Rev Esp Quimioter 2019; 32(5):458-464

Determination of a cutoff value for medication regimen complexity index to predict polypharmacy in HIV+ older patient 

RAMÓN MORILLO-VERDUGO, Mª DE LAS AGUAS ROBUSTILLO-CORTÉS, LAILA ABDEL-KADER MARTÍN, MARÍA ÁLVAREZ DE SOTOMAYOR PAZ, FERNANDO LOZANO DE LEÓN NARANJO, CARMEN VICTORIA ALMEIDA-GONZÁLEZ

Introduction. HIV+ patients have increased their life expectancy with a parallel increase in age-associated co-morbidities and pharmacotherapeutic complexity. The aim of this study was to determine an optimal cutoff value for Medication regimen complexity index (MRCI) to predict polypharmacy in HIV+ older patients
Patients and methods. A transversal observational single cohort study was conducted at a tertiary Hospital in Spain, between January 1st up to December 31st, 2014. Patients included were HIV patients over 50 years of age on active antiretroviral treatment. Prevalence of polypharmacy and it pattern were analyzed. The pharmacotherapy complexity value was calculated through the MRCI. Receiver operating characteristic curve analyses were used to calculate the area under the curve (AUC) for the MRCI value medications to determine the best cutoff value for identifying outcomes including polypharmacy. Sensitivity and specificity were also calculated.
Results. A total of 223 patients were included. A 56.1% of patients had polypharmacy, being extreme polypharmacy in 9.4% of cases. Regarding the pattern of polypharmacy, 78.0% had a cardio-metabolic pattern, 12.0% depressive-psychogeriatric, 8.0% mixed and 2.0% mechanical-thyroidal. The ROC curve demonstrated that a value of medication complexity index of 11.25 point was the best cutoff for predict polypharmacy (AUC=0.931; sensitivity= 77.6%; specificity=91.8%).
Conclusions. A cut-off value of 11.25 for MRCI is proposed to determine if a patient reaches the criterion of polypharmacy. In conclusion, the concept of polypharmacy should include not only the number of prescribed drugs but also the complexity of them.

Rev Esp Quimioter 2019; 32(5):458-464 [Full-text PDF]