Background. The diagnosis of non-adherence is complex and there are no completely reliable methods that can be used widely in daily practice. The aim of this study was to validate electronic prescriptions as a method of measuring treatment adherence in patients with mild-moderate hypertension.
Methods. We conducted a prospective, longitudinal, multicenter study in primary care centers. The study involved 120 patients treated for hypertension and included in the electronic prescription program of the centers. Five visits were made: initial, 6, 12, 18 and 24 months. Adherence was measured using an electronic monitor [medication event monitoring system (MEMS)] and through the electronic prescription program. We calculated the adherence rate (AR) using the MEMS and the electronic prescriptions, with the Medication Possession Ratio (MPR). Adherent patients were considered to be those whose AR was 80-100%. To validate the electronic prescription, its data were compared to the pill count by MEMS (Method of certainty). Sensitivity, specificity, positive (PPV) and negative (NPV) predictive values, and positive (LR+) and negative (LR-) likelihood ratios were calculated. The Kappa concordance index and the area under the ROC curve were also determined.
Results. The study was completed by 108 patients (mean age 61.06 years, SD 9.08). Adherence was 77.4% by MEMS (95% CI: 66.8-88%) and 80.4% (95% CI=70.3-90.5) by MPR. At 24 months the sensitivity was 87%, specificity 93.7%, PPV 80%, NPV 96.1%, LR+ 13.8 and LR- 0.1. The K was 0.782 and the AUC was 0.903 (95% CI: 0.817-0.989). Therapeutic complexity was associated with pharmacological non-adherence (OR=1.35, p<0.01).
Discussion. The electronic prescription was an excellent method to measure non-adherence in hypertensive patients included in this program for over two years. Therapeutic simplicity improved treatment adherence.