ABSTRACT: Hospitalization of HIV-infected adults in Africa is often the last opportunity to initiate and maintain life-saving HIV care. Despite the roll out of ART, HIV remains the most common cause of admission to medical wards in Africa. Our data in Tanzania and other data from Africa indicate that 25-40% of hospitalized HIV- infected adults will die within 1 year of hospital discharge. This high post-hospital mortality is strongly associated with failure to link to primary HIV care after hospital discharge. In collaboration with our Tanzanian partners, we have conducted formative research and have developed a social worker intervention to address this critical gap between hospital and HIV clinic. Utilizing the Gelberg and Andersen Behavioral Model of Health Services Utilization for Vulnerable Population, we identified 8 factors that are associated with poor post-hospital clinic linkage including: unemployment, traditional health beliefs, low self-efficacy, lack of transportation, lack of social support, stigma, low perceived need for HIV care, and physical weakness. Based on these factors, we adapted an evidence-based social worker intervention (ARTAS) which has been effective in improving linkage in the United States. A pilot of the social worker intervention in 31 HIV-infected patients at Bugando Medical Center in Mwanza, Tanzania showed that linkage improved from 63% to 100% and survival from 75% to 90% when compared to historical controls. Based on these pilot data, we propose a randomized clinical trial at the public Bugando Medical Center (BMC) and its associated HIV clinic in Mwanza, Tanzania. Our primary aim is to evaluate the efficacy of the social worker intervention to increase the 1-year survival rate in 500 HIV infected adults consecutively discharged from the Bugando Medical Center. We hypothesize that the one-year survival will be 90% in the intervention group vs. 75% in the standard care group. We will have >80% power to detect this 15% absolute difference in survival at p<0.05. Secondary aims will include: 1) to compare linkage to the HIV clinic and subsequent steps in the HIV care continuum, 2) to validate the Gelberg and Andersen Model, 3) to determine the acceptability of the intervention, and 4) to calculate the incremental cost and cost per life saved. The goal of this proposal is to develop a model of care for hospitalized HIV-infected adults to improve their linkage to outpatient care and post-hospital survival. Improving outcomes for HIV-infected adults could save several hundred thousand lives across Africa each year. After the successful completion of this trial, we will conduct implementation studies with our partners in the Tanzanian Ministry of Health to demonstrate scalability and cost-effectiveness.