Longitudinal Care Cascade Outcomes Among People Eligible for Antiretroviral Therapy Who Are Newly Linking to Care in Zambia: A Multistate Analysis

Abstract

Background: Retention in human immunodeficiency virus (HIV) care is dynamic, with patients frequently transitioning in and out of care. Analytical approaches (eg, survival analyses) commonly used to assess HIV care cascade outcomes fail to capture such transitions and therefore incompletely represent care outcomes over time.

Methods: We analyzed antiretroviral therapy (ART)-eligible adults newly linking to care at 64 clinics in Zambia between 1 April 2014 and 31 July 2015. We used electronic medical record data and supplemented these with updated care outcomes ascertained by tracing a multistage random sample of patients lost to follow-up (LTFU, >90 days late for last appointment). We performed multistate analyses, incorporating weights from sampling, to estimate the prevalence of 9 care states over time since linkage with respect to ART initiation, retention in care, transfers, and mortality.

Results: In sum, 23 227 patients (58% female; median age 34 years [interquartile range 28-41]) were ART-eligible at enrollment. At 1 year, 75.2% had initiated ART and were in care: 61.8% were continuously retained, 6.1% had reengaged after LTFU, and 7.3% had transferred. Also, 10.1% were LTFU within 7 days of enrollment, and 15.2% were LTFU at 1 year (6.7% prior to ART). One year after LTFU, 51.6% of those LTFU prior to ART remained out of care compared to 30.2% of those LTFU after initiating ART. Overall, 6.9% of patients had died by 1 year with 3.0% dying prior to ART.

Conclusion: Multistate analyses provide more complete assessments of longitudinal HIV cascade outcomes and reveal treatment gaps at distinct timepoints in care that will still need to be addressed even with universal treatment.

Keywords: HIV care cascade; Zambia; mortality; multistate analysis; retention in care.

Figure 1.

State transitions framework for multistate analysis. At each time point, patients were categorized into 1 of 9 mutually exclusive and exhaustive states: (1) not initiated on ART and in care, (2) out of care prior to initiating on ART, (3) transfer to another clinic prior to initiating ART, (4) death prior to initiating ART, (5) initiated on ART and consistently in care (ie, no LTFU after initiating ART), (6) out of care after initiating ART, (7) reengaged in care at the original clinic after LTFU on ART, (8) transfer to another clinic after initiating ART, and (9) death after initiating ART. This figure depicts all the possible transitions patients could make from each state. Abbreviations: ART, antiretroviral therapy; LTFU, lost to follow-up.

Figure 2.

Patient flowchart. In sum, 23 227 patients eligible for ART at the time of enrollment at 64 sites were included in this analysis. As of 1 July 2015, 5516 were considered LTFU, and we randomly selected 501 patients from 32 sites for active tracing. We ascertained updated vital status in 395 (78.8%) of the patients and updated care status in 241 of the 299 patients known to be alive (80.6%). Abbreviations: ART, antiretroviral therapy; LTFU, lost to follow-up.

Figure 3.

Longitudinal outcomes among people eligible for ART newly linking to care.This figure represents longitudinal outcomes among 23 227 people eligible for ART newly linking to care in Zambia based on results from the overall multistate analysis. The figure depicts the proportion of patients estimated to be in each care state at any given time point accounting for the transitions patients made between different care states over time. Abbreviations: ART, antiretroviral therapy; LTFU, lost to follow-up.

Figure 4.

Longitudinal outcomes among patients newly entering a specific care state. These figures represent longitudinal outcomes among the patients after newly entering 3 specific care states. A, Outcomes after patients initiate ART. B, Outcomes after patients become LTFU prior to initiating ART. C, Outcomes after patients become LTFU after previously initiating ART. Abbreviations: ART, antiretroviral therapy; LTFU, lost to follow-up.

Figure 5.

Forest plot of results from multistate analyses stratified by patient subgroups. This figure depicts the proportion of patients in 1 of 4 composite care states at 1 year after performing stratified multistate analysis. A, Proportion in each subgroup who are in care and on ART (a composite of ART in care [state 5], reengaged on ART [state 7], and transferred on ART [state 8]). B, Proportion of each subgroup who are LTFU (a composite of LTFU prior to ART [state 2] and LTFU after ART [state 6]). C, Proportion in each subgroup who transferred to a new clinic (a composite of transfer prior to ART [state 3] and transfer after ART [state 8]). D, Proportion who have died (a composite of died prior to ART [state 4] and died after ART [state 9]). Abbreviations: ART, antiretroviral therapy; LTFU, lost to follow-up.