Prediction models for post-discharge mortality among under-five children with suspected sepsis in Uganda: A multicohort analysis

Background: In many low-income countries, over five percent of hospitalized children die following hospital discharge. The lack of available tools to identify those at risk of post-discharge mortality has limited the ability to make progress towards improving outcomes. We aimed to develop algorithms designed to predict post-discharge mortality among children admitted with suspected sepsis. Methods: Four prospective cohort studies of children in two age groups (0–6 and 6–60 months) were conducted between 2012–2021 in six Ugandan hospitals. Prediction models were derived for six-months post-discharge mortality, based on candidate predictors collected at admission, each with a maximum of eight variables, and internally validated using 10-fold cross-validation. Findings: 8,810 children were enrolled: 470 (5.3%) died in hospital; 257 (7.7%) and 233 (4.8%) post-discharge deaths occurred in the 0-6-month and 6-60-month age groups, respectively. The primary models had an area under the receiver operating characteristic curve (AUROC) of 0.77 (95%CI 0.74–0.80) for 0-6-month-olds and 0.75 (95%CI 0.72–0.79) for 6-60-month-olds; mean AUROCs among the 10 cross-validation folds were 0.75 and 0.73, respectively. Calibration across risk strata was good: Brier scores were 0.07 and 0.04, respectively. The most important variables included anthropometry and oxygen saturation. Additional variables included: illness duration, jaundice-age interaction, and a bulging fontanelle among 0-6-month-olds; and prior admissions, coma score, temperature, age-respiratory rate interaction, and HIV status among 6-60-month-olds. Data Processing Methods: The post-processed data files were created using R version 4.2.2. (R Foundation for Statistical Computing, Vienna, Austria) and briefly involved renaming columns from the different datasets so that they are consistent, converting categories coded as “unknown”, “don’t know”, or “missing” to NA, creating new columns, calculating z-scored variables, and converting relevant columns to factors or dates. Ethics Declaration: These studies were approved by the Mbarara University of Science and Technology (No. 15/10-16), the Uganda National Council for Science and Technology (HS 2207), and the University of British Columbia (H16-02679).研究背景：在众多低收入国家中，超过5%的住院儿童会在出院后发生死亡。目前缺乏能够精准识别出院后死亡风险人群的有效工具，这制约了改善患儿预后相关工作的推进。本研究旨在开发针对因疑似脓毒症入院儿童的出院后死亡预测算法。 研究方法：本研究于2012至2021年间，在乌干达6家医院开展了四项针对两个年龄组（0~6月龄与6~60月龄）儿童的前瞻性队列研究。以入院阶段收集的候选预测变量（单模型最多包含8个变量）为基础，构建出院后6个月死亡率预测模型，并采用10折交叉验证开展内部验证。 研究结果：本研究共纳入8810名儿童：其中470名（5.3%）在住院期间死亡；0~6月龄组与6~60月龄组分别出现257例（7.7%）与233例（4.8%）出院后死亡病例。针对两个年龄组的基础预测模型，受试者工作特征曲线下面积（area under the receiver operating characteristic curve, AUROC）分别为0.77（95%置信区间0.74~0.80）与0.75（95%置信区间0.72~0.79）；10折交叉验证得到的平均AUROC分别为0.75与0.73。不同风险分层的模型校准效果良好：布里尔分数（Brier score）分别为0.07与0.04。模型中最重要的预测变量包括人体测量学指标与血氧饱和度。其余重要预测变量包括：0~6月龄组的病程时长、黄疸-年龄交互项及囟门膨出；6~60月龄组的既往住院史、昏迷评分、体温、年龄-呼吸频率交互项及HIV感染状态。 数据处理方法：后处理数据文件采用R 4.2.2版本（奥地利维也纳R统计计算基金会开发）生成，具体处理流程包括：统一不同数据集的列命名规则以实现一致性，将编码为"unknown""don’t know"或"missing"的分类值转换为NA，创建新数据列，计算标准化z分数变量，并将相关列转换为因子类型或日期类型。 伦理声明：本研究已通过姆巴拉拉科技大学（Mbarara University of Science and Technology，审批编号15/10-16）、乌干达国家科学技术委员会（审批编号HS 2207）以及不列颠哥伦比亚大学（审批编号H16-02679）的伦理审查批准。