Skip to main content
NCBI home page
Frontiers in Aging logoLink to Frontiers in Aging
. 2026 Sep 17;7:1870256. doi: 10.3389/fragi.2026.1870256

Association of smoking cessation with late-life frailty risk: the Singapore Chinese health study

Ainsley Ryan Yan Bin Lee 1,*, Huiqi Li 2, Kevin Yiqiang Chua 2, Woon-Puay Koh 2,3
PMCID: PMC13627434  PMID: 42824194

Abstract

Objective

To examine whether smoking cessation, in terms of duration and age at quitting, was associated with late-life physical frailty.

Design

Prospective population-based cohort study.

Setting

Chinese living in Singapore.

Participants

We analyzed data from 4,574 men (mean baseline age 52.7 years, range 45–65 years) enrolled in the Singapore Chinese Health Study.

Measurements

Participants self-reported smoking behaviors at four time-points over a 20-year follow-up. Physical frailty was assessed in late life (mean age 73.6 years, range 65–88 years) using a modified Cardiovascular Health Study phenotype, incorporating weakness, slowness, exhaustion and weight loss. Multivariable logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for association with frailty.

Results

Persistent smokers had higher odds of frailty compared to never smokers (OR: 1.34; 95% CI: 1.01–1.76). Those who had quit smoking within 10 years before frailty assessment also had increased odds (OR = 1.33; 95% CI: 0.86–2.01), albeit not statistically significant. The OR (95% CI) was attenuated to 1.08 (0.76–1.51) for those quitting 10–20 years and 0.97 (0.59–1.55) for those quitting >40 years. Those who quit smoking by age 60 years had similar odds relative to never smokers: OR was 1.05 (0.74–1.48) for quitting at ages 50–60 years and further reduced to 0.86 (0.57–1.25) for quitting at ages <40 years.

Conclusion

While persistent smoking was associated with increased odds of late-life physical frailty, smoking cessation for at least 10 years, or before the age of 60 years, could mitigate this risk substantially.

Keywords: frailty, healthy ageing, older adults, smoking cessation, tobacco use

1. Introduction

Frailty is a multifaceted syndrome characterized by a decline in physical strength and function, physiologic equilibrium, and the ability to maintain homeostasis. This age-related risk is influenced by a combination of genetic, biological, and environmental factors (Fried et al., 2004; Clegg et al., 2013). Since the landmark paper by Fried et al., who defined the “frailty phenotype” based on specific clinical criteria (unintentional weight loss, weakness, self-reported exhaustion, slow walking speed, and low physical activity) (Morley et al., 2013; Kojima, 2015; Fried et al., 2001), frailty has been found to be a significant predictor of disability, hospitalization, and mortality (Morley et al., 2013; Kojima, 2015; Fried et al., 2001).

Smoking is a well-established risk factor for numerous chronic diseases, including cardiovascular disease (Lakier, 1992; Burns, 2003), chronic obstructive pulmonary disease (Forey et al., 2011), and numerous cancers (Caliri et al., 2021; Warren and Cummings, 2013; Gandini et al., 2008). Recently, it has also been recognized to play a role in the initiation and progression of frailty. The physiological mechanisms linking smoking to frailty are multifaceted, encompassing systemic inflammation (Goncalves et al., 2011; Gale et al., 2013), oxidative stress (Caliri et al., 2021), and declines in skeletal muscle mass and function (Degens et al., 2015).

Several observational studies have reported the association between smoking and physical frailty. A systematic review and meta-analysis of prospective studies found that current smokers had a 63% higher risk of incident frailty than lifetime abstainers, whereas former smokers did not show a statistically significant excess risk (Amiri and Behnezhad, 2019). What remains unclear, however, is whether frailty risk among former smokers varies by how long ago they quit or when they quit (e.g., earlier vs. later in life), because prospective evidence addressing cessation duration and age at cessation is limited. One analysis from the English Longitudinal Study of Ageing (ELSA) followed frailty trajectories over 16 years in 2,600 former smokers (grouped by time since cessation) and 2,061 never smokers (Kojima et al., 2025). Frailty levels were generally lower among those who had quit for longer, and those with ≥41 years since cessation showed the lowest frailty trajectory; yet former smokers still had higher frailty than never smokers. Notably, the rate of change in frailty over time did not differ by cessation duration, leaving uncertainty about whether quitting earlier altered frailty progression rather than simply being associated with lower baseline frailty (Kojima et al., 2025). In contrast, data from the Survey on Health, Ageing, and Retirement in Europe (SHARE) suggested that the odds of worsening frailty were 16% higher in current smokers versus never smokers, but only a nonsignificant 7% higher in former smokers (Etman et al., 2015). However, SHARE did not assess whether the timing or duration of cessation modified frailty risk, so the extent to which quitting earlier or remaining abstinent longer could meaningfully reduce frailty risk remains unresolved.

We therefore investigated, in a prospective population-based cohort of Chinese men in Singapore, whether the odds of late-life physical frailty differed according to duration since smoking cessation and age at smoking cessation. Repeated assessment of smoking status over four study visits allowed us to systematically classify never smokers, persistent smokers, and former smokers prospectively, thereby effectively addressing this research gap.

2. Methods

2.1. Study population

This study was nested within the Singapore Chinese Health Study, a population-based prospective cohort study that recruited 63,257 participants (27,959 men and 35,298 women) aged 45–74 years old between April 1993 and December 1998. All participants were of Chinese ethnicity, belonging to either the Hokkien or Cantonese dialect groups, and resided in government-built housing estates, which accommodated 86% of Singapore’s population at the time of recruitment. Further details regarding this cohort study have been published previously (Hankin et al., 2001).

After recruitment, all consenting participants were re-contacted for follow-up interviews every five to 6 years, with the first follow-up interviews being conducted from 1999 to 2004, the second follow-up interviews from 2006 to 2010, and the third follow-up interviews from 2014 to 2017. Ethical approval to conduct this study was granted by the Institutional Review Board at the National University of Singapore. Informed written consent was obtained from all study participants.

2.2. Assessment of smoking habits and other covariates

At baseline, in-person interviews were conducted by trained interviewers using a structured questionnaire to collect information about sociodemographic factors (age, sex, dialect group, and education level), height, weight, lifestyle factors (physical activity, alcohol drinking, usual diet, and sleep duration), and physician-diagnosed medical history of chronic diseases (hypertension, ischemic heart disease, stroke and diabetes). Dietary intake over the past year was measured using a validated food-frequency questionnaire, and diet quality was assessed based on adherence to the Dietary Approaches to Stop Hypertension (DASH) diet, focusing on high intake of fruits, vegetables, nuts and legumes, low-fat dairy, and whole grains, while limiting red/processed meats, and sugary beverages (Talaei et al., 2019; Lai et al., 2024). Body mass index (BMI) was derived using each participant’s self-reported height and weight.

Data related to smoking habits were collected by asking participants whether they ever smoked “at least one cigarette a day for 1 year or longer”. Participants who answered “no” were categorized as never smokers, those who answered “yes, but I quit smoking” were categorized as former smokers, and those who answered “yes, and I currently smoke” were categorized as current smokers. At follow-up interviews, participants were asked similar questions to update their smoking status. In all surveys, we also obtained information on the number of cigarettes smoked per day and age of smoking initiation for smokers, and if they quit, the number of years since quitting.

Participants were classified as never smokers if they reported never smoking at baseline and at all three follow-up interviews. Persistent smokers were participants who reported current smoking at baseline and at all three follow-up interviews. Former smokers were participants who indicated a history of smoking either at baseline or during a follow-up interview, and did not report resuming smoking at any later time point. Participants who relapsed after an initial quitting attempt and resumed smoking were excluded from the analysis. For former smokers, age at quitting was determined by subtracting years since quitting from their reported age at the interview; duration since cessation was calculated as the difference between age at frailty assessment and age at quitting smoking.

2.3. Assessment of physical frailty in late life

Physical frailty was assessed during follow-up 3 interviews (2014–2017) using a modified Cardiovascular Health Study frailty phenotype (Morley et al., 2013; Fried et al., 2001). Four criteria were included: weight loss, exhaustion, slowness, and weakness; low physical activity was excluded as we did not collect relevant data on level of physical activity. We had previously published risk factors of physical frailty using this definition from these four criteria in this cohort (Ch et al., 2026; Chua et al., 2024): 1) weight loss was defined as having lost ≥10% of body weight between follow-up 2 (2006–2010) and follow-up 3 (2014–2017); 2) exhaustion was defined by “No” to the question, “Are you full of energy?”; 3) slowness was defined by being in the slowest sex-specific quintile in the Timed Up-and-Go test; 4) weakness was determined by being in the weakest sex-specific quintile of handgrip strength. Participants meeting at least two criteria were classified as physically frail. Compared with the original five-component phenotype, this definition could classify some participants who would otherwise be considered pre-frail as frail. We therefore repeated a sensitivity analysis using a stricter definition of at least 3 of the 4 available criteria.

2.4. Statistical analyses

We did not include women in this study due to the low prevalence of smoking among the women in this cohort (only 3.0% were smokers at baseline). A total of 5,186 men in this cohort had complete data on smoking habits at all surveys and completed assessment for physical frailty at follow-up 3 interviews. Of the initial sample, 175 men were excluded for exceeding the age of 65 years at baseline, 142 men were excluded for being younger than 65 years at the time of frailty assessment, and 295 men were excluded due to resuming smoking after initially quitting. This resulted in a total of 4,574 men included in the present analysis. The final sample represented 16.4% of the 27,959 men recruited at baseline.

Differences in characteristics across groups of smoking cessation were tested using one-way ANOVA for continuous variables or chi-square test for categorical variables. Multivariate logistic regression models were used to calculate odds ratios (ORs) and 95% confidence intervals (95% CIs) of the associations between elements of smoking cessation and physical frailty at follow-up 3. Model 1 was adjusted for age at assessment of physical frailty, dialect group (Hokkien, Cantonese), and levels of education (no formal, primary, secondary school and higher). Model 2 was additionally adjusted for factors that were either established risk factors of physical frailty in the literature and/or in our cohort (Lai et al., 2024; Chua et al., 2023; Dai et al., 2025; Abd Ghafar et al., 2022; Song et al., 2024). These factors collected at baseline included BMI (<18.5, 18.5–22.9, 23–27.4, ≥27.5 kg/m2), alcohol consumption (non-drinker, monthly drinker, weekly drinker, daily drinker), sleep duration (≤5, 6–8, ≥9 h/day), DASH scores (quartiles), physical activity (no physical activity, 0.5–3.9 h a week, ≥4 h/week), caffeine intake (quartiles), and medical history of hypertension, diabetes, cardiovascular diseases at baseline. Additionally, we also adjusted for data on BMI, alcohol, sleep duration, DASH scores, and the same medical history collected at follow-up 3. To examine potential health-related smoking cessation, we reclassified former smokers who had quit within 2 years before frailty assessment as persistent smokers and repeated the analyses after excluding participants with hypertension, diabetes, ischemic heart disease, or stroke at baseline. A P-value for trend was calculated among former smokers by assigning ordinal scores to categories of increasing duration since cessation (<10, 10–20, 20–30, 30–40, >40 years). Finally, we repeated the analyses using frailty defined as at least 3 of the 4 available criteria.

All analyses were conducted in R 4.5.0. All P-values were two-sided, and a P-value of <0.05 was considered statistically significant.

3. Results

Among the 4,574 men included in the analysis, 2,527 (55.3%) were never smokers, 1,400 (30.6%) were former smokers, and 647 (14.1%) were persistent smokers. Mean age was 52.7 years at baseline, 58.3 years at follow-up 1, 65.1 years at follow-up 2, and 73.6 years at follow-up 3, when frailty was assessed. Baseline characteristics (Table 1) differed across smoking groups for all variables except sleep duration. Compared to never smokers, persistent smokers had lower education levels, lower baseline DASH scores, higher caffeine intake, more daily alcohol use, and less physical activity. Former smokers were the oldest at both baseline and frailty assessment, and had the highest prevalence of baseline hypertension, diabetes, and cardiovascular disease compared to never smokers and persistent smokers.

TABLE 1.

Baseline characteristics of participants according to smoking status (n = 4,574).

Characteristic Total Never smokers Former smokers Persistent smokers P value
Number of participants 4574 2527 1400 647 -
Age at baseline, y 52.7 ± 5.1 52.4 ± 5.1 53.6 ± 5.2 52.3 ± 5.0 <0.001
Age at frailty assessment, y 73.6 ± 5.3 73.2 ± 5.3 74.4 ± 5.4 73.2 ± 5.0 <0.001
BMI, kg/m2 23.2 ± 3.2 23.3 ± 3.2 23.4 ± 3.2 22.4 ± 3.2 <0.001
Dialect group ​ ​ ​ ​ <0.001
 Hokkien 2325 (50.8) 1210 (47.9) 726 (51.9) 389 (60.1) ​
 Cantonese 2249 (49.2) 1317 (52.1) 674 (48.1) 258 (39.9) ​
Level of education ​ ​ ​ ​ <0.001
 No formal education 169 (3.7) 62 (2.5) 62 (4.4) 45 (7.0) ​
 Primary 1979 (43.3) 886 (35.1) 742 (53.0) 351 (54.3) ​
 Secondary or higher 2426 (53.0) 1579 (62.5) 596 (42.6) 251 (38.8) ​
DASH score 20.5 ± 4.7 21.3 ± 4.7 20.0 ± 4.6 18.7 ± 4.3 <0.001
Caffeine intake 158.0 ± 112.1 138.7 ± 103.5 169.2 ± 108.7 209.4 ± 130.6 <0.001
Alcohol consumption ​ ​ ​ ​ <0.001
 Non-drinker 3009 (65.8) 1759 (69.6) 899 (64.2) 351 (54.3) ​
 Monthly drinker 601 (13.1) 334 (13.2) 187 (13.4) 80 (12.4) ​
 Weekly drinker 736 (16.1) 368 (14.6) 231 (16.5) 137 (21.2) ​
 Daily drinker 228 (5.0) 66 (2.6) 83 (5.9) 79 (12.2) ​
Sleep duration ​ ​ ​ ​ 0.391
 ≤5 h 261 (5.7) 151 (6.0) 74 (5.3) 36 (5.6) ​
 6–8 h 4111 (89.9) 2277 (90.1) 1254 (89.6) 580 (89.6) ​
 ≥9 h 202 (4.4) 99 (3.9) 72 (5.1) 31 (4.8) ​
Physical activity ​ ​ ​ ​ <0.001
 No physical activity 2243 (49.0) 1142 (45.2) 731 (52.2) 370 (57.2) ​
 0.5–3.9 h a week 1437 (31.4) 872 (34.5) 405 (28.9) 160 (24.7) ​
 ≥4 h a week 894 (19.5) 513 (20.3) 264 (18.9) 117 (18.1) ​
Hypertension ​ ​ ​ ​ <0.001
 No 3695 (80.8) 2012 (79.6) 1090 (77.9) 593 (91.7) ​
 Yes 879 (19.2) 515 (20.4) 310 (22.1) 54 (8.3) ​
Diabetes ​ ​ ​ ​ <0.001
 No 4351 (95.1) 2417 (95.6) 1304 (93.1) 630 (97.4) ​
 Yes 223 (4.9) 110 (4.4) 96 (6.9) 17 (2.6) ​
Cardiovascular diseases ​ ​ ​ ​ <0.001
 No 4449 (97.3) 2470 (97.7) 1343 (95.9) 636 (98.3) ​
 Yes 125 (2.7) 57 (2.3) 57 (4.1) 11 (1.7) ​
Open in a new tab

Abbreviations: BMI, body mass index; DASH, dietary approaches to stop hypertension.

Data are mean ± SD, or n (%). Between-group differences were assessed using one-way ANOVA, for continuous variables and the chi-square test for categorical variables.

After an average follow-up of 20.9 years, a total of 629 (13.8%) men met the criteria of physical frailty. Participants were first analyzed by the duration of time elapsed since smoking cessation (Table 2). In fully adjusted Model 2, persistent smokers had an elevated OR (95% CI) of 1.34 (1.01–1.76) for frailty compared to never smokers. Among former smokers, quitting for less than 10 years before frailty assessment was linked to a higher OR (95% CI) of 1.33 (0.86–2.01), though this was not statistically significant. Among participants who had quit smoking for more than 10 years, the ORs (ranging from 0.85 to 1.08) were attenuated and close to null across all time-since-quitting categories, with risk estimates that were no longer statistically distinguishable from those of never smokers. In addition, there was no statistically significant dose-dependent or stepwise trend of lower risk with longer duration since cessation (P for trend = 0.112) (Supplementary Table S1).

TABLE 2.

Association of duration since smoking cessation with late-life physical frailty (n = 4,574).

Smoking status Cases/total Mean age at baseline (years) Mean age at frailty assessment (years) Mean age at quitting smoking (years) Mean duration quit (years) Model 1 Model 2
Or (95% CI) Or (95% CI)
Never smoker 300/2527 52.4 73.2 Not applicable Not applicable 1.00 1.00
Quitting >40 years 26/146 54.4 76.2 34.3 42.0 1.02 (0.63–1.60) 0.97 (0.59–1.55)
Quitting 30–40 years 51/385 54.1 74.2 38.6 35.6 0.93 (0.66–1.29) 0.85 (0.60–1.19)
Quitting 20–30 years 49/328 53.5 74.4 49.4 25.1 1.01 (0.71–1.42) 0.91 (0.63–1.29)
Quitting 10–20 years 55/340 53.4 74.5 59.8 14.7 1.12 (0.80–1.55) 1.08 (0.76–1.51)
Quitting <10 years 36/201 52.4 73.4 67.5 5.8 1.47 (0.97–2.16) 1.33 (0.86–2.01)
Persistent smoker 112/647 52.3 73.2 Not applicable Not applicable 1.43 (1.11–1.84) 1.34 (1.01–1.76)
Open in a new tab

Model 1 was adjusted for age at frailty assessment, dialect group, and level of education.

Model 2 was further adjusted for body mass index (baseline, follow-up 3, categorical), alcohol drinking (baseline, follow-up 3, categorical), physical activity (baseline), sleep duration (baseline, follow-up 3, categorical), DASH (baseline, follow-up 3 in ordinal quartiles), caffeine (baseline, in ordinal quartiles), and medical history of hypertension, diabetes, cardiovascular diseases (baseline, follow-up 3).

Participants were then analyzed by their age at smoking cessation (Table 3). In these analyses, those who quit smoking at age 60 years or older had a non-significantly increased OR (95% CI) of 1.20 (0.87–1.63) for late-life physical frailty compared to never smokers. Conversely, those who quit smoking before age 60 years had risk estimates near or below null (ORs ranging from 0.84 to 1.05), with all corresponding 95% CIs encompassing 1.00.

TABLE 3.

Association of age at smoking cessation with late-life physical frailty (n = 4,574).

Age at quitting Cases/total Mean age at baseline (years) Mean age at frailty assessment (years) Mean age at quitting smoking (years) Mean duration quit (years) Model 1 Model 2
Or (95% CI) Or (95% CI)
Never smoker 300/2527 52.4 73.2 Not applicable Not applicable 1.00 1.00
Quitting at age <40 years 35/366 51.3 71.9 33.9 37.9 0.94 (0.63–1.35) 0.86 (0.57–1.25)
Quitting at age 40–50 years 50/340 54.4 75.1 44.9 30.2 0.93 (0.66–1.29) 0.84 (0.59–1.18)
Quitting at age 50–60 years 53/336 53.3 74.2 54.9 19.3 1.12 (0.79–1.55) 1.05 (0.74–1.48)
Quitting at age >60 years 79/358 55.3 76.6 66.6 10.0 1.27 (0.95–1.70) 1.20 (0.87–1.63)
Persistent smoker 112/647 52.3 73.2 Not applicable Not applicable 1.43 (1.11–1.84) 1.34 (1.01–1.76)
Open in a new tab

Model 1 was adjusted for age at frailty assessment, dialect group, and level of education.

Model 2 was further adjusted for body mass index (baseline, follow-up 3, categorical), alcohol drinking (baseline, follow-up 3, categorical), physical activity (baseline), sleep duration (baseline, follow-up 3, categorical), DASH (baseline, follow-up 3 in ordinal quartiles), caffeine (baseline, in ordinal quartiles), and medical history of hypertension, diabetes, cardiovascular diseases (baseline, follow-up 3).

Sensitivity analyses produced similar overall findings. After reclassifying participants who had quit within 2 years and combining with persistent smokers, the adjusted OR (95% CI) was 1.35 (1.03–1.77) for the combined group and 1.28 (0.81–1.97) for the remaining participants who had quit for <10 years (Supplementary Table S2).

After exclusion of participants with hypertension, diabetes, ischemic heart disease, or stroke at baseline, the corresponding ORs were 1.31 (0.96–1.78) for persistent smokers and 1.49 (0.90–2.38) for those who had quit for <10 years (Supplementary Table S3). With frailty defined as at least 3 of 4 components, the OR for persistent smokers was 1.60 (0.97–2.59), while estimates among former smokers remained imprecise (Supplementary Table S4).

4. Discussion

In this prospective population-based cohort study with about 20 years of follow-up, persistent smokers had higher risk of late-life physical frailty than never smokers. While former smokers who had quit for less than 10 years still had a non-significant increased risk estimate close to that of persistent smokers, quitting for at least 10 years before frailty assessment was associated with little to no excess frailty risk. Stopping smoking before age of 60 years brought frailty risk close to that of never smokers, while those who quit after age 60 years still had a non-significant 20 percent increase in risk. Overall, the findings suggest that earlier and sustained smoking cessation may mitigate the elevated risk of late-life physical frailty.

Although previous case-control and cross-sectional studies have documented a link between smoking cessation and physical frailty (Song et al., 2024; Thompson et al., 2018), these designs are susceptible to inherent recall and temporal biases. To date, only a limited number of prospective studies have investigated this relationship between smoking cessation and the risk of late-life frailty.

Findings from 909 community-dwelling older adults in the North West Adelaide Health Study showed that, compared with never smokers, current smokers had substantially higher odds of frailty, whereas the association was attenuated and not evident among former smokers (Thompson et al., 2018). Similarly, the Survey on Health, Ageing, and Retirement in Europe reported worsening frailty among current smokers relative to never smokers, but not among former smokers (Etman et al., 2015). However, neither study further examined whether frailty risk among former smokers differed by age at cessation or by time since quitting before frailty assessment.

The English Longitudinal Study of Ageing investigated the associations of smoking cessation with the subsequent trajectory of frailty status among 2,600 former smokers, and reported that those with a longer duration of quitting smoking had a significantly lower degree of frailty at baseline and over time. Interestingly, although those who quit 41 years earlier or more had the lowest frailty trajectory, there was still a gap between them and never smokers (Kojima et al., 2025). However, smoking status was measured only once at baseline, and former smokers were classified according to their self-report at that time. As such, participants who quit before baseline but later relapsed could have been misclassified as former smokers during the study, which may have inflated frailty risk estimates among former smokers compared to never smokers.

There are several postulated mechanisms underlying the relationship between cigarette smoking and physical frailty. Numerous studies have demonstrated a higher level of inflammatory markers in chronic smokers, including C-reactive protein and the pro-inflammatory cytokines interleukin-1β and interleukin-6 (Ugur et al., 2018; Toker et al., 2012). Recently, the term inflamm-ageing (Pan and Ma, 2024; Fulop et al., 2017; Franceschi et al., 2017) has been coined to describe the link between chronic inflammation and ageing-related biomechanistic processes, including cellular senescence, systemic organ dysfunction, reduced physiological reserves (Pan and Ma, 2024; Fulop et al., 2017; Ferrucci and Fabbri, 2018; Leonardi et al., 2018; Salimi et al., 2018; Gorelick, 2010; Lee et al., 2026), and comorbidities such as cardiovascular (Ambrose and Barua, 2004) and respiratory diseases (Aghapour et al., 2018), all of which have been implicated in the development of frailty. These inflammatory markers have been shown to decline after smoking exposure is eliminated (Peres et al., 2017; Elisia et al., 2020), which may explain our finding that individuals who had quit smoking for a sufficiently long period had frailty risk comparable to that of never smokers. Finally, since early quitting automatically reduces total lifetime smoking, the health benefits observed after early cessation could simply be explained by reduced cumulative exposure. As such, one of the reasons quitting early is beneficial could be that it reduces the total amount of smoking a person accumulates over life.

The strengths of our study are its long follow-up duration, comprehensive adjustment of potential confounding covariates, and meticulous and repeated ascertainment of smoking status at baseline and over three follow-up visits spanning average 20 years, which allowed us to accurately classify participants as persistent, former and never smokers, and exclude those who relapsed after quitting.

However, this study had several limitations. First, the modified frailty phenotype excluded low physical activity and classified participants meeting at least 2 of 4 available criteria as frail. This could have classified some participants who would have been considered pre-frail using the original five-component phenotype as frail. In the sensitivity analysis using a stricter definition of at least 3 of 4 criteria, the estimate for persistent smoking remained elevated, although the risk estimate did not reach statistical significance due to reduced number in frailty outcome. Second, we acknowledge that our study only included a portion of the original populations who survived and consented to follow-up 3 interviews, which would have introduced a selection bias of only including the healthier participants from the initial study sample. As shown in Supplementary Table S5, compared with the 23,380 men who were not included in this study, the 4,574 men in this study were younger, had higher educational attainment, greater physical activity, and lower baseline prevalence of hypertension, diabetes, and heart attack at enrolment (1993–1998) than those who were not included. Such inclusion of participants with a lower baseline frailty risk could have biased the exposure-outcome association toward the null and led to an underestimation of the risk estimates. Third, since we only conducted frailty assessment at the third follow-up, reverse causality could not be precluded since former smokers could have developed frailty before smoking cessation. Alternatively, long-term abstainers may represent a selectively healthier or more health-conscious group. Finally, a previous Chinese cohort reported a stronger association between smoking and frailty in men than in women (Wang et al., 2013), and since we only studied men, our findings may not be generalized to women.

5. Conclusion

In conclusion, our study adds prospective evidence that smoking is an important and potentially modifiable risk factor for physical frailty in later life, and that quitting smoking for at least 10 years (or before age 60 years) may substantially mitigate this excess frailty risk to levels close to never smokers. The results are biologically plausible in light of smoking-related systemic inflammation and its downstream effects on age-related physiological vulnerability, and they underscore the potential benefits of quitting earlier and sustaining cessation. Future studies with earlier and repeated frailty assessments, careful handling of time-varying smoking behaviors, and more diverse populations are needed to further clarify temporality, reduce residual bias, and strengthen the generalizability of these observations.

Acknowledgments

We would like to thank Ms. Siew-Hong Low of the National University of Singapore for supervising the field work of the Singapore Chinese Health Study; this contributor did not receive additional compensation.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. This study was supported by grants from the National Medical Research Council, Singapore (NMRC/CSA/0055/2013 and NMRC/CSA-SI MOH-000434) and the Saw Swee Hock School of Public Health, National University of Singapore. The Singapore Chinese Health Study was supported by the U.S. National Institutes of Health (grant numbers R01 CA144034 and UM1 CA182867) and the Singapore Ministry of Health’s National Medical Research Council under its Strategic Cohorts Funding (Project No. P2022-02).

Footnotes

Edited by: Mário Cunha Espada, Instituto Politecnico de Setubal (IPS), Portugal

Reviewed by: Jiafeng Zhang, Shanghai Changzheng Hospital, China

Shuang Deng, Shanghai Normal University, China

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by National University of Singapore Institutional Review Board. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

AL: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review and editing. HL: Data curation, Formal Analysis, Investigation, Resources, Software, Validation, Visualization, Writing – review and editing. KC: Data curation, Investigation, Methodology, Software, Validation, Visualization, Writing – review and editing. W-PK: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Writing – review and editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fragi.2026.1870256/full#supplementary-material

Table1.docx (37.4KB, docx)

References

  1. Abd Ghafar M. Z. A., O'Donovan M., Sezgin D., Moloney E., Rodríguez-Laso Á., Liew A., et al. (2022). Frailty and diabetes in older adults: overview of current controversies and challenges in clinical practice. Front. Clin. Diabetes Healthc. 3, 895313. 10.3389/fcdhc.2022.895313 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aghapour M., Raee P., Moghaddam S. J., Hiemstra P. S., Heijink I. H. (2018). Airway epithelial barrier dysfunction in chronic obstructive pulmonary disease: role of cigarette smoke exposure. Am. J. Respir. Cell Mol. Biol. 58 (2), 157–169. 10.1165/rcmb.2017-0200TR [DOI] [PubMed] [Google Scholar]
  3. Ambrose J. A., Barua R. S. (2004). The pathophysiology of cigarette smoking and cardiovascular disease: an update. J. Am. Coll. Cardiol. 43 (10), 1731–1737. 10.1016/j.jacc.2003.12.047 [DOI] [PubMed] [Google Scholar]
  4. Amiri S., Behnezhad S. (2019). Systematic review and meta-analysis of the association between smoking and the incidence of frailty. Neuropsychiatr 33 (4), 198–206. 10.1007/s40211-019-0315-4 [DOI] [PubMed] [Google Scholar]
  5. Burns D. M. (2003). Epidemiology of smoking-induced cardiovascular disease. Prog. Cardiovasc. Dis. 46 (1), 11–29. 10.1016/s0033-0620(03)00079-3 [DOI] [PubMed] [Google Scholar]
  6. Caliri A. W., Tommasi S., Besaratinia A. (2021). Relationships among smoking, oxidative stress, inflammation, macromolecular damage, and cancer. Mutat. Res. Rev. Mutat. Res. 787, 108365. 10.1016/j.mrrev.2021.108365 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chua K. Y., Chua R. Y., Li H., Lim W.-S., Koh W.-P. (2026). Association between sleep duration from midlife and the risk of physical frailty in late life. Sci. Rep. 16 (1), 8426. 10.1038/s41598-026-39228-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chua K. Y., Li H., Lim W. S., Koh W. P. (2023). Consumption of coffee, tea, and caffeine at midlife, and the risk of physical frailty in late life. J. Am. Med. Dir. Assoc. 24 (11), 1655. 10.1016/j.jamda.2023.06.015 [DOI] [PubMed] [Google Scholar]
  9. Chua K. Y., Li H., Sheng L.-T., Lim W.-S., Koh W.-P. (2024). Intake of vegetables and fruits at midlife and the risk of physical frailty in later life. J. Nutrition, Health Aging 28 (11), 100374. 10.1016/j.jnha.2024.100374 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Clegg A., Young J., Iliffe S., Rikkert M. O., Rockwood K. (2013). Frailty in elderly people. Lancet 381 (9868), 752–762. 10.1016/S0140-6736(12)62167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dai L., Tang Y., Guo Y., Lai X., Wang X., Li B. (2025). The association between exercise, activities, and frailty in older Chinese adults: a cross-sectional study based on the chinese longitudinal healthy longevity survey (CLHLS) data. BMC Geriatr. 25 (1), 131. 10.1186/s12877-025-05802-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Degens H., Gayan-Ramirez G., van Hees H. W. (2015). Smoking-induced skeletal muscle dysfunction: from evidence to mechanisms. Am. J. Respir. Crit. Care Med. 191 (6), 620–625. 10.1164/rccm.201410-1830PP [DOI] [PubMed] [Google Scholar]
  13. Elisia I., Lam V., Cho B., Hay M., Li M. Y., Yeung M., et al. (2020). The effect of smoking on chronic inflammation, immune function and blood cell composition. Sci. Rep. 10 (1), 19480. 10.1038/s41598-020-76556-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Etman A., Kamphuis C. B., van der Cammen T. J., Burdorf A., van Lenthe F. J. (2015). Do lifestyle, health and social participation mediate educational inequalities in frailty worsening? Eur. J. Public Health 25 (2), 345–350. 10.1093/eurpub/cku093 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ferrucci L., Fabbri E. (2018). Inflammageing: chronic inflammation in ageing, cardiovascular disease, and frailty. Nat. Rev. Cardiol. 15 (9), 505–522. 10.1038/s41569-018-0064-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Forey B. A., Thornton A. J., Lee P. N. (2011). Systematic review with meta-analysis of the epidemiological evidence relating smoking to COPD, chronic bronchitis and emphysema. BMC Pulm. Med. 11 (1), 36. 10.1186/1471-2466-11-36 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Franceschi C., Garagnani P., Vitale G., Capri M., Salvioli S. (2017). Inflammaging and 'Garb-aging. Trends Endocrinol. Metab. 28 (3), 199–212. 10.1016/j.tem.2016.09.005 [DOI] [PubMed] [Google Scholar]
  18. Fried L. P., Tangen C. M., Walston J., Newman A. B., Hirsch C., Gottdiener J., et al. (2001). Frailty in older adults: evidence for a phenotype. J. Gerontol. A Biol. Sci. Med. Sci. 56 (3), M146–M156. 10.1093/gerona/56.3.m146 [DOI] [PubMed] [Google Scholar]
  19. Fried L. P., Ferrucci L., Darer J., Williamson J. D., Anderson G. (2004). Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J. Gerontol. A Biol. Sci. Med. Sci. 59 (3), 255–263. 10.1093/gerona/59.3.m255 [DOI] [PubMed] [Google Scholar]
  20. Fulop T., Larbi A., Dupuis G., Le Page A., Frost E. H., Cohen A. A., et al. (2017). Immunosenescence and inflamm-aging as two sides of the same coin: friends or foes? Front. Immunol. 8, 1960. 10.3389/fimmu.2017.01960 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gale C. R., Baylis D., Cooper C., Sayer A. A. (2013). Inflammatory markers and incident frailty in men and women: the English longitudinal study of ageing. Age (Dordr) 35 (6), 2493–2501. 10.1007/s11357-013-9528-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gandini S., Botteri E., Iodice S., Boniol M., Lowenfels A. B., Maisonneuve P., et al. (2008). Tobacco smoking and cancer: a meta-analysis. Int. J. Cancer 122 (1), 155–164. 10.1002/ijc.23033 [DOI] [PubMed] [Google Scholar]
  23. Goncalves R. B., Coletta R. D., Silverio K. G., Benevides L., Casati M. Z., da Silva J. S., et al. (2011). Impact of smoking on inflammation: overview of molecular mechanisms. Inflamm. Res. 60 (5), 409–424. 10.1007/s00011-011-0308-7 [DOI] [PubMed] [Google Scholar]
  24. Gorelick P. B. (2010). Role of inflammation in cognitive impairment: results of observational epidemiological studies and clinical trials. Ann. N. Y. Acad. Sci. 1207, 155–162. 10.1111/j.1749-6632.2010.05726.x [DOI] [PubMed] [Google Scholar]
  25. Hankin J. H., Stram D. O., Arakawa K., Park S., Low S. H., Lee H. P., et al. (2001). Singapore Chinese health study: development, validation, and calibration of the quantitative food frequency questionnaire. Nutr. Cancer 39 (2), 187–195. 10.1207/S15327914nc392_5 [DOI] [PubMed] [Google Scholar]
  26. Kojima G. (2015). Frailty as a predictor of future falls among community-dwelling older people: a systematic review and meta-analysis. J. Am. Med. Dir. Assoc. 16 (12), 1027–1033. 10.1016/j.jamda.2015.06.018 [DOI] [PubMed] [Google Scholar]
  27. Kojima G., Taniguchi Y., Aoyama R., Urano T. (2025). Association between time since smoking cessation and frailty trajectory among community-dwelling older people: english longitudinal study of ageing. J. Am. Med. Dir. Assoc. 26 (1), 105328. 10.1016/j.jamda.2024.105328 [DOI] [PubMed] [Google Scholar]
  28. Lai J. S., Chua K. Y., Li H., Koh W. P. (2024). Diet quality from mid to late life and its association with physical frailty in late life in a cohort of Chinese adults. Nutr. J. 23 (1), 57. 10.1186/s12937-024-00964-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lakier J. B. (1992). Smoking and cardiovascular disease. Am. J. Med. 93 (1, Suppl. 1), S8–S12. 10.1016/0002-9343(92)90620-q [DOI] [PubMed] [Google Scholar]
  30. Leonardi G. C., Accardi G., Monastero R., Nicoletti F., Libra M. (2018). Ageing: from inflammation to cancer. Immun. Ageing 15, 1. 10.1186/s12979-017-0112-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lee A. R. Y. B., Vidhya S. N., Hong A., Tan W. A., Yau C. E., Low C. E., et al. (2026). Growth differentiation factor-15 as a clinical biomarker of frailty, sarcopenia and functional decline: a systematic literature review. Ageing Res. Rev. 118, 103085. 10.1016/j.arr.2026.103085 [DOI] [PubMed] [Google Scholar]
  32. Morley J. E., Vellas B., van Kan G. A., Anker S. D., Bauer J. M., Bernabei R., et al. (2013). Frailty consensus: a call to action. J. Am. Med. Dir. Assoc. 14 (6), 392–397. 10.1016/j.jamda.2013.03.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pan Y., Ma L. (2024). Inflammatory markers and physical frailty: towards clinical application. Immun. and Ageing 21 (1), 4. 10.1186/s12979-023-00410-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Peres F. S., Barreto S. M., Camelo L. V., Ribeiro A. L. P., Vidigal P. G., Duncan B. B., et al. (2017). Time from smoking cessation and inflammatory markers: new evidence from a cross-sectional analysis of ELSA-brasil. Nicotine Tob. Res. 19 (7), 852–858. 10.1093/ntr/ntx032 [DOI] [PubMed] [Google Scholar]
  35. Salimi S., Shardell M. D., Seliger S. L., Bandinelli S., Guralnik J. M., Ferrucci L. (2018). Inflammation and trajectory of renal function in community-dwelling older adults. J. Am. Geriatr. Soc. 66 (4), 804–811. 10.1111/jgs.15268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Song Y., Liu H., Gu K., Liu Y. (2024). U-shaped association between sleep duration and frailty in Chinese older adults: a cross-sectional study. Front. Public Health 12, 1464734. 10.3389/fpubh.2024.1464734 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Talaei M., Koh W. P., Yuan J. M., van Dam R. M. (2019). DASH dietary pattern, mediation by mineral intakes, and the risk of coronary artery disease and stroke mortality. J. Am. Heart Assoc. 8 (5), e011054. 10.1161/JAHA.118.011054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Thompson M. Q., Theou O., Yu S., Adams R. J., Tucker G. R., Visvanathan R. (2018). Frailty prevalence and factors associated with the frailty phenotype and frailty index: findings from the north west Adelaide health study. Australas. J. Ageing 37 (2), 120–126. 10.1111/ajag.12487 [DOI] [PubMed] [Google Scholar]
  39. Toker H., Akpinar A., Aydin H., Poyraz O. (2012). Influence of smoking on interleukin-1beta level, oxidant status and antioxidant status in gingival crevicular fluid from chronic periodontitis patients before and after periodontal treatment. J. Periodontal Res. 47 (5), 572–577. 10.1111/j.1600-0765.2012.01468.x [DOI] [PubMed] [Google Scholar]
  40. Ugur M. G., Kutlu R., Kilinc I. (2018). The effects of smoking on vascular endothelial growth factor and inflammation markers: a case-control study. Clin. Respir. J. 12 (5), 1912–1918. 10.1111/crj.12755 [DOI] [PubMed] [Google Scholar]
  41. Wang C., Song X., Mitnitski A., Yu P., Fang X., Tang Z., et al. (2013). Gender differences in the relationship between smoking and frailty: results from the beijing longitudinal study of aging. J. Gerontol. A Biol. Sci. Med. Sci. 68 (3), 338–346. 10.1093/gerona/gls166 [DOI] [PubMed] [Google Scholar]
  42. Warren G. W., Cummings K. M. (2013). Tobacco and lung cancer: risks, trends, and outcomes in patients with cancer. Am. Soc. Clin. Oncol. Educ. Book, 33, 359–364. 10.14694/EdBook_AM.2013.33.359 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table1.docx (37.4KB, docx)

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Articles from Frontiers in Aging are provided here courtesy of Frontiers Media SA

RESOURCES