A cross sectional study on hypertension was done on 294 subjects aged 30 years and above. 150 households were selected randomly representing 50 households from each locality inhabited exclusively by the rural Mizos, indigenous rural Assamese and the tea-garden workers respectively, in the northeastern region of India. Blood pressure was measured by sphygmo-manometer in sitting posture. Anthro-pometric measurements were taken using standard procedure for measuring height, weight, waist and hip girth. Information on age, sex, ethnicity, literacy, alcohol intake, smoking pattern, physical activity, occupation, amount of salt consumption was collected using a standard and pre-tested questionnaire. Significant differences were observed in both the systolic and diastolic blood pressure levels among the three different ethnic groups selected for this study (p < 0.0001). Multiple regression analyses indicated that in Mizos, age, waist circumference and alcohol intake were independently associated with increase in systolic blood pressure whereas smoking was found to be negatively associated with systolic blood pressure (R2 = 0.391, p < 0.001). Factors, which were the best predictors of diastolic blood pressure, were age and body mass index [(kg/m2) (R2 = 0.227, p < 0.001)]. In the rural Assamese population, the best predictors of systolic blood pressure were age and waist circumference (R2 = 0.263, p = 0.018). For the diastolic blood pressure, age, alcohol intake and body mass index were important correlates (R2 = 0.131, p < 0.001). In the tea garden community, important predictors of systolic blood pressure were age, gender and marital status (R2 = 0.187, p < 0.001). On the other hand, age and alcohol intake were best predictors for diastolic blood pressure (R2 = 0.09, p < 0.001).