BACKGROUND-AIM:
Tobacco use is the leading cause of preventable death worldwide, due to its high prevalence and association with various diseases. Monitoring the smoking epidemic is essential for guiding public health policies, with attributable mortality (AM) being a key indicator. Estimating AM allows a robust assessment of the impact of smoking on a population. The objective of this study was to estimate AM to tobacco use in Spain in 2023 by sex, age, and specific cause of death using different lag times between smoking prevalence and observed mortality in population aged 35-84 years.

METHODS:
AM was estimated using a prevalence-dependent method based on the calculation of the population attributable fraction. This approach requires data on observed mortality, smoking prevalence, and relative risks. Observed mortality, for the year 2023, was obtained by sex and age group, including all causes of death associated with tobacco use. These causes were classified into three major groups: cardiovascular diseases, tumors, and respiratory diseases. Smoking prevalence was derived from the 2023- Spanish National Health Survey and from reconstructed prevalence based on the historical data from the Spanish National Health Survey (doi: 10.1093/eurpub/ckaf015). Relative risks associated with each cause of death, stratified by sex and age group, derived from five cohort studies conducted in the United States of America. AM due to tobacco use was estimated by applying no-lag specification and several lag times, representing the temporal delay between tobacco use and its associated mortality. So, two scenarios were defined: Scenario 1: no lag between smoking prevalence and observed mortality Scenario 2: 5-year lag for cardiovascular diseases, 15-year lag for cancer and 30-year lag for respiratory diseases

RESULTS:
Scenario 1 In the no–lag-time scenario, it was estimated that tobacco use caused 40350 deaths in Spain among individuals aged 35–84 in 2023. AM was higher in men (31991deaths) than in women (8359 deaths). By cause of death, tumors accounted for the highest proportion of AM [58.69% (n= 23680 deaths)], followed by cardiovascular diseases [25.47% (n= 10276 deaths)] and respiratory diseases [15.85% (n= 6394 deaths)]. Scenario 2 When lag times were applied, it was estimated that tobacco use caused 47056 deaths in Spain among individuals aged 35–84 in 2023. Regarding causes of death, tumors accounted for the highest percentage of deaths [58.80% (n= 27668 deaths)], followed by cardiovascular diseases [25.22% (n= 11868 deaths)] and respiratory diseases [15.98% (n= 7521 deaths)].

CONCLUSIONS:
The estimated AM to tobacco use varies depending on whether lag times are considered. This is particularly relevant in the case of tumors and respiratory diseases, where differences between scenarios are more pronounced. Considering lag times between prevalence and mortality allows for a more accurate estimation of the impact of tobacco use on population’s mortality.