Sleep Research Data: 12 Evidence-Based Tables
A canonical collection of original data tables from across our articles — every figure traceable to a peer-reviewed primary source. Use this page as a quick reference for sleep cycles, REM, naps, mortality risk, chronotype, caffeine, and sleep inertia.
1. Sleep Duration & All-Cause Mortality
Cappuccio meta-analysis pooled 16 prospective studies (1.38 million participants) showing a U-shaped relationship between sleep duration and mortality. Both short and long sleepers carry elevated risk relative to the 7–8 hour reference group.
| Sleep duration | Mortality (RR) | Excess risk |
|---|---|---|
| ≤5 hours | 1.12 | +12% |
| 6 hours | 1.07 | +7% |
| 7–8 hours ⭐ | 1.00 | Baseline |
| 9 hours | 1.17 | +17% |
| ≥10 hours ⚠️ | 1.30 | +30% |
Source: Cappuccio et al. (2010), Sleep · Read full analysis: Is 10 hours of sleep too much?
2. Cognitive Decline at 4-Hour Sleep (Penn Sleep Lab)
Van Dongen and colleagues tracked Psychomotor Vigilance Task (PVT) lapses across 14 nights of restricted sleep. Subjects on 4 hours performed like someone awake for 72 hours — but consistently rated themselves only "a bit tired."
| Day | PVT lapses | Equivalent to | Self-rating |
|---|---|---|---|
| Day 1 ⭐ | ~1 | Baseline | "Fine" |
| Day 3 | ~6 | 24h awake | "A bit tired" |
| Day 7 ⚠️ | ~14 | 48h awake (BAC ~0.10%) | "A bit tired" |
| Day 14 ⚠️ | ~22 | 72h awake | "A bit tired" |
Source: Van Dongen et al. (2003), Sleep · Read full analysis: Is 4 hours of sleep enough?
3. Nap Frequency & Cardiovascular Events (Heart, BMJ)
The Lausanne CoLaus cohort followed 3,462 adults for 5.3 years. Napping 1–2 times per week halved cardiovascular event risk vs. non-nappers, after adjusting for traditional risk factors and nighttime sleep duration.
| Nap frequency | Relative risk | Notes |
|---|---|---|
| 0 naps/week | 1.00 (baseline) | Reference group |
| 1–2 naps/week ⭐ | 0.52 (−48%) | Strongest protective effect |
| 3–5 naps/week | 0.67 (−33%) | Still protective |
| 6–7 naps/week | 0.94 | Often masks poor night sleep |
Source: Häusler et al. (2019), Heart · Read full analysis: Is it bad to nap every day?
4. Nap Length vs. Alertness (NASA Pilot Study)
The classic NASA "fatigue countermeasure" study on long-haul pilots quantified the alertness boost — and the grogginess penalty — of different nap durations.
| Nap length | Alertness boost | Grogginess |
|---|---|---|
| 10–20 min ⭐ | +34% (2h) | None |
| 30–60 min ⚠️ | Net negative first hour | Severe (sleep inertia) |
| 90 min ⭐ | +40% (4h+) | Minimal |
Source: Rosekind et al. (1995), NASA Technical Memorandum · Read full analysis: Nap calculator
5. Sleep Inertia Duration by Wake Stage
Sleep inertia — that 5-to-60-minute fog after waking — depends almost entirely on which sleep stage the alarm interrupts. This is why cycle-aligned bedtimes matter more than total duration.
| Wake stage | Inertia duration | Subjective feel |
|---|---|---|
| Light sleep (N1/N2) ⭐ | 5–10 min | Refreshed |
| REM | 10–20 min | Disoriented but clears fast |
| Deep sleep (N3) ⚠️ | 30–60 min | "Sleep drunkenness" |
Source: Hilditch & McHill (2019), Nature & Science of Sleep · Read full analysis: Why do I wake up tired?
6. Caffeine Half-Life Math
Caffeine has a 5–6 hour half-life in healthy adults. That means a standard 200 mg coffee at 3 PM still has 50 mg in your bloodstream at 9 PM — enough to delay sleep onset and fragment deep sleep.
| Hours after 200mg coffee | Caffeine remaining | Effect |
|---|---|---|
| 0h | 200 mg | Peak alertness |
| 5–6h | 100 mg ⚠️ | Still strongly active |
| 10–12h | 50 mg ⚠️ | Suppresses deep sleep |
| 15–18h ⭐ | 25 mg | Mostly cleared |
Source: Drake et al. (2013), J Clin Sleep Med · Read full analysis: Why can't I fall asleep?
7. Chronotype Distribution (Munich ChronoType Questionnaire)
Roenneberg's MCTQ analysis of 53,689 respondents shows natural mid-sleep time is normally distributed. About 6% of the population sits at the extremes — true morning larks or strong evening owls.
| Chronotype | Natural mid-sleep | Wake window | % population |
|---|---|---|---|
| Extreme morning | Before 2:00 AM | 4–6 AM | ~3% |
| Moderate morning | 2:00–3:30 AM | 6–7 AM | ~22% |
| Intermediate ⭐ | 3:30–5:00 AM | 7–8 AM | ~50% |
| Moderate evening | 5:00–6:30 AM | 8–10 AM | ~22% |
| Extreme evening ⚠️ | After 6:30 AM | 10 AM+ | ~3% |
Source: Roenneberg et al. (2007), Curr Biol · Read full analysis: What time should I sleep for an 11 AM wake-up?
8. REM Distribution Across a Night
REM density is not uniform. Early cycles are dominated by deep N3 sleep; REM lengthens progressively, peaking in cycles 4 and 5 — which together carry ~60% of the night's total REM.
| Cycle | REM duration | Dominant stage |
|---|---|---|
| Cycle 1 | ~5 min | N3 deep sleep |
| Cycle 2 | ~10 min | N3 deep sleep |
| Cycle 3 | ~20 min | Mixed |
| Cycle 4 ⭐ | ~30 min | REM-dominant |
| Cycle 5 ⭐ | ~35 min | REM-dominant |
Source: Walch et al. (2016), Sci Adv · Read full analysis: The science of REM sleep
9. Phase-Shift Speed by Method
The human circadian pacemaker shifts at a limited rate. Earlier bedtimes (phase advance) are harder than later ones. Bright morning light roughly doubles the natural shift speed.
| Method | Daily phase advance | Time to shift 3h earlier |
|---|---|---|
| Schedule change only | ~15 min | ~12 days |
| + Bright light at wake ⭐ | ~30 min | ~6 days |
| + Evening melatonin (0.5 mg) | ~45 min | ~4 days |
| Forcing it (no light cues) ⚠️ | 0 min | Insomnia + crash |
Source: Burgess et al. (2003), J Biol Rhythms · Read full analysis: What time should I sleep for a 3 AM wake-up?
10. Shift-Work Health Risks (IARC Monograph 124)
The International Agency for Research on Cancer classifies persistent night-shift work as "probably carcinogenic to humans"(Group 2A), driven by chronic circadian misalignment.
| Outcome | Relative risk | Exposure |
|---|---|---|
| Breast cancer | 1.32 | ≥20 yrs night shifts |
| Cardiovascular disease | 1.40 | Long-term shift work |
| Type 2 diabetes | 1.42 | Rotating shifts |
| Daytime workers ⭐ | 1.00 | Reference |
Source: IARC Monograph 124 (2020) · Read full analysis: Sleep calculator for 2 AM
11. Natural Wake Windows by Bedtime
Cycle-aligned wake times for a typical adult with 14-minute sleep latency. Each window is ±10 minutes wide because cycles vary slightly between individuals and across the night.
| Bedtime | 5 cycles (7.5h) | 6 cycles (9h) |
|---|---|---|
| 10:00 PM | 5:46 AM | 7:16 AM |
| 10:30 PM ⭐ | 6:16 AM | 7:46 AM |
| 11:00 PM ⭐ | 6:46 AM | 8:16 AM |
| 11:30 PM | 7:16 AM | 8:46 AM |
| 12:00 AM | 7:46 AM | 9:16 AM |
Source: Carskadon & Dement (2011), Principles & Practice of Sleep Medicine · Read full analysis: What time should I wake up if I sleep at 11 PM?
12. Long-Sleep Mortality Risk Detail
Long sleep (≥9 hours) is consistently associated with elevated mortality — but the relationship is complex. Long sleep is often a symptom of underlying inflammation, depression, or undiagnosed sleep-disordered breathing rather than a direct cause.
| Sleep duration | All-cause mortality (RR) | Likely confounders |
|---|---|---|
| 7–8 hours ⭐ | 1.00 | Reference |
| 9 hours | 1.17 | Depression, low SES |
| 10 hours ⚠️ | 1.30 | Inflammation markers |
| 11+ hours ⚠️ | 1.40+ | Undiagnosed sleep apnea |
Source: Cappuccio et al. (2010), Sleep · Read full analysis: Why do I feel more tired the more I sleep?
Citation Policy
Every figure on this page is traceable to a peer-reviewed primary source. We do not cite secondary aggregators or popular-science summaries. If you find a figure that doesn't reconcile with the cited source, please email the editorial team and we'll review it.
You may reuse these tables for non-commercial educational purposes with attribution to SnapSleepCalc and a link back to this page.