OBJECTIVES: The aim of this study was to investigate whether working married women in different occupational classes affected diverse health outcomes. METHODS: We used data for married women aged 25-59 (N=2,273) from the 2005 National Health and Nutritional Examination Survey. Outcome measures included physical/mental and subjective/objective indicators (self-rated poor health, chronic diseases, depression, and suicidal ideation from reported results; metabolic syndrome and dyslipidemia from health examination results). Age-standardized prevalence and logistic regression were employed to assess health status according to three types of working groups (housewives, married women in manual jobs, married women in non-manual jobs). Sociodemographic factors (age, numbers of children under 7, education, household income) and health behaviors (health examination, sleep, rest, exercise, smoking, drinking) and a psychological factor (stress) were considered as covariates. RESULTS: Non-manual married female workers in Korea showed better health status in all five health outcomes than housewives. The positive health effect for the non-manual group persisted in absolute (age-adjusted prevalence) and relative (odds ratio) measures, but multivariate analyses showed an insignificant association of the non-manual group with dyslipidemia. Manual female workers showed significantly higher age-adjusted prevalence of almost all health outcomes than housewives except chronic disease, but the associations disappeared after further adjustment for covariates regarding sleep, rest, and stress. CONCLUSIONS: Our results suggest that examining the health impact of work on married women requires the consideration of occupational class.