Differences in health risk behavior clusters by occupational group among Korean adult workers: a sex-stratified latent class analysis using the Korea National Health and Nutrition Examination Survey (KNHANES) 2022–2024

Health risk behaviors do not occur in isolation but tend to co-occur in clusters, amplifying occupational health risk. However, few studies have examined the association between multiple health-risk-behavior clustering patterns and occupational group among Korean workers using a sex-stratified approach. This is a cross-sectional study using pooled data from the 9th cycle of the Korea National Health and Nutrition Examination Survey (KNHANES, 2022–2024) (final analytic sample, n = 8,246; 3,285 men, 4,961 women). Sex-stratified latent class analysis (LCA) was performed using five binary health-risk-behavior indicators: current smoking, high-risk drinking, short sleep duration (< 6 h), skipping breakfast, and insufficient physical activity. The association between occupational group (clerical, service/sales, agriculture/forestry/fishery, and craft/labor) and latent class membership was examined using complex-sample weighted logistic regression (svyglm), adjusting for age, education level, household income, and employment status. A 3-class model was optimal for both sexes. Men were classified into a low-risk, physically inactive class (68.8%), a high-risk, smoking/drinking class (25.6%), and a moderate-risk, drinking/breakfast-skipping/smoking class (5.6%); women were classified into a low-risk, physically inactive class (85.5%), a moderate-risk, drinking/breakfast-skipping class (8.7%), and a high-risk, multiple-behavior class (5.7%). Among men, the craft/labor group had a significantly higher risk of belonging to the high-risk, smoking/drinking class than the clerical group even after full adjustment (adjusted OR = 1.31; 95% CI: 1.01–1.69). Men in the agriculture/forestry/fishery group also showed a significantly increased risk (adjusted OR = 1.51; 95% CI: 1.01–2.26). Among women, the service/sales group had a significantly higher risk of belonging to the high-risk, multiple-behavior class after adjustment for age (adjusted OR = 2.66; 95% CI: 1.50–4.71), but this association attenuated after further adjustment for socioeconomic factors. Health risk behaviors among Korean adult workers show distinct clustering patterns by sex, and occupational group was significantly associated with cluster membership, particularly among men in the craft/labor and agriculture/forestry/fishery groups. These findings may help inform occupation- and sex-specific workplace health promotion strategies that address multiple co-occurring health risk behaviors, including among women in the service/sales group.