학술논문

An assessment of the relationship between the World Health Organization HIV drug resistance early warning indicators and HIV drug resistance acquisition.
Document Type
Article
Source
HIV Medicine. May2017, Vol. 18 Issue 5, p342-353. 12p.
Subject
*HIV prevention
*CLINICAL medicine
*CONFIDENCE intervals
*DRUG resistance in microorganisms
*DOSE-effect relationship in pharmacology
*HIV infections
*PUBLIC health
*RISK assessment
*LOGISTIC regression analysis
*PROTEASE inhibitors
*KEY performance indicators (Management)
*HIGHLY active antiretroviral therapy
*PREDICTIVE tests
*LAMIVUDINE
*EMTRICITABINE
*NON-nucleoside reverse transcriptase inhibitors
*NUCLEOSIDE reverse transcriptase inhibitors
*ODDS ratio
Language
ISSN
1464-2662
Abstract
The World Health Organization (WHO)'s HIV drug resistance (HIVDR) early warning indicators (EWIs) measure antiretroviral therapy (ART)-site factors associated with HIVDR prevention, without HIVDR laboratory testing. We assessed the relationship between EWIs and HIVDR acquisition using data from British Columbia, Canada. Methods Eligible patients were ART-naïve, were ≥ 19 years old, had initiated ART between 1 January 2000 and 31 December 2012, had ≥ 15 months of follow-up, and were without transmitted HIVDR. Patients were followed for acquired HIVDR until 31 March 2014, the last contact date, or death. We built logistic regression models to assess the associations and predictive ability of individual indicators and of the EWI Score (the number of indicators for which a patient did not meet the criteria) on HIVDR acquisition (to any class of HIVDR, lamivudine (3TC)/emtricitabine (FTC), nonnucleoside reverse transcriptase inhibitors (NNRTIs), nucleoside reverse transcriptase inhibitors (NRTIs) or protease inhibitors (PIs)]). Results All explored EWIs were associated with at least one class of HIVDR, with the exception of 'ART prescribing practices'. We observed a dose--response relationship between acquiring HIVDR to any antiretroviral class and an increasing EWI score in our predictive logistic regression model. The area under the curve was 0.848 (excellent discrimination). The adjusted odds ratios for acquiring any class of HIVDR for an EWI score of 1, 2 and ≥ 3 versus 0 were 2.30 [95% confidence Interval (CI) 1.21-4.38], 3.35 (95% CI: 1.86-6.03) and 7.26 (95% CI: 4.18-12.61), respectively. Conclusions Several EWIs were associated with and predictive of HIVDR, supporting the WHO EWIs as a component of the HIVDR prevention method in settings where HIVDR testing is not routinely or widely available. [ABSTRACT FROM AUTHOR]