Changes in tobacco sales before, during, and after the COVID-19 pandemic in Japan: An interrupted time series analysis

Objectives: This study aimed to longitudinally examine nationwide changes in smoking behavior among the Japanese population in response to the COVID-19 pandemic. Design and setting: A secondary analysis was conducted using cigarette sales data from the Tobacco Institute of Japan, monthly tobacco expenditures from the Family Income and Expenditure Survey conducted by the Ministry of Internal Affairs and Communications, and smoking prevalences from the National Database of Open Data Japan. An interrupted time-series segmented regression model adjusted for seasonality and autocorrelation was used to examine changes in cigarette sales and monthly tobacco expenditures before and after the first declaration of a state of emergency. Participants: Depending on the data source, data from 2014 to 2022, from 2015 to 2025, or from 2015 to February 2026 were used. Primary and secondary outcome measures: Changes in level and slope were evaluated before and after the first state of emergency and after COVID-19 was downgraded to a common infectious disease. Results: Cigarette sales declined at a significantly slower rate after the first state of emergency than before. Price-adjusted monthly tobacco expenditures increased significantly by 132 JPY compared with the previous month, although the expenditure findings were less robust in the quadratic sensitivity analysis. The smoking prevalence among men declined slightly each year, from 34.2% in 2014 to 31.9% in 2022. In contrast, the smoking prevalence among women remained relatively stable, ranging between 9.4% and 9.9%. Conclusions: Although cigarette sales declined more slowly and household tobacco expenditures increased during the first state of emergency in Japan, smoking prevalences did not change substantially. These findings suggest changes in tobacco purchasing patterns, including stockpiling, and possible changes in smoking intensity among continuing smokers, rather than a meaningful population-level decline in smoking behavior. There was no robust evidence of a clear reversal after May 2023.