To achieve global goals for the treatment of HIV, most high-prevalence countries are experimenting with and scaling up differentiated service delivery models (DSD). A handful of efforts have been formally described and evaluated in the literature; many others are being implemented formally or informally under routine care, without a research or evaluation goal. For most countries, however, there is little evidence on the big picture-the proportion of clinics offering alternative models, eligibility criteria and the proportion of patients considered eligible, the number of patients actually participating, health outcomes such as viral suppression, empirical resource utilization compared to traditional care, variations among the models, duration of patient participation, fidelity to model guidelines, effects on clinic efficiency, and sustainability without external donor support. AMBIT a set of data synthesis, data collection, and data analysis activities aimed at generating information for near- and long-term decision making and creating an approach and platform for ongoing evaluation of differentiated models of HIV treatment delivery in the future. The project will collect and analyze a wide range of existing data sets pertinent to DSD. This protocol is for an analysis of existing medical record data collected by the Ministry of Health, implementing partners, and other completed, ongoing, or new evaluations, trials, and observational studies. Outcomes to be reported include coverage/uptake of DSD, patients' outcomes, and distribution of each model. There will be no study interaction with individual patients, providers, caregivers, or others for this analysis.