM hardest.

This is why quantity without quality does not count. Someone who sleeps seven hours but has untreated sleep apnoea (pauses in breathing during sleep) will never reach enough deep or REM sleep, and will wake unrefreshed — even though the clock says they slept long enough.

The J-curve — why more isn't always better

📊
Both too little AND too much sleep are associated with increased risk
Large meta-analyses of millions of people consistently show a J-shaped or U-shaped curve for sleep duration and health outcomes: lowest risk sits around 7 hours per night, with risk rising on either side. Persistent short sleep (under 6 hours) is linked to cardiovascular disease, type 2 diabetes, and dementia. Persistent long sleep (over 9 hours) is associated with higher mortality too — though here the long sleep is often a marker of underlying illness rather than the cause itself.

Finally — chronotype matters (your genetic tendency to be a morning lark or a night owl). Around half of individual variation in sleep timing is heritable. ⚑ Forcing a natural night owl into an early schedule produces chronic sleep restriction even if the total hours look fine on paper.

📖 Key Terms
CBT-I
Cognitive Behavioural Therapy for Insomnia — a structured psychological treatment, usually 5–8 sessions. NICE first-line treatment for chronic insomnia in adults of any age.
Chronotype
Your natural tendency toward being a morning person ("lark") or evening person ("owl"). Partly genetic, partly age-related.
Circadian rhythm
The body's internal 24-hour clock, governed by a small region of the brain (the suprachiasmatic nucleus) and synchronised primarily by daylight.
Insomnia
Persistent difficulty falling asleep, staying asleep, or waking unrefreshed — causing daytime impairment. Short-term (under 3 months) or long-term (3 months or more).
Melatonin
A hormone released by the brain (pineal gland) in response to darkness, helping signal the body that it is time to sleep.
REM sleep
Rapid Eye Movement sleep — the dreaming stage. Brain highly active, body temporarily paralysed. Critical for emotional processing and memory.
Sleep apnoea (obstructive)
Repeated pauses in breathing during sleep caused by the airway narrowing or closing. Common, often undiagnosed. Treatable with CPAP (continuous positive airway pressure) — a mask delivering pressurised air overnight.
Sleep hygiene
Everyday behaviours that support sleep — regular schedule, dark cool bedroom, limiting caffeine, winding down before bed. Helpful, but not on its own a treatment for chronic insomnia.
Slow-wave sleep
Also called N3 or deep sleep — the deepest, most restorative stage. Key for physical repair and waste clearance in the brain.
Z-drugs
Non-benzodiazepine sleeping tablets — zopiclone (Zimovane), zolpidem (Stilnoct), zaleplon. Work similarly to benzodiazepines. Licensed for short-term use only.
2

Why does it matter?

Sleep is often treated as a lifestyle preference — nice if you can get it, not essential. The evidence tells a different story. Persistent poor sleep affects nearly every major body system, and the consequences compound over decades.

~7h
the lowest-risk sleep duration for cardiovascular disease, stroke, and all-cause mortality — confirmed across a dose-response meta-analysis of 3.5 million people1
Yin et al., JAHA 2017
+30%
higher risk of developing dementia with persistent short sleep (≤6h) starting from midlife — Whitehall II study, 7,959 participants followed for 25 years4
Sabia et al., Nature Comms 2021
+12%
higher all-cause mortality with short sleep (<7h) — and +30% with long sleep (>8h) — both independent of other risk factors in 16 cohort studies2
Cappuccio et al., SLEEP 2010
1st line
CBT-I (Cognitive Behavioural Therapy for Insomnia) — NICE's first-line treatment for chronic insomnia; effects persist long after treatment ends, unlike sleeping tablets5
van Straten et al., SMR 2018

What makes sleep unusual as a health behaviour is its upstream position. Diet, exercise, and stress responses all depend on how well you slept the night before. When sleep improves, several other things tend to improve with it — almost free of charge.

3

What your doctor might do

Sleep problems are one of the most common reasons people see their GP. Here is what a more thorough assessment looks like, and what NHS treatment pathways are available.

📔
A sleep diary — usually the first step
For at least two weeks, recording when you go to bed, roughly how long it took to fall asleep, any night-time waking, when you got up, and how rested you felt. This is simple but powerful — it often reveals patterns (erratic bedtimes, napping, worry-driven waking) that are not obvious from memory alone.
🫁
Screening for sleep apnoea
Obstructive sleep apnoea is extraordinarily common — recent UK estimates suggest around 1 in 5 adults may have it, and the large majority are undiagnosed. ⚑ Classic features are loud snoring, witnessed pauses in breathing, gasping or choking at night, and persistent daytime sleepiness. GPs may use the Epworth Sleepiness Scale or STOP-Bang questionnaire and refer for a sleep study if screening is positive. Treatment — most commonly CPAP (continuous positive airway pressure) — can be genuinely life-changing.
🧠
CBT-I — the NHS first-line treatment for chronic insomnia
Cognitive Behavioural Therapy for Insomnia (CBT-I) is recommended by NICE as the first-line treatment for chronic insomnia in adults of any age.6 It combines sleep restriction (spending only the hours you actually sleep in bed), stimulus control (breaking the association between bed and wakefulness), and cognitive work on worries about sleep. It works at least as well as sleeping tablets short-term and much better in the long term. Digital CBT-I programmes (including NHS-approved apps) are increasingly available and also effective. ⚑
💊
Sleeping tablets — the honest picture
Benzodiazepines and Z-drugs (zopiclone/Zimovane, zolpidem/Stilnoct) are licensed by NICE for short-term use only — typically up to 4 weeks.7 They work quickly but carry meaningful downsides: tolerance, dependence, rebound insomnia on stopping, and significantly increased risk of falls and fractures, particularly in older adults. They have a place for a short crisis, but are not a solution for long-term insomnia. A gradual taper supported by a GP or pharmacist is a well-established clinical route where long-term use has developed — this is a routine situation in practice.
🌙
Special situations — menopause, shift work, older age
Sleep difficulties around the menopause (especially frequent night-time waking with night sweats) are extremely common and worth raising specifically — hormone replacement therapy (HRT) may help if other menopausal symptoms are present. Shift workers have specific challenges (circadian misalignment, social disruption) and may benefit from structured scheduling advice. In older adults, sleep patterns naturally change — more fragmented and lighter — and clinical focus typically shifts toward supporting quality rather than chasing the same totals as in younger life.
🚩 Clinically time-sensitive features
  • Loud snoring with witnessed pauses in breathing, or waking gasping or choking
  • Persistent daytime sleepiness — falling asleep during meetings, while reading, or (dangerously) while driving
  • Persistent sleep problems affecting daily life — mood, concentration, work, or relationships
  • Low mood, anxiety, or thoughts of self-harm alongside poor sleep
  • New severe morning headaches, or waking with a racing heart
  • Acting out dreams physically (shouting, kicking, jumping out of bed) — can indicate REM sleep behaviour disorder, which warrants assessment
4

What the research shows

Five core pieces of evidence anchor most of what is known — covering mortality, cardiovascular disease, type 2 diabetes, dementia, and the best treatment for chronic insomnia.

~7h the lowest-risk sleep duration across cardiovascular disease, coronary heart disease, stroke, and all-cause mortality — consistent J-shaped relationship across 3.5 million people Yin et al · JAHA 2017
The J-curve at scale — 7 hours is where the cardiovascular data converges

Who was studied: A dose-response meta-analysis pulling together prospective cohort studies of generally healthy populations — more than 3.5 million people followed for multiple years, examining sleep duration against four outcomes: all-cause mortality, total cardiovascular disease, coronary heart disease, and stroke.

Key finding: A consistent U-shaped (J-shaped) relationship for all four outcomes — lowest risk at approximately 7 hours per day, with risk rising on both sides. The pattern held across men and women.

Why it matters: This is among the strongest evidence that 7 hours is not just a rule of thumb — it is where the combined cardiovascular and mortality risk is lowest across very large populations.

+12% / +30% increased all-cause mortality with short sleep (<7h) and long sleep (>8h) respectively — both independent of other risk factors, in 16 prospective studies Cappuccio et al · SLEEP 2010
Both short and long sleep independently predict higher mortality

Who was studied: A systematic review and meta-analysis by Francesco Cappuccio's group at the University of Warwick, pooling 16 prospective studies following 1.38 million people for longer than 3 years, assessing sleep duration at baseline and all-cause mortality.

Key finding: Short sleepers (under 7 hours) showed a 12% increased risk of death (RR 1.12, 95% CI 1.06–1.18). Long sleepers (over 8 hours) showed a 30% increased risk (RR 1.30, 95% CI 1.22–1.38) — although much of the long-sleep signal likely reflects underlying illness driving the longer sleep, rather than sleep itself causing harm.

Why it matters: One of the first large meta-analyses establishing that sleep duration independently tracks with mortality risk in the general population.

+9% / +14% higher type 2 diabetes risk per each hour below 7h (+9%) and per each hour above 8h (+14%) — a U-shaped dose-response across 482,502 participants Shan et al · Diabetes Care 2015
Sleep duration and type 2 diabetes — the U-shaped dose-response

Who was studied: 10 prospective studies, 482,502 participants, with 18,443 new cases of type 2 diabetes occurring over follow-up periods of 2.5 to 16 years.

Key finding: A U-shaped dose-response relationship. Compared with 7 hours per night, each hour less sleep below that was associated with a 9% increase in type 2 diabetes risk, and each hour more above 8 hours was associated with a 14% increase.

Why it matters: Sleep deprivation has direct effects on insulin sensitivity and appetite-regulating hormones. This meta-analysis shows those short-term effects translate into long-term disease risk at population scale.

+30% higher dementia risk with persistent short sleep (≤6h) from midlife — the longest-follow-up cohort study on sleep and dementia, 25 years Sabia et al · Nature Comms 2021
Midlife short sleep and dementia — the Whitehall II 25-year finding

Who was studied: 7,959 participants in the Whitehall II study — a long-running UK cohort of British civil servants — followed for 25 years, with sleep duration recorded at ages 50, 60, and 70, and dementia diagnoses captured through NHS records.

Key finding: People who consistently slept 6 hours or less per night in midlife had approximately 30% higher risk of developing dementia over the following 25 years — independently of cardiovascular risk factors, mental health, and socioeconomic status. Hazard ratios were 1.22 at age 50 and 1.37 at age 60.

Why it matters: The 25-year follow-up rules out the possibility that early brain changes were simply disrupting sleep. The study adds weight to the idea that chronically short sleep in midlife may be a contributor to later dementia, not only a symptom.

Lasting improvement in sleep — CBT-I effects persist long after treatment ends, unlike sleeping tablets, across 87 randomised controlled trials and 3,724 patients van Straten et al · SMR 2018
CBT-I — the evidence base for NICE's first-line recommendation

What was studied: A meta-analysis pooling 87 randomised controlled trials (118 separate treatment arms, 3,724 CBT-I patients vs 2,579 controls) comparing CBT-I against non-treated control groups.

Key finding: CBT-I produced consistent improvements across the core insomnia outcomes — how long it takes to fall asleep, how much time is spent awake during the night, and subjective sleep quality. Effects were robust across age groups, and whether or not participants were taking sleep medication.

Why it matters: This is the evidence base behind NICE's recommendation of CBT-I as first-line treatment for chronic insomnia. Unlike sleeping tablets, the benefits persist long after the treatment ends — because CBT-I works by changing the behaviours and thinking patterns that maintain insomnia, rather than by sedating the brain.

Putting it all together

Sleep sits alongside diet and exercise as one of the three genuine pillars of practical health — yet it is usually the first to get squeezed, and the last to be taken seriously. The evidence is now unambiguous: persistent short sleep is associated with higher risk of cardiovascular disease, type 2 diabetes, dementia, depression, and earlier death. Long sleep is not the answer either — the data point to a J-shaped relationship, with the lowest risk sitting around 7 hours for most adults.

The encouraging news is that much of what causes chronic poor sleep is treatable. Obstructive sleep apnoea is extremely common, widely under-diagnosed, and genuinely life-changing when treated. Chronic insomnia responds well to CBT-I — which works better than sleeping tablets in the long term and is now the NHS first-line recommendation.

For anyone whose sleep is genuinely disrupted, the most useful starting point is a two-week sleep diary and a conversation for you to have with your GP or healthcare professional — with a willingness to look for treatable causes rather than simply accepting poor sleep as inevitable. Good sleep is not a luxury. It is medicine you make yourself, every night.

About the author — Dr Paul spent over twenty years as an NHS GP before retiring in 2019. helf.school exists to give every person access to clear, honest, evidence-based health education. Read more about Dr Paul →

References

1 — Sleep duration, cardiovascular events & all-cause mortality
Yin J, Jin X, Shan Z, et al. Relationship of sleep duration with all-cause mortality and cardiovascular events: a systematic review and dose-response meta-analysis of prospective cohort studies. Journal of the American Heart Association. 2017;6(9):e005947. DOI: 10.1161/JAHA.117.005947
JAHA Full Text ✓PubMed ✓
2 — Sleep duration & all-cause mortality meta-analysis
Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. SLEEP. 2010;33(5):585–592. DOI: 10.1093/sleep/33.5.585
SLEEP Full Text ✓PubMed ✓
3 — Sleep duration & risk of type 2 diabetes
Shan Z, Ma H, Xie M, et al. Sleep duration and risk of type 2 diabetes: a meta-analysis of prospective studies. Diabetes Care. 2015;38(3):529–537. DOI: 10.2337/dc14-2073
Diabetes Care Full Text ✓PubMed ✓
4 — Sleep duration in midlife & dementia (Whitehall II)
Sabia S, Fayosse A, Dumurgier J, et al. Association of sleep duration in middle and old age with incidence of dementia. Nature Communications. 2021;12(1):2289. DOI: 10.1038/s41467-021-22354-2
Nature Comms Full Text ✓PubMed ✓
5 — CBT-I meta-analysis
van Straten A, van der Zweerde T, Kleiboer A, et al. Cognitive and behavioral therapies in the treatment of insomnia: a meta-analysis. Sleep Medicine Reviews. 2018;38:3–16. DOI: 10.1016/j.smrv.2017.02.001
SMR Full Text ✓PubMed ✓
6 — NICE guidance · CBT-I first-line for chronic insomnia
NICE Clinical Knowledge Summary — Insomnia. CBT-I recommended as first-line treatment for chronic insomnia in adults of any age. Last revised 2025.
NICE CKS · Insomnia
7 — NICE guidance · short-term hypnotics
National Institute for Health and Care Excellence (NICE). Guidance on the use of zaleplon, zolpidem and zopiclone for the short-term management of insomnia. Technology appraisal guidance [TA77]. 2004.
NICE TA77
8 — Recommended sleep duration · Joint Consensus Statement
Watson NF, Badr MS, Belenky G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. SLEEP. 2015;38(6):843–844.
PMC Full Text ✓

This article is for health education only. It is not a substitute for medical advice, diagnosis, or treatment. Anything personally relevant is a conversation for you to have with your GP or healthcare professional.