Sponsor
This research was supported by the Doris Duke Charitable Foundation Grant #2011042, CFAR P30-AI50410, and the multi-site adherence collaboration in HIV (MACH14) grant R01MH078773 from the National Institute of Mental Health Office on AIDS. The original grants of individual participating studies are: R01DA11869, R01MH54907, R01NR04749, R01NR04749, R01MH068197, R01DA13826, K23MH01862, K08 MH01584, R01AI41413, R01MH61173, AI38858, AI069419, K02DA017277, R01DA15215, P01MH49548, R01MH58986, R01MH61695, CC99-SD003, CC02-SD-003 and R01DA015679. JH was supported by K23MH087228.
Published In
Journal of Acquired Immune Deficiency Syndromes
Document Type
Post-Print
Publication Date
5-2015
Subjects
HIV infections -- Treatment, Antiretroviral therapy, Medical informatics, Patient compliance -- Effect of behavior-driven testing on
Physical Description
20 pages
Abstract
Objective—Regular HIV RNA testing for all HIV positive patients on antiretroviral therapy (ART) is expensive and has low yield since most tests are undetectable. Selective testing of those at higher risk of failure may improve efficiency. We investigated whether a novel analysis of adherence data could correctly classify virological failure and potentially inform a selective testing strategy.
Design—Multisite prospective cohort consortium.
Methods—We evaluated longitudinal data on 1478 adult patients treated with ART and monitored using the Medication Event Monitoring System (MEMS) in 16 United States cohorts contributing to the MACH14 consortium. Since the relationship between adherence and virological failure is complex and heterogeneous, we applied a machine-learning algorithm (Super Learner) to build a model for classifying failure and evaluated its performance using cross-validation.
Results—Application of the Super Learner algorithm to MEMS data, combined with data on CD4+ T cell counts and ART regimen, significantly improved classification of virological failure over a single MEMS adherence measure. Area under the ROC curve, evaluated on data not used in model fitting, was 0.78 (95% CI: 0.75, 0.80) and 0.79 (95% CI: 0.76, 0.81) for failure defined as single HIV RNA level >1000 copies/ml or >400 copies/ml, respectively. Our results suggest 25–31% of viral load tests could be avoided while maintaining sensitivity for failure detection at or above 95%, for a cost savings of $16–$29 per person-month.
Conclusions—Our findings provide initial proof-of-concept for the potential use of electronic medication adherence data to reduce costs through behavior-driven HIV RNA testing.
DOI
10.1097/QAI.0000000000000548
Persistent Identifier
http://archives.pdx.edu/ds/psu/18632
Publisher
Lippincott, Williams & Wilkins
Citation Details
Petersen ML, LeDell E, Schwab J, Sarovar V, Gross R, Reynolds N, Haberer JE, Goggin K, Golin C, Arnsten J, Rosen MI, Remien RH, Etoori D, Wilson IB, Simoni JM, Erlen JA, van der Laan MJ, Liu H, Bangsberg DR. Super Learner Analysis of Electronic Adherence Data Improves Viral Prediction and May Provide Strategies for Selective HIV RNA Monitoring. J Acquir Immune Defic Syndr. 2015 May 1; 69(1):109-118
Description
Author's version of an article that subsequently appeared in Journal of Acquired Immune Deficiency Syndromes, 2015 May 1; 69(1): 109–118. doi:10.1097/QAI.0000000000000548. Published by Lippincott, Williams & Wilkins.
At the time of writing, David Bangsberg was affiliated with Massachusetts General Hospital, Center for Global Health; Harvard School of Public Health, Department of Global Health and Population.