As HIV treatment is scaled-up in resource-poor settings, the timely identification of persons with HIV infection remains an important challenge. Most people with HIV are unaware of their status, and those who are often present late in the course of their illness. Free-standing voluntary counseling and testing sites often have poor uptake of testing. We aimed to evaluate a 'providerinitiated' HIV testing strategy in a primary care clinic in rural resource-poor Haiti by reviewing the number of visits made to clinic before an HIV test was performed in those who were ultimately found to have HIV infection. In collaboration with the Haitian Ministry of Health, a nongovernmental organization (Partners In Health) scaled up HIV care in central Haiti by reinforcing primary care clinics, instituting provider-initiated HIV testing and by providing HIV treatment in the context of primary medical care, free of charge to patients. Among a cohort of people with HIV infection, we assessed retrospectively for delays in or 'missed opportunities' for diagnosis of HIV by the providers in one clinic. Of the first 117 patients diagnosed with HIV in one clinic, 100 (85%) were diagnosed at the first medical encounter. Median delay in diagnosis for the remaining 17 was only 62 days (IQR 19 – 122; range 1 – 272). There was no statistical difference in CD4 cell count between those with and without a delay. 3787 HIV tests were performed in the period reviewed. Provider-initiated testing was associated with high volume uptake of HIV testing and minimal delay between first medical encounter and diagnosis of HIV infection. In scale up of HIV care, providerinitiated HIV testing at primary care clinics can be a successful strategy to identify patients with HIV infectio