Pediatric post-discharge mortality in resource-poor countries: a systematic review and meta-analysis

Abstract

Background: Under-five mortality remains concentrated in resource-poor countries. Post-discharge mortality is becoming increasingly recognized as a significant contributor to overall child mortality. With a substantial recent expansion of research and novel data synthesis methods, this study aims to update the current evidence base by providing a more nuanced understanding of the burden and associated risk factors of pediatric post-discharge mortality after acute illness.

Methods: Eligible studies published between January 1, 2017 and January 31, 2023, were retrieved using MEDLINE, Embase, and CINAHL databases. Studies published before 2017 were identified in a previous review and added to the total pool of studies. Only studies from countries with low or low-middle Socio-Demographic Index with a post-discharge observation period greater than seven days were included. Risk of bias was assessed using a modified version of the Joanna Briggs Institute critical appraisal tool for prevalence studies. Studies were grouped by patient population, and 6-month post-discharge mortality rates were quantified by random-effects meta-analysis. Secondary outcomes included post-discharge mortality relative to in-hospital mortality, pooled risk factor estimates, and pooled post-discharge Kaplan-Meier survival curves. PROSPERO study registration: #CRD42022350975.

Findings: Of 1963 articles screened, 42 eligible articles were identified and combined with 22 articles identified in the previous review, resulting in 64 total articles. These articles represented 46 unique patient cohorts and included a total of 105,560 children. For children admitted with a general acute illness, the pooled risk of mortality six months post-discharge was 4.4% (95% CI: 3.5%-5.4%, I² = 94.2%, n = 11 studies, 34,457 children), and the pooled in-hospital mortality rate was 5.9% (95% CI: 4.2%-7.7%, I² = 98.7%, n = 12 studies, 63,307 children). Among disease subgroups, severe malnutrition (12.2%, 95% CI: 6.2%-19.7%, I² = 98.2%, n = 10 studies, 7760 children) and severe anemia (6.4%, 95% CI: 4.2%-9.1%, I² = 93.3%, n = 9 studies, 7806 children) demonstrated the highest 6-month post-discharge mortality estimates. Diarrhea demonstrated the shortest median time to death (3.3 weeks) and anemia the longest (8.9 weeks). Most significant risk factors for post-discharge mortality included unplanned discharges, severe malnutrition, and HIV seropositivity.

Interpretation: Pediatric post-discharge mortality rates remain high in resource-poor settings, especially among children admitted with malnutrition or anemia. Global health strategies must prioritize this health issue by dedicating resources to research and policy innovation.

Funding: No specific funding was received.

Keywords: Child health; Child mortality; Global health; Meta-analysis; Post-discharge mortality.

Fig. 1

PRISMA flowdiagram.

Fig. 2

Pooled post-discharge mortality curves (1 minus survival) by population sub-group.

Fig. 3

6-month post-discharge mortality random effects estimates among population sub-groups. Abbreviations: RE = random effects.

Fig. 4

Percentage of total deaths among population sub-groups, in-hospital versus six-month post-discharge. Abbreviations: IH = in-hospital; PD = post-discharge.

Fig. 5

Pooled hazard ratios (random-effects estimate) for post-discharge mortality across all population sub-groups. Red: indicates confidence interval did not include 1 and increased the risk of post-discharge mortality. Green: indicates confidence interval did not include 1 and reduced risk of mortality. Black: indicates confidence interval includes 1 and the effect on mortality is inconclusive. Note: Increased respiratory rate included studies that defined it as a respiratory rate >30 breaths per minute or tachypnea. The number of patients was estimated based on the number of children discharged.