Table 1.

Policy Summary

Background	Chewing tobacco has historically received less public health attention than its smoked counterparts despite its inclusion in the WHO Framework Convention for Tobacco Control and high rates of use in some regions and communities where it is associated with important cultural practices. Chewing tobacco is often categorized under the vast umbrella of smokeless tobacco. Previous meta-analyses have found that studies on the health risks related to the broader category of smokeless tobacco conducted in south Asian countries—where chewing tobacco products dominate—typically report greater associated health risks. The present meta-analysis synthesizes the available evidence on chewing tobacco use and its association with seven health outcomes. We applied a meta-regression framework to data extracted through three comprehensive systematic reviews, spanning 53 years of peer-reviewed literature indexed by three major databases. Using Burden of Proof methodology, we generated conservative estimates of health risks associated with chewing tobacco use consistent with existing evidence—rigorously quantifying and incorporating measures of between-study heterogeneity, accounting for potential within-study correlation, and testing and adjusting for potential systematic bias.
Main findings and limitations	Our conservative interpretation of available data finds weak-to-moderate evidence of harmful associations between chewing tobacco use and esophageal cancer and stroke. The risk of these outcomes is at least 2-16% higher among tobacco chewers than non-chewers. However, we also detected a large degree of between-study heterogeneity in the evidence underpinning chewing tobacco’s association with esophageal cancer. High estimates of between-study heterogeneity were also observed in the weak existing evidence suggesting a relationship between chewing tobacco and lip and oral cavity cancer, laryngeal cancer, nasopharynx cancer, and other pharynx cancer. Ischemic heart disease has insufficient evidence of a significant risk-outcome relationship with chewing tobacco. There is a need for large high-quality prospective cohort studies to further our understanding of the health burden of chewing tobacco, particularly for risk-outcome pairs with limited or inconsistent evidence. Some limitations of our approach include the variability of exposure and outcome definitions. We included studies that reported on any chewing tobacco products because local chewing tobacco products vary depending on study location. However, the composition of different local chewing tobacco products may impact their health effects. We also did not restrict the use of aggregate outcome definitions in input data for the head and neck cancer outcomes analyzed because of the frequency with which cancer sub-types were grouped together. We evaluated the risk of systematic bias introduced by these two limitations. More broadly, we evaluated the health risks associated with chewing tobacco compared to not chewing tobacco without considering dosage. Accordingly, the dose-response relationship of these risk-outcome associations is an important area for future work.
Policy implications	In contrast to global trends of reduced smoking prevalence, the use of chewing tobacco has increased in recent decades, especially among youth and older women in some areas of the world. Despite the relative paucity of data, our analysis indicates that chewing tobacco use may increase the risk of stroke, esophageal cancer, lip and oral cavity cancer, nasopharynx cancer, other pharynx cancer, and laryngeal cancer. Our research further highlights the need for more large prospective cohort studies on the risks associated with chewing tobacco to bolster our understanding of its potential health consequences. These findings highlight the urgent need to better incorporate chewing tobacco into new and existing tobacco control efforts and to expand research efforts investigating the health burden of chewing tobacco.