This 29-year heart study tracked what happens after smokers quit

By Tarun Sai LomteReviewed by Susha Cheriyedath, M.Sc.Aug 20 2026

Decades of follow-up reveal how cumulative smoking exposure and the years after quitting shape the long-term risk of sudden cardiac death.

Long-term risk of sudden cardiac death according to cumulative smoking exposure and smoking cessation in the Copenhagen City Heart Study. Image Credit: Orawan Pattarawimonchai / Shutterstock

In a recent study published in the journal Scientific Reports, researchers investigated associations between smoking exposure and the risk of sudden cardiac death (SCD).

Smoking is among the most prevalent cardiovascular risk factors and is a central target of various prevention strategies. While its association with cardiovascular disease (CVD) and all-cause mortality (ACM) is well characterized, its relationship with SCD remains incompletely understood. SCD is a distinct form of mortality and may have mechanisms different from those involved in non-sudden death associated with atherosclerotic disease.

About the study

In the present study, researchers from Denmark and Japan investigated associations between smoking and SCD risk. They used data from the population-based Copenhagen City Heart Study (CCHS). The CCHS cohort was established in 1976 and underwent examinations around every decade. The analysis included participants from the third CCHS examination (CCHS3; baseline), conducted in 1991–94, and followed them until death or December 2021. Smoking status was assessed at CCHS3 and reassessed at CCHS4; the latter measurement was incorporated into sensitivity analyses examining time-updated smoking exposure.

Smoking exposure was determined using smoking status, years since cessation of smoking (in former smokers), and cumulative smoking exposure (in pack-years). SCDs were identified following standardized adjudication criteria. Cause-specific Cox proportional hazards models, with non-SCD deaths treated as competing events, estimated hazard ratios (HRs) for associations between smoking exposure and SCD. The primary analytic model was adjusted for sex, age, education, alcohol intake, leisure-time physical activity (LTPA), and income.

A secondary analytic model was further adjusted for diabetes, CVD, and chronic obstructive pulmonary disease. Population-standardized cumulative incidence (of SCD) was also estimated. The team decomposed cumulative mortality by cause of death (into SCD, non-cardiac, and other cardiac deaths). Further, they assessed the contribution of SCDs to smoking-associated excess ACM and explored dose-response relationships of cumulative exposure with SCD risk.

Exploratory analyses assessed cross-sectional associations of smoking status with select metabolic and cardiovascular markers using linear regression models adjusted for sex, age, education, alcohol intake, income, and LTPA. These markers were high-density lipoprotein cholesterol (HDL-C), body mass index (BMI), systolic blood pressure (SBP), resting heart rate (RHR), creatinine, and total cholesterol.

Findings

The study included 10,038 CCHS participants, of whom 9,106 were included in the primary complete-case analysis. Overall, 890 SCD events were recorded over a median follow-up of 28.6 years. At baseline, former smokers had a greater comorbidity prevalence and were older than never smokers, while current smokers had higher alcohol intake and lower LTPA levels and socioeconomic status. The median time since smoking cessation was 14.5 years among former smokers. The median cumulative exposure was 26.2 pack-years among current smokers.

The team found associations of both former and current smoking with an increased SCD risk compared to never smoking. HRs for SCD were 1.34 and 1.88 for former and current smokers, respectively, which were comparable in the secondary model. Population-standardized cumulative incidence over 25 years indicated progressively higher ACM risks across smoking categories. At 20 years, ACM was 25.5%, 20.4%, and 17.4% among current, former, and never smokers, respectively.

Cardiac deaths exhibited a graded increase across smoking categories, with SCDs consistently accounting for 13% to 16% of ACM across smoking categories. Notably, smoking-associated excess ACM at 20 years, relative to never smokers, was 8.2 percentage points in current smokers and 3.0 percentage points in former smokers, with SCDs contributing to 17.1% and 26.8% of excess ACM among them, respectively. Further, there was a dose-response relationship, with SCD risk increasing progressively with cumulative exposure before largely plateauing at the highest exposure levels.

Specifically, HR was 2.50 for participants in the highest-exposure category (40–159 pack-years) compared with never smokers. SCD risk declined with an increase in time since cessation among former smokers, with the risk approaching that of never smokers in those with the longest abstinence (26–67 years). In exploratory analyses, current smokers showed lower creatinine, BMI, SBP, and HDL-C, and higher RHR than never smokers. Total cholesterol did not differ significantly between smoking groups, and none of the examined markers differed significantly between former and never smokers.

The authors noted that smoking exposure was self-reported, pack-years were assessed only at baseline, and smoking behavior was captured at only two examinations. As an observational study conducted in an urban Danish population, the findings cannot establish causality and may not fully generalize to contemporary smoking patterns or other populations.

Conclusions

Collectively, smoking was strongly associated with an elevated SCD risk in this large, population-based cohort. SCD risk generally decreased in former smokers with time since cessation of smoking, although it remained elevated during the first decades after cessation.

Further, a dose-response relationship was evident, with SCD risk approximately doubling at higher cumulative exposure levels. Overall, the results underscore the importance of smoking prevention and early cessation to reduce SCD burden.