CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

regular schedule (80%), after HIV exposure (29%), if they have HIV symptoms (22%), and between new partners (32%); this differed by testing history (Table). Most thought they should test at least annually (88%), including 85% of never testers. Conclusion: Regular testing and perceived HIV exposures were important drivers of HIV testing among MSM, but one-fifth reported never having tested. Messages regarding frequent, regular HIV testing have reached these men but have not necessarily resulted in desired testing behaviors. Strategies for translating knowledge into practice, particularly for never testers, are needed. 902 INCREASES IN HIV TESTING FREQUENCY AMONG MEN WHO HAVE SEX WITH MEN, UNITED STATES Qian An , Riuguang Song, Teresa Finlayson, Cyprian Wejnert, Gabriela Paz-Bailey CDC, Atlanta, GA, USA Background: Since 2006, CDC has recommended annual HIV testing among men who have sex with men (MSM). Knowing the HIV testing frequency and its temporal changes can inform the adherence to HIV testing recommendations. The objective of this analysis is to estimate the mean HIV inter-test interval (ITI) and assess its temporal trends among HIV-negative black and white MSM who reported previous HIV tests. Methods: Using National HIV Behavioral Surveillance System (NHBS) data collected in 2008, 2011, and 2014, we estimated the mean HIV ITI for each year by age among black/ African American (black) and white MSM with a negative NHBS HIV test who reported ever tested for HIV. We used separate statistical models based on renewal process theory by fitting exponential distributions using PROC NLIN with Newton-Raphson method. We compared the mean HIV ITI lengths by assessing whether the 95% confidence intervals overlap. Results: Among 4406 HIV-negative black MSM and 8500 HIV-negative white MSM, 383 (8.7%) black MSM and 558 (6.6%) white MSM reported no previous HIV test. Among MSM who reported the most recent HIV test date (black: n=3945 and white: n=7776), the estimated mean HIV ITI decreased from 2008 to 2014 in each age group for both black (from 9.3 months to 6.5 months) and white (from 10.6 months to 7.7 months, table). There were differences in ITI by age. In 2014, the estimated HIV ITI in months among black MSM was: 5.5 among 18-29 years, 7.6 among 30-39 years, and 9.7 months among 40 years and older. Among white MSM it was: 6.0 among 18-29 years, 7.3 among 30-39 years and 10.5 months among 40 years and older. Conclusion: Black and white MSM in NHBS cities who previously tested for HIV on average adhere to CDC HIV testing recommendations and the average HIV testing frequency has increased since 2008. Young MSM aged 18-29 years had a shorter mean testing interval than other age groups. Young black MSM tested more frequently than young white MSM.

Poster and Themed Discussion Abstracts

903 UNDIAGNOSED HIV AND HCV IN A NEW YORK CITY EMERGENCY ROOM, 2015 Lucia V. Torian 1 , Uriel Felsen 2 , Qiang Xia 3 , Eric Rude 3 , Herbert Rose 4 , Angelica Bocour 3 , Gary J. Williams 5 , Bob F. Bridgforth 5 , Lisa A. Forgione 3 , Barry S. Zingman 4 1 New York City DHMH, New York, NY, USA, 2 Albert Einstein Coll of Med, Bronx, NY, USA, 3 New York City DHMH, Long Island City, NY, USA, 4 Montefiore Med Cntr, Bronx, NY, USA, 5 Quest Diagnostics, San Clemente, CA, USA Background: Undiagnosed HIV and HCV infection are missed opportunities for care and prevention of secondary transmission. CDC estimates that 13% of HIV-infected and 50% of HCV-infected persons nationwide are undiagnosed/unaware of their infection; a blinded Emergency Room (ER) HIV serosurvey in New York City (NYC) in 2010 found 14% undiagnosed/unaware. We sought to measure the prevalence of HIV and HCV, the proportion of undiagnosed/unaware, and the proportion coinfected in persons presenting to a high-volume NYC ER in 2015. Methods: We conducted a blinded cross-sectional serosurvey using remnant serum from specimens originally drawn for clinical indications. Serumwas deduplicated and matched to (1) the hospital’s electronic medical record for demographic and clinical data and (2) the HIV and HCV surveillance registries for evidence of previous diagnosis prior to being de-identified and tested using standard 2- and 3-step clinical testing protocols. Results: Among unique individuals successfully tested for HIV, 250/4990 (5.0%, 95% CI 4.4,5.7) were positive for HIV infection. Among patients tested for HCV, 372/4989 (7.5%, 95% CI 6.7,8.2) were anti-HCV-antibody positive; 196 (3.9%, 95% CI 2.8,5.1) were positive by HCV RNA PCR, indicating current infection. Overall, 12/250 patients testing positive for HIV (4.8%, 95% CI 2.5,8.2) were undiagnosed; 148/372 persons testing positive for anti-HCV antibody (39.8%, 95% CI 34.8,45.0) were undiagnosed; and 38/196 (19.2%, 95% CI 11.4,27.0) patients with detectable HCV RNA were undiagnosed. Among 250 HIV-positive individuals, 246 had sufficient serum for HCV testing; 79 (32.1%; 95% CI 26.3,38.3) were HCV-antibody positive, and 39 (15.7%, 95% CI 7.4,24.0) were currently HCV infected. Among 372 anti-HCV-positive patients, 79 (21.2%, 95% CI 17.2,25.8) were HIV-positive; among 196 HCV-infected patients, 39 (19.4%, 95% CI 12.8,26.0) were HIV-positive. Conclusion: A reduction in the proportion of HIV-infected but undiagnosed/unaware persons presenting to the NYC ER setting between 2010 and 2015 was coterminous with legislation, funding and aggressive programming to increase HIV testing and diagnosis in NYC. Undiagnosed HCV was high, suggesting that initiatives similar to those directed toward HIV should be mounted to improve HCV diagnosis and linkage to care and treatment. 904 EXAMINATION OF UNRECOGNIZED AND MISREPORTED HIV STATUS IN BALTIMORE MSM AND PWID Danielle German 1 , Kate Shearer 1 , Colin Flynn 2 , Carl A. Latkin 1 , Oliver Laeyendecker 3 , Thomas Quinn 1 , William Clarke 1 1 The Johns Hopkins Univ, Baltimore, MD, USA, 2 Maryland DHMH, Baltimore, MD, USA, 3 NIAID, Baltimore, MD, USA Background: Identifying individuals with unrecognized HIV infection is critical to deploying HIV testing, prevention, and surveillance resources. This measure is most commonly assessed using self-reported data about prior HIV testing and results. The accuracy of this measure can vary depending on the survey. In a previous study (2008), 33% of men

CROI 2017 392