HIV Care Trajectories as a Novel Longitudinal Assessment of Retention in Care

Use of machine learning approaches to predict transition of retention in care among people living with HIV in South Carolina: a real-world data study

Maintaining retention in care (RIC) for people living with HIV (PLWH) helps achieve viral suppression and reduce onward transmission. This study aims to identify the best machine learning model that predicts the RIC transition over time. Extracting from the enhanced HIV/AIDS reporting system, this study included 9765 PLWH from 2005 to 2020 in South Carolina. Transition of RIC was defined as the change of RIC status in each two-year time window. We applied seven classifiers, such as Random Forest, Support Vector Machine, eXtreme Gradient Boosting and Long-short-term memory, for each lagged response to predict the subsequent year's RIC transition. Classification performance was assessed using balanced prediction accuracy, the area under the curve (AUC), recall, precision and F1 scores. The proportion of the four categories of RIC transition was 13.59%, 29.78%, 9.06% and 47.57%, respectively. Support Vector Machine was the best approach for every lag model based on both the F1 score (0.713, 0.717 and 0.719) and AUC (0.920, 0.925 and 0.928). The findings could facilitate the risk augment of PLWH who are prone to follow-up so that clinicians and policymakers could come up with more specific strategies and relocate resources for intervention to keep them sustained in HIV care.

Intersectionality-informed analysis of durable viral suppression disparities in people with HIV

OBJECTIVE

The aim of this study was to examine drivers of durable viral suppression (DVS) disparities among people with HIV (PWH) using quantitative intersectional approaches.

DESIGN

A retrospective cohort analysis from electronic health records informed by intersectionality to better capture the concept of interlocking and interacting systems of oppression.

METHODS

We analyzed data of PWH seen at a LGBTQ federally qualified health center in Chicago (2012-2019) with at least three viral loads. We identified PWH who achieved DVS using latent trajectory analysis and examined disparities using three intersectional approaches: Adding interactions, latent class analysis (LCA), and qualitative comparative analysis (QCA). Findings were compared with main effects only regression.

RESULTS

Among 5967 PWH, 90% showed viral trajectories consistent with DVS. Main effects regression showed that substance use [odds ratio (OR) 0.56, 0.46-0.68] and socioeconomic status like being unhoused (OR: 0.39, 0.29-0.53), but not sexual orientation or gender identity (SOGI) were associated with DVS. Adding interactions, we found that race and ethnicity modified the association between insurance and DVS ( P for interaction <0.05). With LCA, we uncovered four social position categories influenced by SOGI with varying rates of DVS. For example, the transgender women-majority class had worse DVS rates versus the class of mostly nonpoor white cisgender gay men (82 vs. 95%). QCA showed that combinations, rather than single factors alone, were important for achieving DVS. Combinations vary with marginalized populations (e.g. black gay/lesbian transgender women) having distinct sufficient combinations compared with historically privileged groups (e.g. white cisgender gay men).

CONCLUSION

Social factors likely interact to produce DVS disparities. Intersectionality-informed analysis uncover nuance that can inform solutions.

The Cooperative Re-Engagement Controlled Trial (CoRECT): Durable Viral Suppression Assessment

BACKGROUND

A collaborative, data-to-care strategy to identify persons with HIV (PWH) newly out-of-care, combined with an active public health intervention, significantly increases the proportion of PWH re-engaged in HIV care. We assessed this strategy's impact on durable viral suppression (DVS).

METHODS

A multisite, prospective randomized controlled trial for out-of-care individuals using a data-to-care strategy and comparing public health field services to locate, contact, and facilitate access to care versus the standard of care. DVS was defined as the last viral load, the viral load at least 3 months before, and any viral load between the 2 were all <200 copies/mL during the 18-month postrandomization. Alternative definitions of DVS were also analyzed.

RESULTS

Between August 1, 2016-July 31, 2018, 1893 participants were randomized from Connecticut (n = 654), Massachusetts (n = 630), and Philadelphia (n = 609). Rates of achieving DVS were similar in the intervention and standard-of-care arms in all jurisdictions (all sites: 43.4% vs 42.4%, P = 0.67; Connecticut: 46.7% vs 45.0%, P = 0.67; Massachusetts: 40.7 vs 44.4%, P = 0.35; Philadelphia: 42.4% vs 37.3%, P = 0.20). There was no association between DVS and the intervention (RR: 1.01, CI: 0.91-1.12; P = 0.85) adjusting for site, age categories, race/ethnicity, birth sex, CD4 categories, and exposure categories.

CONCLUSION

A collaborative, data-to-care strategy, and active public health intervention did not increase the proportion of PWH achieving DVS, suggesting additional support to promote retention in care and antiretroviral adherence may be needed. Initial linkage and engagement services, through data-to-care or other means, are likely necessary but insufficient for achieving DVS for all PWH.

Trajectories of re-engagement: factors and mechanisms enabling patient return to HIV care in Zambia

INTRODUCTION

While disengagement from HIV care threatens the health of persons living with HIV (PLWH) and incidence-reduction targets, re-engagement is a critical step towards positive outcomes. Studies that establish a deeper understanding of successful return to clinical care among previously disengaged PLWH and the factors supporting re-engagement are essential to facilitate long-term care continuity.

METHODS

We conducted narrative, patient-centred, in-depth interviews between January and June 2019 with 20 PLWH in Lusaka, Zambia, who had disengaged and then re-engaged in HIV care, identified through electronic medical records (EMRs). We applied narrative analysis techniques, and deductive and inductive thematic analysis to identify engagement patterns and enablers of return.

RESULTS

We inductively identified five trajectories of care engagement, suggesting patterns in patient characteristics, experienced barriers and return facilitators that may aid intervention targeting including: (1) intermittent engagement;(2) mostly engaged; (3) delayed linkage after testing; (4) needs time to initiate antiretroviral therapy (ART); and (5) re-engagement with ART initiation. Patient-identified periods of disengagement from care did not always align with care gaps indicated in the EMR. Key, interactive re-engagement facilitators experienced by participants, with varied importance across trajectories, included a desire for physical wellness and social support manifested through verbal encouragement, facility outreach or personal facility connections and family instrumental support. The mechanisms through which facilitators led to return were: (1) the promising of living out one's life priorities; (2) feeling valued; (3) fostering interpersonal accountability; (4) re-entry navigation support; (5) facilitated care and treatment access; and (6) management of significant barriers, such as depression.

CONCLUSIONS

While preliminary, the identified trajectories may guide interventions to support re-engagement, such as offering flexible ART access to patients with intermittent engagement patterns instead of stable patients only. Further, for re-engagement interventions to achieve impact, they must activate mechanisms underlying re-engagement behaviours. For example, facility outreach that reminds a patient to return to care but does not affirm a patient's value or navigate re-entry is unlikely to be effective. The demonstrated importance of positive health facility connections reinforces a growing call for patient-centred care. Additionally, interventions should consider the important role communities play in fostering treatment motivation and overcoming practical barriers.

Longitudinal trajectories of HIV care engagement since diagnosis among persons with HIV in the Florida Ryan White program

HIV care engagement is a dynamic process. We employed group-based trajectory modeling to examine longitudinal patterns in care engagement among people who were newly diagnosed with HIV and enrolled in the Ryan White program in Florida (n = 9,755) between 2010 and 2015. Five trajectories were identified (47.9% "in care" with 1-2 care visit(s) per 6 months, 18.0% "frequent care" with 3 or more care visits per 6 months, 11.0% "re-engage", 11.0% "gradual drop out", 12.6% "early dropout") based on the number of care attendances (including outpatient/case management visits, viral load or CD4 test) for each six-month during the first five years since diagnosis. Relative to "in care", people in the "frequent care" trajectory were more likely to be Hispanic/Latino and older at HIV diagnosis, whereas people in the three suboptimal care retention trajectories were more likely to be younger. Area deprivation index, rurality, and county health rankings were also strongly associated with care trajectories. Individual- and community-level factors associated to the three suboptimal care retention trajectories, if confirmed to be causative and actionable, could be prioritized to improve HIV care engagement.

Novel Longitudinal Methods for Assessing Retention in Care: a Synthetic Review

PURPOSE OF REVIEW

Retention in care is both dynamic and longitudinal in nature, but current approaches to retention often reduce these complex histories into cross-sectional metrics that obscure the nuanced experiences of patients receiving HIV care. In this review, we discuss contemporary approaches to assessing retention in care that captures its dynamic nature and the methodological and data considerations to do so.

RECENT FINDINGS

Enhancing retention measurements either through patient tracing or "big data" approaches (including probabilistic matching) to link databases from different sources can be used to assess longitudinal retention from the perspective of the patient when they transition in and out of care and access care at different facilities. Novel longitudinal analytic approaches such as multi-state and group-based trajectory analyses are designed specifically for assessing metrics that can change over time such as retention in care. Multi-state analyses capture the transitions individuals make in between different retention states over time and provide a comprehensive depiction of longitudinal population-level outcomes. Group-based trajectory analyses can identify patient subgroups that follow distinctive retention trajectories over time and highlight the heterogeneity of retention patterns across the population. Emerging approaches to longitudinally measure retention in care provide nuanced assessments that reveal unique insights into different care gaps at different time points over an individuals' treatment. These methods help meet the needs of the current scientific agenda for retention and reveal important opportunities for developing more tailored interventions that target the varied care challenges patients may face over the course of lifelong treatment.

Annual and durable HIV retention in care and viral suppression among patients of Peter Ho Clinic, 2013-2017

Objectives

To determine rates of annual and durable retention in medical care and viral suppression among patients enrolled in the Peter Ho Clinic, from 2013–2017.

Methods

This is a retrospective review of medical record data in an urban clinic, located in Newark, New Jersey, a high prevalence area of persons living with HIV. Viral load data were electronically downloaded, in rolling 1-year intervals, in two-month increments, from January 1, 2013 to December 31, 2019. Three teams were established, and every two months, they were provided with an updated list of patients with virologic failure. Retention and viral suppression rates were first calculated for each calendar-year. After patients were determined to be retained/suppressed annually, the proportion of patients with durable retention and viral suppression were calculated in two, three, four, five and six-year periods. Descriptive statistics were used to summarize sample characteristics by retention in care, virologic failure and viral suppression with Pearson Chi-square; p-value <0.05 was statistically significant. Multiple logistic regression models identified patient characteristics associated with retention in medical care, virologic failure and suppression.

Results

As of December 31, 2017, 1000 (57%) patients were retained in medical care of whom 870 (87%) were suppressed. Between 2013 and 2016, decreases in annual (85% to 77%) and durable retention in care were noted: two-year (72% to 70%) and three-year (63% to 59%) periods. However, increases were noted for 2017, in annual (89%) and durable retention in the two-year period (79%). In the adjusted model, when compared to current patients, retention in care was less likely among patients reengaging in medical care (adjusted Odds Ratio (aOR): 0.77, 95% CI: 0.61–0.98) but more likely among those newly diagnosed from 2014–2017 (aOR: 1.57, 95% CI: 1.08–2.29), compared to those in care since 2013. A higher proportion of patients re-engaging in medical care had virologic failure than current patients (56% vs. 47%, p < 0.0001). As age decreased, virologic failure was more likely (p<0.0001). Between 2013 and 2017, increases in annual (74% to 87%) and durable viral suppression were noted: two-year (59% to 73%) and three-year (49% to 58%) periods. Viral suppression was more likely among patients retained in medical care up to 2017 versus those who were not (aOR: 5.52, 95% CI: 4.08–7.46). Those less likely to be suppressed were 20–29 vs. 60 years or older (aOR: 0.52, 95% CI: 0.28–0.97), had public vs. private insurance (aOR: 0.29, 95% CI: 0.15–0.55) and public vs. private housing (aOR: 0.59, 95% CI: 0.40–0.87).

Conclusions

Restructuring clinical services at this urban clinic was associated with improved viral suppression. However, concurrent interventions to ensure retention in medical care were not implemented. Both retention in care and viral suppression interventions should be implemented in tandem to achieve an end to the epidemic. Retention in care and viral suppression should be measured longitudinally, instead of cross-sectional yearly evaluations, to capture dynamic changes in these indicators.

What we talk about when we talk about durable viral suppression

: As policies built on 'Undetectable = Untransmittable' become more popular, use of durable viral suppression (DVS) as an outcome in analyses is increasing. We identified a case series of recent HIV-related publications that study the DVS outcome. The majority did not distinguish between a definition of DVS and the operationalization of that definition. Clearer discussion of DVS, including a formal definition, is needed to ensure better comparability across studies and ultimately better public health outcomes.