Secondhand smoke, a study estimates 2.71 billion exposed and 1.66 million deaths in 2023
The Global Burden of Disease analysis published on September 8 in The Lancet Public Health covers 204 countries from 1990 to 2023. These are model-based estimates, with the limitations the authors disclose.
A study published on Tuesday, September 8, 2026 in The Lancet estimates that 2.71 billion people were exposed to secondhand smoke in 2023. Among those exposed are 767 million children aged 0 to 14 and 1.49 billion women. The authors associate this exposure with 1.66 million deaths and 44.8 million disability-adjusted life years. The work was picked up by the AFP news agency in reports by Free Malaysia Today and the Jamaica Observer, and reconstructed from the study by the Nigerian Tribune.
How the burden is distributed
The leading attributable cause of death, in the estimate, is ischemic heart disease, with approximately 537,000 deaths. Chronic obstructive pulmonary disease follows with 354,000, then stroke. It is a hierarchy that shifts the focus from the condition most intuitively associated with smoke — respiratory disease — to the cardiovascular system.
| Cause | Attributed deaths (2023) |
|---|---|
| Ischemic heart disease | approximately 537,000 |
| COPD | 354,000 |
| Stroke | not quantified in the excerpts |
The second figure to hold alongside the first is that of healthy life years lost: 44.8 million DALYs. This is an indicator that sums premature mortality and disability, and serves to prevent a chronic, non-fatal disease from disappearing from the count.
The limitations, disclosed
This is a Global Burden of Disease analysis: not a case count, but an estimate obtained from models. The analysis covers 204 countries and territories between 1990 and 2023 and is based on a review of studies indexed on PubMed and on population surveys. Some explicit limitations follow from this.
First: coverage of primary data is not uniform across countries, and where population surveys are rare the estimate rests on models more than on direct measurements. Second: this is a statistical association between exposure and mortality, not case-by-case causal proof; the attribution of 1.66 million deaths is the result of an attributable-risk calculation, not the outcome of autopsies. Third: exposure to secondhand smoke is almost always self-reported in surveys, with the margins of error this entails.
The authors state in the published text that exposure to secondhand smoke is a globally recognized risk and that no level of inhalation is free of risk: this is their conclusion, reported as such. The excerpts available to us do not report the conflict-of-interest disclosures contained in the publication, which remain in the full text on The Lancet.
Why a study like this is not enough on its own
A global estimate of this scale serves to guide public health policies — regulation of indoor environments, monitoring, surveillance — not to establish a definitive truth about a single disease determinant. Global Burden of Disease estimates are revised with each cycle: the values published today for 2023 are comparable with previous series only if methodological revisions are taken into account.
What we don’t know
We do not know, from the available excerpts, the breakdown by country or by WHO region, nor the uncertainty interval around the central values: in a model-based estimate, the interval is an integral part of the result. We do not know what the variation is compared with previous estimates from the same group, that is, whether the burden attributed to secondhand smoke is growing or decreasing over time.
The two sources reporting the figure partly trace back to the same news agency; the third refers directly to the study published in The Lancet Public Health.
← Archive · Front page · Past editorials · Report an error · Original article (in Italian)