Super learner analysis of electronic adherence data improves viral prediction and may provide strategies for selective HIV RNA monitoring

Details

• Alternative Title:
  J Acquir Immune Defic Syndr
• Personal Author:
  Petersen, Maya L. ; LeDell, Erin ; Schwab, Joshua ; Sarovar, Varada ; Gross, Robert ; Reynolds, Nancy ; Haberer, Jessica E. ; Goggin, Kathy ; Golin, Carol ; Arnsten, Julia ; Rosen, Marc ; Remien, Robert ; Etoori, David ; Wilson, Ira ; Simoni, Jane M. ; Erlen, Judith A. ; van der Laan, Mark J. ; Liu, Honghu ; Bangsberg, David R
• Description:
  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.
  
  
• Subjects:
  Adult Anti-HIV Agents Artificial Intelligence Cohort Studies Decision Support Techniques Drug Monitoring Female HIV Infections Humans Longitudinal Studies Male Medication Adherence Middle Aged Prospective Studies RNA, Viral Treatment Failure

  More +
• Keywords:
  Adherence Antiretroviral Therapy HIV HIV RNA Monitoring Medication Event Monitoring System Super Learner Virological Failure
• Source:
  J Acquir Immune Defic Syndr. 69(1):109-118
• Pubmed ID:
  25942462
• Pubmed Central ID:
  PMC4421909
• Document Type:
  Journal Article
• Funding:
  R01 MH058986/MH/NIMH NIH HHSUnited States/ ; K08 MH01584/MH/NIMH NIH HHSUnited States/ ; R01 MH078773/MH/NIMH NIH HHSUnited States/ ; R01MH61695/MH/NIMH NIH HHSUnited States/ ; K23MH087228/MH/NIMH NIH HHSUnited States/ ; AI069419/AI/NIAID NIH HHSUnited States/ ; R01MH078773/MH/NIMH NIH HHSUnited States/ ; R01 MH061695/MH/NIMH NIH HHSUnited States/ ; R01 MH054907/MH/NIMH NIH HHSUnited States/ ; CC99-SD003/CC/ODCDC CDC HHSUnited States/ ; CC02-SD-003/CC/ODCDC CDC HHSUnited States/ ; P30 MH062294/MH/NIMH NIH HHSUnited States/ ; R01MH61173/MH/NIMH NIH HHSUnited States/ ; R01 DA015215/DA/NIDA NIH HHSUnited States/ ; AI38858/AI/NIAID NIH HHSUnited States/ ; P30 MH043520/MH/NIMH NIH HHSUnited States/ ; R01 NR004749/NR/NINR NIH HHSUnited States/ ; R01MH068197/MH/NIMH NIH HHSUnited States/ ; R01DA13826/DA/NIDA NIH HHSUnited States/ ; K24 MH092242/MH/NIMH NIH HHSUnited States/ ; K23 MH087228/MH/NIMH NIH HHSUnited States/ ; R01MH58986/MH/NIMH NIH HHSUnited States/ ; K02DA017277/DA/NIDA NIH HHSUnited States/ ; P30 AI045008/AI/NIAID NIH HHSUnited States/ ; R01 DA015679/DA/NIDA NIH HHSUnited States/ ; R01 MH061173/MH/NIMH NIH HHSUnited States/ ; R01DA015679/DA/NIDA NIH HHSUnited States/ ; P30-AI50410/AI/NIAID NIH HHSUnited States/ ; K23MH01862/MH/NIMH NIH HHSUnited States/ ; U01 AI038858/AI/NIAID NIH HHSUnited States/ ; K08 MH001584/MH/NIMH NIH HHSUnited States/ ; R01AI41413/AI/NIAID NIH HHSUnited States/ ; K02 DA017277/DA/NIDA NIH HHSUnited States/ ; P01MH49548/MH/NIMH NIH HHSUnited States/ ; R01 AI074345/AI/NIAID NIH HHSUnited States/ ; R01DA11869/DA/NIDA NIH HHSUnited States/ ; R01 MH068197/MH/NIMH NIH HHSUnited States/ ; R01MH54907/MH/NIMH NIH HHSUnited States/ ; P30 AI042853/AI/NIAID NIH HHSUnited States/ ; R01DA15215/DA/NIDA NIH HHSUnited States/ ; UM1 AI069419/AI/NIAID NIH HHSUnited States/ ; R01NR04749/NR/NINR NIH HHSUnited States/ ; R01 DA013826/DA/NIDA NIH HHSUnited States/ ; U01 AI069419/AI/NIAID NIH HHSUnited States/ ; P30 AI050410/AI/NIAID NIH HHSUnited States/
• Volume:
  69
• Issue:
  1
• Collection(s):
  CDC Public Access
• Main Document Checksum:
  urn:sha256:5099e45a522be2fb55d2f625bf4905e251a0781d0c9c7886dcac1b6026d40680
• Download URL:
  https://stacks.cdc.gov/view/cdc/31990/cdc_31990_DS1.pdf
• File Type:
  [PDF - 1.78 MB ]