Air pollution · Mortality
Fine-particle mortality risk persisted below the air-quality guideline
Save this development or follow its topic to track what changes.
Short answer
At exposures averaging 6.3 μg/m³, PM2.5 remained associated with mortality, with no evident threshold below the 10 μg/m³ guideline.
What happened
Pinault and colleagues linked Canadian Community Health Survey respondents to three-year satellite and land-use estimates of residential PM2.5 and followed deaths for up to 12 years with Cox models adjusted for socioeconomic, ecological and behavioural covariates. The fully adjusted non-accidental hazard ratio was 1.26 (95% CI 1.19–1.34) per 10 μg/m³; respiratory mortality was highest at 1.52. The association did not flatten below the guideline value.
Why it matters
Guideline values are often read as thresholds below which exposure is safe. A cohort whose entire exposure range sits under the guideline, still showing a dose-response, is direct evidence against that reading — and it is why later guidance moved downward.
Evidence
- Study type
- Prospective cohort with Cox proportional-hazards models on satellite and land-use estimated PM2.5
- Sample
- 299,500 Canadian Community Health Survey respondents, followed up to 12 years
- Journal
- Environmental Health · peer reviewed
- Replication
- Consistent with other low-exposure cohorts; the shape of the curve at the lowest concentrations remains contested
- Limitations
- Confidence intervals widen at the low end of the exposure spline. Exposure is assigned by residence, so mobility is only partly captured, and residual confounding from smoking intensity cannot be excluded.
What this connects to
Sources
The 2 studies this explanation is built from, by the role each plays. Every source links to PaperFren’s explanation of it and to the original paper.
Primary study
- Low PM2.5 still tracks Canadian mortality
In 299,500 Canadians, a 10 μg/m³ rise in PM2.5 was linked to HR 1.26 for non-accidental death even at mean 6.3 μg/m³.
What it does not showLimitations
Wide confidence intervals on the low-concentration spline; residual confounding from unmeasured smoking intensity or indoor sources cannot be ruled out, and mobility was only partly captured.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Supporting evidence
- Short-term PM2.5 raises deaths and Parkinson’s admissions
Two-day average PM2.5 is linked to higher all-cause mortality and to hospital admissions for Parkinson’s disease and diabetes among US Medicare enrollees.
What it does not showLimitations
Medicare captures only hospitalized neurological disease, so milder dementia is missed. Differences versus people never hospitalized for these conditions were not statistically significant. The study does not prove particles enter the brain or cause disease onset.
PaperFren explanationStudy with cards and a quizOriginal paper (DOI)cc by
Before
Most fine-particle cohort evidence came from higher-exposure populations, leaving open whether risk continues below regulatory guideline concentrations.
Now
A large low-exposure cohort finds elevated risk throughout its range. Confidence intervals widen at the bottom of the spline, and smoking intensity and indoor sources are imperfectly captured.