Outcomes of the three-month weekly isoniazid with rifapentine (3HP) versus the six-month isoniazid preventive therapy (6H) among people newly enrolled in HIV care in western Kenya

BackgroundIn trials, three-month weekly rifapentine and isoniazid (3HP) showed higher adherence and completion than the six-month daily isoniazid (6H) regimen for TB preventive treatment (TPT). However, programmatic outcome data remain limited. MethodologyWe evaluated the TPT cascade among people with HIV (PWH) aged >15 years newly enrolled in HIV care in western Kenya. Initiation and completion of 6H (Jan to Sept 2022) were compared to 3HP (Oct 2022-Sept 2023) using Chi-square tests. Correlates of non-initiation and non-completion were assessed using Poisson regression with generalized linear models. Mortality within 24 months was evaluated using Cox proportional hazards regression. ResultsOf 1,930 PWH (median age 33 years [IQR=27-41]), 65.8% were female, and 19.5% had AHD at enrolment. Overall, 1,922 (99.6%) were screened for active TB, of whom 1,790 (97.5%) were TPT-eligible; 1577 (88.1%) of these initiated TPT. TPT initiation was higher with 3HP than 6H (89.8% vs. 84.2%; p<0.001). TPT completion was similar for 3HP and 6H (89.2% vs. 88.8% p=0.77). TB incidence (per 1,000 person-months) was lower among TPT-completers (0.22; 95% CI 0.15-0.35) than those who neither initiated (4.25; 95% CI 1.77-10.23) nor completed TPT (3.75; 95% CI 2.49-5.64). AHD was associated with higher risk of TPT non-initiation (aRR=2.14; 95% CI 1.59-2.87) and non-completion of both 6H (aRR=2.56; 95% CI: 1.55-4.23) and 3HP (aRR=1.68; 95% CI 1.07-2.63). Conclusions3HP increased TPT initiation but did not improve completion rates compared to 6H. Targeted interventions are needed to support 3HP completion, particularly in PWH with AHD Key pointsWe compared 3HP and 6H for TB prevention in people with HIV in western Kenya. 3HP led to better initiation and both had high completion rates. Advanced HIV disease affected participation and non-completers faced significant mortality.