Background: On a given night around 40,000 Veterans are homeless and another 300,000 are at risk. Many of these Veterans have complex health and social needs, and they experience frequent residential transitions which can disrupt access to healthcare and treatment adherence. Many persons who are housing unstable experience frequent residential transitions, such as from transitional housing to shelter, or from doubled up with family/friend to living out of a car. Disruptions caused by such transitions likely contribute to this population's poor health by interfering with access to care and treatment adherence. Despite the potential harm, there is a major gap in detailed, contextual knowledge of these transitions, especially from the perspective of the individuals experiencing them – such as their in-the-moment emotions, behaviors, geographic movements, and social support. Filling this gap would improve knowledge of Veterans' trajectories into and out of homelessness and their day-to-day barriers to health care and other services. New approaches are needed. Research has shown that most persons experiencing homelessness have mobile phones, and increasingly they are smartphones. This provides an opportunity to gather near real-time information, at relatively low cost, that would help improve understanding of Veterans' changes in housing, health, mood, and use of services. Significance/Impact: Homelessness among Veterans represents one of the worst failures of our national and VA health care and social service systems. Health is poor, and life expectancy for homeless persons is 10 to 20 years shorter than for housed populations. The work proposed here will contribute to VA priorities of improving access to care, increasing virtual care/telehealth, and improving the health of homeless Veterans. Innovation: We propose the first significant test of passive and active mobile phone data collection among homeless Veterans, including the use of global positioning system (GPS) location and ecological momentary assessment (EMA) to improve understanding of context, mobility, and distance to services. These methods can change how health services researchers think about collecting data from marginalized and hidden populations. These techniques can identify sequences of micro-temporal events, for example teasing apart the events and experiences that immediately precede (and follow) the transition from one housing type to the next, or the events that led to a missed healthcare opportunity (e.g. a missed appointment). Specific Aims: This study assesses the feasibility of smartphone data collection from homeless Veterans. Aim 1: Characterize the real-time lived experience of homeless Veterans, including day-to-day activities, interactions with services, and residential transitions using a formative ethnographic approach; Aim 2: Refine and tailor smartphone data collection methods; Aim 3: Conduct a 4-week demonstration of smartphone data collection to evaluate its acc...