Tuberculosis (TB), the leading cause of death in people living with HIV, is the archetypal airborne disease, spread person-to-person through the inhalation of aerosolized bacilli. In endemic regions, most TB disease results from recently acquired infection with M. tuberculosis (Mtb), and studies support that a minority of individuals are responsible for most TB transmission events. Although focusing TB control interventions, such as TB preventive treatment (TPT), on contacts of these highly infectious individuals is cost-effective, tests that accurately determine the infectiousness of patients with TB are lacking. The overall objective of this proposal is to prevent TB-related morbidity and mortality among PLHIV by developing a novel cough-based diagnostic of TB infectiousness. We previously developed a research tool to measure Mtb cough aerosolization, the cough aerosol sampling system (CASS), that predicts TB infectiousness. Recently, we investigated cough frequency as a marker of TB infectiousness, identified unique TB-related cough signatures and developed a mobile privacy-preserving cough detection system. In the proposed study, we will enroll HIV-positive and HIV-negative Kenyan adults who are newly diagnosed with pulmonary TB to assess their level of TB infectiousness using CASS-measured aerosolized Mtb counts (the outcome). In Aim 1, we will investigate cough frequency and/or cough spectrogram (collectively “cough signature”) to estimate TB infectiousness and investigate the impact of HIV co-infection on TB infectiousness. In Aim 2, we will identify the household contacts of Aim 1 participants (index cases) to investigate for latent TB infection using interferon-gamma release assays (IGRAs) and determine whether index case cough signatures associated with high Mtb aerosol counts are associated with a greater frequency of positive IGRAs in household contacts. Additionally, at 6 months after enrollment we will re-test household contacts with IGRAs to understand if the durability of positive responses is associated with index case cough signatures. We hypothesize that individuals with high Mtb aerosol counts will have high cough frequencies and unique cough spectrograms, and that cough spectrograms will differ by HIV status despite similar Mtb aerosol counts. Furthermore, we hypothesize that household contacts to index cases with “highly infectious” cough signatures are more likely to have positive IGRAs at baseline and at 6-month follow- up. Our team has developed the tools to not only robustly determine infectiousness (CASS), but to evaluate both cough frequency and intensity (spectrogram analyses) in a high HIV/TB burden setting. By identifying cough signatures predictive of TB infectiousness, we will develop tools for clinicians and public health programs to support the targeting of TB control interventions.