Association of Smoking and Alcohol Consumption with the Presence of Sarcopenia at Follow-up in Community-Dwelling People Aged 40–74 Years

Article information

Ann Geriatr Med Res. 2026;30(3):379-389

Publication date (electronic) : 2026 July 21

doi : https://doi.org/10.4235/agmr.26.0074

Takumi Kaneko^1,3, Keiko Kabasawa^,2, Kaori Kitamura¹, Yumi Watanabe¹, Tomoyo Komata², Yumi Ito², Akemi Takahashi³, Toshiko Saito¹, Ryosaku Kobayashi³, Rieko Oshiki³, Ribeka Takachi⁴, Shoichiro Tsugane⁵, Kei Watanabe⁶, Junta Tanaka^2,7, Suguru Yamamoto⁸, Kazutoshi Nakamura^,1

¹Division of Preventive Medicine, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan

²Department of Health Promotion Medicine, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan

³Department of Rehabilitation, Niigata University of Rehabilitation, Murakami, Niigata, Japan

⁴Department of Food Science and Nutrition, Nara Women’s University Graduate School of Humanities and Sciences, Kitauoya-Higashimachi, Nara, Japan

⁵International University of Health and Welfare Graduate School of Public Health, Tokyo, Japan

⁶Department of Orthopaedic Surgery, Niigata University Medical and Dental Hospital, Niigata, Japan

⁷Health and Wellness Center, University of Niigata Prefecture, Niigata, Japan

⁸Division of Clinical Nephrology and Rheumatology, Kidney Research Center, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan

Corresponding Authors: Kazutoshi Nakamura, MD, MPH Division of Preventive Medicine, Niigata University Graduate School of Medical and Dental Sciences, 1-757 Asahimachi-dori, Niigata 951-8510, Japan E-mail: kazun@med.niigata-u.ac.jp

Keiko Kabasawa, MD, PhD Department of Health Promotion Medicine, Niigata University Graduate School of Medical and Dental Sciences. 1-757 Asahimachi-dori, Niigata 951-8510, Japan E-mail: keikoyk@med.niigata-u.ac.jp

Received 2026 April 7; Revised 2026 June 1; Accepted 2026 July 17.

Abstract

INTRODUCTION

Sarcopenia is an age-related musculoskeletal disorder characterized by a progressive decline in muscle mass and strength.1) It can adversely affect health through muscle weakness, increased risk of falls, reduced physical function, disability, and frailty, primarily due to inflammation and hormonal changes.2,3) Although its prevalence varies depending on the diagnostic criteria used and population studied, it reportedly ranges from 1% to 29% among community-dwelling older adults.4) The economic burden associated with sarcopenia is also substantial. In the United States, the annual healthcare costs attributable to this condition are estimated at $18.5 billion, and a 10% reduction in its prevalence could result in an annual healthcare cost saving of approximately $1.1 billion.5) Given these considerations, sarcopenia prevention should be recognized as a public health priority in aging societies.

Smoking and alcohol consumption are major lifestyle factors that influence health and reportedly have a negative impact on the development of sarcopenia. Smoking-induced oxidative stress and inflammation may contribute to muscle loss through pathways involving myostatin, impaired muscle protein synthesis, and other catabolic mechanisms.6) A meta-analysis reported higher odds of smoking for sarcopenia (odds ratio [OR]=1.12) compared to non-smoking.7) More recently, a cohort study reported that smokers were 2.4 times more likely to develop sarcopenia compared to non-smokers.8) Furthermore, a recent cross-sectional study reported that smokers had a 1.8-fold higher risk of sarcopenia compared to non-smokers.9) However, a dose-dependent relationship between the number of cigarettes smoked and sarcopenia has not been established.7-9) In addition, most previous studies were cross-sectional, highlighting the need for large-scale longitudinal studies.

The association between alcohol consumption and sarcopenia is complex. A meta-analysis involving 62 observational studies (60 were cross-sectional studies) revealed no linear association between alcohol consumption and sarcopenia; rather, a nonlinear (J-shaped) relationship was observed.10) A UK Biobank study suggested that high alcohol consumption has a harmful effect on the development of sarcopenia based on a cross-sectional analysis, but found no association between alcohol consumption and developing sarcopenia in a follow-up analysis.11) Similar to studies on smoking, few longitudinal studies have addressed this topic and with inconsistent findings. Furthermore, smoking and heavy alcohol consumption cause oxidative stress in humans, and their interactive effects on numerous chronic diseases have been suggested.12) In this context, their interaction may also contribute to sarcopenia.

This study aimed to examine associations of smoking levels and alcohol consumption with sarcopenia among middle-aged and older adults, and to explore potential interactions between smoking and alcohol consumption on the risk of sarcopenia. While many previous studies on sarcopenia have focused on older individuals, the present study targeted people in earlier life stages, i.e., between the ages of 40 and 74 years, potentially providing insights into modifiable factors for the early prevention of sarcopenia. This approach may also provide a life-course perspective on how pre-old-age lifestyle factors influence later muscle health.

MATERIALS AND METHODS

Design and Participants

This study examined the association between smoking levels and alcohol consumption at baseline (2011–2014) and the presence of sarcopenia at follow-up physical examinations conducted in 2021 and 2022. Participants were selected from individuals who participated in the baseline surveys of the Murakami Cohort Study (n=14,364)13) and the Uonuma Cohort Study (n=30,707).14) All eligible participants for the physical examinations were invited by mail—12,246 from the Murakami cohort and 21,534 from the Uonuma cohort. Of these, 2,199 individuals from the Murakami cohort and 4,772 from the Uonuma cohort, totaling 6,971 participants, took part in the physical examinations. The participant selection process is illustrated in Fig. 1. The following participants were excluded from the statistical analysis: 124 individuals with missing handgrip strength data; 87 individuals with asymmetrical appendicular skeletal muscle mass values considered outliers, possibly due to hemiplegia or residual metal implants; and 10 individuals with outlier values for appendicular skeletal muscle mass. We further excluded participants with exclusively high or low body mass index (BMI) or energy intake (n=88), as well as those with multiple missing values (n=8), leaving 6,654 participants (3,005 men and 3,649 women) for the final analysis population.

Fig. 1.

Flowchart of participant selection. SD, standard deviation; BMI, body mass index.

Since this study was based on a subset of the original cohort, we compared the baseline characteristics of the analytic sample with those of the entire cohort. Baseline percent distributions of age, sex, smoking, and alcohol consumption in the present sample (n=6,654) and the original total sample (45,071) are summarized in Supplementary Table S1.

Written informed consent was obtained from all participants, and the study protocol was approved by the Niigata University Ethics Committee (Approval Nos. 2019-0376, 2019-0409, and 2022-0240).

RESULTS

The mean age of participants was 62.4 years (standard deviation [SD]=7.1). Participant characteristics at baseline according to smoking levels and alcohol consumption are shown in Table 1, those according to levels of smoking and alcohol consumption by sex are shown in Supplementary Tables S3 and S4, and detailed information on marital status, education level, smoking, and alcohol consumption according to levels of smoking and alcohol consumption are shown in Supplementary Tables S5 and S6. At a mean of 9.1 years (SD=1.6) after baseline, the presence of sarcopenia was noted in 1/423 (0.2%) among participants in their 40s, 21/1,484 (1.4%) among those in their 50s, 133/3,635 (3.7%) among those in their 60s, 117/1,112 (10.5%) among those in their 70s, and 272/6,654 (4.1%) overall. Participant characteristics for the 2021–2022 physical examinations by sex are shown in Supplementary Table S7.

Table 1.

Participant characteristics at the 2011–2014 baseline survey according to levels of smoking and alcohol consumption

Higher smoking levels were associated with higher odds of the presence of sarcopenia at follow-up (multivariable-adjusted p for trend=0.0367), with the ≥20-cigarettes/day group having a higher OR than non-smokers (multivariable-adjusted OR=1.95, 95% confidence interval [CI] 1.17–3.24) (Table 2). When the ≥20-cigarettes/day group was subdivided into the 20-cigarettes/day group (23/365 cases) and the >20-cigarettes/day group (8/126 cases, median: 30 cigarettes/day), the respective multivariable-adjusted ORs were 1.86 (95% CI 1.07–3.23) and 2.29 (95% CI 0.99–5.28), with an overall multivariable-adjusted p for trend of 0.0213 (Fig. 2). A similar trend was observed (multivariable-adjusted p for trend=0.0397) in the male subgroup, with the ≥20-cigarettes/day group having significantly higher odds of the presence of sarcopenia than non-smokers (multivariable-adjusted OR=1.98, 95% CI 1.09–3.58) (Table 2). Although a statistically significant association was not observed in the female subgroup (multivariable-adjusted p for trend=0.5551), the ≥20-cigarettes/day group had significantly higher odds of the presence of sarcopenia than non-smokers (multivariable-adjusted OR=4.30, 95% CI 1.38–13.38). Sensitivity analyses for the association between smoking levels and odds of the presence of sarcopenia without imputation, using the original diagnostic criteria of AWGS 2019 and inverse-probability weighting, are shown in Table 3. The pattern of these associations was similar to that in Table 2.

Table 2.

Presence of sarcopenia at follow-up and odds ratios (ORs) according to smoking levels

Fig. 2.

Multivariable-adjusted odds ratios (ORs) for sarcopenia according to smoking levels, adjusted for sex, age, marital status, education, occupation, BMI, total physical activity, energy intake, alcohol consumption, extreme sleep duration, and histories of heart disease, stroke, and diabetes. The overall association was statistically significant (p for trend=0.0213), although the >20 cigarettes/day group did not reach statistical significance. Participants who smoked 20 or more cigarettes per day were divided into two subgroups according to the median. The numbers of participants were 3,786 for non-smokers, 1,952 for past smokers, 416 for the 1–19-cigarettes/day group, 365 for the 20-cigarettes/day group, and 126 for the >20-cigarettes/day group.

Table 3.

Sensitivity analyses of odds ratios (ORs) for the presence of sarcopenia at follow-up according to smoking levels

Significant associations were observed between alcohol consumption and odds of the presence of sarcopenia in the overall study population and women (multivariable-adjusted p for trend=0.0492 and 0.0013, respectively); however, no association was observed in men (multivariable-adjusted p for trend=0.9872) (Table 4). Sensitivity analyses for the association between drinking levels and the presence of sarcopenia without imputation, using the original diagnostic criteria of AWGS 2019 or inverse-probability weighting, are shown in Supplementary Table S8. The pattern of these associations was similar to that in Table 4.

Table 4.

Presence of sarcopenia at follow-up and odds ratios (ORs) according to alcohol consumption categories

The presence of sarcopenia and ORs according to smoking levels by drinking status are shown in Table 5. Smoking ≥20 cigarettes/day was associated with higher odds compared with non-smoking among drinkers (multivariable-adjusted OR=2.14, 95% CI 1.13–4.04), with a multivariable-adjusted p for trend of 0.0289, whereas no such association was observed among non-drinkers (multivariable-adjusted p for trend=0.7435). The p for interaction between levels of smoking and alcohol consumption on the multivariable-adjusted odds of the presence of sarcopenia was 0.0483.

Table 5.

Presence of sarcopenia at follow-up and odds ratios (ORs) according to smoking levels by drinking status

DISCUSSION

The main findings of the present study were as follows: (1) smoking levels were associated with increased odds of the presence of sarcopenia at follow-up; in particular, individuals who smoked ≥20 cigarettes/day had the highest odds; (2) an inverse association was observed between alcohol consumption and the presence of sarcopenia in the overall population, but not among men; and (3) the association between smoking and the presence of sarcopenia may be stronger among individuals with higher alcohol consumption, suggesting an interaction between smoking and alcohol consumption.

Smoking levels were found to be associated with increased odds of the presence of sarcopenia, particularly among individuals who smoked ≥20 cigarettes/day compared to non-smokers, suggesting that there may be a dose-response relationship between smoking and sarcopenia (Fig. 2). However, the OR in the >20 cigarettes/day group was not statistically significant and should therefore be interpreted with caution. These findings imply that smoking during middle age and early older adulthood adversely affects the musculoskeletal system, leading to subsequent declines in muscle mass and muscle strength.

Previous studies have compared sarcopenia status between smokers and non-smokers. According to a meta-analysis of cross-sectional studies,7) smokers had a significantly higher prevalence of sarcopenia than non-smokers (OR=1.12). More recently, a longitudinal study8) reported that smokers had a higher risk of sarcopenia than non-smokers (risk ratio=2.36). These findings are consistent with another longitudinal study that assessed skeletal muscle strength, showing that smoking is associated with decreased muscle strength in adults.30) These two longitudinal studies suggest a causal association between smoking and the risk of sarcopenia.

A number of studies have proposed possible mechanisms underlying the association between smoking and the risk of sarcopenia. One study31) reported that smoking impairs muscle protein synthesis and upregulates the expression of genes associated with compromised muscle maintenance. Another study32) reported that serum cotinine levels were inversely associated with muscle mass. These findings suggest a causal link in which smoking increases the risk of sarcopenia.

In the sex-stratified analysis, we observed an association between smoking and the presence of sarcopenia in men, but not in women. The lack of a significant association in women may be attributable to insufficient statistical power due to the small number of female smokers. Nevertheless, this finding is consistent with previous evidence indicating that the smoking-related decline in both the quantity and quality of skeletal muscle is greater in men than in women.33,34)

We further examined the association between alcohol consumption and odds of the presence of sarcopenia, and an inverse association was observed in the overall study population, but not among men. A recent meta-analysis found no association between alcohol consumption and sarcopenia overall.10) Similarly, a longitudinal analysis (n=12,298) of the UK Biobank study showed no association between alcohol consumption and 4-year declines in muscle mass/strength.11) In addition, a cohort study among individuals aged ≥65 years reported no association between alcohol consumption and sarcopenia risk.35) The results in men in the present study are generally consistent with those of these previous studies.10,11,35) The overall association between alcohol consumption and sarcopenia was statistically significant; however, this result appears to have been largely driven by the findings in women, whose alcohol consumption was much lower than that of men. The prevalence of sarcopenia among women in groups consuming >300 g ethanol/week was zero, which may be explained by insufficient statistical power or the healthy drinker effect (i.e., healthier individuals may be more likely to continue drinking). In addition, the relatively high prevalence of sarcopenia among female abstainers and past drinkers may be due to the sick quitter effect (i.e., individuals with poorer health may be more likely to stop drinking).36) For these reasons, the results for the overall population do not necessarily provide clear evidence of a dose–response association between alcohol consumption and sarcopenia.

Our results suggest an interaction between smoking and alcohol consumption on the risk of sarcopenia. The mechanisms underlying the alcohol-smoke interaction on muscle health have not been fully elucidated; however, pharmacodynamic and behavioral interactions and additive effects on the antioxidant defense system may play a role. First, chronic consumption of alcohol may lead to an increase in the volume of distribution of nicotine due to the altered activity of cytochrome P450 families,37) which are known to be reduced in sarcopenic older individuals.38) Second, chronic exposure to a combination of ethanol and nicotine results in additive oxidative tissue injuries in rats.12) Third, smoking and alcohol have reciprocal reinforcing effects mediated by common dopaminergic reward pathways.39) Finally, both smoking and alcohol consumption may alter the gastrointestinal microbiome, resulting in dysbiosis associated with systemic inflammation and altered immune responses.40) Nonetheless, further studies will be needed to confirm our observations.

In the present study, younger individuals (<60 years), men, smokers, and heavy drinkers (≥450 g ethanol/week) tended not to participate in the follow-up physical examination (Supplementary Table S1). This may be partly due to the fact that the physical examination was commonly conducted in association with annual health checkups offered by the local government. Thus, individuals with greater health awareness, who are generally more likely to attend health checkups,41) may have been more likely to participate, potentially reducing the representativeness of the findings. Analysis of the overall association between smoking levels and sarcopenia using inverse-probability weighting showed a slightly lower OR (1.84) than that in the main analysis (OR=1.95). This suggests the possibility of a slight overestimation due to selection bias.

A strength of this study is that it provides important evidence on the association of smoking and alcohol consumption with sarcopenia in a large East Asian population. However, this study also has several limitations worth noting. First, sarcopenia status at baseline was not assessed, and some participants may already have had sarcopenia at the time of the baseline survey. Consequently, some cases identified at follow-up may not represent incident sarcopenia, raising the possibility of reverse causation or misclassification. Second, because deterioration in health status, including sarcopenia, may lead individuals to quit smoking or drinking,36,42) individuals with sarcopenia may have been more likely to be included in the smoking cessation and non-drinking groups in the present study, potentially resulting in an underestimation of the ORs for smoking and alcohol consumption. Third, lifestyle information was assessed using self-reported questionnaires, which may have led to misclassification and, consequently, an underestimation of ORs. Fourth, because abstainers were used as the reference group, this group may include individuals who abstain from alcohol due to health problems; therefore, the ORs should be interpreted with caution. Fifth, although we adjusted for energy intake, we did not consider other dietary factors, such as protein intake, as potential residual confounders. Finally, since BIA is influenced by an individual’s hydration status, it may have affected the assessment of muscle mass.

As with many other diseases, smoking has been shown to exert dose-dependent adverse effects on sarcopenia, providing further evidence in support of smoking cessation. With regard to alcohol consumption, no adverse effects on sarcopenia were observed. Nevertheless, in view of the numerous health risks inherent to alcohol, drinking cannot be regarded as a habitual practice for the prevention of sarcopenia.

In conclusion, the present study targeting middle-aged and older individuals demonstrated that smoking levels are positively associated with sarcopenia, particularly among individuals who smoked ≥20 cigarettes/day. Although a statistically significant association with sarcopenia was observed for alcohol consumption, the findings should be interpreted with caution. In addition, an interaction between smoking and alcohol consumption was suggested. However, causality should be interpreted carefully, as the present study did not assess sarcopenia status at baseline.

SUPPLEMENTARY MATERIALS

Article information Continued