Age-Specific Differences in Factors Associated with Obstructive Sleep Apnea among Middle-Aged Men: Analysis of the Korea National Health and Nutrition Examination Survey (2021–2023)

Article information

Korean J Health Promot. 2026;.kjhp.2026.00206

Publication date (electronic) : 2026 April 27

doi : https://doi.org/10.15384/kjhp.2026.00206

Hyun-Ji SONG, RN^1,2, Jiyun KIM, PhD, RN^,3

¹Department of Nursing, Gachon University Gil Hospital, Incheon, Korea

²Graduate School of Nursing, Gachon University, Incheon, Korea

³Research Institute of AI and Nursing Science, College of Nursing, Gachon University, Incheon, Korea

Corresponding author: Jiyun KIM, PhD, RN Research Institute of AI and Nursing Science, College of Nursing, Gachon University, 191 Hambangmoe-ro, Yeonsu-gu, Incheon 21936, Korea Tel: +82-32-820-4226 Fax: +82-32-820-4201 E-mail: jkim@gachon.ac.kr

Received 2026 February 9; Revised 2026 March 27; Accepted 2026 March 27.

Abstract

INTRODUCTION

Obstructive sleep apnea (OSA) is a common sleep disorder characterized by recurrent upper airway obstruction during sleep, leading to intermittent cessation of breathing [1]. OSA affects approximately 9%–38% of adults worldwide, with particularly high prevalence among middle-aged men [2,3]. Beyond being a simple sleep disturbance, OSA is associated with various adverse health outcomes, including hypertension, cardiovascular disease, and cognitive impairment [4,5].

Age is one of the most well-established risk factors for OSA, and numerous studies have reported an increased prevalence with advancing age [6,7]. However, most previous research has compared younger adults with older adults, and few studies have examined age-specific differences within the middle-aged population itself [8].

Understanding the specific risk factors for OSA among men in their 40s is particularly important for several reasons. First, this age group experiences rapid changes in metabolic and cardiovascular health [9]. Second, early detection and intervention during middle age may prevent the progression of OSA-related complications [10]. Third, middle-aged men often have competing health priorities and may not recognize OSA symptoms [11].

Emerging evidence suggests that OSA risk factors may vary depending on age, sex, and ethnicity [12,13]. However, research specifically examining how risk factors differ between younger and older middle-aged men remains limited [14].

In Korea, several epidemiological studies have examined OSA, but most have focused either on the general adult population or elderly individuals [15-17]. To our knowledge, no study has systematically compared OSA prevalence and risk factors between middle-aged men in their 40s versus those in their 50s and older using nationally representative data.

Therefore, this study aimed to (1) compare the prevalence of high-risk OSA between middle-aged men <50 years and those ≥50 years, and (2) identify age-specific risk factors associated with OSA in each age group using data from the Korea National Health and Nutrition Examination Survey (KNHANES).

METHODS

This cross-sectional study analyzed data from the KNHANES, conducted from 2021 to 2023. KNHANES is a nationally representative survey that employs a complex, multistage, probability sampling design.

The study population comprised middle-aged men aged ≥40 years who completed the STOP-BANG questionnaire. After excluding participants with missing data on key variables, the final analytical sample consisted of 11,761 men.

The STOP-BANG questionnaire is a validated screening tool for OSA consisting of eight yes/no questions: Snoring, Tiredness, Observed apnea, high blood Pressure, Body mass index (BMI) >35 kg/m², Age >50 years, Neck circumference >40 cm, and male Gender [16]. Total STOP-BANG scores range from 0 to 8. While the original scoring system categorizes risk into low (0–2), intermediate (3–4), and high (5–8) [16], we dichotomized the scores into low risk (0–2) and high risk (3–8) for the purpose of this study. Furthermore, since the ‘Age’ item (≥50 years) is a constituent component of the score, we conducted a sensitivity analysis by excluding the age criteria to address potential endogeneity issues when comparing OSA risk across different age groups.

Participants were categorized into two age groups: <50 years and ≥50 years. Independent variables included demographic characteristics (education, marital status), health behaviors (smoking, high-risk drinking, aerobic physical activity), anthropometric measurements (waist circumference), and psychosocial factors (self-rated health, perceived stress).

Categorical variables were presented as weighted frequencies and Rao-Scott chi-square tests compared categorical variables between age groups. Under the complex sample design, multiple logistic regression analysis identified factors associated with high-risk OSA in each age group, adjusting for potential confounders. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. All analyses accounted for the complex survey design using sampling weights. Statistical significance was set at P<0.05. Analyses were performed using IBM SPSS Statistics version 31 (IBM Corp.). This study used publicly available secondary data. The KNHANES data were obtained from the Korea Disease Control and Prevention Agency, and all personal identifying information had been fully anonymized. The study was conducted with approval from the Institutional Review Board of Gachon University (approval number: 1044396-202512-HR-287-01).

RESULTS

Prevalence of high-risk obstructive sleep apnea

Among the total study population, 41.8% of individuals had high-risk OSA. By age group, 37.2% of participants in the <50 years group and 43.6% of participants in the ≥50 years group were classified as high risk, showing a statistically significant difference (Rao-Scott χ²=36.698, P<0.001) (Table 1).

Table 1.

Prevalence of high-risk OSA by age group

Comparison of sociodemographic characteristics and health behaviors by age group

Significant differences were observed across all variables between age groups (P<0.05). This study comparing health behaviors between age groups (<50 years vs. ≥50 years) among Korean men aged ≥40 years (n=11,761) revealed that the ≥50 years group had significantly higher prevalence of abdominal obesity (29.0% vs. 31.8%), current smoking (22.0% vs. 14.9%), and high-risk drinking (16.1% vs. 10.1%).

Educational disparities were notable, with the ≥50 years group showing higher proportions of lower education levels (23.4% elementary or less vs. 0.8%). Higher smoking and drinking rates in older adults indicate prolonged exposure to health risk factors. Aerobic physical activity and stress levels also differ by age (Table 2).

Table 2.

Comparison of sociodemographic characteristics and health behaviors by age group (n=11,761)

Factors associated with high-risk obstructive sleep apnea by age group

Weighted multivariable logistic regression analyses revealed that factors associated with high-risk OSA differed by age group (Table 3). In the <50 years group, the strongest risk factor was waist circumference ≥90 cm (OR=6.03, 95% CI=4.75–7.65), indicating that abdominal obesity had a substantial impact on OSA risk in younger middle-aged men. Smoking was also a significant risk factor: current smokers had a 5.07-fold higher risk (95% CI=3.79–6.79), and former smokers had a 3.21-fold higher risk (95% CI=2.44–4.22) compared with never smokers. Participants with poor self-rated health had a 1.50-fold higher risk (95% CI=1.07–2.09) than those reporting good health. High-risk drinking was associated with a 2.04-fold increased risk (95% CI=1.51–2.76) compared with non–high-risk drinking. Participants with high stress level had a 1.89-fold higher risk (95% CI=1.02–3.48) than those reporting no stress. Moderate stress level was associated with a 1.55-fold increased risk (95% CI=1.02–2.34) compared with no stress. In contrast, educational attainment, marital status, and physical activity were not statistically significant in this age group.

Table 3.

Factors associated with high-risk OSA by age group: multivariable logistic regression

In the ≥50 years group, former smoking emerged as the strongest risk factor, with a 5.69-fold higher risk (95% CI=4.98–6.49) compared with never smokers; current smokers also showed a 3.88-fold higher risk (95% CI=3.28–4.59). Waist circumference ≥90 cm was associated with a 2.94-fold higher risk (95% CI=2.61–3.31) compared with <90 cm, although the magnitude of the effect was smaller than that observed in the <50 years group. Poor self-rated health was associated with a 2.09-fold higher risk (95% CI=1.77–2.46), and fair self-rated health with a 1.17-fold higher risk (95% CI=1.02–1.35), compared with good health. Psychosocial factors showed significant associations in the ≥50 years group. Being married was associated with a 1.30-fold higher risk (95% CI=1.14–1.48) compared without being married. With no perceived stress as the reference category, those reporting high stress had a high risk (OR=1.85, 95% CI=1.30–2.63).

Regarding educational attainment, having a high school education was associated with a lower risk of high-risk OSA (OR=0.78, 95% CI=0.68–0.89) compared with elementary school education or less. High-risk drinking was associated with a 1.63-fold higher risk (95% CI=1.32–2.00) compared with non–high-risk drinking.

DISCUSSION

This nationwide study demonstrates age-specific differences in OSA prevalence and risk factors among middle-aged Korean men. High-risk OSA was more prevalent in men ≥50 years (43.6%) compared to those <50 years (37.2%), demonstrating an age gradient within the middle-aged population itself. The observed age-related increase aligns with previous reports [6-8] and may be related to pathophysiological changes including upper body fat redistribution, reduced upper airway muscle tone, and hormonal alterations [9,10].

Our analysis revealed distinct risk factor profiles between age groups. In younger middle-aged men, abdominal obesity emerged as the predominant risk factor, suggesting a potential role of central adiposity in OSA pathogenesis [18,19]. The strong association may reflect metabolic consequences of central fat accumulation, including increased neck circumference and pharyngeal fat deposition [20-25].

Smoking showed significant associations in both groups but with different patterns. Among younger men, current smoking (OR=5.07) showed a stronger association with OSA than past smoking (OR=3.21). In contrast, among older men, past smoking was the most significant risk factor (OR=5.69), outweighing the effect of current smoking (OR=3.88). While smoking intensity was not quantified in this study, the robust association between past smoking and OSA in older men suggests a possible lasting influence of smoking history. This observation may point toward a potential link between long-term smoking and age-related increases in airway collapsibility [26,27].

Interestingly, marital status was associated with an increased OSA risk specifically in older men. While the exact mechanism remains unclear, this association may be attributed to spousal observation and reporting of symptoms rather than a true increase in disease prevalence. Since the STOP-BANG score includes subjective items such as witnessed apnea and loud snoring, participants with spouses are more likely to be informed of these nocturnal symptoms, potentially leading to higher scores. This suggests that the presence of a bed partner may play a critical role in the clinical screening of OSA in older populations.

Regarding psychological factors, perceived stress showed a significant association with OSA risk in both age groups, but the patterns of sensitivity differed. In men aged <50 years, even moderate stress levels were associated with significantly higher odds of OSA risk, whereas in the older group, this association was more pronounced at higher stress levels. These results may indicate that younger individuals are more responsive to stress, or that stress-related physiological changes, such as increased sympathetic activity, could contribute more noticeably to OSA risk in this group.

The findings have important clinical implications. For younger middle-aged men, interventions may benefit from prioritizing aggressive abdominal obesity management alongside smoking cessation. Although abdominal obesity was a significant risk factor in both age groups [14], its stronger association in younger middle-aged men may reflect earlier pathophysiological vulnerability before age-related structural changes become predominant. The finding that past smoking history—rather than current smoking—emerged as the most potent risk factor among older men may indicate a possible long-term influence of smoking on airway physiology [20]. This pattern suggests that smoking-related respiratory effects could persist or become more apparent with age, even after cessation. Therefore, our results highlight the potential importance of early smoking cessation; quitting at a younger age may help reduce the cumulative physiological burden on the upper airway and may be associated with a lower risk of developing OSA later in life.

Study strengths include large nationally representative data, validated screening tool, and age-stratified analysis within middle-aged men. Age-specific approaches to screening and intervention may enhance OSA detection and management effectiveness in middle-aged men. Limitations include cross-sectional design precluding causal inference, self-reported data potentially introducing bias, and lack of polysomnography confirmation. Future longitudinal studies with objective OSA measures are needed.

This study identified age-specific differences in factors associated with high-risk OSA among middle-aged men. Study findings suggest that age-tailored public health strategies may be beneficial. Further longitudinal studies using objective diagnostic measures are needed to clarify causal relationships and to inform more effective intervention strategies [28].

Article information Continued