Background: Although timely tuberculosis (TB) diagnosis and treatment are essential to TB elimination, care delays remain a challenge. Understanding factors associated with delayed TB diagnosis and treatment may inform interventions. Methods: The TB Sentinel Research Network of the International epidemiology Databases to Evaluate AIDS (IeDEA) is a prospective study of people aged ≥15 years with pulmonary TB. At sites in Eldoret, Kenya, and Mbarara, Uganda, questionnaires ascertained timing of recalled TB symptom onset, number/setting of healthcare visits prior to TB diagnosis, symptoms, demographics, food insecurity, and TB- and HIV-related stigma. Robust negative binomial regression was used to examine factors associated with longer overall duration from TB symptom onset to documented treatment initiation. Results: Among 264 participants—median age 33 years (IQR, 25–43 years), 67% male, 31% people with HIV, 51% urban residents—median duration from TB symptom onset to treatment initiation was 65.5 days (IQR, 33–131 days). Longer time to TB treatment was associated with female sex (adjusted incidence rate ratio [aIRR], 1.27 [95% CI, 1.00–1.61]); previous care-seeking at a pharmacy (aIRR, 1.53 [95% CI, 1.22–1.92]), and presence of dyspnea (aIRR, 1.43 [95% CI, 1.09–1.86]) or fatigue (aIRR, 1.63 [95% CI, 1.11–2.37]) at initiation. Shorter duration to TB treatment was associated with HIV-positive status (aIRR, 0.50 [95% CI, .39–.64]). Among people with HIV, shorter duration to TB treatment was associated with CD4 count <200 cells/μL (aIRR, 0.63 [95% CI, .43–.93]). Conclusions: Interventions are needed to increase community-level TB diagnostic access, including at pharmacies and lower levels of care. Decentralized TB testing strategies and tailored interventions for groups with lower care access should be pursued.