Summer ozone tracks with heart-death risk

Ozone is the pollutant that builds up on hot, sunny days, and unlike fine particles its long-term cardiovascular effects have been poorly characterised. Earlier reviews were either incomplete or lumped all cardiovascular outcomes together, leaving open whether ozone tracks with death, with disease onset, or with both. Also unresolved was whether the exposure window matters — a year-round average and a warm-season average describe quite different realities for the same person. This analysis set out to close those gaps and look at who might be more affected and where.

The authors systematically retrieved cohort studies from PubMed, Web of Science and Embase and pooled them to estimate the association between long-term ozone exposure and cardiovascular disease mortality and morbidity. Risks were expressed as relative risks per 10 µg/m3 increase in ozone concentration, so results from different settings could be compared on a common scale. Separate analyses examined exposure windows, comparing warm-season with annual average ozone. Further stratified analyses looked at sex and at geographical region, splitting Europe and the USA from Asia.

Each 10 µg/m3 rise in ozone was associated with a relative risk of 1.0105 for cardiovascular mortality (95% CI 1.0027 to 1.0185) and 1.0578 for cardiovascular morbidity (95% CI 1.0289 to 1.0874). Exposure window mattered: warm-season ozone carried a mortality relative risk of 1.0186 (95% CI 1.0119 to 1.0253), whereas the annual average estimate was 0.9971 (95% CI 0.9781 to 1.0164) and did not exclude no effect. By sex, the mortality relative risk was 1.1135 in females (95% CI 1.0057 to 1.2328) and 1.0707 in males (95% CI 0.9855 to 1.1633), the latter crossing no effect. By region, estimates were 1.0129 in Europe and the USA and 1.0326 in Asia, with the Asian estimate also spanning no effect.

This adds structure to a literature the authors themselves describe as sparse, separating cardiovascular subtypes and exposure windows rather than pooling everything. The warm-season finding is the most notable feature, because it suggests peak-period exposure may drive the association in a way an annual mean obscures — a caution against averaging pollution over a year. The wide confidence intervals in the male and Asian subgroups mean the differences between subgroups are not firmly established. Because all included studies were observational, confounding by other pollutants, temperature and behaviour cannot be excluded.

The increases here are modest per unit of ozone, but they apply across whole populations rather than to individuals, which is how air-quality risks usually work. The clearest signal is that hot-season ozone is the part of the exposure worth watching, since the annual average showed no association with cardiovascular death. None of this establishes that ozone causes the outcomes; cohort data can only show that higher exposure and higher risk travel together. It is a reason to view summer air quality as a cardiovascular consideration rather than only a respiratory one.