Air pollution and Parkinson's

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Maria Kasdagli, Klea Katsouyanni, Konstantina Dimakopoulou and Evangelia Samoli (meta-analysis; institutions not rechecked); Brittany Krzyzanowski, Susan Searles Nielsen, Jay Turner and Brad Racette, Washington University in St. Louis (Medicare study).

What happened

• 2019 meta-analysis (Kasdagli et al., Int J Hyg Environ Health 2019;222:402–409). Searches to November 2018 found 102 unique studies; 15 went into the meta-analyses. Per 10 µg/m³ rise in long-term PM2.5 the pooled relative risk of Parkinson's was 1.06 (95% CI 0.99–1.14, not significant); for NO2 it was 1.01 (0.98–1.03). The authors called the evidence "weak" and noted high heterogeneity.
• 2023 US study (Krzyzanowski et al., Neurology 2023;101:e2058–e2067). 21,639,190 Medicare beneficiaries aged 66–90; 89,390 had Parkinson's newly diagnosed in 2009. People living at the median PM2.5 level had a 56% higher relative risk than those at the lowest level (95% CI 47–66%). Risk rose roughly 4.2% per extra µg/m³ up to 13 µg/m³. There was a hot spot in the Mississippi–Ohio River Valley (risk 19% higher than the rest of the US), and the strongest PM2.5 link was in a low-risk Rocky Mountain region. The authors verified no confounding by indicators of socioeconomic status or neurologist density. The county-level analysis is geographic, with an individual-level check using zip+4 locations.

Why it moved us toward a cure

Pollution may be a modifiable cause that builds on the chemical risks in landmark 25. It could act by inflammation, or by particles carrying toxins to the brain. Cleaner air would be prevention, not a cure.

Outcome

Open. The studies are observational and geographic, so they cannot show causation. A 2025 meta-analysis of 42 studies (Xie et al., Toxics 13:139) found an odds ratio of 1.10 (95% CI 1.03–1.17) for Parkinson's per 5 µg/m³ of long-term PM2.5, with very high heterogeneity (I² 95%). Neurology published an author response to the Medicare paper in 2024 and a journal-wide correction to received dates; I did not read the responses and found no change to the results.

Sources

• Krzyzanowski et al., Neurology 2023, PMC10663024 · WashU record
• Kasdagli et al., Int J Hyg Environ Health 2019, DOI 10.1016/j.ijheh.2018.12.006 (PMID 30606679)
• Xie et al., Toxics 2025;13:139, DOI 10.3390/toxics13020139 (PMC11861764)
• Krzyzanowski et al., Neurology 2023, DOI 10.1212/WNL.0000000000207871
• Not yet verified: funders of both papers; the content of the 2024 author responses; full text of the Medicare paper (abstract read).