Latitude and rheumatoid arthritis onset track a country’s wealth more than its clinic population
A global analysis of the METEOR registry found a striking geographic pattern in rheumatoid arthritis onset: patients living farther from the equator were diagnosed at older ages. The pattern was present across 17 countries and 93 hospitals, yet it nearly disappeared once country-level socioeconomic indicators were added to the model. That result shifts attention away from the patient alone and toward the broader conditions in which autoimmune disease emerges, is recognized, and is recorded.
The study included 37,981 people with rheumatoid arthritis. Mean age at diagnosis varied substantially by country, from 39 years in Iran to 55 years in the Netherlands. Across the full range of country latitude, from 9.9° to 55.8°, every 1-degree increase in latitude was associated with a 0.23-year increase in mean age at diagnosis, with a 95% credibility interval of 0.095 to 0.38 years. On its face, that looks small on a per-degree basis. Over a broad geographic span, however, it adds up to more than a decade of difference in age at onset, a gap large enough to matter in clinic, epidemiology, and health planning.
The more interesting finding emerged when the analysis was layered. Within a country, latitude did not meaningfully track age at diagnosis across hospitals. The gradient lived at the country level, not the hospital level. Once patient-specific factors such as sex and anticitrullinated protein antibody status were entered into the model, the latitude effect became larger rather than smaller, increasing from 0.23 to 0.36 years per degree. That pattern argues against a simple explanation based on the mix of patients seen at different sites. The stronger effect after adjustment suggests that individual characteristics were not driving the geographic trend and may even have been masking part of it.
Country-level socioeconomic status absorbed the geographic signal
The analysis changed direction when country-level socioeconomic indicators were included. Gross domestic product per capita and related national measures almost eliminated the latitude association, reducing the estimate from 0.23 years to 0.051 years per degree, with a credibility interval spanning -0.37 to 0.38. In practical terms, the apparent effect of geography on rheumatoid arthritis onset was largely accounted for by differences in national wealth and development.
This is the central clinical message of the study. Latitude itself is not a biologic mechanism; it is a marker for environmental exposure, health system structure, diagnostic access, population nutrition, infectious burden, occupational conditions, and other features that travel with geography. A country’s socioeconomic position can shape how early inflammatory arthritis is recognized, how quickly symptoms reach specialist care, and which populations enter registries in the first place. The result also fits a broader epidemiologic reality: autoimmune disease patterns often reflect social structure as much as biology.
The investigators used Bayesian multilevel structural equation models to separate effects at the patient, hospital, and country levels. That design matters because it prevents the analysis from flattening local variation into a single national average. The absence of a meaningful hospital-within-country latitude effect suggests that the observed gradient was not a matter of one referral center seeing younger or older patients than another. Instead, the divergence appeared to arise from differences between countries themselves, and those differences were closely tied to socioeconomic status.
Why geography may shape age at rheumatoid arthritis diagnosis
Rheumatoid arthritis is a chronic autoimmune disease in which inflammation develops through a complex interplay of genetic risk, immune activation, and environmental triggers. Latitude has long been of interest because it can stand in for variation in sunlight exposure, vitamin D status, climate, infection patterns, and social development. This study does not identify a single causal pathway, and it does not claim that geography directly causes earlier or later disease onset. It does, however, show that the latitude gradient in age at diagnosis cannot be explained by the patient’s sex, serologic status, or other individual-level features captured in the registry.
The socioeconomic finding is especially persuasive because it transforms a geographic association into a social one. Wealthier countries may have better diagnostic networks, broader public awareness, earlier access to rheumatology, or more complete registry capture of incident disease. In lower-resource settings, patients may present later, get labeled later, or enter registries only after prolonged symptoms. Any of those pathways could shift the recorded age at diagnosis upward or downward. The data cannot distinguish among these mechanisms, but the country-level mediation strongly suggests that welfare and healthcare infrastructure are key parts of the story.
A one-sentence reality check is warranted here: the latitude gradient was not a biological constant.
The use of age at diagnosis as a proxy for age at onset also deserves careful interpretation. Diagnosis often lags symptom onset, and the length of that lag can vary by country, healthcare access, referral pathways, and disease awareness. If diagnostic delay is longer in some settings than others, measured age at diagnosis may reflect system performance as much as disease biology. That limitation is built into the endpoint, yet it may be part of what makes the finding so relevant to real-world rheumatology.
What the registry data can and cannot tell clinicians
The METEOR registry offers breadth, but cross-sectional registry data cannot prove causality. The analysis shows association, mediation, and multilevel clustering; it does not show which specific socioeconomic factor is responsible, nor does it isolate environmental exposures such as ultraviolet radiation, infection, or nutrition. The country-level indicators were broad measures, and broad measures often absorb several mechanisms at once. Even so, the near-disappearance of the latitude effect after those indicators were introduced is difficult to ignore.
For clinicians, the practical implication is not to treat latitude as a risk score. It is to recognize that age at rheumatoid arthritis onset may be shaped by social and structural conditions before it ever reaches the exam room. In countries with lower socioeconomic resources, younger age at diagnosis may partly reflect earlier inflammatory disease biology, but it may also reflect the ways health systems surface disease sooner or later. In wealthier settings, older age at diagnosis may reflect other exposures, different referral patterns, or better case ascertainment. The same disease label can therefore sit on top of different population histories.
The study leaves a narrow but important question on the table: which component of country welfare is doing the explanatory work? Access to care, educational attainment, income distribution, diagnostic delay, and environmental exposures remain plausible candidates. The registry analysis cannot separate them cleanly, and that restraint is part of its value. It shows where the signal lives, even if it cannot yet identify every mechanism inside the signal.
For now, the strongest conclusion is simple and data-driven: patients living closer to the equator were diagnosed with rheumatoid arthritis at younger ages, and country-level socioeconomic status largely explained that relationship. The geography remained visible until welfare entered the model, then it faded almost completely. Future work will need to determine whether that pattern reflects access to diagnosis, true differences in disease onset, or both.