OBJECTIVE: The prevalencies of hypertension are analysed by sex and age group, in social groupings established in accord with social criteria. With a view better to understanding the social dimension of the disease, prevalencies were characterised by type of occupation. MATERIAL AND METHODS: The sample consisted of 1,041 people and corresponds to the sum of the samples representing the "study areas" established by the use of socio-economic and geographical criteria. Four social strata were defined in obedience to a socioeconomic gradient. Hypertension was defined by the Joint National Committee (JNC), 140/90 mmHg, and of the World Health Organization (WHO), 160/95 mmHg, standard references. RESULTS: According to the JNC and WHO standard references the prevalencies of hypertension, age adjusted, were of approximately the following: stratum (I+II) 60 and 37%; stratum III 50 and 39%; stratum IV 55 e 46%. Among women the prevalencies were 40 and 38% (stratum I+II); 56 and 47% (stratum III) and 55 and 46% (stratum IV). For the men belonging to the economically active population, classified by occupation, it was showed that the freelance professionals, consisting of businessmen of small firms, small traders and liberal professionals, presented a prevalence of about 60 and 37%; the skilled workers, employed in factories of 35 and 14%; the daily freelance workers, unskilled laborers and unemployed, of 59 and 40%. The women were divided by occupation as belonging or not to the economically active population (EAP) and presented the following prevalencies: 39 and 47%, respectively, according to the JNC standard, and 27 and 45%, respectively, according to the WHO standard. Thus it may be seen that these results run counter to the hypothesis that women integrated into the labour market are more exposed to the risk factors for non-transmissible diseases. CONCLUSION: Thus it may be concluded that the categories most affected by the present economic were those most affected by hypertension. On the other hand the possibility of there being and a intense social determination in the etiology of hypertension in this population is demonstrated.