Adherence to oral tamoxifen: a comparison of patient self-report, pill counts, and microelectronic monitoring.

PURPOSE Recent innovations allow the integration of microelectronics into drug packaging, providing a continuous record of the interactions of the patient with the drug package. We hypothesized that adherence to oral tamoxifen, as measured by a pressure-activated microelectronic monitoring device, would be significantly discrepant from traditional measures of patient adherence, ie, patient self-report (SR) and pill counts (PCs). PATIENTS AND METHODS Twenty-six patients receiving oral tamoxifen therapy were assessed by patient SR, PCs, and Medication Event Monitoring System (MEMS; Aprex Corp, Fremont, CA) microelectronic monitoring. A microprocessor in the MEMS cap recorded each opening as a presumptive dose, listing the date, time, and duration of opening for later retrieval on a microcomputer. Patients were not informed that their adherence was to be monitored electronically or that PCs would be performed. RESULTS A total of 2,102 days (70.1 months) of tamoxifen therapy were monitored; patients were monitored for a mean of 2.92 months of tamoxifen therapy. SR adherence to oral tamoxifen was significantly higher than that suggested by either PCs (SR missed doses only v PC, P = .008) or MEMS adherence monitoring (SR missed doses only v MEMS missed doses only, P = .005; SR dosing-interval errors only v MEMS dosing-interval errors only, P < .0001; SR all dosing errors v MEMS all dosing errors, P < .0005). PC data also suggested significantly higher adherence rates than MEMS monitoring. CONCLUSION Microelectronic adherence monitoring provides both confirmatory and complimentary data regarding adherence behavior, while also allowing for the evaluation of patterns of nonadherence. Patient SRs and PCs likely overestimate the degree to which patients adhere to their tamoxifen regimen.