CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Conclusion: DMmay play among persons with HIV a role in TB occurrence and presentation similar to that documented in non-HIV infected persons. Increasing prevalence of diabetes among persons with HIV linked to the aging of this population may foster a recrudescence of HIV-associated TB. 735 RISK PROFILES IN MONOINFECTED TB PATIENTS AND THOSE COINFECTED WITH HCV AND/OR HIV David Roth 1 , James Johnston 1 , Jason Wong 1 , Mel Krajden 1 , Zahid A. Butt 2 , Nabin Shrestha 1 , Mark Gilbert 1 , Victoria Cook 1 , Naveed Z. Janjua 1 1 BC Cntr for Disease Control, Vancouver, British Columbia, Canada, 2 Univ of British Columbia, Vancouver, British Columbia, Canada Background: Acquisition risk factors for Hepatitis C Virus (HCV), HIV and tuberculosis (TB) are often shared or may differ between populations. Groups with different infection combinations may represent subpopulations with unique demographic and risk characteristics. Understanding the shared and differing risk factors between populations is key to developing targeted prevention strategies. Methods: We used diagnostic testing, co-infection and risk factor data for active TB cases diagnosed in BC between 1990-2013 in the BC-Hepatitis Testers Cohort (BCHTC). The BCHTC includes ~1.5 million individuals tested for HCV or HIV, or reported as a case of HCV, HIV, HBV, or TB. This cohort is linked with medical visits, hospitalizations, prescription drugs, cancers and deaths. Demographic and acquisition risks were compared between: 1) TB mono infected, 2) TB/HIV coinfected, 3) TB/HCV coinfected, and 4) tri-infected (TB/ HIV/HCV). Coinfection is defined over the clients’ lifespan, and is not necessarily concurrent in time. Results: A total of 5927 (90.0%) individuals were identified with TB only, 144 (2.2%) with TB/HIV, 294 (4.5%) with TB/HCV, and 222 (3.4%) with TB/HIV/HCV. TB mono infected cases were mostly foreign-born (FB) (73.5%) with low injection drug use (IDU) (3.2%), alcohol use (10.1%), and mental illness (12.4%). TB/HIV cases were evenly distributed between Canadian born (CB) (47.9%) and FB (47.9%), and showed moderate IDU (22.2%), alcohol use (27.8%), and mental illness (29.2%). In contrast, those with TB/HCV or TB/HIV/ HCV infection were mostly CB (74.5% and 88.3%, respectively), with high proportions of IDU (TB/HCV: 45.2%, TB/HIV/HCV: 86.5%) and alcohol use (TB/HCV: 59.5%, TB/HIV/HCV: 64.4%). Mental illness was more common in the TB/HIV/HCV (58.1%) than in TB/HCV (33.0%). Approximately 20.3% of TB only cases belonged to the lowest quintile of the social deprivation index, compared to 45.8% of TB/HIV, 48.6% of TB/HCV, and 60.4% of TB/HIV/HCV. Conclusion: Populations with TB only differ from those coinfected with HCV and/or HIV. Findings confirm a commonality of social disparities, mental illnesses and substance use across those with coinfections, especially in those with HCV. The TB/HIV/HCV population is particularly socially deprived with a greater proportion of comorbidities. For those TB/ HIV/HCV, low threshold support services such as harm reduction and mental health interventions may help control both TB transmission and underlying co-morbid conditions. 736 PROBLEM DRUG USE PREDICTS HIGHER HIV PREVALENCE IN UK TUBERCULOSIS CASES, 2010–2014 Joanne R. Winter 1 , Helen R. Stagg 1 , Colette J. Smith 1 , Maeve K. Lalor 2 , Alison E. Brown 2 , H. L. Thomas 2 , Valerie Delpech 2 , Ibrahim Abubakar 1 1 Univ Coll London, London, UK, 2 Pub Hlth England, London, UK Background: HIV co-infection in tuberculosis (TB) patients is associated with poorer outcomes; including higher rates of recurrence, adverse events, treatment interruption or non-completion, and death. Black African ethnicity and being born in sub-Saharan Africa are known predictors of HIV co-infection. Methods: We identified adults aged ≥15 years diagnosed with TB from 2010-2014 in England, Wales and Northern Ireland from national TB surveillance data. HIV status was determined by record linkage to national HIV surveillance data. We calculated the proportion of TB cases diagnosed with HIV and used logistic regression to identify associations between social risk factors (problem drug use, alcohol misuse, imprisonment, homelessness and index of multiple deprivation [IMD] score) and HIV co-infection. Results: 30,064 adults diagnosed with TB were included, 931 (3.1%) were co-infected with HIV. The median age at TB diagnosis was 37 years (inter-quartile range 28-53). Among all TB cases, 4,756 (16.0%) were of black African ethnicity and 14,605 (48.6%) were of Indian/Pakistani/Bangladeshi ethnicity. 6,754 (22.5%) were UK-born and 3,467 (11.5%) were born in countries with high (>1%) HIV prevalence. Among all TB cases, 780 (2.6%) had a history of problem drug use, 855 (2.8%) alcohol misuse, 723 (2.4%) homelessness, and 772 (2.6%) incarceration. 6,298 TB cases (20.1%) were in the lowest IMD decile. In univariable analyses, increased odds of HIV co-infection were associated with drug use (odds ratio [OR] 2.31, 95% confidence interval 1.73-3.08), homelessness (OR 2.51, 1.88-3.35) and imprisonment (OR 1.55, 1.10-2.18), whilst increasing IMD decile was associated with lower odds of co-infection (OR 0.96 per decile increase, 0.93-0.99). Alcohol misuse was not significantly associated with HIV co-infection. In a multivariable model adjusted for age, sex, ethnicity, country of birth and year of diagnosis, drug use remained associated with HIV co-infection (OR 2.44, 1.71-3.49). Conclusion: Problem drug use was associated with a higher prevalence of HIV among TB cases. Greater focus on screening and preventive activities for both TB and HIV may reduce the prevalence of HIV co-infection and improve outcomes for TB patients.

Poster and Themed Discussion Abstracts

737 OPPORTUNISTIC INFECTIONS AMONG US HIV-INFECTED ADULTS IN CARE, 2009–2014 Alejandro Pérez , John Weiser, Heather Bradley CDC, Atlanta, GA, USA

Background: Despite improvements in antiretroviral therapy and rates of viral suppression, opportunistic infections (OIs) remain an important source of morbidity and mortality among people living with HIV. National estimates of the prevalence of OIs and trends over time among HIV-infected persons in care are limited. Addressing these gaps in data may help highlight OIs of special concern as well as guide clinical care. Methods: We analyzed 2009–2014 medical record data from the Medical Monitoring Project (MMP), a surveillance system that uses an annual cross-sectional survey to produce estimates of the characteristics of HIV-infected adults receiving medical care in the US We reviewed medical record data for evidence of a diagnosis within the past 12 months of any of the stage-3-defining OIs according to CDC’s HIV case definition. We estimated the weighted prevalence of diagnosis of any OI by year and assessed temporal change using

CROI 2017 321