Knee Osteoarthritis Statistics 2026: How Many People Are Affected Worldwide
Share
The largest population-based meta-analysis to date — pooling 88 studies and over 10 million participants worldwide — found that 16.0% of people aged 15 and older have knee osteoarthritis, rising to 22.9% of people aged 40 and older. Applied to the 2026 global population, that works out to roughly 950 million to 1 billion people aged 15+ living with knee osteoarthritis worldwide — though this is a rough extrapolation, not a directly published 2026 figure (see caveats below).
The most extensive evidence base to date
A systematic review and meta-analysis published in eClinicalMedicine (The Lancet) searched PubMed, Embase and Scopus without language restriction and pooled data from 88 population-based studies covering 10,081,952 participants across 34 countries — by far the largest participant pool used to estimate knee osteoarthritis prevalence to date. It found a pooled global prevalence of 16.0% (95% CI, 14.3%–17.8%) in individuals aged 15 and over, rising to 22.9% (95% CI, 19.8%–26.1%) in individuals aged 40 and over. The same analysis estimated roughly 654.1 million individuals aged 40 and older were living with knee osteoarthritis worldwide in 2020, and calculated a global incidence of about 203 new cases per 10,000 person-years in people aged 20 and over — equivalent to an estimated 86.7 million new cases annually.
What this looks like applied to today's global population
Scaling this study's 16.0% prevalence rate (ages 15+) to the current world population gives a useful, if approximate, present-day picture. The global population in 2026 is approximately 8.28 billion; roughly 75% of people worldwide are aged 15 or older (about 25% of the global population is under 15), giving an estimated 6.2 billion people in the 15-and-over age group. Applying the study's 16.0% prevalence rate to that figure gives an estimate of roughly 990 million to 1 billion people worldwide currently living with knee osteoarthritis.
This figure should be read with three important caveats. First, it is an extrapolation, not a number the study itself published — the 16.0% rate is based on data collected up to 2020, applied here to the 2026 population total. A fully precise estimate would need current, region-specific age-structure data, since knee OA prevalence varies substantially by country and tends to be higher in higher-income regions. Second, this figure covers the knee only — total osteoarthritis burden across all joints (hips, hands, and others combined) is considerably larger. Third, prevalence generally rises steeply with age, so the true figure depends heavily on how each region's population is distributed by age, which a flat global rate applied uniformly does not capture.
Who is most affected
The same meta-analysis found prevalence and incidence were both consistently higher in women than men, with a female-to-male prevalence ratio of 1.69 (95% CI, 1.59–1.80) and an incidence ratio of 1.39 (95% CI, 1.24–1.56). Prevalence increased substantially with age and varied considerably between individual countries, generally trending higher in higher-income regions.

Where the numbers are heading
For long-term trend data, the Global Burden of Disease Study — a separate, model-based estimation approach — offers useful trajectory context: knee osteoarthritis cases are projected to increase by roughly 74.9% by 2050 compared with 2020 levels, reaching an estimated 642 million cases by 2050, driven primarily by population growth and aging rather than a rising rate of disease within any given age group. It's worth noting the GBD's own knee-specific prevalence estimates (around 375 million cases in 2021) are considerably lower than the population-based meta-analysis above — a known feature of GBD's modeling approach compared with pooling raw survey data directly, which is one reason the 88-study meta-analysis is considered by many researchers the more direct empirical estimate of current-day prevalence.
Scientific sources and experts
-
Cui et al., 'Global, regional prevalence, incidence and risk factors of knee osteoarthritis in population-based studies', 88 studies, 10,081,952 participants, eClinicalMedicine (The Lancet) 2020 — https://pmc.ncbi.nlm.nih.gov/articles/PMC7704420/
-
Osteoarthritis burden across 204 countries and territories, Global Burden of Disease Study 2021 — https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12727419/
-
Global, regional, and national burden of osteoarthritis, 1990–2020 and projections to 2050, GBD 2021 Collaborators — https://pmc.ncbi.nlm.nih.gov/articles/PMC10477960/
-
The global burden of osteoarthritis knee: secondary data analysis of a population-based study (knee-specific 1990–2021 figures) — https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11993502/
FAQ
How common is knee osteoarthritis globally?
The largest available meta-analysis (88 studies, 10+ million participants) found 16.0% of people aged 15 and over have knee osteoarthritis, rising to 22.9% of people aged 40 and over — scaled to 2026's global population, that's roughly 950 million to 1 billion people, though this scaled figure is an estimate rather than a directly published number.
Who is most at risk of developing knee osteoarthritis?
Women consistently show higher prevalence than men (a 1.69-times higher rate, per the largest pooled analysis), and risk rises substantially with age; higher-income regions also tend to show higher rates in the data.
Is knee osteoarthritis increasing worldwide?
Yes — Global Burden of Disease projections estimate a roughly 74.9% increase in cases by 2050 compared with 2020, mainly due to population growth and aging rather than the underlying rate of disease changing dramatically.
Why do different studies report different global numbers?
Different methodologies produce different estimates: large pooled meta-analyses of raw prevalence surveys (like the 88-study, 10-million-participant analysis) versus the Global Burden of Disease Study's statistical modeling approach. Both are legitimate, peer-reviewed methods, but they can produce meaningfully different absolute numbers — always cite which methodology a figure comes from.