Background: Isoniazid preventive therapy (IPT) prevents active TB in people living with HIV (PLHIV), but prior studies found no evidence of benefit in PLHIV with a negative tuberculin skin test (TST) and non-significant impact on mortality. We conducted an individual participant data meta-analysis of randomized controlled 56 trials to estimate the effect of IPT given with ART to prevent TB and death among PLHIV across population subgroups. Methods: We searched PubMed, Embase, the Cochrane database, and conference abstracts. Eligible studies included trials of HIV-positive adults taking ART randomized to daily IPT versus no IPT. We performed a single-stage individual participant data meta-analysis of the outcomes of incident TB disease and all- cause mortality using stratified Cox-proportional hazards models. Registration: PROSPERO (CRD42019121400). Findings: We found 838 citations and included 2611 participants from three trials in Cote D’Ivoire, Malawi, and South Africa. IPT with ART was associated with lower risk for TB than ART alone (hazard ratio=0·68, 95% CI 0·49–0·95, p=0·02) and non-significantly lower all -cause mortality (hazard ratio=0·69, 95% CI 0·43–1·10, p=0.12). TB risk differed by baseline CD4 <200 cells/mm3, but there was no evidence of varying benefit of IPT with ART by sex, baseline CD4, or results of TST or interferon gamma release assays. Elevated alanine aminotransferase occurred in 65/2611 (2.5%) of participants, but data were insufficient to calculate a hazard ratio. Interpretation IPT with ART prevents TB across demographic and HIV- and TB-specific subgroups, which supports efforts to further increase use of IPT with ART broadly among PLHIV.