OBJECTIVES: In a cross-sectional study of 157 Turkish women attending outpatient clinics of a university hospital during April-May 2003, association between various subject characteristics and menopause timing was investigated. METHODS: Characteristics were self-reported by women aged 45-60. Of the lifestyle factors, sun exposure, physical activity, food intake and dressing with headscarf were obtained as recalled average lifelong practices up to time of menopause. Cox proportional hazard modeling was used, censoring for hysterectomy, oopherectomy and HRT use. RESULTS: Median age at natural menopause was 52 years. In multivariate analysis, earlier natural menopause was associated with low level of lifelong sun exposure (HR=6.381, 95% CI: 2.996-13.588, p< or =0.0001), heavy physical activity (HR=2.335, 95% CI: 1.305-4.177, p=0.0043), current calcium supplement use (HR=3.191, 95% CI: 1.361-7.485, p=0.0076), diagnosis of hypertension (HR=2.002, 95% CI: 1.186-3.378, p=0.0093), not owning a house (HR=3.002, 95% CI: 1.148-7.852, p=0.0250) and longer years on oral contraceptives (HR=1.085, 95% CI: 1.000-1.176, p=0.0487). Engagement in farming (HR=2.043, 95% CI: 1.056-3.952, p=0.0339), height (cm) (HR=0.953, 95% CI: 0.907-0.994, p=0.0279) and fish consumption (servings/week) (HR=0.600, 95% CI: 0.375-0.960, p=0.0331) were associated with age at menopause in univariate analysis only. For n=109 women who recalled whether maternal menopausal age was <50 or > or =50, sun exposure (HR=7.221, 95% CI: 2.971-17.547, p<0.0001) was a stronger predictor of age at natural menopause than maternal menopausal age (HR=2.882, 95% CI: 1.477-5.621, p=0.0019). CONCLUSIONS: We identify some previously unrecognized correlates of age at natural menopause, namely self-reported lifelong sun exposure, lifelong physical activity, house-ownership, current use of calcium supplements, and lifelong fish consumption. These findings should be confirmed in larger studies.