Digit preferences observed in the measurement of blood pressure: repercussions on the success criteria in current treatment of hypertension.

In a multicenter, open, noncomparative trial to assess the efficacy of an angiotensin-converting enzyme inhibitor, quinapril (Accupril; Parke-Davis), after 12 weeks of treatment in 667 adult patients 19-83 years of age with stage 1-3 hypertension, conducted by 85 physicians in primary health care, with systolic blood pressure (SBP) < 140 mm Hg and diastolic blood pressure (DBP) < 90 mm Hg as criteria for normalization, the efficacy of the drug was 75.1%. When an analysis was made of the frequency tables of BP recorded by the physicians in the case-report forms, a clear numerical preference was found in which the DBP was expressed in multiples of 5 in 81.3% of the cases and the SBP was expressed in multiples of 10 in 19% of the records, so that when a cutoff point <140/90 mm Hg is chosen in daily practice, 130/85 mm Hg is actually being selected. It suffices to change the criteria to accept as normal values less than or equal to instead of less than 140/90 mm Hg to increase the efficacy of the drug from 75.1% to 90.7% in our trial. Therefore, it is proposed to use multiples of 5 for DBP and multiples of 10 for SBP as cutoff points and the diffusion of clear recommendations on BP measurement.