CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

1042 State-Space Models for Engagement, Retention, and Reentry in the HIV Care Cascade Hana Lee 1 ; Becky L. Genberg 1 ; Monicah Nyambura 2 ; Joseph Hogan 1 ; Paula Braitstein 3 ; Edwin Sang 2 1 Brown Univ, Providence, RI, USA; 2 Academic Model Providing Access to Hlthcare, Eldoret, Kenya; 3 Univ of Toronto, Toronto, ON, Canada

Background: We propose a state space representation of the HIV care cascade and corresponding statistical framework that can describe the longitudinal dynamics along the phases in the cascade. For this representation, each phase in the HIV care cascade is viewed as a ‘state’. Then various types of patient behaviors such as the cyclical process of engagement, disengagement, and re-entry into care (a.k.a. ‘churn’) in the cascade are described in terms of transition from one state to another. We illustrate the proposed framework using data on 57,596 patients enrolled in AMPATH (a partnership between Kenya and North American institutions) between 2008 and 2012. As testing, linkage, and viral load data was highly limited, we focused on retention aspects of care. Methods: We operationalized a 5-state care cascade including engaged in care, disengaged from care, transferred-out, lost-to-follow-up (LTFU), and deceased. Transition probabilities were defined by Pjk(t)=Pr(S(t)=k | S(t-1)=j) where S(t) denotes the state at time t. For example, P11(t) represents probability of transition from engaged to engaged, which is retention; P21(t) is probability of transition from disengaged to engaged, which is re-entry into care. We used a multinomial regression for longitudinal data to capture the effects of covariates on state transitions. Results: On average, in a given 200-day interval, 78% of individuals were engaged in care, 16% disengaged, 0.4% transferred, 5% LTFU, and 1% died. Among those engaged at a given time, probability of retention, disengagement, transfer-out, and death were .84, .13, .001, and .01. Once disengaged, probability of return to care, continued disengagement, LTFU, and death were .10, .54, .35, and .004. Regression modeling identified that among those engaged, patients with CD4>350, age>35 and patients on ARV were more likely to remain engaged in care, while male patients and those whose CD4 was not recently measured were substantially less likely. Among those disengaged, older patients, those previously on ART, and those with CD4<350 were more likely to return to care, and less likely to be permanently LTFU. Conclusions: Our representation of the HIV care cascade is a novel application of state space models, and includes regression formulations. It provides a unified approach to modeling individual-level longitudinal data from a clinical cohort. A simple version of the model is illustrated here, but an extended version includes testing, initial linkage, and viral suppression.

Table. Overall state transition rates (STR) from engaged (or disengaged) state and relative risk ratio (RRR) for effect of covariates on the STR. (*Note: transition from disengaged to transfer-out was removed from the model due to the small sample size (n=44)) State at t-1 Engaged in care Disengaged from care (S(t-1)=1) (S(t-1)=2) State at t Engaged Disengaged Transfer-out Death Engaged Disengaged LTFU Death (S(t)=1) (S(t)=2) (S(t)=3) (S(t)=5) (S(t)=1) (S(t)=2) (S(t)=4) (S(t)=5) Rate of transition from ‘engaged’ at time t – 1 to state at t Rate of transition from ‘disengaged’ at time t – 1 0.84 0.13 0.01 0.01 0.1 0.54 0.35 0.04 Relative risk ratio (RRR) for effect of covariates on state transition rates (relative to disengagement at t) On ARV 2.90 (2.82, 2.97) __ 2.62 1.59 2.34 (2.19, 2.50) __ 0.76 1.15 (2.29, 3.01) (1.47, 1.73) (0.74, 0.78) (0.85, 1.56)

__

0.81

1.40

__

1.05

1.45

Age>=35 at enrollment

1.37 (0.86, 0.90) 0.88 (0.86, 0.90)

1.13 (1.07, 1.20) 1.03 (9.67, 1.09)

(0.73, 0.91) (1.30, 1.51)

(1.02, 1.08) (1.07, 1.96)

__

0.78

1.55

__

0.96

1.09

Male

(0.69, 0.88) (1.44, 1.66)

(0.94, 0.99) (0.81, 1.47)

Most recent CD4 < 350 Most recent CD4>=350 CD4 not measured in last 200 days

ref

ref __

ref

ref

ref

ref __

ref

ref

0.98

0.27

1.14

3.04

1.69 (1.65, 1.73) 0.34 (0.33, 0.36)

0.84 (0.78, 0.90) 0.72 (0.65, 0.79)

(0.87,1.11) (0.24, 0.30)

(1.11, 1.17) (1.99, 4.65)

__

0.41

0.46

__

0.91

1.73

(0.32, 0.51) (0.42, 0.52)

(0.87, 0.96) (1.02, 2.96)

1043 How Far AreWe From Early cART for All? A Nationwide Population-Based Study in France Virginie Supervie 1 ; Jean-Marc Lacombe 2 ; Rosemary Dray-Spira 2 ; Dominique Costagliola 1 ; for the FHDH-ANRS CO4 Study Group 1 Sorbonne Univs, Paris, France; 2 INSERM U1136, Paris, France Background: Ensuring early universal access to combination antiretroviral treatment (cART), and especially within the first year of HIV infection, is critical to reach the end of AIDS and control the HIV epidemic. However, how far or how close we are from early universal cART initiation remains unknown. Methods: We estimated the timing of HIV care in France in 2010 using statistical modeling and two large data sources: the national HIV surveillance system and the French Hospital Database on HIV (FHDH). To estimate the distribution of times from infection to diagnosis, we fitted a back-calculation model to the annual numbers of new HIV diagnoses. To estimate the distribution of times from HIV diagnosis to care entry, from care entry to cART initiation and from cART initiation to reaching undetectable viral load, we used survival methods and data on the dates of HIV diagnosis, care entry, cART initiation and viral suppression of the 6268 HIV-infected individuals who newly engaged in care between 2008 and 2010 and were enrolled onto the FHDH cohort. We summed up the distributions to obtain the distributions of time intervals from HIV infection to cART access. Figures were computed overall and by HIV exposure group. Results: We found that only 8.3% of HIV-infected individuals accessed cART within the first year of infection (see Table). This proportion reached 10.1% among men who have sex with men (MSM). The estimated median time interval from HIV infection to cART initiation was 5.0 years (IQR: 2.7-7.9). MSM had the shortest median time to cART initiation (4.4 years) and injecting drug users (IDUs) the longest (8.1 years). Time lost in accessing cART was mainly due to delays in HIV testing (overall median: 3.2 years), except for IDUs where it was also due to delayed care entry once diagnosed (median of ~1 year versus <1 month for other groups). Times to access cART once in care and times to reaching viral suppression once on cART were short (<6 months in median). Conclusions: Our study shows that even in a country like France, where the health care system offers one of the best environments for HIV care, we are far from early ART for all. Similar gap is likely to exist in other settings and should be investigated. To close this gap, evaluating patient flow-time through the continuum of care will be key to identify what kind of actions is needed to accelerate cART access.

Poster Abstracts

446

CROI 2016