HYNES CONVENTION CENTER

Boston, Massachusetts
March 4–7, 2018

 

Conference Dates and Location: 
February 22–25, 2016 | Boston, Massachusetts
Abstract Number: 
1014

Prospective Multisite Cohort Study of Pre-ART Losses and ART Refusal in South Africa

Author(s): 

Ingrid T. Katz1; Janan Dietrich2; Laura M. Bogart3; Dominick Leone4; Ingrid Courtney5; Gugu Tshabalala2; Garrett M. Fitzmaurice1; Glenda Gray6; Catherine Orrell7; David R. Bangsberg8
1Brigham and Women's Hosp, Harvard Med Sch, Boston, MA, USA;2Perinatal HIV Rsr Unit, Soweto, South Africa;3Boston Children's Hosp, Harvard Med Sch, Boston, MA, USA;4Ragon Inst of MGH, MIT and Harvard, Cambridge, MA, USA;5Desmond Tutu HIV Fndn, Cape Town, South Africa;6South African Med Rsr Council, Cape Town, South Africa;7Univ of Cape Town, Cape Town, South Africa;8Harvard Med Sch, Boston, MA, USA

Abstract Body: 

Failure to initiate antiretroviral therapy (ART) is a significant barrier to population-level viral suppression. We quantified losses in pre-ART care from presentation for voluntary counseling and testing (VCT) through 3 months post-VCT among adults at two urban testing centers in South Africa (RSA). In addition, we measured and identified factors associated with treatment refusal at baseline.  

We prospectively surveyed HIV-infected ART-eligible (CD4<500 cells/mm3) adults who presented for VCT between July 2014-July 2015 in Soweto and Gugulethu over a 3 month period post-VCT. Trained interviewers administered a structured questionnaire to participants to understand psychosocial, clinical and structural factors associated with ART refusal at baseline. Bivariate analyses were performed to identify factors associated with ART refusal. All covariates with p<0.2 were included in a multivariable model. The model was further stratified by social support based on our prior qualitative findings.

Across sites, 1071 individuals presented for testing and were found to be HIV-infected and ART-eligible over the study period. Of those, 360 (33.6%) did not return to collect their CD4 results within 6 weeks of testing (Median CD4: 194 cells/mm3). Among those who returned, 500 (70.3%) were enrolled in the prospective cohort. Median age at testing was 35 years-old, of whom 62.6% were women, 57.2% were unemployed, and 75.6% had not completed high school. Median CD4 was 244 cells/mm3. We found a 6.6% (n=33) refusal rate at baseline. Multivariable analyses showed fatalism was associated with ART refusal at baseline (AOR: 1.16, p=0.02). Additional exploratory stratified analyses found that high social support attenuated the association of fatalism with ART refusal (Table 1). After three months, 77.8% (n=389) of participants were retained. Of those, 1.3% (n=5) died in 3 months prior to ART initiation, and 14.9% (n=58) reported no ART initiation by 3 months post-VCT. 

Among over 1000 HIV-infected, ART-eligible individuals presenting for VCT at 2 sites in RSA, one-third did not return for their CD4 count. Of the 500 ART-eligible individuals enrolled in our cohort, nearly 7% initially refused ART at baseline. Among those tracked for 3 months, 16% (n=63) died or had yet to initiate ART. Future interventions should be designed to focus on decreasing gaps in pre-ART care, and addressing fatalistic beliefs by optimizing social support.

Session Number: 
P-X7
Session Title: 
Access, ART Initiation, Retention, and Outcomes in the Global South
Presenting Author: 
Ingrid Katz
Presenter Institution: 
Brigham and Women's Hospital, Harvard Medical School
Poster: