RESUMEN
PURPOSE: To evaluate the relaibility of the blood pressure measurements in an arterial hypertension study. METHODS: It was evaluated 502 nurse workers of a public hospital. After careful training and selection of the observers for the quality control at the fieldwork, it was analyzed the correlation between the 1st and 2nd measurements, final digit preference and mean systolic (SBP) and diastolic pressure (DBP). 5% of the sample had their measures compared with those checked by an expert. RESULTS: There was a great confiability intra-observer (96.2% in the SBP and 93.4% in the DBP). The final digit distribution was almost identical. At the 5% sample, there was a strong correlation between them (96.0% in the SBP and 93.9% in the DBP). CONCLUSION: This methodology is feasible in national studies and confers validity to the results.