Medication Possession Ratio Predicts Antiretroviral Regimens Persistence in Peru

Academic Article

Abstract

  • Objectives:In developing nations, the use of operational parameters (OPs) in the prediction of clinical care represents a missed opportunity to enhance the care process. We modeled the impact of multiple measurements of antiretroviral treatment (ART) adherence on antiretroviral treatment outcomes in Peru.Design And Methods:Retrospective cohort study including ART naïve, non-pregnant, adults initiating therapy at Hospital Nacional Cayetano Heredia, Lima-Peru (2006-2010). Three OPs were defined: 1) Medication possession ratio (MPR): days with antiretrovirals dispensed/days on first-line therapy; 2) Laboratory monitory constancy (LMC): proportion of 6 months intervals with ≥1 viral load or CD4 reported; 3) Clinic visit constancy (CVC): proportion of 6 months intervals with ≥1 clinic visit.Three multi-variable Cox proportional hazard (PH) models (one per OP) were fit for (1) time of first-line ART persistence and (2) time to second-line virologic failure. All models were adjusted for socio-demographic, clinical and laboratory variables.Results:856 patients were included in first-line persistence analyses, median age was 35.6 years [29.4-42.9] and most were male (624; 73%). In multivariable PH models, MPR (per 10% increase HR=0.66; 95%CI=0.61-0.71) and LMC (per 10% increase 0.83; 0.71-0.96) were associated with prolonged time on first-line therapies.Among 79 individuals included in time to second-line virologic failure analyses, MPR was the only OP independently associated with prolonged time to second-line virologic failure (per 10% increase 0.88; 0.77-0.99).Conclusions:The capture and utilization of program level parameters such as MPR can provide valuable insight into patient-level treatment outcomes. © 2013 Salinas et al.
  • Authors

    Published In

  • PLoS ONE  Journal
  • Digital Object Identifier (doi)

    Author List

  • Salinas JL; Alave JL; Westfall AO; Paz J; Moran F; Carbajal-Gonzalez D; Callacondo D; Avalos O; Rodriguez M; Gotuzzo E
  • Volume

  • 8
  • Issue

  • 10