Impact of Patient Navigators in Linkage to Care for HIV-Positive Patients in an Urban Emergency Department in Newark, NJ

Abstract

Abstract Background Despite CDC recommendations, areas with high HIV prevalence have not implemented routine HIV testing, stating among other concerns, inability to effectively link them to care. We implemented a routine HIV testing program in the Emergency Department (ED) at University Hospital in Newark, NJ that had 46,164 visits from July 2015 to November 2016 and looked at the impact of patient navigators (PN) on linkage to care (LTC) rates. Methods This was a retrospective study of all patients newly diagnosed (ND) with HIV or previously positive (PP) but lost to follow-up (LTFU) in select areas of the ED from July 2015 to November 2016. We collected information on demographics, HIV risk factor, and looked at the impact of PN on LTC by comparing months the PN was able to make personal contact compared with months when the PN was unavailable for substantial periods of time. Results A total of 9,511 individuals were screened, and 151 (1.6%) had a positive HIV test; 8 died and 2 were incarcerated. Of the remaining 141, 102 (72%) were LTC. The mean age was 49, 57% Male, 77% Black, 14% Hispanic, and 6% White. The reported HIV risk factors were 67% Heterosexual, 9% MSM, 6% IV drug use (IDU) and 18% Other. Of the patients with a positive HIV test, 60 (43%) were ND and 81 (57%) were PP. Only 52% ND patients were LTC, while 88% PP patients were LTC. Black and Hispanic patients tended to be PP (60% of both groups), while White patients tended to be ND (75% of white patients were ND). The risk factors for ND were 44% Heterosexual, 39% MSM, and 25% IDU. Average LTC while the PN was unavailable decreased from 78% to 56%. There were no demographic differences in the LTC group compared with the LTFU group. IDU had the highest rate of being LTFU at 37% followed by MSM and Heterosexual at approximately 23% each. The primary reason for LTFU was incorrect contact information in the medical record such as wrong address or phone number. PN would make 3 phone calls, send 2 letters and 1 outreach attempt. If all of those failed, the PN notified the state health department. Conclusion PN have a positive impact on LTC even in busy ED settings. Given limitations of staffing a busy ED 24/7, we need to develop strategies to link patients even if the PN is not present. To address this limitation, we plan on looking at the impact of involving medical residents to help with linkage to care after business hours. Disclosures S. Swaminathan, Gilead Sciences: Grant Investigator and Scientific Advisor, Consulting fee and Research grant