Open Collections

Description

Background:Mortality following hospital discharge remains a significant threat to child health, particularly in resource-limited settings. In Uganda, the Smart Discharges risk-prediction models have demonstrated success in their ability to predict those at highest risk of death after discharge and use this to guide a risk-based approach to post-discharge care in children admitted with suspected sepsis. Respective prediction models for post-discharge mortality in ages 0-6 months and ages 6-60 months were developed in this cohort but have not yet been validated outside of Uganda. This study aimed to externally validate existing risk prediction models for pediatric post-discharge mortality within the Rwandan context.

Methods: Prospective cohort of children 0d-60 mos admitted with suspected sepsis at two hospitals in Rwanda: Ruhengeri Referral Hospital in Musanze (rural) and University Hospital of Kigali in Kigali (urban) from May 2022 to February 2023. Vital status follow up was conducted at 2-, 4- and 6-months post-discharge.

Five existing models from Smart Discharges Uganda were validated in this cohort: two models for children 0-6 months, and three for children 6-60 months. Models were applied to each participant in the Rwanda cohort to obtain a risk score which was then used to calculate predicted probability of post-discharge death. Model performance was evaluated by comparing to observed outcomes and to determine sensitivity, specificity, and AUROC. Threshold was set at 80% sensitivity. .

Findings:In a cohort of 1218 children, 1123 children (96.7%) completed follow up. The overall rate of post-discharge mortality was 4.8% (n=58). The highest performing models had an AUROC of 0.75 (0-6 mos) and 0.74 (6-60mos), respectively. All five prediction models tested achieved an AUROC greater than 0.7 (range 0.706 - 0.738). Model degradation (determined by the percent reduction in AUC between the original model and the derived model) was relatively low, ranging from from 1.1% to 7.7%. Calibration plots showed good calibration for all models at predicted probabilities below 10%. There were too few outcomes to assess calibration among those at higher levels of predicted risk.

Data Processing Methods:

Ethics Declaration: Ethical approval was obtained from the University of Rwanda College of Medicine and Health Sciences (No 411/CMHS IRB/2021); University Teaching Hospital of Kigali (EC/CHUK/005/2022), University of California San Francisco (381688) and the University of British Columbia (H21-02795).


NOTE for restricted files: If you are not yet a CoLab member, please complete our membership application survey to gain access to restricted files within 2 business days.
Some files may remain restricted to CoLab members. These files are deemed more sensitive by the file owner and are meant to be shared on a case-by-case basis. Please contact the CoLab coordinator at sepsiscolab@bcchr.ca or visit our website.