Patient-reported Reasons for Stopping Care or Switching Clinics in Zambia: A Multisite, Regionally Representative Estimate Using a Multistage Sampling-based Approach in Zambia

Abstract

Background: Understanding patient-reported reasons for lapses of retention in human immunodeficiency virus (HIV) treatment can drive improvements in the care cascade. A systematic assessment of outcomes among a random sample of patients lost to follow-up (LTFU) from 32 clinics in Zambia to understand the reasons for silent transfers and disengagement from care was undertaken.

Methods: We traced a simple random sample of LTFU patients (>90 days from last scheduled visit) as determined from clinic-based electronic medical records from a probability sample of facilities. Among patients found in person, we solicited reasons for either stopping or switching care and predictors for re-engagement. We coded reasons into structural, psychosocial, and clinic-based barriers.

Results: Among 1751 LTFU patients traced and found alive, 31% of patients starting antiretroviral therapy (ART) between 1 July 2013 and 31 July 2015 silently transferred or were disengaged (40% male; median age, 35 years; median CD4 level, 239 cells/μL); median time on ART at LTFU was 480 days (interquartile range, 110-1295). Among the 544 patients not in care, median prevalences for patient-reported structural, psychosocial, and clinic-level barriers were 27.3%, 13.9%, and 13.4%, respectively, and were highly variable across facilities. Structural reasons, including, "relocated to a new place" were mostly cited among 289 patients who silently transferred (35.5%). We found that men were less likely to re-engage in care than women (odds ratio, .39; 95% confidence interval, .22-.67; P = .001).

Conclusions: Efforts to improve retention of patients on ART may need to be tailored at the facility level to address patient-reported barriers.

Keywords: HIV; Zambia; disengagement; reasons; retention.

Figure 1.

Flowchart depicting tracing outcomes among those categorized as “lost” by electronic medical record. Abbreviations: ART, antiretroviral therapy; CIDRZ, Centre for Infectious Disease Research in Zambia.

Figure 2.

Reasons for disengagement. n = 255. Abbreviation: ARV, antiretroviral threapy.

Figure 3.

Reasons for silent transfer. n = 289. Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus.

Figure 4.

Reasons to return to care if disengaged. n = 255. Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus. tx, treatment

Figure 5.

Venn diagrams depicting overlap between barrier domains.

Figure 6.

Facility-level reasons for silent transfer, n = 289 (A); for disengagement (participants who were found to be alive but out of care), n = 255 (B); and patient-reported changes required to return to care, among those disengaged, n = 255 (C).

Figure 7.

Estimated probability of re-engagement at a new facility (ie, silent transfer) among patients lost to follow-up from their original care facility (adjusted as per Table 2). Abbreviation: CI, confidence interval.