Assessing Healthcare Cost Changes Associated with Transitioning away from Cigarette Smoking Using Healthcare Claims Data: an Exploratory Study among Adult Male Patients with COPD

Background: The assessment of potential health effects of switching from cigarette smoking to non-combustible tobacco products has important implications for public health and regulatory decisions. Robust epidemiological evidence requires long-term follow-up of a large number of individuals. Real-world evidence derived from health records has the potential to help fill the gap in the interim. To our knowledge, this is the first study using individual-level healthcare claims data to assess the potential impact of transitioning from cigarette smoking to smokeless tobacco on short-term direct healthcare costs. Methods: We conducted a retrospective cohort study of adult male patients with COPD who smoked cigarettes at baseline using the MarketScan ^® Databases. We compared changes in direct healthcare costs between the 12-month periods before (baseline) and after the index date (follow-up) across three cohorts: continued smoking (CS), quit all tobacco (QT), or switched to smokeless tobacco (SW), using a non-linear difference-in-differences model with average marginal effects. Results: A total of 23,427 COPD patients were included (CS: 11,167; QT: 12,013; SW: 247). At baseline, the QT cohort had the highest total average healthcare costs ($43,771), followed by SW ($38,419), and CS ($27,149). The unadjusted difference-in-differences model revealed no statistically significant differences in total healthcare cost changes when comparing the QT or SW cohorts to the CS cohort (-$1,532 [95% CI: -$3,671, $608] for the QT cohort, and -$452 [95% CI: -$15,415, $14,511] for the SW cohort). After adjusting for Deyo-Charlson Comorbidity Index and COPD exacerbation, the QT cohort had greater reduction in total healthcare costs compared to the CS cohort (-$2,910 dollars [95% CI: -$4,485, $-1,335]) assuming a two-unit increase in comorbidities and COPD exacerbations. The same trend was observed for the SW cohort, although the estimate was not statistically significant (-$5,312 [95%CI: -$11,067, $442], p=0.08). Conclusions: This study demonstrated the feasibility of using administrative claims to conduct real-world evidence studies on the harm-reduction potential of non-combustible tobacco products and found evidence suggesting reductions in direct healthcare costs after quitting tobacco or switching to smokeless tobacco among patients with COPD. Based on the learnings and limitations identified during the study, we propose concrete recommendations to improve future observational studies by integrating additional real-world healthcare data from multiple data sources.