Longevity & AgingResearch PaperOpen Access

Global Pain Atlas Reveals How Burden Rises With Age Across 118 Countries

The largest-ever pain dataset — 6.1 million people — maps how pain prevalence climbs steeply before 55 and hits hardest in low-HDI nations.

Wednesday, October 7, 2026 0 views
Published in Nat Med
An elderly woman holding her lower back while seated on a wooden bench outdoors, with a blurred urban park background, photographed in natural afternoon light

Summary

Researchers pooled self-reported pain data from more than 6.1 million people in 118 countries to build global reference curves for pain from childhood to age 100 and older. Back pain was the most common site at about 40%, and facial pain the least at about 2%. Women reported more pain at every one of 11 body sites. Pain rose fastest before age 55, and several types, including headache, abdominal, neck and shoulder pain, peaked in earlier adulthood and then declined. Older adults in the lowest-development countries reported far more pain than those in the wealthiest, a pattern that runs against many existing estimates. Smoking, obesity and low income explained only about 18% of pain globally, and even less in lower-income regions. The team released a free online tool for benchmarking other datasets against these norms.

Detailed Summary

Pain is a leading cause of years lived with disability worldwide, yet it has no objective biomarker and can only be measured by self-report. Major surveillance efforts such as the Global Burden of Disease (GBD) study rely on clinically defined conditions and modeled estimates, which can miss pain in people who never reach a diagnosis, especially where health care access is limited. Published pain prevalence ranges from 8% to 64% depending on the study. This paper set out to build a harmonized, global reference framework for how pain is distributed across ages, sexes and world regions.

The team pooled individual-level data from 6,125,459 participants across 902 population-based data sources, 138 study programs and 118 countries, collected between 1990 and 2025. Participants were aged 5 to over 100, and 55% were female. Study-specific wording was mapped to binary pain indicators at 11 anatomical sites. Age trajectories were estimated with generalized linear mixed-effects logistic models using natural cubic splines (3 degrees of freedom, chosen by study-clustered five-fold cross-validation), standardized to a past-month recall period. The curves are available in an open-access Global Pain Benchmarking Tool.

Prevalence ranged from about 2% for facial pain to 40% for back pain. Women reported more pain at all 11 sites (risk ratios 1.08 to 1.83), with the largest gaps for facial pain (1.83), headache (1.74) and abdominal pain (1.60). Seven of 11 sites followed an inverted-U pattern, peaking at age 34 for headache, 23 for abdominal pain, 45 for facial pain, 59 for elbow pain and 62 for neck or shoulder pain. Back, hip and knee pain kept rising into old age. Any bodily pain climbed from about 30% at age 5 to 70% at 100 or older, while high-intensity pain peaked near 50 and generalized pain near 70. Women had higher rates of both (RR 1.52 and 1.50).

Regions looked similar in early adulthood but diverged after about age 50 to 65. Eastern Europe sat roughly 21 percentage points above the global reference at 80 and older. Across the Human Development Index (HDI) gradient, adults 80 and older in low-HDI countries had 30.6 percentage points more any bodily pain (95% CI 26.9 to 34.3), 21.0 more joint pain and 20.2 more back pain than those in high-HDI countries. Low back pain approached twice the prevalence. This contradicts GBD estimates, which generally show higher back pain, osteoarthritis and migraine burden in high-income regions.

Risk patterns were site-specific. Current smoking was linked to facial pain (RR 1.67). Obesity was tied to foot or ankle pain (1.91), knee pain (1.86) and hip pain (1.61). The lowest income group had double the risk of chest pain (2.09). Together, smoking, obesity and low income accounted for 18.3% of pain burden globally. That share ranged from 12.6% in sub-Saharan Africa to 27.1% in eastern Europe. Obesity's share ranged from 2.7% in central and southern Asia to 14.6% in northern America, so the drivers of pain in lower-HDI settings remain poorly understood.

The findings suggest the prevention window opens earlier than usually assumed, since most pain burden accumulates before midlife. Most data were cross-sectional, so the curves describe population age patterns rather than within-person change. Sampling was not uniformly nationally representative, and cohort and reporting effects may shape the late-life declines. Data from children and from low-HDI regions remain sparse.

Key Findings

  • Pooled 6,125,459 participants from 902 data sources in 118 countries, aged 5 to 100+, to build global pain reference curves
  • Pain prevalence ranged from about 2% (facial) to 40% (back); women reported more pain at all 11 sites (RR 1.08–1.83), highest for facial pain (RR 1.83)
  • Seven of 11 sites showed inverted-U trajectories; headache peaked at age 34, abdominal at 23, and neck/shoulder at 62, with the steepest rises before age 55
  • Any bodily pain rose from about 30% at age 5 to 70% at 100+, while high-intensity pain peaked near age 50 and generalized pain near age 70
  • Adults 80+ in low-HDI countries had 30.6 percentage points more any bodily pain (95% CI 26.9–34.3) than in high-HDI countries; low back pain approached twice the prevalence
  • Obesity was linked to foot/ankle (RR 1.91), knee (RR 1.86) and hip pain (RR 1.61); lowest income roughly doubled chest pain risk (RR 2.09)
  • Smoking, obesity and low income explained 18.3% of pain burden globally, from 12.6% in sub-Saharan Africa to 27.1% in eastern Europe

Methodology

The authors harmonized individual participant data from 902 population-based surveys and cohorts (138 study programs, 1990–2025) into binary pain indicators at 11 anatomical sites, plus any bodily pain, high-intensity pain (NRS ≥7) and generalized pain (2016 ACR fibromyalgia spatial criteria). Generalized linear mixed-effects logistic models with natural cubic splines (3 df, chosen by study-clustered five-fold cross-validation) estimated age trajectories, with random effects for cohorts and world regions. Trajectories were standardized to a past-month recall period and general case definition, with sensitivity analyses of seven methodological features. Regional and country-level HDI effects and population attributable fractions for smoking, obesity and low income were also estimated.

Study Limitations

Most contributing datasets were cross-sectional, so the curves reflect population age patterns rather than within-person change, and cohort or reporting effects may contribute to the late-life declines. Sampling designs varied and were not uniformly nationally representative, though results held in restricted-sample analyses except for stomach or abdominal pain. Data from children, adolescents and lower-HDI regions are sparse, and the available text does not state conflicts of interest.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: