CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Results: Among 1,617 new HIV diagnoses, we identified statistically significant spatial clustering of new diagnoses in multiple metro Atlanta counties, smaller clusters of poorly linked individuals in two counties, and a single significant poor viral suppression cluster in a single county. In the single virally unsuppressed cluster, there were 5 cases of virally unsuppressed persons per square mile (vs 0 in other areas). Within this cluster, there were 60-200 homeless persons per census block group (vs <60 outside cluster). From 2009- 2014, there were 20-30 cases of active TB per square mile (vs 1-4 cases/square mile outside cluster). We also found significant spatial clustering of excess HIV/TB co-infection, and overlap with homeless service organizations. Conclusions: The local spatial cluster we identified of absolute number of virally unsuppressed persons within a single county serves as a prime target for focused service delivery, potentially with collaboration of geographically proximal service organizations. The spatial overlap with TB disease and with a high density of homeless underscores that control of chronic HIV infection remains inextricably linked with TB and homelessness, and also suggests opportunities may exist to partner with TB and homeless service providers to improve local HIV outcomes.

Poster Abstracts

996 Factors AssociatedWith Retention and Engagement in HIV Care (the REACH Survey)

Alison R. Howarth 1 ; Fiona Burns 1 ; Caroline Sabin 1 ;Vanessa Apea 2 1 Univ Coll London, London, UK; 2 Barts Hlth NHS Trust, London, UK

Background: The life expectancy for successfully treated people living with HIV in the UK is now similar to that in the general population but patients who engage poorly with care are at risk of poorer health outcomes and death. Engaging patients in HIV care remains a major challenge with little evidence available on the factors that need to be addressed. Methods: We conducted a cross-sectional survey on experience of care and living with HIV for REACH (Retention and Engagement Across Care services for HIV in the UK). Patients attending seven London HIV clinics (May 2014-August 2015) completed the survey (N=990). We systematically recruited 557 regular attenders (RA: all appointments attended in past year), 269 irregular attenders (IA: one or more missed appointments in past year) and 164 non-attenders (NA: returned to care in past year after absence of a year or more). The sample was stratified to over-represent IA and NA. Results: The median age of patients was 44 years (IQR 37-51). 27.4%were female; 36.8% identified as heterosexual and 57.0% as homosexual; 53.4%were white, 28.1%were black African and 18.5%were from other ethnic groups; 41.0%were born in the UK; and 16.0% had no post-16yrs education. Women were more likely to be NA (34.0%) or IA (29.3%) than RA (24.6%, p=0.05). Older people (>45yrs) were more likely to be RA (49.9%) than IA (37.6%) or NA (36.6%, p<0.001). Those who identified as homosexual were more likely to be RA (59.9%) or IA (56.9%) than NA (47.1%) whereas those who identified as heterosexual were more likely to be NA (40.8%) or IA (38.5%) than RA (34.9%, p=0.003). NA were more likely to have no post-16yrs education (20.5% vs IA=13.8%, RA=15.8%, p=0.02). There were no significant differences (p>0.05) in attendance pattern by ethnic group, country of birth, language or relationship status. Table 1 shows the proportion of IA and NA reporting the listed reasons for ever missing appointments (sometimes or often) at the HIV clinic. IA and NA also reported missing appointments because of drinking alcohol (IA=6.4%, NA=6.5%) and taking drugs (IA=13.2%, NA=12.6%). IA and NA with caring responsibilities sometimes or often missed appointments because of this (IA=41.2%, NA=52.3%).

424

CROI 2016