Paperonium

Air pollution · JAMA Network Open · 2025

Trimming pollution above the US limit clearly cuts strokes, the paper says. Its table can’t tell.

The summaryall fewer, yet leaves out itsbest guessThe paper’s own tablecould be fewer, none, or more−2.26−20.64 to −8.97should be inside−8.97 to 20.64−20 per 1,000−1001020no changefewer strokesmore
The summaryall fewer, yetleaves out itsbest guessshould be insideThe paper’sown tablecould be fewer,none, or more−2.26−20.64−8.9720.64−8.97−20−1001020no changeper 1,000fewer strokesmore

Change in stroke hospitalizations per 1,000 people over 5 years

The paper says

The summary: cutting fine-particle air pollution in the UK by 5% wherever it tops the US yearly limit would mean 2.26 fewer stroke hospitalizations per 1,000 people over 5 years (range: 8.97 to 20.64 fewer). Its key-points box says this would clearly reduce strokes.

Its own data show

The paper’s main results table, and an extra table published with it, give the same 2.26 fewer but a range of 8.97 fewer to 20.64 more. The table’s own risks agree: 12.32 in 1,000 without the cut, 3.37 to 32.97 with it; subtracting gives 8.95 fewer to 20.65 more.

A range is the span of results the data can’t rule out. It must contain the best guess, and a drop is clear only if the whole span is fewer strokes. The summary’s range misses 2.26; the table’s runs from fewer to more. The clear stroke cut claimed at the US limit, the threshold tied to current policy, is unsupported.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

A Hypothetical PM2.5 Intervention for the Risk of Hospitalization for Cardiovascular Diseases · JAMA Network Open (2025) · doi:10.1001/jamanetworkopen.2025.39862