To explore the effect of occupational stress on hypertension. 498 workers whose accumulative length of service was more than two years were investigated with questionnaire by method of cluster sampling from a thermal power plant in Henan province in China; 446 respondents returned qualified questionnaire including 281 male and 165 female Han People. After the patients with secondary hypertension, diabetes patients, and patients with liver or kidney disease were excluded, 84 workers (58 males and 26 females) were diagnosed as hypertension. The occupational stressors, personalities, buffering factors and occupational strain were measured by using the Job Demand-control Model, the Effort-reward Imbalance Model questionnaires and Occupational Stress Measurement Scale. Main risk factors for the development of hypertension such as heredity, body mass index, high salt diet, alcohol use, smoking habit and lack of physical activity were investigated. 498 whole blood samples were collected from workers in field epidemiologic survey. All of the samples were detected TG, CHO and FPG by common biochemistry methods. Multivariate logistic regression analysis were used to determine the relationship between occupational stressors and prevalence rate of hypertension. The difference of morbidity of hypertension between different stress level subjects was analyzed by chi2 test. (1) Logistic regression analysis of the hypertension by all occupational stressors and risk factors of hypertension indicated that not only some common factors such as parents' hypertensive history, BMI, alcohol use and TG, but also responsibility for person, work locus of control and social support were significantly correlated with elevated risks of hypertension. (2) Logistic regression analysis of the hypertension by main dimensions of effort-reward imbalance model and risk factors of hypertension indicated that parents' hypertensive history, BMI, alcohol use, TG, and effort were significantly correlated with elevated risks of hypertension. Logistic regression analysis indicated the risk of hypertension had an effect on the FRI and effort (OR was 1.71 and 2.43 respectively). (3) For the job strain model, results indicated that parents' hypertensive history, UMI, alcohol use, TG, work locus of control and social support were significantly correlated with elevated risks of hypertension. But the main dimensions of job strain model (job demands and decision latitude) didn't enter regression equation. (4) The difference of prevalence of hypertension between high- and low stress level groups in male was statistically significant (OR = 3.13, P < 0.01), but the case was not the same in female (P > 0.05). Occupational stress might be risk factor of hypertension; The predictive power of effort-reward imbalance model for the development of hypertension would be larger than that of job strain model.