Background and objective:
Smoking caused 8.96 million deaths in 2022 worldwide, representing the second largest modifiable risk factor. Although global smoking prevalence declined by 34.8% since 1990, it remains high in many countries. The objective of our study is to forecast cause-specific smoking attributable deaths, under reference and alternative scenarios.

Methods:
Our forecasts of cause-specific mortality to 2050 for 204 countries by age and sex are based on past trends in risk factors and other drivers of health from the Global Burden of Disease Study. Estimates of smoking attributable deaths use dose-response relative risk curves and continuous exposure distributions among both current and former smokers. In the reference scenario, we forecasted current and former smoking prevalence given past trends. To estimate avoidable deaths, we implemented a scenario eliminating smoking immediately. In this scenario, forecasted smoking attributable deaths are entirely due to past smoking exposure; the difference in deaths compared to the reference are deaths avoidable through tobacco control. Finally, we model an accelerated progress scenario that simulates linear elimination of smoking prevalence by 2050.

Results:
If past trends persist, we estimate 12.2 million deaths due to smoking in 2050, with a cumulative 310 million deaths over the next 28 years. Of these, 162 million deaths are potentially avoidable. We estimate 84.0 million lives could be saved by 2050 with accelerated progress, 66.7% of which would be in lower- and middle-income countries. In this scenario, deaths due to cardiovascular disease and lung cancer will be reduced by 25.9% and 23.0%, respectively.

Conclusion:
Smoking will continue to cause enormous health burden in the coming decade. Accelerated progress in tobacco control is necessary to maximize prevention of smoking deaths.