<sec><title>BACKGROUND</title>WHO guidance to defer isoniazid preventive therapy (IPT) among those with regular alcohol use because of hepatotoxicity concerns may exclude many people living with HIV (PLWH) at high TB risk in these settings.</sec><sec><title>OBJECTIVE</title>To evaluate hepatotoxicity during TB preventive therapy (TPT) in PLWH who report alcohol use in Uganda over 10 years.</sec><sec><title>METHODS</title>We developed a Markov model of latent TB infection, isoniazid preventive therapy (IPT - a type of TPT), and TB disease using data from the Alcohol Drinkers' Exposure to Preventive Therapy for TB (ADEPTT) study. We modeled several treatment scenarios, including no IPT, IPT with liver enzyme monitoring (AST/ALT) during treatment, and IPT with pre-screening using the tuberculin skin test (TST).</sec><sec><title>RESULTS</title>The no IPT scenario had 230 TB deaths/100,000 population over 10 years, which is more than that seen in any IPT scenario. IPT, even with no monitoring, was preferred over no IPT when population TB disease incidence was >50 in 100,000.</sec><sec><title>CONCLUSIONS</title>For PLWH who report alcohol use in high TB burden settings, IPT should be offered, ideally with regular AST/ALT monitoring. However, even if regular monitoring is not possible, IPT is still preferable to no IPT in almost every modeled scenario.</sec>.