Clinical predictors of treatment reduction in hypertensive patients.

<bold>Objective: </bold>To demonstrate that some hypertensive patients under good blood pressure (BP) control can reduce medications, and to identify predictors of successful reduction.<bold>Design: </bold>Observational study with 11-month follow-up.<bold>Setting: </bold>Outpatient hypertension clinic at the Seattle Veterans Administration Hospital.<bold>Patients: </bold>59 males (51% of those eligible) with diastolic BP less than 95 mm Hg for greater than or equal to 6 months; 57 patients (97%) completed the study.<bold>Intervention: </bold>Gradual reduction of medications unless diastolic BP rose above 95 mm Hg.<bold>Measurements and Main Results: </bold>Intensity of treatment with BP medications was assessed using a scale of their comparative "vigors." 35 patients (59%) reduced medications successfully. By the end of the study, systolic BP had risen by 8.2 +/- 12.3 mm Hg (mean +/- SD) in successful patients, while diastolic BP did not change significantly. Two predictors of treatment reduction were statistically significant in both univariate and multivariate analyses: successful patients had been treated more intensively (2.7 +/- 1.7 vs. 1.3 +/- 0.5 "vigor units," p = 0.0001), and they had been enrolled in the clinic longer (5.5 +/- 3.0 vs. 3.1 +/- 2.3 years, p = 0.003). Lower systolic BP, higher urinary sodium excretion, lower compliance, and younger age were significant predictors of treatment reduction on univariate analysis only. Age less than or equal to 65 years had the highest sensitivity (86%) for treatment reduction, while treatment with two or more "vigor units" had the highest specificity (79%) and likelihood ratio (3.3).<bold>Conclusions: </bold>Treatment reduction is feasible in many well-controlled hypertensives, though systolic BP rises. Patients with high intensity and long duration of treatment are most likely to reduce medications successfully. [ABSTRACT FROM AUTHOR]