Passive smoking associated with significantly increased risk of nine health outcomes

In a recent study published in natural medicine, researchers assessed and quantified the adverse health effects of secondhand smoke (SHS) exposure.

survey: Health effects associated with exposure to second-hand smoke: a burden-of-evidence study. Image credit: Namning / Shutterstock.com

The constant threat of passive smoking

Tobacco use, a leading global health risk, contributed to more than 229.8 million disability-adjusted life years and 8.7 million deaths in 2019. Exposure to SHS, which affects about 37% of the world’s population, particularly harms non-smokers, with women and children often at greater risk of exposure.

Despite reduced smoking rates, the health impact of SHS remains significant, particularly in low- and middle-income countries. In fact, the 2019 Global Burden of Disease (GBD) study attributed 1.3 million deaths to SHS.

Further research is needed to address gaps in evidence quality and study heterogeneity, to better understand the full health impact of SHS, and to effectively inform and improve global tobacco control policies and interventions for public health.

About the research

The researchers used the weight-of-evidence function (BPRF) methodology to estimate the association between SHS exposure and nine health outcomes while assessing the strength of the supporting evidence. SHS was defined as current exposure among nonsmokers to smoke from any combustible tobacco product, consistent with definitions used in previous GBD studies.

The BPRF framework, which has previously been used to estimate the health effects of smoking and dietary factors, uses a meta-regression-Bayesian adjusted and shortened (MR-BRT) tool to estimate pooled relative risks (RRs) and uncertainty intervals. This approach accounts for systematic bias, within-study correlation, and unexplained heterogeneity between studies.

A systematic review was used to extract data from relevant studies, estimate pooled RRs comparing SHS exposure risks while adjusting for biases, quantify unexplained heterogeneity between studies, assess publication and reporting biases, and assess BPRF to generate conservative risk estimates and corresponding risk-outcome scores (ROS).

The BPRF reflects the smallest harmful effect of an exposure that is consistent with the available evidence. The ROS, which is a signed value of the log RR, reflects the effect size and strength of evidence for each risk–outcome association, which is then translated into a star rating scale for interpretation.

The study did not disaggregate RRs by sex, geography, or age, except for breast cancer and asthma, which focused on female-only and child populations, respectively. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Guidelines for Reporting Accurate and Transparent Health Evaluations (GATHER) guidelines were followed, with the approval of the University of Washington Institutional Review Board.

The systematic review process involved searching PubMed and Web of Science for studies published between January 1970 and July 2022, in which researchers screened studies based on inclusion criteria and extracted data from selected publications. Effect sizes that closely matched the GBD risk definition were prioritized and the MR-BRT tool was subsequently used for meta-regression analysis, thereby generating pooled RRs for health outcomes in those exposed to SHS.

Biases in study design and characteristics were tested and adjusted using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach. Any remaining between-study heterogeneity was quantified using a linear mixed-effects model. Publication and reporting biases were assessed using funnel plots and Egger’s regression tests.

Sensitivity analyzes were performed to determine the strength of the primary findings. These analyzes include applying restrictions on data entry and ensuring reproducibility by providing data and code.

Survey results

A total of 410 publications from a pool of 9081 records were used for the systematic review. Of the included studies, 125 were conducted on asthma, 104 on lung cancer (104), as well as 21 on chronic obstructive pulmonary disease (COPD) and nine studies on type 2 diabetes. This resulted in 623 observations from different sites.

For cardiovascular disease, 37 studies or 59 observations assessed the association between SHS exposure and ischemic heart disease (CHD), while 20 studies or 26 observations assessed its association with stroke. RRs of 1.26 and 1.16 were reported for CHD and stroke, respectively, suggesting that SHS exposure increased the risk of CHD and stroke by 8% and 5%, respectively.

Cancer-related outcomes had a weak association between SHS exposure and lung cancer, with an RR of 1.37, while the RR for breast cancer was 1.22. Both associations were rated weak, with lung cancer receiving two stars and breast cancer one star in the BPRF framework. Sensitivity analyzes supported these weak associations and no significant publication bias was detected.

For respiratory conditions such as asthma, lower respiratory tract infections and COPD, the evidence is consistently rated as weak. The RRs for these conditions were 1.21, 1.34, and 1.44, respectively, with adjustments made for self-reported diagnoses and other biases. Sensitivity analyzes and tests of publication bias confirmed these weak associations.

Other health outcomes assessed included type 2 diabetes and otitis media. Weak deleterious effects of SHS exposure on the risk of type 2 diabetes and otitis media were reported, with RRs of 1.16 and 1.12, respectively. Both outcomes were associated with a one-star rating, indicating weak evidence of an association.