EACS2023: 983 Predictive capacity of the 7-item I-Score, a patient-reported measure of barriers to ART adherence for HIV care K. Engler , S. Vicente , D. Lessard , K. Lacombe , H. Rougier , L. Delvallez , D. Chu , J.-P. Routy , A. de Pokomandy , M. Klein , B. Lebouché Research Institute of the McGill University Health Centre, Centre for Outcomes Research and Evaluation, Montreal, Canada, McGill University, Department of Family Medicine, Montreal, Canada, Sorbonne University, Faculty of Medicine, Paris, France, Institut de Médecine et d'Épidémiologie Appliquée (IMEA), Paris, France, University of Lille, Faculty of Engineering and Health Management, Lille, France, McGill University Health Centre, Chronic Viral Illness Service, Montreal, Canada General data Abstract category: Antiretroviral therapy – observational studies Abstract body Purpose: To identify adherence barriers to antiretroviral therapy (ART) in routine HIV care, we created a 7-item patient-reported measure (Interference-Score). It assesses seven barrier domains (Thoughts/Feelings, Habits/Activities, Social situation, Economic status, Medication, Care, and Health) and has undergone cognitive testing. Method: To evaluate the measure’s construct validity and predictive capacity, we recruited people with HIV (PWH) on ART at our HIV clinic in Montreal, Canada, from Jan-Aug 2022. Participants completed the measure along with 4 dependent variables (DVs) at baseline (Time 1) and 4 weeks later (Time 2). DVs were self-reported and dichotomized measures of adherence (past 7 days, past 4 weeks), intention to adhere, and viral load. Analyses included: a) inter-item correlations (Spearman’s coefficients) to check item redundancy (Time 1); b) logistic regressions, with one model per DV, to assess the significance of each item (covariate); and c) Receiver operating characteristic (ROC) curve analyses with corresponding areas under curves (AUCs), to determine the 7-item models’ predictive capacity. Analyses b) and c) were performed using I-Score Time 1 items to predict DVs at Time 1 and Time 2, respectively. Results: Analyses included 127 PWH at Time 1 and 107 at Time 2. Correlation coefficients ranged from 0.26 to 0.69. The items (covariates) of “Habits/Activities”, “Health” and “Economic status” were significantly and independently associated with from 1 to 3 DVs. AUCs showed the 7-item models’ predictive capacity to be “excellent” for viral load, correctly classifying ≥84% of respondents. The models were also “acceptable” for adherence in the past 4 weeks and past 7 days, correctly classifying ≥70% of respondents. Conclusions: The 7-item I-Score is a simple, comprehensive tool to evaluate ART adherence barriers in HIV care. It is statistically associated with self-reported adherence and viral load and has useful predictive capabilities. General conditions 1. I confirm that I previewed this abstract and that all information is correct. I accept that the content of this abstract cannot be modified or corrected after the submission deadline and I am aware that it will be published exactly as submitted.: Yes 2. I confirm that the submission of the abstract constitutes my consent to publication (e.g. conference website, programmes, other promotions, etc.). : Yes 3. I herewith confirm that the contact details saved in this system are those of the corresponding author, who will be notified about the status of the abstract. The corresponding author is responsible for informing the other authors about the status of the abstract.: Yes 4. I agree that all data provided may be used (saved, stored, processed, transmitted and deleted) in compliance with the privacy policy to provide the services described.: Yes 1 2 1 3 4 5 2 6 2 6 2 1 2 3 4 5 6 187
RkJQdWJsaXNoZXIy Mzc2ODc=