Insomnia Statistics: Key Facts and Data

Figures come from the published research and official sources listed below. This page is for information, not medical advice.

How common is insomnia globally?

An estimated 852 million adults worldwide — about 16.2% of the global adult population — experience clinically relevant insomnia, based on a 2025 systematic review that combined prevalence data with UN population statistics (Sleep Medicine Reviews, 2025). When clinician-administered diagnostic interviews are used — the stricter, gold-standard method — the pooled figure is 12.4% (Journal of Sleep Research, 2025).

• 16.2% (≈852 million adults): global adult prevalence by population modelling — covers adults ≥20 years worldwide; 2025 systematic review combining 18 studies (n=262,582) (Sleep Medicine Reviews, 2025)
• 7.9%: global prevalence of severe insomnia specifically, from the same modelling study (Sleep Medicine Reviews, 2025)
• 12.4% (95% CI 9.0–16.8%): pooled prevalence using clinician-administered DSM diagnostic interview (gold standard); meta-analysis of 14 studies; general population adults, predominantly DSM-IV criteria (Journal of Sleep Research, 2025)
• 16.3% (95% CI 11.3–23.0%): pooled prevalence when self-reported DSM criteria are used instead of clinician interview — same meta-analysis, different measurement method (Journal of Sleep Research, 2025)
• 30–40% of adults: experience insomnia symptoms (difficulty initiating or maintaining sleep) at some point — a broader symptom-level figure, not full diagnostic criteria (Circulation Research, 2026)
• 10–15%: estimated range for chronic insomnia disorder specifically, using stricter definitional criteria (Circulation Research, 2026)
• Only 31 of 237 countries had nation-specific insomnia prevalence studies included in the 2025 global modelling review — meaning most regional estimates rely on extrapolation (Sleep Medicine Reviews, 2025)

Prevalence by country and region

In the Americas, an estimated 16.8% of adults have insomnia disorder — accounting for roughly 37 million people in the US, 29 million in Brazil, and 16 million in Mexico (SLEEP, 2024). In the US specifically, self-reported sleep difficulty data from 2020 shows 14.5% of adults had trouble falling asleep most or every day, and 17.8% had trouble staying asleep most or every day (CDC/NCHS, 2022).

• 16.8% (Americas overall): estimated prevalence of insomnia disorder in adults across the Americas; 2024 modelling study (SLEEP, 2024)
• ~37 million (USA): estimated adults with insomnia disorder; ~29 million (Brazil); ~16 million (Mexico) — all from 2024 modelling (SLEEP, 2024)
• 14.5% of US adults: reported trouble falling asleep most or every day — 2020 NHIS national survey data (CDC/NCHS, 2022)
• 17.8% of US adults: reported trouble staying asleep most or every day — 2020 NHIS national survey data (CDC/NCHS, 2022)
• 22.1% vs 16.9%: trouble staying asleep among US adults below 100% of the federal poverty level vs those at ≥200% FPL — a meaningful income gradient in 2020 NHIS data (CDC/NCHS, 2022)

Who is most affected? Age, sex, income, and pregnancy

Women are consistently more likely than men to have insomnia across all age groups and in every region studied. In the Americas, the gap is 19.5% (women) vs 14.0% (men) for insomnia disorder (SLEEP, 2024). Pregnancy sharply raises the risk further, with a pooled insomnia prevalence of 38.2% in pregnant women (PMC / Frontiers in Psychiatry, 2024).

• 19.5% (women) vs 14.0% (men): insomnia disorder prevalence by sex in the Americas — 2024 modelling study; total ~123 million people across both sexes (SLEEP, 2024)
• OR 1.45 (95% CI 1.32–1.58): women’s odds of insomnia symptoms vs men among US older adults (60+), controlling for age and other factors — 2025 NHIS secondary analysis, 2005–2020 data (PubMed / Frontiers, 2025)
• 38.2%: pooled insomnia prevalence in pregnant women — a systematic review and meta-analysis; pregnancy-specific drivers include physical discomfort, hormonal changes, and emotional distress (PMC / Frontiers in Psychiatry, 2024)
• 19.6% (95% CI 12.3–28.3%): DSM-defined insomnia disorder in older adults (mean age 65.9–83.1 years) — global systematic review and meta-analysis; associated with female sex, depression, anxiety, and cardiovascular/painful conditions (European Geriatric Medicine, 2023)
• 13–37.8%: range of insomnia prevalence reported for adults aged 65+ across epidemiological studies — the wide range reflects the same measurement variation seen in the general adult literature (PMC, 2026)
• 21.8% vs 13.8%: trouble staying asleep among US adults aged 45–64 vs 18–44 — 2020 NHIS data; note this is symptom-level self-report, not full insomnia disorder (CDC/NCHS, 2022)
• 22.1% vs 16.9%: trouble staying asleep among US adults below vs above the federal poverty level — 2020 NHIS; lower income is consistently associated with higher insomnia rates (CDC/NCHS, 2022)

What does insomnia do to health?

Insomnia is associated with substantially higher risks of cardiovascular disease, depression, and anxiety. A 2023 meta-analysis of 21 real-world data studies found that people with insomnia had a 53% higher risk of cardiovascular mortality and a 48% higher risk of myocardial infarction compared to those without insomnia (PMC / PLOS One, 2023).

• RR 1.53 (p<0.01): relative risk of cardiovascular mortality in people with insomnia vs those without — 2023 meta-analysis, 21 real-world studies, adults (PMC / PLOS One, 2023)
• RR 1.48 (p=0.03): relative risk of myocardial infarction — same 2023 meta-analysis (PMC / PLOS One, 2023)
• RR 1.31 (p<0.01): relative risk of all-cause mortality — same 2023 meta-analysis (PMC / PLOS One, 2023)
• OR 2.60 (95% CI 1.98–3.42): odds ratio for insomnia predicting future depression — a meta-analysis of prospective cohort longitudinal studies; this describes risk over time, not a cross-sectional snapshot (BMC Psychiatry, 2016)
• OR 1.53 (95% CI 1.18–1.98): odds of developing depression or anxiety within 5 years of an insomnia diagnosis, in UK primary care patients (n=1,535 insomnia vs n=1,502 controls); data 2010–2020 (Family Practice / Oxford Academic, 2026)
• Mendelian randomisation evidence: large-scale genetic studies involving millions of participants support a likely causal link between insomnia and multiple cardiometabolic outcomes — not just an association (Circulation Research, 2026)
• Note on causality: A 2019 systematic review found that specific insomnia symptom subtypes (notably difficulty falling asleep and non-restorative sleep) were associated with mortality, but insomnia disorder overall and difficulty staying asleep were not consistently linked across studies (Sleep Medicine Reviews, 2019). The relationship is real and important, but the specific pathways are still being untangled.

What does insomnia cost?

Chronic insomnia is associated with an average loss of 45–54 workdays of productivity per year, translating into estimated annual GDP losses of $1.8–$207.5 billion across countries (in 2019 US dollars), according to a 2023 RAND international modelling study (RAND, 2023).

• 45–54 lost workdays per year: average workplace productivity loss associated with chronic insomnia — international modelling, 2023 (RAND, 2023)
• $1.8–$207.5 billion annually: range of national GDP losses attributable to chronic insomnia across studied countries (2019 USD), equivalent to 0.64%–1.31% of GDP (RAND, 2023)
• ~14% of annual per-capita income: what people with insomnia say they would be willing to trade to avoid its consequences — a well-being valuation estimate from the same RAND study (RAND, 2023)
• Note on comparability: The RAND figures cover chronic insomnia and are modelled rather than directly measured. An older US-specific Harvard/AASM study (2011 data) estimated $63.2 billion in net annual US workplace productivity losses from insomnia — a different scope, population, and method. These two figures are not directly comparable.

Why do insomnia estimates range from 6% to 32%+?

The wide range of insomnia prevalence figures — from around 6% to over 32% — is not a sign that the science is confused. It reflects a well-documented measurement problem: the choice of diagnostic instrument is the single biggest driver of the estimate (Frontiers in Public Health, 2026).

• 12.4%: pooled estimate using clinician-administered DSM diagnostic interview — the most rigorous method, least subject to over-reporting (Journal of Sleep Research, 2025)
• 16.3%: pooled estimate using self-reported DSM criteria questionnaires — less rigorous but more scalable for large surveys (Journal of Sleep Research, 2025)
• 32.3%: estimate produced when the Athens Insomnia Scale cut-off of ≥6 is used — a broader screening tool that captures more people with sleep difficulties that fall short of a full insomnia disorder diagnosis (Frontiers in Public Health, 2026)
• 5–10%: estimates produced when the most conservative, stricter chronic disorder criteria are applied (Sleep Medicine, 2023)
• The best range for a general adult population statement: 12.4% (clinician interview, gold standard) to 16.2% (population modelling) — these figures use the most transparent methods and are the most defensible for a lay summary.

Treatment evidence in numbers

Cognitive behavioural therapy for insomnia (CBT-I) is the most evidence-rich treatment in the insomnia literature. Across all delivery settings — in-person, digital, and self-help — 52 randomised controlled trials (RCTs) show significant improvements in insomnia compared with wait-list control (PMC / Scientific Reports, 2023).

• 52 RCTs: evidence base for CBT-I across all settings (in-person, digital, self-help); all showed significant improvement vs wait-list control — network meta-analysis (PMC / Scientific Reports, 2023)
• 29 RCTs, n=9,475: evidence base for fully automated digital CBT-I specifically; showed moderate to large effects on insomnia severity vs control — 2025 systematic review and meta-analysis (npj Digital Medicine, 2025)
• Digital CBT-I and depression/anxiety: a meta-analysis of 22 RCTs found that digital CBT-I also produced small-to-moderate reductions in depressive symptoms (SMD −0.42) and anxiety symptoms (SMD −0.29) alongside its large effect on sleep (SMD −0.76) (npj Digital Medicine, 2023)
• Note on scope: These figures cover CBT-I efficacy on insomnia severity scores. They do not compare CBT-I to pharmacotherapy head-to-head, and individual response varies. Talk to a clinician about which approach suits your situation.

Insomnia statistics at a glance

Where the numbers disagree — and why

Disagreement 1: Global prevalence (12.4% vs 16.2% vs 32.3% vs 5–10%)
A 2025 meta-analysis using clinician-administered DSM interviews found a pooled prevalence of 12.4% (Journal of Sleep Research, 2025). A 2025 systematic review applying population modelling produced 16.2% (Sleep Medicine Reviews, 2025). A 2026 article using the Athens Insomnia Scale ≥6 found 32.3% (Frontiers in Public Health, 2026), and stricter chronic disorder criteria produce estimates of 5–10% (Sleep Medicine, 2023).
Reason (sourced): The choice of diagnostic instrument is the documented driver of this variation; a 2026 public health review states this explicitly. It is not a sign of scientific disagreement about insomnia’s reality — it is a measurement difference.
What this means for you: For a general adult population, the most defensible range is 12.4%–16.2%. Figures above 30% reflect symptom screening, not diagnosed disorder. All figures confirm insomnia is genuinely common.

Disagreement 2: Pregnant women (38.2%) vs general adult rate (12.4–16.2%)
The 38.2% figure in pregnant women (PMC / Frontiers in Psychiatry, 2024) looks dramatically different from the general adult rate. This is not a conflict — they measure different populations. The review attributes the elevated rate to pregnancy-specific physical, hormonal, and emotional factors. These are not directly comparable because populations, mechanisms, and contributing factors differ.

Disagreement 3: Mortality risk estimates vary across studies
The 2023 meta-analysis found insomnia associated with RR 1.31 for all-cause mortality (PMC / PLOS One, 2023), while a 2019 systematic review found that only specific insomnia symptom subtypes (difficulty falling asleep, non-restorative sleep) were consistently linked to mortality — not insomnia disorder overall (Sleep Medicine Reviews, 2019). The reason is partly methodological: the studies differed in how they defined insomnia, what populations they included, and how long they followed participants. This area of the evidence is still developing; causality has not been fully established, and the association likely varies by insomnia severity and presence of comorbidities.

Definitions and limits of the data

Insomnia disorder vs insomnia symptoms: Insomnia disorder requires difficulty initiating or maintaining sleep, or early morning awakening, occurring at least 3 nights per week for at least 3 months, plus significant daytime impairment (fatigue, mood disturbance, concentration difficulties, or similar). Insomnia symptoms — experienced by roughly one-third of adults — do not necessarily meet this full clinical threshold. Most of the figures on this page specify which definition was used, and you should not mix them in comparisons.

Self-report vs clinician interview: Almost all large-scale prevalence studies rely on self-reported sleep difficulty. Clinician-administered interviews consistently produce lower estimates than self-report questionnaires and screening scales, because they apply diagnostic rules more precisely and can clarify ambiguous answers.

Geographic gaps: Only 31 of 237 countries had nation-specific insomnia prevalence data in the most recent global review. Estimates for large parts of Africa, South Asia, and the Middle East rely on modelling or extrapolation.

Paediatric insomnia: All figures on this page are for adults (18+). Childhood and adolescent insomnia have different epidemiology and are not included here.

Economic burden: Economic modelling figures (GDP losses, productivity days) are estimates based on extrapolation from survey data. They carry wide ranges and should be read as orders of magnitude, not precise measurements.

Trend data: The US NHIS trend data (2002–2012) use self-reported “trouble sleeping,” not clinical insomnia disorder. A 2023 Sleep Medicine paper cites a roughly 60% increase in sleep problems since 2010, attributing this to a referenced source; the primary source for that 60% figure was not independently verified for this page, so it is not listed as a confirmed figure here.

How we compiled this page: we gathered figures from the published studies and official sources listed below, recorded each figure’s population and year, and did not run new studies or surveys. The original figures belong to their publishers; our compilation and wording may be reused with credit under CC BY 4.0. Spotted an error? Tell us.

Figures checked September 2026 · 30 figures from 21 sources · Next review due March 2027