CROI 2019 Abstract eBook

Abstract eBook

Oral Abstracts

147 OPIOID OVERDOSE DEATHS AMONG PERSONS WITH HIV INFECTION, UNITED STATES, 2011-2015 Karin A. Bosh 1 , Nicole Crepaz 1 , Xueyuan Dong 2 , Sheryl Lyss 1 , Maria Mendoza 1 , Andrew J. Mitsch 1 1 CDC, Atlanta, GA, USA, 2 ICF International, Atlanta, GA, USA Background: The opioid epidemic is a nationwide public health emergency. Persons with HIV might be at increased risk for drug overdose deaths, including overdoses involving an opioid. We examined characteristics of unintentional drug overdose deaths involving an opioid (hereafter, opioid overdose deaths) during 2011-2015 among persons with diagnosed HIV infection in the United States. Methods: We used National HIV Surveillance System data reported through December 2017 to summarize opioid overdose deaths between 2011 and 2015 among persons with diagnosed HIV in the 50 states and District of Columbia. Opioid overdose deaths were selected by using the International Classification of Disease, Tenth Revision (ICD-10). Death rates were calculated per 100,000 persons with diagnosed HIV. We examined death rates by demographic, geographic, and HIV transmission categories. Results: There were 1,363 opioid overdose deaths among persons with diagnosed HIV during 2011-2015. Although the rate of all deaths among persons with diagnosed HIV was 12.7% less in 2015 (1630.6 per 100,000) than in 2011 (1,868.8 per 100,000), the opioid overdose death rate among persons with diagnosed HIV was 42.7% greater in 2015 (33.1 per 100,000) than in 2011 (23.2 per 100,000). Rates of opioid overdose deaths were higher in 2015 than 2011 for all subgroups examined by age, sex, race/ethnicity, transmission category, and US Census region of residence at death, with the exception of the West US Census region. In 2015, the rate of opioid overdose deaths was highest among persons aged 50–59 years at death (41.9 per 100,000), females (35.2 per 100,000), whites (49.1 per 100,000), persons who inject drugs (137.4 per 100,000), and the Northeast US Census region (60.6 per 100,000), compared to their respective counterparts. Conclusion: Opioid overdose death rates were higher in 2015 than in 2011 among nearly all demographic, transmission, and geographic categories examined despite the decreased rate of total deaths among persons with diagnosed HIV during 2011–2015. Differences in opioid overdose deaths among subgroups of persons with diagnosed HIV call for targeted prevention efforts. Intensified overdose prevention is needed for achieving optimal care of persons with diagnosed HIV and to further decrease mortality. 148 EARLY MORTALITY IN HIV-INFECTED PATIENTS INITIATING ART WITHOUT A PRETHERAPY CD4 Kombatende Sikombe 1 , Ingrid Eshun-Wilson 2 , Aybuke Koyuncu 1 , Sandra Simbeza 1 , Aaloke Mody 2 , Nancy Czaicki 2 , Laura K. Beres 3 , Carolyn Bolton Moore 4 , Nancy Padian 5 , Izukanji Sikazwe 1 , Charles B. Holmes 6 , Elvin Geng 2 1 Centre for Infectious Disease Research in Zambia, Lusaka, Zambia, 2 University of California San Francisco, San Francisco, CA, USA, 3 Johns Hopkins University, Baltimore, MD, USA, 4 University of Alabama at Birmingham, Birmingham, AL, USA, 5 University of California Berkeley, Berkeley, CA, USA, 6 Georgetown University, Washington, DC, USA Background: In the treat-all era, CD4 levels are no longer required to determine treatment eligibility, resulting in some programs phasing out CD4 tests altogether. Pre-therapy CD4, however, can play a crucial role in informing screening and prophylaxis for opportunistic infections, which are contributors to HIV-related mortality. We assessed the association between presence of a pre-therapy CD4 and early mortality among patients in Zambia starting ART. Methods: We evaluated patients starting ART between August 1, 2013 and July 31, 2015 in Zambia. We obtained pre-therapy CD4 (most recent determination within 6 months of treatment initiation), socio-demographic and clinical data from the electronic medical record. We identified a probability sample of patients lost to follow-up for intensive tracing to determine vital status. Findings from tracing were incorporated into Kaplan-Meier estimates and multivariate proportional hazards regression through inverse probability- weights. Estimates were adjusted for potential common causes of CD4 determination and survival (e.g. WHO stage, calendar time, facility type, etc.). Results: Of 39,556 patients starting ART (63%women, median age 35.64 (IQR 29.88 – 42.41)), 31,895 (76%) had a pre-therapy CD4 on record (median CD4 270 cells/μl (IQR 145-396)). The cumulative incidence of mortality after ART

Oral Abstracts

146 MORTALITY REDUCTION IN WESTERN KENYA DURING SCALE-UP OF HIV TREATMENT, 2011-2016 Martien W. Borgdorff 1 , George O. Otieno 2 , Y. O. Whiteside 3 , Thomas Achia 4 , Daniel Kwaro 2 , Sylvia Ojoo 5 , Maquins Sewe 2 , Paul K. Musingila 6 , Victor Akelo 6 , David Obor 2 , Amek Nyaguara 2 , Kevin M. De Cock 4 , Emily C. Zielinski- Gutierrez 3 1 University of Amsterdam, Amsterdam, Netherlands, 2 Kenya Medical Research Institute, Kisumu, Kenya, 3 CDC, Atlanta, GA, USA, 4 US CDC Nairobi, Nairobi, Kenya, 5 University of Maryland, Baltimore, MD, USA, 6 US CDC Kisumu, Kisumu, Kenya Background: In the early years after ART introduction in Africa, there were marked declines in annual mortality, with reductions of 10-20% observed in various settings. There is limited information on the impact of the current rapidly expanding HIV treatment access on general population mortality in sub-Saharan Africa. Methods: From 2011 to 2016, ART coverage in western Kenya increased from 34% to 60%. Data from a health and demographic surveillance system (HDSS) measured mortality and migration for the period; HIV home-based counselling and testing (HBCT) surveys took place in 2011, 2012, 2013, and 2016. Mortality trends were assessed in a closed cohort of residents. Results: Seventy percent of HDSS residents in Gem, western Kenya, (22,688/32,467, aged 15-64 years) participated in the 2011 survey and comprised the cohort followed over time. All-cause mortality was 10.0 (95% confidence interval (CI) 8.4-11.7) per 1000 person-years (PY) in 2011, and declined to 7.5 (95% CI 5.8-9.1) per 1000 PY in 2016. Mortality was stable over the study period, at 5.7 per 1,000 PY among the non-HIV infected. Among HIV- infected persons, mortality declined from 30.5 per 1000 PY in 2011 to 15.9 per 1000 PY in 2016 (average decline 6% per year). Individuals on ART experienced higher mortality rates than non-HIV-infected individuals (rate ratio 2.8, 95% CI 2.2-3.4). Conclusion: This study suggests mortality among HIV infected individuals declined substantially during ART expansion between 2011 and 2016, though less than the declines reported during early ART introduction. Mortality trends among HIV positive persons are critical to understanding epidemic dynamics. As ART use continues to expand, HDSS platforms offer a unique opportunity to monitor mortality alongside trends in HIV prevalence and incidence.

55

CROI 2019