Temporal proximity to counseling is associated with longitudinal bedaquiline and antiretroviral therapy adherence in multidrug-resistant tuberculosis and HIV treatment
10 affiliations
- 1Division of Pulmonary, Allergy, and Critical Care Medicine, Columbia University Irving Medical Center, New York, New York, United States of America
- 2Graduate School of Social Work, Touro University, New York, New York, United States of America
- 3School of Applied Human Sciences, University of KwaZulu-Natal, Durban, South Africa
- 4Department of Industrial and Systems Engineering, Lehigh University, Bethlehem, Pennsylvania, United States of America
- 5CAPRISA MRC- HIV-TB Pathogenesis and Treatment Research Unit, Durban, South Africa
- 6Department of Pulmonology and Critical Care, School of Clinical Medicine, College of Health Sciences, University of KwaZulu-Natal, Durban, South Africa
- 7University of Michigan School of Public Health, Ann Arbor, Michigan, United States of America
- 8Department of Medicine, University of Washington, Seattle, Washington, United States of America
- 9School of Global Health and Dahdaleh Institute of Global Health Research, York University, Toronto, Canada
- 10Department of Epidemiology, Columbia University Irving Medical Center, New York, New York, United States of America
Abstract
Background: Medication adherence is a persistent challenge in the treatment of multidrug-resistant tuberculosis (MDR-TB) and HIV. Routine counseling notes are an underutilized source of information on patient adherence context. Methods: We applied natural language processing (NLP) to routine counseling notes from a randomized controlled trial of MDR-TB/HIV treatment in South Africa. Topics were derived using topic modeling; notes were semantically matched to topics and clustered using K-means. We fitted linear mixed-effects (LME) regression models with participant-level random intercepts and five-fold cross-validation to model adherence to bedaquiline (BDQ) and antiretroviral therapy (ART), measured using cellular-enabled electronic dose monitor (EDM). Cosine similarity scores, cluster assignments, and temporal proximity to counseling sessions were included as covariates. Model fit was assessed using mean absolute error (MAE) and root mean square error (RMSE). Results: We analyzed 327 counseling notes from 76 participants. Six thematic clusters were derived: substance use, logistical barriers, limited social support, side effect management, financial stability, and financial insecurity/low support. LME showed modest fit. For BDQ adherence, MAE was 0.158 (95% CI 0.113–0.204) and RMSE 0.208 (95% CI 0.133–0.283). For ART, MAE was 0.180 (95% CI 0.170–0.191) and RMSE 0.224 (95% CI 0.202–0.246). Cluster membership was not associated with adherence. In contrast, temporal proximity to counseling was more consistently associated with adherence. Adherence was highest immediately following counseling sessions. Both BDQ and ART adherence declined during the intermediate period (β=−0.048, p = 0.001; β=−0.031, p = 0.022, respectively), and ART adherence declined further in the period preceding the next session (β=−0.032, p = 0.020). Conclusion: NLP methods structured routine counseling notes into interpretable, adherence-related themes. Temporal proximity to counseling sessions was consistently associated with longitudinal BDQ and ART adherence. These findings offer preliminary insights into adherence dynamics and generate hypotheses about the role of counseling session timing in adherence support.
Related
- Contribution of the novel oxazolidinone TBD09 (MK-7762) to regimens with bedaquiline and pretomanid in a mouse model of tuberculosis
- Empiric tuberculosis treatment and 12-month mortality among sputum GeneXpert-negative adults living with HIV in Uganda in the era of widespread antiretroviral therapy: A prospective cohort study
- The HIV care continuum and determinants of antiretroviral therapy initiation among transgender women in France (1997–2023): a longitudinal cohort study
- Viral immune profiles in multiple sclerosis and HIV co-infection providing insights into antiretroviral treatment-associated immuno-remodeling
- HIV-1 drug resistance and associated risk factors in antiretroviral therapy failure: a 2022–2024 surveillance in Huzhou, China
BibTeX
@article{lu2026temporal,
title = {{Temporal proximity to counseling is associated with longitudinal bedaquiline and antiretroviral therapy adherence in multidrug-resistant tuberculosis and HIV treatment}},
author = {Xuan Lu and Jennifer Zelnick and Meng Zhao and Matthew Cummings and Allison K. Wolf and Kevin Guzman and Hlengiwe Nyilana and Rubeshan Perumal and Kathleen Rivet Amico and Karl Reis and Mbali Zulu and Amrita Daftary and Boitumelo Seepamore and Kogieleum Naidoo and Max O’Donnell},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0355696},
url = {https://doi.org/10.1371/journal.pone.0355696},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Temporal proximity to counseling is associated with longitudinal bedaquiline and antiretroviral therapy adherence in multidrug-resistant tuberculosis and HIV treatment
AU - Xuan Lu
AU - Jennifer Zelnick
AU - Meng Zhao
AU - Matthew Cummings
AU - Allison K. Wolf
AU - Kevin Guzman
AU - Hlengiwe Nyilana
AU - Rubeshan Perumal
AU - Kathleen Rivet Amico
AU - Karl Reis
AU - Mbali Zulu
AU - Amrita Daftary
AU - Boitumelo Seepamore
AU - Kogieleum Naidoo
AU - Max O’Donnell
PY - 2026
DA - 2026/10/05
JO - PLOS One
DO - 10.1371/journal.pone.0355696
UR - https://doi.org/10.1371/journal.pone.0355696
AB - Background: Medication adherence is a persistent challenge in the treatment of multidrug-resistant tuberculosis (MDR-TB) and HIV. Routine counseling notes are an underutilized source of information on patient adherence context. Methods: We applied natural language processing (NLP) to routine counseling notes from a randomized controlled trial of MDR-TB/HIV treatment in South Africa. Topics were derived using topic modeling; notes were semantically matched to topics and clustered using K-means. We fitted linear mixed-effects (LME) regression models with participant-level random intercepts and five-fold cross-validation to model adherence to bedaquiline (BDQ) and antiretroviral therapy (ART), measured using cellular-enabled electronic dose monitor (EDM). Cosine similarity scores, cluster assignments, and temporal proximity to counseling sessions were included as covariates. Model fit was assessed using mean absolute error (MAE) and root mean square error (RMSE). Results: We analyzed 327 counseling notes from 76 participants. Six thematic clusters were derived: substance use, logistical barriers, limited social support, side effect management, financial stability, and financial insecurity/low support. LME showed modest fit. For BDQ adherence, MAE was 0.158 (95% CI 0.113–0.204) and RMSE 0.208 (95% CI 0.133–0.283). For ART, MAE was 0.180 (95% CI 0.170–0.191) and RMSE 0.224 (95% CI 0.202–0.246). Cluster membership was not associated with adherence. In contrast, temporal proximity to counseling was more consistently associated with adherence. Adherence was highest immediately following counseling sessions. Both BDQ and ART adherence declined during the intermediate period (β=−0.048, p = 0.001; β=−0.031, p = 0.022, respectively), and ART adherence declined further in the period preceding the next session (β=−0.032, p = 0.020). Conclusion: NLP methods structured routine counseling notes into interpretable, adherence-related themes. Temporal proximity to counseling sessions was consistently associated with longitudinal BDQ and ART adherence. These findings offer preliminary insights into adherence dynamics and generate hypotheses about the role of counseling session timing in adherence support.
SN - 1932-6203
ER -