Because people with undiagnosed HIV will not receive treatment and may unknowingly infect others, the investigators propose a mobile computer-based video intervention to increase HIV test rates in high volume urban hospital emergency departments (EDs). EDs offer important points of contact for many of those at greatest risk for HIV. Unfortunately, when ED patients are offered routine HIV testing, most decline. Our proposed intervention builds upon initial findings from a trial our research team conducted with patients who declined HIV testing. The intervention, grounded in the Information-Motivation and Behavioral Skills model (IMB), showed an onscreen physician explaining the importance of HIV testing (to build knowledge and motivation) and modeling a rapid HIV test (to increase motivation and behavioral skill). This brief intervention had a potent effect: a third accepted HIV testing post-intervention. While this preliminary study is highly encouraging, it revealed a number of other critical research questions. First, it remains unclear what intervention component most strongly contributed to patients' decisions to test: the video content or the offer of an HIV test by a computer rather than a person. Second, consistent with the literature, participants indicated a community member disclosing positive HIV status onscreen would increase the proportion of patients who test. Third, results suggest there is individual variation in the extent to which behavior is more strongly influenced by onscreen community members or experts (e.g. physicians). Therefore, the goal of the present study, guided by the IMB model, is to determine how the investigators can refine mobile computer-based interventions to maximize HIV testing rates among patients who initially decline to test in the ED. At the end of the computerized intervention, onscreen text will ask patients if they would agree to an HIV test. Those who agree will be tested by ED staff. The study's endpoint will be post-intervention HIV test rates. The investigators' study will inform scalable interventions for underserved populations nationwide.

Inclusion Criteria:

• Age 18 or over
• Capable of providing informed consent
• Reads English
• Declined HIV test offered by hospital staf at triage

Exclusion Criteria:

• intoxicated
• a prisoner
• known to be HIV positive
• presenting to the hospital for a psychiatric problem