Use of electronic pill boxes to assess risk of poor treatment compliance: results of a large-scale trial.

The objective of the present study was to determine the predictive factors of treatment compliance in hypertensive patients. This was an open large-scale multicenter study where mild to moderate essential hypertensive patients received trandolapril (2 mg) once daily for 30 to 60 days in addition to their usual treatment. Trandolapril was packed in electronic pill boxes that registered date and time of each opening. The main compliance parameters were the percentage of missed doses, the percentage of delayed doses, and the percentage of correct dosing periods. Predictive factors of poor compliance (correct dosing periods < 80%) were determined using a multivariate stepwise logistic regression analysis. Two thousand one hundred seventy-three patients aged 60 +/- 12 years were analyzed. Of the total patients 37% were poor compliers; 29% of patients forgot more than 10% of doses and 36% of patients delayed more than 10% of doses. Ranked predictive factors of poor compliance were: age < 60 years (odds ratio [OR], 1.80 [1.49 to 2.17], P = .0001), the Paris area (OR, 1.70 [1.32 to 2.19], P = .0001), smokers (OR, 1.65 [1.29 to 2.11], P = .0001), monotherapy (OR, 1.40 [1.14 to 1.72], P = .0012), and baseline diastolic blood pressure > or = 100 mm Hg (OR, 1.21 [1.01 to 1.46], P = .044). Therefore, we conclude that young hypertensives, large city dwellers, and smokers are more likely to be poor compliers. The presence of some of these characteristics might incite the physician either to encourage patient compliance or to prescribe antihypertensive drugs that have an effect that persists even beyond 24 h.