Fine particle pollution (PM2.5) worsened all 4 measures of Rheumatoid Arthritis activity - cohort

Effects of air pollution on disease activity in patients with rheumatoid arthritis

Annals of the Rheumatic Diseases (in press), 2026, https://doi.org/10.1016/j.ard.2026.06.018

Sung Cheol Jung, Jeffrey R Curtis, Seojin Yang, Yunhee Choi, Min Hyuk Lim, Saram Lee, Sung Ik Cho, Jin Kyun Park, Jun Won Park, Eun Bong Lee

Summary by Claude Opus 5 - August, 2026

For people with rheumatoid arthritis, it appears to be the fine-particle pollution accumulated over the previous 2 to 4 weeks - not today's bad-air day - that tracks with a higher chance of a flare.

This is a prospective cohort study (not a trial): 1,070 RA patients at a Seoul tertiary hospital, 12,583 outpatient visits, 2021-2024. Pollution was assigned from government monitoring data for each patient's residential region - no one wore a personal monitor. Average PM2.5 was 18.5 +/- 6.0 ug/m3.

Of the 6 pollutants tested (SO2, NO2, O3, CO, PM10, PM2.5), only PM2.5 was consistently linked to disease. Each 1-SD rise - about 6 ug/m3 - raised the odds of a flare 11% (adjusted OR 1.113, 95% CI 1.017-1.218), and PM2.5 was the only pollutant significantly associated with all four activity measures (DAS28-CRP +0.031, CDAI +0.171, tender joints +0.064, swollen joints +0.046 per SD). A within-patient case-crossover analysis found nothing at 1 day or 1 week, but rising risk once cumulative exposure passed roughly 14 days - the authors conclude that "cumulative exposure over time may be more relevant than transient peaks." The association was stronger in women, nonsmokers, and patients with more tender joints at baseline.

What this does not show / limitations

Observational - no proof that pollution causes flares, and no evidence that cleaning the air reduces them. The effect sizes are modest, and the E-value of 1.47 means a fairly ordinary unmeasured confounder could erase the association entirely.
Vitamin D is a candidate for exactly that: it was never measured, Korean PM2.5 peaks in winter and spring precisely when 25(OH)D hits its yearly low, and flares peaked in January. The models adjusted for temperature and humidity but not for season or vitamin D status.
Exposure was assigned by home region, ignoring workplace, commuting, mask use, and air purifiers. Most patients had well-controlled RA (median DAS28-CRP 2.4), so the effect in more active disease is unknown. Korean patients only.

PDF

Related in VitaminDWiki

Air Pollution