CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

824 HIGH HIV BURDEN AMONG CHILDREN IN LESOTHO: FINDINGS FROM A POPULATION BASED SURVEY Koen Frederix 1 , Stefania Davia 2 , Keletso Ntene-Sealiete 3 , Stephen M. Arpadi 4 , Stephen McCracken 2 , Kyaw Thin 3 , Hetal Patel 2 , Mosilinyane Letsie 3 , Andrea Low 4 1 ICAP at Columbia University–Lesotho, Maseru, Lesotho, 2 CDC, Atlanta, GA, USA, 3 Ministry of Health, Maseru, Lesotho, 4 ICAP at Columbia University, New York, NY, USA Background: Lesotho has the second highest HIV prevalence in the world. In 2013 the country became one of the first African countries to implement Option B+ nationwide. However Lesotho does not have reliable data on HIV prevalence in children, previously relying on estimates modeled from program data. The 2016 Lesotho Population-based HIV Impact Assessment (LePHIA) aimed, among other objectives, to determine HIV prevalence among children 0-14 years in order to assess the efficacy of the prevention of mother to child transmission (PMTCT) program and guide future policy. Methods: A nationally representative sample of children under 15 years underwent household-based, two-stage rapid HIV testing from November 2016-May 2017. Children <18 months with a reactive screening test were tested for HIV infection using DNA PCR. Parents or legal guardians provided information on children’s clinical history. Children aged 10-14 years also answered a questionnaire which included socio-demographic and behavioral questions. National weighted pediatric prevalence was estimated accounting for complex survey design, with Jackknife replication to calculate 95% confidence intervals (CI). We used bivariate statistics to compare characteristics across demographic groups. Results: A total of 3,966 children were tested for HIV, and 1,601 10-14 year olds completed interviews. Overall, HIV prevalence was 2.1% (95% CI: 1.5-2.6%), corresponding to roughly 13,300 children living with HIV. Prevalence varied across districts, with the highest prevalence in Mokhotlong (4.8%, 95% CI: 2.2- 7.3%), and the lowest in Berea (0.5%; 95%CI: 0-1.2%). Prevalence in 10-14 year olds (3.2%; 95% CI: 2.1%, 4.2%) was significantly greater compared to 0-4 year olds (1.0%; 95% CI: 0.5%, 1.6%). Sexual activity was reported in 9.3% (95%CI: 7.3-11.3%) of 10-14 year olds but there was no association with HIV infection. Overall HIV prevalence in females and males was 2.6% (95% CI: 1.8% – 3.3%) and 1.5% (95% CI: 1.0% – 2.1%), respectively. Conclusion: Substantial progress has been made in the reduction of vertical transmission in Lesotho. But despite the early roll-out of Option B+ in Lesotho, pediatric HIV prevalence remains high, with girls disproportionately affected. Further research is required to understand the greater prevalence among girls, barriers to PMTCT and the possible contribution of horizontal transmission in older children. 825 IMPACT OF BIRTH TESTING ON EARLY INFANT DIAGNOSIS IN KWAZULU- NATAL, SOUTH AFRICA Shayla Smith 1 , Louise Kuhn 1 , Moherndran Archary 2 , Pravi Moodley 2 , Phillip LaRussa 1 , Kerusha Govender 2 1 Columbia University Medical Center, New York, NY, USA, 2 University of KwaZulu- Natal, Durban, South Africa Background: In 2015, early infant diagnosis (EID) guidelines in South Africa shifted to recommending birth HIV PCR testing and a follow up test at 10 weeks of age. Prior to this, initial PCR testing was recommended at 6 weeks of age. Here we examine parameters of EID performance in the KwaZulu-Natal Province before and after this change. Methods: Data on all HIV diagnostic PCR tests conducted by the National Health Laboratory Service for the province between Jan 2013 and April 2016 were assembled. Tests undertaken on birth cohorts of children born in 2013, 2014 and 2015 were analyzed. Laboratory barcodes allowed identification of repeat tests on the same child. We evaluated coverage, positivity rates, age at testing and frequency of repeat tests across birth cohorts. Results: The total number of PCR tests and (total number of unique children PCR-tested) increased from 92,226 (78,453) in birth cohort 2013 and 89227

(75,667) birth cohort 2014 to 132,645 (112,533) in birth cohort 2015. The number of unique children PCR-tested <6 weeks of age increased from 67,681 in 2013 and 61,851 in 2014 to 103,298 in birth cohort 2015. Based on numbers of registered births and an assumed 40% HIV prevalence in the Province, these numbers were 79.4% and 73.9% of the estimated HIV-exposed births in 2013 and 2014, respectively, and exceeded the estimated number of HIV-exposed births in 2015. The proportion of positive tests decreased from 3.08% in 2013 and 2014 to 1.81% in 2015. In birth cohorts 2013 and 2014, 62.1% and 61.8%, respectively, of tests <16 weeks were done in children 6 to 8 weeks of age. In birth cohort 2015, 41.3% of tests <16 weeks were done earlier at less than 2 weeks of age. The percentage of children with a positive result who had at least one follow up test increased from 11.5% and 13.1% in birth cohorts 2013 and 2014 to 24.8% in 2015. The percentage of non-positive infants who received at least one follow up test did not appreciably change from 15.0% and 14.4% in 2013 and 2014 to 14.7% in 2015. Conclusion: Shifts to recommending birth testing has led to greater coverage of HIV-exposed infants and earlier PCR testing. This allows for earlier identification of HIV-infected infants who urgent ART initiation. Although follow-up testing rates may be under-estimated in this data source, repeat testing rates remained consistently low. More effort is needed to ensure infants tested at birth continue to be engaged in care and undergo follow-up testing.

Poster Abstracts

826 AGE AT HIV DIAGNOSIS WITHIN SOUTH AFRICA’S EARLY INFANT DIAGNOSIS PROGRAM, 2010-2015 Ahmad Haeri Mazanderani 1 , William B. MacLeod 2 , Jacob Bor 2 , Gayle G. Sherman 3 1 National Institute for Communicable Diseases, Johannesburg, South Africa, 2 Boston University, Boston, MA, USA, 3 University of the Witwatersrand, Johannesburg, South Africa Background: Among infants aged <2 months, South Africa’s National Health Laboratory Service’s (NHLS) data warehouse has provided effective monitoring of early infant diagnosis (EID) within the national prevention of mother-to-child transmission (PMTCT) program demonstrating improved EID coverage (54%- >85%) alongside decreasing early MTCT rates (4.3%-1.5%) between 2010-2015. However, only a third of all positive HIV PCR results occurred in <2 month olds. For older infants, the number of positive HIV PCR tests could not be used to quantify the number of HIV-infected children since the absence of a unique patient identifier precluded de-duplication. Older infants may test PCR positive due to late presentation for testing, postnatal infection or failure to link to care. We describe age at HIV diagnosis in South Africa’s EID program for the first time. Methods: HIV PCR data from 2010-2015, including patient demographics, year of testing and PCR result, were extracted from the NHLS’ data warehouse. A patient linking-algorithm using probabilistic matching of date of birth, names, file number and facility was applied to assign multiple tests to a single patient. ‘Age at HIV diagnosis’ was defined as age at first positive PCR; ‘early diagnosis’

CROI 2018 311