https://orcid.org/0009-0007-7803-4357
Figures
Abstract
Background
Uncertainties persist regarding the precise shape of the smoking-outcome curves across various cardiovascular and mortality endpoints. This study aims to elucidate the relationships among smoking burden, intensity, and cessation duration across multiple cardiovascular outcomes.
Methods and findings
Cox proportional hazard models were constructed to evaluate the association between pack-years, cigarettes per day (CPD), and years since cessation with cardiovascular outcomes in participants from 22 prospective cohort studies within the Cross-Cohort Collaboration Tobacco Working Group. We evaluated myocardial infarction (MI), stroke, coronary heart disease (CHD; MI, coronary revascularization, or coronary death), cardiovascular disease (CVD; stroke or cardiovascular death), heart failure (HF), atrial fibrillation (AFib), CHD mortality, CVD mortality, and all-cause mortality. Median follow-up varied across outcomes, with 14.4 years for MI (17,570 events), 19.3 years for CHD (30,625 events), 18.6 years for CVD (54,078 events), and approximately 19.4–19.9 years for mortality outcomes (CHD mortality: 17,429 events; CVD mortality: 33,120 events; all-cause mortality: 125,044 events). Spline terms were used to investigate the nonlinear association of continuous smoking/cessation measures with the examined outcomes. Models were adjusted for demographic, socioeconomic, and other cardiovascular risk factors. The study included 323,826 adults (148,635 non-mortality and 176,396 mortality outcomes with 25 and 16 million person-years at risk, respectively). Compared to never-smokers, current smokers had significantly increased risks for CVD (hazard ratio (HR) 1.74, 95% confidence intervals (CIs) [1.66,1.83] in men; HR 2.07, 95% CI [2.00,2.14] in women) and all-cause mortality (HR 2.17, 95% CI [2.09,2.25] in men; HR 2.43, 95% CI [2.38,2.48] in women; all p < 0.001). Compared with never-smokers, participants with 2–5 CPD demonstrated substantially elevated cardiovascular risks, with HR ranging from 1.26 (95% CI [1.09,1.45], p = 0.002) for AFib to 1.57 (95% CI [1.39,1.78], p < 0.001) for HF. Smoking 2–5 CPD was associated with increased CVD mortality (HR 1.57, 95% CI [1.41,1.75]), and all-cause mortality (HR 1.60, 95% CI [1.52,1.69]; both p < 0.001). Smoking 11–15 CPD conferred a higher risk of CVD (HR 1.87, 95% CI [1.69,2.06]) and all-cause mortality (HR 2.30, 95% CI [2.14,2.47]; both p < 0.001). The increased risk associated with the evaluated outcomes was steeper for the initial 20 pack-years and 20 CPD, respectively, compared to further smoking exposure. The most substantial reduction in risk across all outcomes was observed within the first 10 years after smoking cessation. However, the progressive risk reduction continues over extended time periods, with former smokers demonstrating over 80% lower relative risk than those of current smokers within 20 years of cessation. Limitations include potential exposure misclassification due to reliance on single baseline self-reported smoking measurements with extended follow-up periods, which may underestimate true risk associations, and lack of data on other tobacco products and electronic nicotine delivery systems, preventing analysis of dual- and poly-use patterns.
Conclusion
Lower-intensity smoking is associated with cardiovascular risk and the primary public health message for current smokers should be early cessation, rather than reducing the amount of smoking. Cessation provides substantial immediate risk reduction, although risk continues to decrease significantly for the following two decades.
Author summary
Why was this study done?
- Previous research established that cigarette smoking increases cardiovascular disease risk, but the exact relationship between smoking intensity and health outcomes remained unclear, especially for low-intensity smoking.
- Studies reported inconsistent timeframes—ranging from 2 to 29 years—for how long former smokers need to remain smoke-free before their risk approaches that of never-smokers.
- As smoking patterns shift toward low-intensity smoking in many populations, with more people smoking fewer cigarettes per day, there is a need to understand the cardiovascular risks of low-intensity smoking and the long-term benefits of quitting.
What did the researchers do and find?
- We analyzed data from 22 prospective cohort studies with 323,826 adults (76% women), following participants for up to 19.9 years and documenting over 125,000 deaths and 54,000 cardiovascular events.
- Low-intensity smoking (2–5 cigarettes per day) was associated with a 50% higher risk of cardiovascular disease and a 60% higher risk of all-cause mortality compared to never smoking.
- The most substantial risk reduction across all outcomes occurred within the first decade following smoking cessation. However, continued risk diminution persisted beyond this initial period.
What do these findings mean?
- Even occasional or low-intensity smoking significantly increases cardiovascular and mortality risks.
- The primary public health message of our paper is the importance of complete smoking cessation at younger ages
- Smoking cessation is associated with substantial risk reduction that begins immediately upon quitting and continues with a steep decline in risk during the first 10 years following smoking cessation.
- Limitations include the possibility that smoking habits changed over time since we only measured smoking at the study’s start, which may have led us to underestimate the true health risks. Moreover, we lacked information about other tobacco products like e-cigarettes, preventing us from studying people who use multiple tobacco products.
Citation: Tasdighi E, Yao Z, Dardari ZA, Jha KK, Osuji N, Rajan T, et al. (2025) Association between cigarette smoking status, intensity, and cessation duration with long-term incidence of nine cardiovascular and mortality outcomes: The Cross-Cohort Collaboration (CCC). PLoS Med 22(11): e1004561. https://doi.org/10.1371/journal.pmed.1004561
Academic Editor: Emily Banks, Australian National University, AUSTRALIA
Received: February 17, 2025; Accepted: September 11, 2025; Published: November 18, 2025
Copyright: © 2025 Tasdighi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data used in this study were collected from 22 different cohorts. Data cannot be shared, as for any researcher to access the data, there needs to be an official proposal request and a proposal review process approval from the respective cohort committee. The website URL link for each of the cohorts has been included in S1 Table. The code used in the analysis is available from Zenodo [https://doi.org/10.5281/zenodo.15477700].
Funding: This research was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (NIH) and the FDA Center for Tobacco Products (CTP) under Award U54HL120163 to MJB. E.T. has received support from the NIH (grant T32 HL007227). ACS has received funding from the NIH under the award 1K01HL154130. EJB has received funding from the American Heart Association under awards AF AHA_18SFRN34110082 and 2U54HL120163 and Framingham Heart Study NIH 75N92019D00031. APDF has received funding from the National Institutes of Health under award 2U54HL120163. Authors have included a detailed overview of the funding for all cohorts appearing in this study in the Acknowledgements section. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: R.T. is an advisor/consultant to Astellas Pharma, Bayer, and Hello Therapeutics. K.M. received personal fees from Fukuda Denshi, RhythmX AI, Kowa Company outside of the submitted work.
Abbreviations: AFib, atrial fibrillation; BMI, body mass index; CCC, Cross-Cohort Collaboration; CHD, coronary heart disease; CI, confidence interval; CPD, cigarettes per day; CVD, cardiovascular disease; HF, heart failure; HR, hazard ratio; MI, myocardial infarction
Introduction
Tobacco use is the leading preventable cause of cardiovascular disease (CVD) and mortality globally, contributing to more than 8 million deaths annually worldwide [1–3]. The relationship between tobacco smoking and cardiovascular disease has been firmly established through decades of landmark epidemiological research. Large-scale prospective investigations have consistently documented the profound cardiovascular impact of tobacco use, with the Australian 45 and Up Study demonstrating that current smokers have at least double the risk of developing most major cardiovascular diseases across 36 different disease subtypes [4]. Meta-analytic evidence from the CHANCES consortium, analyzing 503,905 older adults across 25 cohorts, revealed that smoking advances cardiovascular mortality by more than five years, with current smokers exhibiting a summary hazard ratio (HR) of 2.07 compared to never smokers [5]. Sex-specific analyses have shown that women may be disproportionately affected, with women who smoke having a 25% higher relative risk of coronary heart disease (CHD) compared to men with equivalent tobacco exposure [6]. These foundational studies have established smoking as one of the most potent modifiable risk factors for cardiovascular disease, with clear dose-response relationships that persist across diverse populations and the entire spectrum of cardiovascular pathology.
Beyond these well-recognized cardiovascular outcomes, smoking also significantly increases the risk of cardiac arrhythmias, including a 31% increased risk of atrial fibrillation (AFib) and flutter and a 50% increased risk of paroxysmal tachycardia [7]. The recognition of arrhythmic complications as smoking-related outcomes is particularly important given emerging evidence that approximately 17% of excess mortality among current smokers may be due to diseases not yet formally established as smoking-attributable, suggesting that the cardiovascular burden of smoking, including arrhythmic disorders, may be substantially underestimated in current public health assessments [8]. Public education campaigns and stringent regulations over the past few decades have led to a significant reduction in the use of conventional cigarettes, especially in high-income countries [9,10]. Moreover, a significant number of adult cigarette smokers express a desire to discontinue smoking, indicating widespread recognition of smoking’s adverse effects and a potential willingness to embrace cessation efforts [11]. However, data suggest that smokers who do not quit may be shifting their cigarette use to lower levels of smoking. Accordingly, a prospective cohort study in the National Institutes of Health–AARP (formerly known as the American Association of Retired Persons) the percentage of daily smokers consuming fewer than 10 cigarettes per day (CPD) has increased from 16% to 27% between 2005 and 2014, and the proportion of those who do not smoke daily rose from 19% to 23% [12]. A previous study showed that the risk for CHD was reduced in individuals who smoked 1 CPD compared to those smoking 20 CPD [13]. However, data regarding the impact of low-intensity smoking on cardiovascular diseases and death have not been fully explored [13].
Further, while the prevalence of former smokers is increasing, there is inconsistent evidence about how long it takes for their risk levels to align with those of individuals who have never smoked. Research suggests that for individuals who have quit smoking, it can take anywhere from 2 to 29 years for their risk of cardiovascular outcomes, as well as specific and all-cause mortality, to return to the same level as those who have never smoked. However, a recent large-scale meta-analysis demonstrated that smoking cessation confers substantial mortality benefits at any age, with short-term cessation (<3 years) reducing excess death risk by 90%–95% in younger adults and averting approximately 5 years of life lost. Long-term cessation (≥10 years) nearly eliminated excess mortality risk, achieving survival rates comparable to never smokers and averting about 10 years of life lost. This temporal variability in risk reduction underscores the complexity of quantifying smoking cessation benefits and emphasizes the importance of individualized risk assessment approaches [14–19].
The primary objective of the current study was to leverage a large study population with detailed participant phenotyping to provide robust dose–response risk estimates of cardiovascular and mortality outcomes associated with smoking intensity and cessation duration compared to individuals who have never smoked. Additionally, we sought to elucidate the distinctive aspects of the relationships involving cumulative pack-years, CPD, duration of cessation, and the concurrent association between cessation duration and smoking pack-years across nine outcomes. This information could assist the tobacco control community in developing evidence-based public policies, regulations, and clinical practice guidelines.
Methods
Ethics statement
This study was approved by the Johns Hopkins School of Medicine Institutional Review Board (Approval number: IRB00226738; Date: 24 June 2020). Parent cohorts had participants sign informed consent, and were covered by local IRBs. The Johns Hopkins IRB committee granted a waiver regarding informed consent in this study because this study used only data previously collected under individual cohort-level IRBs.
Study population
CCC-Tobacco, a working group of the Cross-Cohort Collaboration (CCC), has created a harmonized dataset from 22 prospective cohort studies, encompassing 322,782 participants, 21 from the U.S. and 1 cohort from Brazil. CCC-Tobacco comprises both traditional cardiovascular cohorts (N = 12) and noncardiovascular cohorts aimed at studying aging (N = 10). The rationale and design of the CCC-Tobacco project have been previously described in detail [20]. Descriptions of these cohorts and their baseline characteristics are shown in S1 and S2 Tables.
Definition of exposures
Participants who reported smoking fewer than 100 cigarettes in their lifetimes were classified as never smokers. Conventional cigarette use was defined as a lifetime use of at least 100 cigarettes. Current conventional cigarette use was characterized by self-reported ongoing use at the time of the baseline examination on individual cohort questionnaires. The Characteristics of smoking parameters by cohorts are shown in S3 Table. Moreover, the timeline of baseline examination of each cohort has been laid out in S4 Table. Former use was defined as self-reported cessation of smoking conventional cigarettes.
Smoking burden and intensity were assessed using two metrics at baseline: pack-years and CPD. Both pack-years and CPD were evaluated as continuous variables (i.e., per 10 increase) and also as categorical variables. Pack-years were categorized into four groups (≤5, 6–10, 11–20, and >20 pack-years). CPD categories were set as (≤1, 2–5, 6–10, 11–15, 16–20, and >20 CPD). While we analyzed the risk estimates for pack-years separately in both former and current smokers, CPD by its nature was evaluated only among current smokers.
The CPD was collected as a continuous variable in most studies. However, if a study collected the CPD as a categorical variable, we harmonized this variable as a continuous variable. For instance, if a cohort measured the CPD as categories of 1–3 CPD, we use the average number of CPD (i.e., 2 CPD) for harmonization purposes.
For pack-years, the cohorts were collected as continuous variables, or we calculated the pack-years as a continuous variable using the duration of smoking and CPD.
For analysis, we utilized both continuous and pre-defined categories for CPD and pack-years. For instance, the categorical analyses used the predefined categories (≤1 CPD, >1 to 5 CPD, >5 to 10 CPD, etc.) The spline models presented in the figures treated CPD as a continuous variable.
Cessation duration among former smokers was defined at the baseline examination of each cohort based on the years since quitting smoking at the baseline examination of each cohort, determined by the self-reported age of quitting. We categorized cessation duration into four groups (≤10, 11–20, 21–30, and 31–40 years since quitting smoking).
Cardiovascular and mortality outcomes
The selection of outcomes was based on those that were systematically collected and formally adjudicated across most of the participating cohorts. Moreover, we included various cardiovascular outcomes to provide comprehensive insights into the potential cardiovascular risks associated with conventional cigarette tobacco products.
A total of nine outcomes relevant to cardiovascular health were collected and harmonized in CCC-Tobacco: myocardial infarction (MI), stroke, heart failure (HF), AFib), CHD, cardiovascular disease (CVD), CHD mortality, CVD mortality, and all-cause mortality. CHD events were defined as a composite of MI, coronary revascularization, or coronary death. CVD events were defined as a composite of all Atherosclerotic CVD events, including CHD, stroke, or cardiovascular death (coronary death, stroke death, other atherosclerotic death, or other CVD death).
Median follow-up by outcome, ranging from 8.7 years (AFib) to approximately 19 years for CHD (19.3), CVD (18.6), and mortality outcomes (19.4–19.9), underscoring the prolonged interval between baseline smoking exposure and observed events.
Cardiovascular outcomes in this study were harmonized across the 22 cohorts by using each cohort’s specific definitions of the outcomes, which in most cases were adjudicated by a dedicated adjudication committee.
In the occasion that an individual cohort had some, but not all, components (i.e., angina) of the CHD or CVD composite events, the components that were present were retained to represent a modified CHD or CVD composite for that cohort.
Harmonization of covariates
The definitions of all demographics, anthropometric, and traditional risk factors, including hypertension, diabetes, hyperlipidemia, and dyslipidemia, have been previously described [20]. Data harmonization in CCC-Tobacco adhered to published best practices in the field and has been coded into a master file to enable replication.
Harmonization of covariates for race/ethnicity was based on self-reported data and adhered to the data collection protocols of each cohort into American Indian or Alaskan, Asian, Black/African American, Hispanic, White, and Other. Participants with missing information on race/ethnicity were categorized as “other” in this study.
In the case of missing supportive risk factor data in <10% of total participants, multiple imputations were conducted using the remaining non-missing risk factors within each individual cohort [21]. For missing continuous data related to diastolic blood pressure measurements, rule-based imputation was used, utilizing non-missing systolic blood pressure and binary risk factor data from the rest of the dataset, along with average diastolic blood pressure from comparable subgroups in the CCC-tobacco dataset. Missing data on blood pressure-lowering and lipid-lowering medication and hyperlipidemia were imputed based on age, gender, hypertension, diabetes, and presence of CHD at baseline. We use hyperlipidemia to impute blood pressure and lipid-lowering medication as well. Imputations were done based on the records with at most one missing risk factor, except for lipid-lowering medication in the SOF cohort, which allowed for at most two risk factors. Within the CCC-Tobacco dataset, these methods produce nearly identical mean and median values for all risk factors.
Statistical analysis
These baseline characteristics are presented by baseline smoking status (conventional cigarette smoking: never, former, current). Age- and sex- adjusted absolute risk of each outcome in the total population and according to smoking status was based on the 2000 U.S. Census data.
Baseline variables were utilized to conduct adjusted Cox proportional hazards models in the pooled cohort to evaluate the association between smoking and the 9 study outcomes. The first multivariable model was adjusted for age, sex, race, and ethnicity (White, Black/African American, Asian, Hispanic, American Indian or Alaska Native, and Other), and education status (did not complete high school or less than 12 years of full-time education, completed high school or 12 years of full-time education, college degree or higher, or more than 12 years of full-time education). The second multivariable model was additionally adjusted for harmonized covariates, including body mass index (BMI), systolic blood pressure, diastolic blood pressure, diabetes, hyperlipidemia, antihypertensive medication use, lipid-lowering medication use, history of CHD at baseline, and any alcohol use. The reference group for all analyses was individuals who had never smoked. Of note, to consider the intra-group correlation within cohorts, we incorporated a shared frailty component into our Cox model, represented by the variable “cohort”, which includes 22 unique cohort identifiers. This approach enhances the robustness of our inference by acknowledging and adjusting for the nonindependence of survival times within cohorts, thereby providing a more accurate estimation of the association of covariates on survival.
Cubic splines for pack-years and CPD were utilized to allow for nonlinear associations between these continuous predictors and the HR of the 9 outcomes and these continuous predictors. The cubic splines were included in the Cox proportional hazards models, adjusting for age, sex, education, and cohort status.
Using a prediction model, we also evaluated the joint association of the cessation duration and smoking pack-years among former smokers. The Cox model’s predictions were used to compute the HR for each graded combination of cessation duration and smoking pack-years strata. The final heat plots illustrate the relative hazards across varying pack years and cessation periods, with the lines connecting areas of equivalent risk, providing insights into smoking cessation’s long-term effects and the risk of smoking intensity.
To examine temporal trends in smoking-related cardiovascular risks, we stratified participating cohorts by enrollment year using 2001 as the median cutpoint, comparing earlier (≤2,001) versus later (>2,001) enrollment periods. The WHI cohort was excluded from this analysis due to its exceptionally large sample size (n = 161,808), uniform enrollment year, and inclusion of only women, which created an analytical artifact that obscured true chronological trends in smoking epidemiology. This stratification approach allowed for balanced temporal comparison and assessment of how smoking-related cardiovascular HRss have evolved over time. Sensitivity analyses examining standardized rates by time period and censoring follow-up at 6 and 10 years produced results identical to the primary analysis using complete follow-up, confirming the robustness of our temporal trend findings. To address potential bias from the “sick quitter effect,” where individuals may be more likely to quit smoking after developing health conditions, additional methodological approaches were incorporated into the analysis. First, we leveraged the exclusion criteria of the cohort studies, which typically eliminated participants with pre-existing serious conditions (cancer, advanced kidney disease, nursing home residence, or other conditions preventing long-term participation). Second, we conducted sensitivity analyses reclassifying recent quitters (≤2 years) as current smokers to account for illness-induced smoking cessation and excluded participants with CHD at baseline. Third, we performed age-stratified analyses comparing participants younger than or equal to 60 years versus those older than 60 years to examine whether the relationships between smoking cessation and outcomes differed by age groups, as the “sick quitter effect” tends to be more pronounced in older populations.
This study was conducted according to a prospective analysis plan developed as part of the CCC framework. Due to the multi-cohort nature of this collaboration, each participating cohort received a slightly differently formatted proposal tailored to their specific data structure and governance requirements, though all followed the same core analytical approach and objectives. The primary statistical analysis plan was established prior to data analysis and focused on examining associations between smoking status, intensity, and cessation duration with cardiovascular and mortality outcomes.
Several modifications to the original analysis plan were made during the peer review process to strengthen the study’s methodological rigor. Specifically, the examination of temporal trends stratified by enrollment year was added in response to reviewer comments requesting an investigation of how smoking-related risks have evolved over time. The exclusion of the Women’s Health Initiative cohort from temporal trend analyses was implemented after identifying that its large sample size and uniform enrollment year created analytical artifacts that obscured true chronological trends. Additionally, methodological approaches to address the “sick quitter effect” were enhanced during peer review.
This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 STROBE Checklist).
Results
The total analytic sample size was 323,826, with a mean (SD) age of 59.7 (11.9) years; 76% were women. A total of 46,125 (14.08%) individuals reported current smoking, 119,049 (36.4%) were never smokers, and 158,652 (49%) were former smokers of conventional cigarettes.
In general, compared to never users of conventional cigarettes, former smokers were older, and the prevalence of women was lower among current smokers. White and Black or African American race/ethnicities were the most prevalent among all smoking status groups. Never and former smokers had higher educational attainment compared to current smokers. Alcohol use was more prevalent among current and former smokers compared to never users. The current smokers showed a more adverse cardiometabolic profile compared to other groups (Table 1).
The median (25–75 IQR) smoking pack-years among former and current smokers was 13 (3.5–30) and 25 (10.9–45), respectively. The mean (SD) CPD amongst current smokers was 19.8 (15.14). The mean (SD) cessation duration among former smokers was 20.8 (13.1) years since quitting (Table 1).
The total number of events, age-adjusted incidence rate, and HR for each of the nine study outcomes were shown separately in men and women in Tables 2 and 3, respectively. In men, after full adjustment in Model 2, current male smokers exhibited significantly elevated risks with HRs of 1.70 (95% CI [1.58,1.82]) for MI, 1.55 (95% CI [1.41,1.71]) for stroke, 1.70 (95% CI [1.60,1.81]) for CHD, 1.74 (95% CI [1.66,1.83]) for CVD, 1.89 (95% CI [1.75,2.04]) for HF, 1.57 (95% CI [1.45,1.70]) for AFib, 1.87 (95% CI [1.71,2.04]) for CHD mortality, 1.87 (95% CI [1.75,2.00]) for CVD mortality, and 2.17 (95% CI [2.09,2.25]) for all-cause mortality (all p < 0.001) (Table 2). Women current smokers, compared with never smokers counterparts, demonstrated HRs of 2.17 (95% CI [2.05,2.30]) for MI, 1.77 (95% CI [1.67,1.87]) for stroke, 2.23 (95% CI [2.13,2.33]) for CHD, 2.07 (95% CI [2.00,2.14]) for CVD, 2.09 (95% CI [1.97,2.22]) for HF, 1.54 (95% CI [1.42,1.66]) for AFib, 2.48 (95% CI [2.33,2.63]) for CHD mortality, 2.32 (95% CI [2.22,2.42]) for CVD mortality, and 2.43 (95% CI [2.38,2.48]) for all-cause mortality (all p < 0.001). Former smokers in both sexes showed intermediate risk profiles, with incidence rates consistently falling between never and current smokers, and statistically significant HRs (all p < 0.001) typically ranging from 4% to 27% increased risk across outcomes after full adjustment in Model 2 (Table 3). Additional sensitivity analyses were conducted to assess the robustness of our findings. These included analyses excluding participants with reported CHD at baseline and stratifying participants by age (≤60 years versus >60 years), all of which yielded comparable results to the primary analysis. (S5–S7 Tables).
Smoking intensity and cardiovascular and mortality outcomes
The association between continuous and categorical values of smoking pack-years and evaluated outcomes among former and current smokers compared to never smokers is presented in S8 and S9 Tables. For every 10 pack-year increments, there was a further increased risk of between 2.4% and 4.6% across the outcomes. Former smokers with up to 5 pack-years of smoking did not show a significant association with any of the outcomes. Compared with never-smokers, current smokers with ≤5 pack-years showed an increased risk for all the outcomes except for AFib. There was an increased risk for all the outcomes for current smokers with >6 pack-years.
Fig 1 illustrates the HRs (95% CI) of each pack-year category and cardiovascular and mortality outcomes for former and current smokers compared to never smokers. This figure shows that current smokers consistently had a higher risk than former smokers, irrespective of the pack-year category, for all the outcomes except for AFib.
Groups have been created based on the smoking status (i.e., F: former smokers, C: current smokers) and the pack year count category. The hazard ratio (HR) and 95% CI have been revived from the comprehensive adjusted model. The Cox models were adjusted for age, sex, race and ethnicity, education status, cohort, body mass index (BMI), systolic blood pressure, diastolic blood pressure, diabetes, hyperlipidemia, antihypertensive medication use, lipid-lowering medication use, history of coronary heart disease at baseline and any alcohol use.
The cubic splines analysis (illustrated in Fig 2) revealed a nonlinear dose-response relationship between pack-years of smoking and the risk of cardiovascular and mortality outcomes. Initially, the curve began with a steep ascent, indicating that the initial 20 pack years were associated with a pronounced increase in evaluated outcomes. After approximately 20 pack years, the curves still increased but at a decelerating rate.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure; AFib, atrial fibrillation; 95% confidence interval has been shown with dashed line.
The association between continuous and categorical values of CPD and evaluated outcomes among current smokers is presented in Table 4. Participants with CPD ≤1 showed significantly increased risks across most cardiovascular outcomes, with HRs ranging from 1.16 (95% CI [0.87,1.53]; p = 0.30) for AFib to 2.07 (95% CI [1.69,2.54]; p < 0.001) for HF. Those smoking 2–5 CPD demonstrated consistently elevated risks compared to never-smokers, with HRs ranging from 1.26 (95% CI [1.09,1.45]; p = 0.002) for AFib to 1.60 (95% CI [1.52,1.69]; p < 0.001) for all-cause mortality, indicating substantial cardiovascular harm even at very low levels of smoking exposure. Of note, compared to other outcomes, AFib estimates showed wider confidence intervals, which demonstrate more heterogeneity in the effects of smoking on AFib. Using Cox proportional hazards models with cubic spline functions indicated a significant risk increase associated with the first 20 CPD across all outcomes assessed. Beyond this point, the risk escalation began to level off for most outcomes. However, the risk associated with AFib continued to climb even after exceeding 20 CPD, diverging from the pattern seen with other conditions (Fig 3 and S10 Table).
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure; AFib, atrial fibrillation; 95% confidence interval has been shown with a dashed line.
A sensitivity analysis was performed to compare the association of smoking status and outcomes based on the individuals’ enrollment year. Among participants enrolled after 2001 compared to those enrolled in 2001 or earlier, current smokers demonstrated consistently higher HRs across all measured outcomes. In the adjusted Model 2, current smokers enrolled after 2001 showed markedly elevated risks for MI (HR 1.92, 95% CI [1.72,2.13] versus HR 1.74, 95% CI [1.64,1.85]), stroke (HR 1.70, 95% CI [1.50,1.93] versus HR 1.55, 95% CI [1.44,1.66]), CHD (HR 1.97, 95% CI [1.79,2.16] versus HR 1.73, 95% CI [1.65,1.82]), and overall mortality (HR 2.44, 95% CI [2.32,2.57] versus HR 2.05, 95% CI [1.99,2.11]; all p < 0.001). Former smokers also exhibited modestly increased risks in the later enrollment period for most cardiovascular outcomes, though the temporal differences were less pronounced than those observed for current smokers (Table 5).
Cessation duration and cardiovascular and mortality outcomes
The association among former smokers between cessation duration categories and each evaluated outcome compared with never-smokers is illustrated in Fig 4. The higher risk for MI with HR 1.09 (95% CI [1.02,1.16]; p = 0.012), AFib with HR 1.08 (95% CI [1.01,1.17]; p = 0.049), and CVD mortality with HR 1.04 (95% CI [1.00,1.09]; p = 0.018) remained significant among former smokers with 21–30 years since quitting. The age-stratified analyses showed that both age groups demonstrated substantial benefits from smoking cessation across all cardiovascular outcomes and mortality endpoints. (S2 and S3 Figs). For most outcomes, risk normalization (HR approaching 1.0, similar to never smokers) occurred more rapidly in the younger age group. For instance, MI, stroke, and CHD HRs declined to near or below 1.0 by the 20–30 year mark for participants ≤60 years old.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease, HF, heart failure, AFib, atrial fibrillation. Current smokers presented with a red dot-dashed line. The reference group is never-smokers presented with a dotted black line.
Fig 5 illustrates the nonlinear relationship between the duration of cessation and the risk reduction for various outcomes. The curve demonstrates a more pronounced decrease in risk during the initial 15 years following cessation, transitioning to a plateau phase after approximately 20 years for most outcomes except for AFib, for which the risk continues to descend even beyond 40 years of cessation.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure; AFib, atrial fibrillation; 95% confidence interval has been shown with a dashed line.
Fig 6 depicts predicted HRs for each outcome based on concurrent consideration of smoking cessation duration and the number of pack-years among former smokers. Notably, for cardiovascular and mortality outcomes, the change in predicted HR (i.e., change in the color gradient) along the time since cessation y-axis exceeds the change in predicted HR across the cumulative pack-years x-axis. A visual analysis of the lines connecting areas of equivalent risk indicates that 5–10 additional years of time since smoking cessation may be roughly comparable to an incremental 30–50 pack-years of smoking burden in terms of total risk. Participants with the shortest time since cessation and with the largest number of pack-years were at the highest risk, while those with the longest time since cessation with the smallest number of pack-years were at the lowest risk.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure; AFib, atrial fibrillation.
Our sensitivity analyses examining the association between cessation duration and evaluated outcomes were conducted to assess potential confounding from the sick-quitter effect. Reclassification of recent quitters (≤2 years) as current smokers to account for illness-induced smoking cessation yielded results consistent with the primary analysis (S1 Fig). Additionally, age-stratified analyses were performed, given that the sick-quitter effect is more pronounced in older populations (S2 and S3 Figs), which confirmed the immediate risk reduction following cessation and demonstrated continued steep risk attenuation for at least 20 years post-cessation.
Discussion
This comprehensive study, encompassing 22 predominantly U.S.-based cohorts with follow-up for cardiovascular outcomes spanning a median of 24 years, presents a large and detailed analysis of the impacts of different smoking measures on cardiovascular, cause-specific, and all-cause mortality outcomes. Our analysis delivers point estimates of risk for former and current smoking, with an emphasis on low-intensity smoking behaviors. In particular, we delineate the specific dynamics of the relationships between smoking pack-years, CPD, and cessation duration and cardiovascular and mortality outcomes. This study also highlights the complex interplay and nonlinear associations between cumulative smoking exposure and the time since smoking cessation in former smokers, underscoring the critical need to educate the public and promote early smoking cessation.
The findings of our research underscore the assertion that no level of smoking is without risk. Our findings indicate that current smokers using up to 1 CPD are associated with elevated risks of cardiovascular and mortality incidents, except for stroke and AFib. Notably, the 2–5 CPD range was associated with an increased risk for all assessed outcomes. Consistent with these findings, it has been reported that men and women smoking 3–5 CPD exhibited HRs for all-cause mortality of 45% increased risk (95% CI [33, 59]; p < 0.001) and 49% increased risk (95% CI [34, 66]; p < 0.001), respectively [22]. Additionally, another study investigating 505,500 nationally representative U.S. adults revealed that lifelong non-daily smokers exhibited higher all-cause mortality risks 82% increased risk (95% CI [65, 101]; p < 0.001)compared to never-smokers [23]. The increased risk of low-intensity and other outcomes other than cardiovascular outcomes, such as cancer, has also been reported [24].
Comparing the associations within pack-year categories between former and current smokers, it was observed that the magnitude of health risk for former smokers within the highest pack-year group (>20 pack-years) was lower than the magnitude of health risk for current smokers within the lowest pack-year group (≤5 pack-years). This finding reinforces the argument that pack-years alone may not sufficiently capture the complexities of smoking-related cardiovascular risk [25]. Incorporating both smoking status and pack-years as combined measures of smoking intensity offers a more nuanced understanding of the relationship between smoking behaviors and cardiovascular outcomes. This approach acknowledges the significant differences in risk between current and former smokers and underscores the importance of considering both the quantity and duration of smoking in risk assessments. Notably, the dose-response curve for smoking intensity and its associated health outcomes did not plateau until reaching the thresholds of approximately 20 CPD and 20 pack-years.
Quitting smoking is a critical step for individuals who smoke to lower their cardiovascular risk [25]. Although evidence varies regarding the timeline for excess cardiovascular risk reduction following smoking cessation, most studies suggest a return to levels comparable to never-smokers after 20 years of cessation [15,17,18]. Our study findings showed that former smokers exhibited risk reductions of greater than 80% compared to current smokers within two decades of smoking cessation. Moreover, it is important to note that the cubic spline analysis showed that the reduction in risk starts immediately after quitting and continues to decrease substantially up to 20 years post-cessation, and every attempt should be taken into account to encourage early quitting. The concurrent evaluation of cessation duration and smoking pack-years indicates that the time elapsed since quitting holds greater importance for risk estimation than the total pack-years. Moreover, quitting five years earlier may compensate for a significant portion of the risk associated with an extensive smoking history. Moreover, our results showed that the effect of increasing pack-years is more prominent in mortality outcomes than cardiovascular outcomes, which emphasizes the strong cumulative effect of smoking on total mortality [26].
Comparison of the two eras shows an increase of the smoking–cardiovascular dose–response relationship over time. The observed temporal trend of increasing smoking-related cardiovascular risks represents a concerning epidemiological phenomenon that warrants careful consideration of multiple contributing factors. The heightened HRs among participants enrolled after 2001 may reflect several interconnected mechanisms, including evolving smoking behaviors and patterns and improvements in baseline cardiovascular health among never-smokers that accentuate the relative harm of smoking exposure. This finding is consistent with multiple papers by leaders in tobacco epidemiology that have demonstrated relative risks of all-cause and cardiovascular disease mortality have increased over time, as documented in the British Doctors Study [27] and analyses of 50-year smoking trends in the United States [17]. These temporal increases in relative risk have been largely attributed to very rapid falls in mortality rates among never smokers and less rapid falls (or even stagnation) in rates among people who smoke. This phenomenon has been linked to the “maturing” of the tobacco epidemic, whereby people who smoke in the most recent birth cohorts have commenced smoking at younger ages and have smoked heavily throughout their lives, whereas people who smoke in earlier cohorts, on average, started smoking at later ages and smoked less on average. The temporal intensification of smoking’s cardiovascular impact suggests that contemporary smokers may face even greater health risks compared to their never-smoking peers than previously estimated from earlier cohort studies, underscoring the critical importance of smoking cessation interventions and reinforcing the public health imperative to prevent smoking initiation in younger populations who may be exposed to these heightened risk profiles throughout their lifetime. Our study has several strengths. First, CCC-Tobacco is an integration of highly detailed, primarily U.S.-based cohorts focusing on cardiovascular health and aging. Additionally, the prolonged observation period of up to 74 years significantly improves our capability to detect even small associations. Moreover, this comprehensive and large study population provided a precise effect size estimate of low-burden smoking and precise shapes of associations that could be utilized as robust evidence for regulatory authorities in making informed decisions. Secondly, by harmonizing various cohorts with respect to specific demographic criteria, as well as racial and ethnic categories, we achieved a representation in which women constituted the majority of the study population. Additionally, we included diverse racial and ethnic groups, including White, Black or African American, Hispanic or Latino, and American Indian or Alaskan Native. Third, to the best of our knowledge, this is the first study to investigate the concurrent influence of cessation duration and smoking pack-years on different outcomes. It is also important to notice that the CCC-Tobacco working group will continue to extend this work with a special focus on age, sex, and race disparities regarding the associations of smoking burden, smoking intensity, and time since quitting with the various cardiovascular and mortality outcomes.
This study has limitations. First, smoking status was determined based on self-reported data from individual cohort surveys. The stigma attached to smoking could lead to underreporting of current smoking status, particularly among women, affecting the accuracy of the data. Nevertheless, it is important to acknowledge that self-reported measures of smoking behavior are the current standard in clinical practice settings [28,29]. Moreover, data regarding other tobacco products and electronic nicotine delivery systems were not available for all the participants of the study population; hence, we could not investigate other use patterns, such as dual- and poly-use, in our analysis. Given the observational study design, we cannot exclude residual confounding, including but not limited to types of cigarettes and time cohort effects.
Our study spans over 50 years, and since smoking-related risks have changed over successive decades, our estimates represent an average across time and may underestimate or overestimate current risks. However, we have conducted a sensitivity analysis based on the baseline enrollment date of the participant, stratifying them into before and after enrollment in 2001 to clarify the distinction of the relative risk in these two different periods, although this analysis should also be interpreted with caution, as different decades are represented by different cohorts, introducing some inherent variability. Additionally, we analyzed data spanning multiple years of follow-up but could not fully explore how the relationship between the use of tobacco products and cardiovascular disease outcomes has evolved over time, especially considering the evolution of the tobacco industry and tobacco regulation.
Due to variability in baseline cardiovascular disease ascertainment across cohorts, we could not consistently exclude all participants with all prevalent cardiovascular conditions (e.g., CHD, stroke, AFib, HF) from the primary analysis. Therefore, our primary estimates could reflect a mixture of incidence and recurrent cardiovascular events. Although we addressed this issue through adjustment and sensitivity analyses, excluding individuals with known prevalent disease at baseline, residual bias from incomplete exclusions might persist. Comparison between Tables 4 and S5, however, shows that there are comparable risk estimates generated from each approach. However, emerging evidence has demonstrated that most predictors, including smoking status, show similar associations with MACE regardless of baseline atherosclerotic CVD status and have emphasized universal risk prediction models [30].
We acknowledge limitations in estimating the relationship between time since quitting and health outcomes, particularly at older ages, where the “sick quitter effect” becomes more pronounced. While our sensitivity analyses provide reassurance regarding the overall robustness of our findings, caution is warranted when interpreting the precise temporal relationship between smoking cessation and risk reduction over extended timeframes. This study’s analysis of “time since quitting” smoking is limited by the challenge of time-varying duration. Younger individuals rarely exhibit long quit durations, violating the positivity assumption and confounding results with age-related biases. Additionally, left-censoring due to mortality skews the distribution, as older individuals with shorter quit times are underrepresented, creating potential selection bias. Moreover, due to the unavailability of data on other causes of mortality, including cancer mortality, a competing risk analysis was not conducted. Lastly, our reliance on a single baseline self-reported measurement of smoking exposure, combined with the substantial time gap between this assessment and outcome measurement, introduces potential exposure misclassification as smoking behaviors likely changed for many participants during follow-up. This regression dilution bias may have attenuated the observed associations. Prior cohort studies suggest that using a single measurement of smoking can underestimate true relative risks by up to 15%, particularly for intensity-related exposures [17,31]. Future work in CCC-Tobacco will seek to harmonize tobacco use data across all visits for all cohorts, which will enable the study of product transitions, including cessation. We reported comprehensive smoking behavior measures and their association with various cardiovascular, cause-specific, and all-cause mortality outcomes. The results of our study showed that although former smokers with cumulative exposure below 5 pack-years did not show statistically significant associations, continuous dose-response analyses and findings among current low-intensity smokers underscore that no threshold of exposure is risk-free. Our results indicate that there is a substantial risk reduction within the first 20 years of smoking cessation. However, even after 21–30 years of cessation, former smokers may still exhibit higher risks compared to those who never smoked. These findings suggest that health authorities should emphasize both smoking cessation and the prevention of smoking initiation. Our findings reinforce well‐established public health guidance—that recent cessation drives substantially greater cardiovascular and mortality risk reductions than intensity reduction alone. Consistent with WHO recommendations, complete quitting should remain the principal goal, with any reductions in smoking intensity viewed only as interim steps toward full cessation. The epidemiological findings from this study can aid in shaping regulatory strategies and intervention guidelines against smoking. Additionally, they provide a framework for future research to delve into the complex dynamics between smoking habits, various population subgroups, and their associations with cardiovascular health outcomes. As part of the CCC-Tobacco project, future projects will focus on age, sex, and race disparities the evaluated association of smoking parameters, and a comprehensive set of cardiovascular and mortality outcomes.
Supporting information
S1 Table. Characteristics of the 23 participating cohorts of the cross-cohort collaboration-tobacco dataset.
https://doi.org/10.1371/journal.pmed.1004561.s001
(DOCX)
S2 Table. Baseline characteristics of the included cohorts (parts 1&2).
https://doi.org/10.1371/journal.pmed.1004561.s002
(DOCX)
S3 Table. Characteristics of smoking parameters by cohorts.
https://doi.org/10.1371/journal.pmed.1004561.s003
(DOCX)
S4 Table. Outcomes follow-up time in each cohort.
https://doi.org/10.1371/journal.pmed.1004561.s004
(DOCX)
S5 Table. Association between smoking status and incidence of cardiovascular and mortality outcomes, excluding patients reported baseline cardiovascular disease.
https://doi.org/10.1371/journal.pmed.1004561.s005
(DOCX)
S6 Table. Association between smoking status and incidence of cardiovascular and mortality outcomes in participants older than 60 years old.
https://doi.org/10.1371/journal.pmed.1004561.s006
(DOCX)
S7 Table. Association between smoking status and incidence of cardiovascular and mortality outcomes in participants aged younger or equal to 60 years old.
https://doi.org/10.1371/journal.pmed.1004561.s007
(DOCX)
S8 Table. Association between pack years with cardiovascular outcomes among former cigarette users.
https://doi.org/10.1371/journal.pmed.1004561.s008
(DOCX)
S9 Table. Association between pack years with cardiovascular outcomes among current cigarette users.
https://doi.org/10.1371/journal.pmed.1004561.s009
(DOCX)
S10 Table. Association between the number of cigarettes per day with cardiovascular outcomes among current cigarette users.
https://doi.org/10.1371/journal.pmed.1004561.s010
(DOCX)
S1 Fig. CVD and mortality outcomes hazard ratios (HRs) by years since quit smoking among former smokers compared with current and never smokers, excluding baseline CHD and considering recent quitters (within 2 years) as current smokers.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure, AFib, atrial fibrillation. Current smokers presented with a red dot-dashed line. The reference group is never-smokers presented with a dotted black line.
https://doi.org/10.1371/journal.pmed.1004561.s011
(DOCX)
S2 Fig. CVD and mortality outcomes hazard ratios (HRs) by years since quit smoking among former smokers compared with current and never smokers among participants equal to or younger than 60 years old.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure, AFib, atrial fibrillation. Current smokers presented with a red dot-dashed line. The reference group is never-smokers presented with a dotted black line.
https://doi.org/10.1371/journal.pmed.1004561.s012
(DOCX)
S3 Fig. CVD and mortality outcomes hazard ratios (HRs) by years since quit smoking among former smokers compared with current and never smokers among participants older than 60 years old.
MI, myocardial infarction; CHD, coronary heart disease; CVD, cardiovascular disease; HF, heart failure, AFib, atrial fibrillation. Current smokers presented with a red dot-dashed line. The reference group is never-smokers presented with a dotted black line.
https://doi.org/10.1371/journal.pmed.1004561.s013
(DOCX)
S1 STROBE Checklist. STROBE checklist. Adapted from the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) Statement checklist, available at https://www.strobe-statement.org/.
Licensed under Creative Commons Attribution 4.0 International License (CC BY 4.0).
https://doi.org/10.1371/journal.pmed.1004561.s014
(DOCX)
Acknowledgments
A portion of the data reported here have been supplied by the United States Renal Data System (USRDS). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy or interpretation of the U.S. government.
The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. This paper has been reviewed and approved by the MESA Publications and Presentations Committee.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NINDS or the NIA. The authors thank the other investigators, the staff, and the participants of the REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at: https://www.uab.edu/soph/regardsstudy/
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Indian Health Service (IHS).
This paper’s content is solely the authors’ responsibility and does not necessarily represent the official views of the NIH or the FDA.
The ARIC study (Atherosclerosis Risk in Communities) has been funded in whole or in part with federal funds from the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH), Department of Health and Human Services, under contract numbers HHSN268201700001I, HHSN268201700002I, HHSN268201700003I, HHSN268201700005I, and HHSN268201700004I.
BLSA was supported in part by the Intramural Research Program of the National Institute on Aging (BLSA protocol number: 03-AG-0325).
The Coronary Artery Risk Development in Young Adults Study (CARDIA) is conducted and supported by the National Heart, Lung, and Blood Institute (NHLBI) in collaboration with the University of Alabama at Birmingham (75N92023D00002 & 75N92023D00005), Northwestern University (75N92023D00004), University of Minnesota (75N92023D00006), and Kaiser Foundation Research Institute (75N92023D00003). This manuscript has been reviewed by CARDIA for scientific content.
CHS-This research was supported by contracts HHSN268201200036C, HHSN268200800007C, HHSN268201800001C, N01HC55222, N01HC85079, N01HC85080, N01HC85081, N01HC85082, N01HC85083, N01HC85086, 75N92021D00006, and grants U01HL080295 and U01HL130114 from the National Heart, Lung, and Blood Institute (NHLBI), with additional contribution from the National Institute of Neurological Disorders and Stroke (NINDS). Additional support was provided by R01AG023629 from the National Institute on Aging (NIA). A full list of principal CHS investigators and institutions can be found at CHS-NHLBI.org.
Funding for the CRIC Study was obtained under a cooperative agreement from National Institute of Diabetes and Digestive and Kidney Diseases (U01DK060990, U01DK060984, U01DK061022, U01DK061021, U01DK061028, U01DK060980, U01DK060963, U01DK060902, and U24DK060990). In addition, this work was supported in part by: the Perelman School of Medicine at the University of Pennsylvania Clinical and Translational Science Award NIH/NCATS UL1TR000003, Johns Hopkins University UL1 TR-000424, University of Maryland GCRC M01 RR-16500, Clinical and Translational Science Collaborative of Cleveland, UL1TR000439from the National Center for Advancing Translational Sciences (NCATS) component of the National Institutes of Health and NIH roadmap for Medical Research, Michigan Institute for Clinical and Health Research (MICHR) UL1TR000433, University of Illinois at Chicago CTSA UL1RR029879, Tulane COBRE for Clinical and Translational Research in Cardiometabolic Diseases P20 GM109036, Kaiser Permanente NIH/NCRR UCSF-CTSI UL1 RR-024131, Department of Internal Medicine, University of New Mexico School of Medicine Albuquerque, NM R01DK119199. A portion of the data reported here have been supplied by the United States Renal Data System (USRDS). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy or interpretation of the U.S. government [32].
The DHS was supported by grants from the Donald W. Reynolds Foundation and the National Center for Advancing Translational Sciences (UL1TR001105).
The Framingham Heart Study is supported by contracts NO1-HC-25195, HHSN268201500001I, and 75N92019D00031 from the National Heart, Lung and Blood Institute.
The GOLDN Study was funded by grant HL091357 (Arnett PI) from the National Heart, Lung and Blood Institute. The HAPI Heart Study was supported by research grants P30DK072488, U01HL072515, and U01HL084756 from the National Institutes of Health.
The Jackson Heart Study (JHS) is supported and conducted in collaboration with Jackson State.
University (HHSN268201800013I), Tougaloo College (HHSN268201800014I), the Mississippi State Department of Health (HHSN268201800015I), and the University of Mississippi Medical Center (HHSN268201800010I, HHSN268201800011I, and HHSN268201800012I) contracts from the National Heart, Lung, and Blood Institute (NHLBI) and the National Institute on Minority Health and Health Disparities (NIMHD). The authors also wish to thank the staffs and participants of the JHS.
The MESA study was supported by contracts 75N92020D00001, HHSN268201500003I, N01-HC-95159, 75N92020D00005, N01-HC-95160, 75N92020D00002, N01-HC-95161, 75N92020D00003, N01-HC-95162, 75N92020D00006, N01-HC-95163, 75N92020D00004, N01-HC-95164, 75N92020D00007, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169 from the National Heart, Lung, and Blood Institute, and by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420 from the National Center for Advancing Translational Sciences (NCATS). The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org. This paper has been reviewed and approved by the MESA Publications and Presentations Committee.
The Multiple Risk Factor Intervention Trial was contracted by the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH), Bethesda, MD. Follow‐up after the end of the trial was supported with NIH/NHLBI grants R01‐HL‐43232 and R01‐HL‐68140.
This research project is supported by cooperative agreement U01 NS041588 co-funded by the National Institute of Neurological Disorders and Stroke (NINDS) and the National Institute on Aging (NIA), National Institutes of Health, Department of Health and Human Service. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NINDS or the NIA. Representatives of the NINDS were involved in the review of the manuscript but were not directly involved in the collection, management, analysis or interpretation of the data. The authors thank the other investigators, the staff, and the participants of the REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at: https://www.uab.edu/soph/regardsstudy/
The Rancho Bernardo Study was funded by research grants AG028507 and AG07181 from the National Institute on Aging and grant DK31801 from the National Institute of Diabetes and Digestive and Kidney Diseases.
The Strong Heart Study has been funded in whole or in part with federal funds from the National Heart, Lung, and Blood Institute, National Institute of Health, Department of Health and Human Services, under contract numbers 75N92019D00027, 75N92019D00028, 75N92019D00029, & 75N92019D00030. The study was previously supported by research grants: R01HL109315, R01HL109301, R01HL109284, R01HL109282, and R01HL109319 and by cooperative agreements: U01HL41642, U01HL41652, U01HL41654, U01HL65520, and U01HL65521. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Indian Health Service (IHS).
SWAN: The SWAN is supported by the National Institutes of Health (NIH), DHHS, through the National Institute on Aging (NIA), the National Institute of Nursing Research (NINR), and the NIH Office of Research on Women’s Health (ORWH) (grants U01NR004061; U01AG012505, U01AG012535, U01AG012531, U01AG012539, U01AG012546, U01AG012553, U01AG012554, U01AG012495, and U19AG063720); The SWAN Repository (U01AG017719).
The WHI program is funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C, and HHSN268201600004C. Funding for this project was supported by NICHD R03 HD102403 (Farland).
The Study of Osteoporotic Fractures (SOF) is supported by National Institutes of Health funding. The National Institute on Aging (NIA) provides support under the following grant numbers: R01 AG005407, R01 AR35582, R01 AR35583, R01 AR35584, R01 AG005394, R01 AG027574, R01 AG027576, and R01 AG026720.
We gratefully acknowledge the dedicated cohort staff whose efforts in establishing and maintaining the individual study cohorts made this research possible.
Cohort representatives: Kunihiro Matsushita, MD, PhD (ARIC), Eleanor M. Simonsick, PhD (BLSA, Health ABC), Joao A. C. Lima, MD, MBA (CARDIA), Donald M. Lloyd-Jones, MD, ScM (Cohort Reviewer), Debbie L. Cohen, MD (CRIC), Lawrence J. Appel, MD, MPH (CRIC), Amit Khera, MD, MSc (DHS), Michael E. Hall, MD (JHS, MESA), Carlos J. Rodriguez, MD, MPH (HCHS-SOL), Suzanne Judd, PhD (REGARDS), Shelley A. Cole, PhD (SHS), Vasan S. Ramachandran, MD (FHS), Emelia J. Benjamin, MD, ScM (FHS), Paulo A. Lotufo, MD, DrPH (ELSA-Brasil), Marcio Sommer Bittencourt, MD (ELSA-Brasil), Samar R. El Khoudary, PhD, MPH (SWAN), Rebecca C. Thurston, PhD (SWAN), Carol A. Derby, PhD (SWAN), Bruce M. Psaty, MD, PhD, MPH (CHS), Charles B. Eaton, MD, MS (WHI), Michael J. LaMonte, PhD, MPH (WHI), Peggy M. Cawthon, PhD, MPH (SOF/MROS), Eric S. Orwoll, MD (SOF/MROS).
References
- 1.
National Center for Chronic Disease Prevention, Health Promotion (US) Office on Smoking and Health. Reports of the surgeon general. The health consequences of smoking—50 years of progress: a report of the surgeon general. Atlanta, GA: Centers for Disease Control and Prevention (US); 2014.
- 2.
World Health Organization. Tobacco; 2023 [cited 2024 Feb 25]. Available from: https://www.who.int/news-room/fact-sheets/detail/tobacco#:~:text=The%20tobacco%20epidemic%20is%20one%20of%20the%20biggest,of%20non-smokers%20being%20exposed%20to%20second-hand%20smoke%20%284%29
- 3.
World Health Organization. Tobacco responsible for 20% of deaths from coronary heart disease; 2020 [cited 2025 Feb 25]. Available from: https://www.who.int/news/item/22-09-2020-tobacco-responsible-for-20-of-deaths-from-coronary-heart-disease#:~:text=Every%20year%2C%201.9%20million%20people%20die%20from%20tobacco-induced,second-hand%20smoke%20increase%20the%20risk%20of%20heart%20disease
- 4. Banks E, Joshy G, Korda RJ, Stavreski B, Soga K, Egger S, et al. Tobacco smoking and risk of 36 cardiovascular disease subtypes: fatal and non-fatal outcomes in a large prospective Australian study. BMC Med. 2019;17(1):128. pmid:31266500
- 5. Mons U, Müezzinler A, Gellert C, Schöttker B, Abnet CC, Bobak M, et al. Impact of smoking and smoking cessation on cardiovascular events and mortality among older adults: meta-analysis of individual participant data from prospective cohort studies of the CHANCES consortium. BMJ. 2015;350:h1551. pmid:25896935
- 6. Huxley RR, Woodward M. Cigarette smoking as a risk factor for coronary heart disease in women compared with men: a systematic review and meta-analysis of prospective cohort studies. Lancet. 2011;378(9799):1297–305. pmid:21839503
- 7. Aune D, Schlesinger S, Norat T, Riboli E. Tobacco smoking and the risk of atrial fibrillation: a systematic review and meta-analysis of prospective studies. Eur J Prev Cardiol. 2018;25(13):1437–51. pmid:29996680
- 8. Carter BD, Abnet CC, Feskanich D, Freedman ND, Hartge P, Lewis CE, et al. Smoking and mortality—beyond established causes. N Engl J Med. 2015;372(7):631–40. pmid:25671255
- 9. Cornelius ME, Loretan CG, Wang TW, Jamal A, Homa DM. tobacco product use among adults—United States, 2020. MMWR Morb Mortal Wkly Rep. 2022;71(11):397–405. pmid:35298455
- 10.
Office for National Statistics (ONS). Adult smoking habits in the UK: 2021. ONS website, statistical bulletin; 2022.
- 11.
Centers for Disease Controle and Prevention FFaFS. Smoking and cigarettes; 2023 [cited 2024 Mar 28]. Available from: https://www.cdc.gov/tobacco/data_statistics/fact_sheets/fast_facts/index.htm
- 12. Inoue-Choi M, Liao LM, Reyes-Guzman C, Hartge P, Caporaso N, Freedman ND. Association of long-term, low-intensity smoking with all-cause and cause-specific mortality in the National Institutes of Health-AARP Diet and Health Study. JAMA Intern Med. 2017;177(1):87–95. pmid:27918784
- 13. Hackshaw A, Morris JK, Boniface S, Tang J-L, Milenković D. Low cigarette consumption and risk of coronary heart disease and stroke: meta-analysis of 141 cohort studies in 55 study reports. BMJ. 2018;360:j5855. pmid:29367388
- 14. Cho ER, Brill IK, Gram IT, Brown PE, Jha P. Smoking cessation and short- and longer-term mortality. NEJM Evid. 2024;3(3). pmid:38329816
- 15. Duncan MS, Freiberg MS, Greevy RA Jr, Kundu S, Vasan RS, Tindle HA. Association of smoking cessation with subsequent risk of cardiovascular disease. JAMA. 2019;322(7):642–50. pmid:31429895
- 16. Cao Y, Kenfield S, Song Y, Rosner B, Qiu W, Sesso HD, et al. Cigarette smoking cessation and total and cause-specific mortality: a 22-year follow-up study among US male physicians. Arch Intern Med. 2011;171(21):1956–9. pmid:22123811
- 17. Thun MJ, Carter BD, Feskanich D, Freedman ND, Prentice R, Lopez AD, et al. 50-year trends in smoking-related mortality in the United States. N Engl J Med. 2013;368(4):351–64. pmid:23343064
- 18. Thomson B, Islami F. Association of smoking cessation and cardiovascular, cancer, and respiratory mortality. JAMA Intern Med. 2024;184(1):110–2. pmid:38010645
- 19. Okorare O, Evbayekha EO, Adabale OK, Daniel E, Ubokudum D, Olusiji SA, et al. Smoking cessation and benefits to cardiovascular health: a review of literature. Cureus. 2023;15(3):e35966. pmid:37041912
- 20. Tasdighi E, Jha KK, Dardari ZA, Osuji N, Rajan T, Boakye E, et al. Investigating the association of traditional and non-traditional tobacco product use with subclinical and clinical cardiovascular disease: The Cross-Cohort Collaboration-Tobacco working group rationale, design, and methodology. Tob Induc Dis. 2023;21:89. pmid:37427074
- 21.
StataCorp. Stata multiple imputation reference manual: MI impute logit. Stata Press; 2025.
- 22. Inoue-Choi M, Freedman ND, Saito E, Tanaka S, Hirabayashi M, Sawada N, et al. Low-intensity cigarette smoking and mortality risks: a pooled analysis of prospective cohort studies in Japan. Int J Epidemiol. 2022;51(4):1276–90. pmid:34718588
- 23. Inoue-Choi M, Christensen CH, Rostron BL, Cosgrove CM, Reyes-Guzman C, Apelberg B, et al. Dose-response association of low-intensity and nondaily smoking with mortality in the United States. JAMA Netw Open. 2020;3(6):e206436. pmid:32492162
- 24. Inoue-Choi M, Hartge P, Liao LM, Caporaso N, Freedman ND. Association between long-term low-intensity cigarette smoking and incidence of smoking-related cancer in the national institutes of health-AARP cohort. Int J Cancer. 2018;142(2):271–80. pmid:28929489
- 25. Prevention CfDCa. Cardiovascular care settings and smoking cessation; 2021 [cited 2024 Mar 26]. Available from: https://www.cdc.gov/tobacco/patient-care/care-settings/cardiovascular/index.htm#:~:text=Reduces%20the%20risk%20of%20coronary%20heart%20disease%2C%20with,risk%20approaching%20that%20of%20never%20smokers%20after%20cessation
- 26. Darden M, Gilleskie DB, Strumpf K. Smoking and mortality: new evidence from a long panel. Int Econ Rev (Philadelphia). 2018;59(3):1571–619. pmid:31274880
- 27. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years’ observations on male British doctors. BMJ. 2004;328(7455):1519. pmid:15213107
- 28. Liber AC, Warner KE. Has underreporting of cigarette consumption changed over time? Estimates derived from US National Health Surveillance Systems between 1965 and 2015. Am J Epidemiol. 2018;187(1):113–9. pmid:28595348
- 29. Park MB, Kim C-B, Nam EW, Hong KS. Does South Korea have hidden female smokers: discrepancies in smoking rates between self-reports and urinary cotinine level. BMC Womens Health. 2014;14:156. pmid:25495192
- 30. Mok Y, Dardari Z, Sang Y, Hu X, Bancks MP, Mathews L, et al. Universal risk prediction for individuals with and without atherosclerotic cardiovascular disease. J Am Coll Cardiol. 2024;83(5):562–73. pmid:38296400
- 31. Pirie K, Peto R, Reeves GK, Green J, Beral V, Million Women Study Collaborators. The 21st century hazards of smoking and benefits of stopping: a prospective study of one million women in the UK. Lancet. 2013;381(9861):133–41. pmid:23107252
- 32. Johansen KL, Chertow GM, Gilbertson DT, Ishani A, Israni A, Ku E, et al. US Renal Data System 2022 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis. 2023;81(3 Suppl1):A8–11. pmid:36822739