Factors Associated With Noncompletion of Latent Tuberculosis Infection Treatment: Experience From the PREVENT TB Trial in the United States and Canada

Background. Overall rates of noncompletion of treatment (NCT) for latent tuberculosis infection (LTBI) in the PREVENT TB trial were 18% for 3 months of directly observed once-weekly rifapentine (maximum dose, 900 mg) plus isoniazid (maximum dose, 900 mg) (3HP-DOT) and 31% for 9 months of daily self-administered isoniazid (maximum dose, 300 mg; 9H-SAT). NCT for LTBI reduces its effectiveness. The study objective was to assess factors associated with NCT for LTBI among adult participants enrolled at US and Canadian sites of the PREVENT TB trial. Methods. This was a post hoc exploratory analysis of the randomized, open-label PREVENT TB trial. Factors were analyzed by univariate and multivariate logistic regression (with enrollment site as a random effect). Results. From 6232 participants analyzed, 1406 (22.6%) did not complete LTBI treatment (317 NCT attributed to an adverse event [NCT-AE] and 1089 NCT attributed to reasons other than an adverse event [NCT-O]). The proportion of NCT-AE was similar with both regimens (3HP-DOT= 6.4% vs 9H-SAT= 5.9%; P=.23); NCT-O was higher among participants enrolled in 9H-SAT (9H-SAT= 24.5% vs 3HP-DOT= 12.7%; P=.02). Among those in the NCT-AE group, being non-Hispanic and receiving 3HPDOT, having cirrhosis and receiving 9H-SAT, alcohol consumption among men, and use of concomitant medication were associated with NCT-AE. Among those in the NCT-O group, receiving 9H-SAT, missing >= 1 early visit, men receiving 9H-SAT, men with a history of incarceration, alcohol abuse, use ever of intravenous drugs, younger age receiving 9H-SAT, and smoking were associated with NCT-O. Conclusions. Factors associated with NCT, such as missing a clinic visit early during treatment, might help identify persons for whom tailored interventions could improve completion of LTBI treatment.