Systematic review of risk assessment tools for post-discharge mortality among children in sub-Saharan Africa

Post-discharge mortality is increasingly recognized as a major contributor to the high burden of childhood mortality in sub-Saharan Africa. Accurate identification of children at risk for post-discharge mortality is critically important to inform interventions to reduce deaths following hospital discharge. Our objective was to describe the current state of development, validation, or implementation for risk assessment tools for post-hospital discharge mortality (PDM) in sub-Saharan Africa. We conducted a systematic review of publications on risk assessment tools for PDM among children aged 0-18 years in sub-Saharan Africa. We searched CABI Global Health, Cochrane Reviews, Cochrane Trials, ProQuest Dissertations and Theses, Embase, PubMed, and Web of Science with no date or language restriction. We included publications if they described a tool/model with weights assigned to variables to quantify risk of PDM, included children, and were conducted in sub-Saharan Africa. We determined the level of evidence for tools using the Evidence-Based-Medicine Working Group hierarchy. Of 4,893 publications screened, 289 full texts were reviewed, and seven publications that reported 23 risk assessment tools for PDM among children in sub-Saharan Africa were identified. These studies enrolled 49,669 total participants (3.6%, n = 1,795 experienced PDM). There was substantial heterogeneity in identified risk factors, although all identified malnutrition as a risk factor for PDM. All risk assessment tools had fair (i.e., area under the receiver operating characteristic curve [AUC] ≥0.70) or good (AUC ≥ 0.80) discriminatory value in internal validation. Only two risk assessment tools had been externally validated, and none were implemented. Existing risk assessment tools to identify children at risk for PDM in sub-Saharan Africa lack broad validation and implementation. Malnutrition is a common risk factor for PDM. Further studies are needed to validate and implement such tools to reduce PDM among children.