Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia

Background The time after hospital discharge carries high rates of mortality in neonates and young children in sub-Saharan Africa. Previous work using logistic regression to develop risk assessment tools to identify those at risk for postdischarge mortality has yielded fair discriminatory value. Our...

Full description

Saved in:
Bibliographic Details
Main Authors: Rodrick Kisenge, Todd A. Florin, Cynthia G. Whitney, Michelle Niescierenko, Julia Kamara, Abraham Samma, Evance Godfrey, Chris A. Rees, Readon C. Ideh, Ye-Jeung G. Coleman-Nekar, Hussein K. Manji, Christopher R. Sudfeld, Adrianna L. Westbrook, Claudia R. Morris, Karim P. Manji, Christopher P. Duggan, Rishikesan Kamaleswaran
Format: Article
Language:English
Published: BMJ Publishing Group 2025-06-01
Series:BMJ Paediatrics Open
Online Access:https://bmjpaedsopen.bmj.com/content/9/1/e003547.full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850113436545974272
author Rodrick Kisenge
Todd A. Florin
Cynthia G. Whitney
Michelle Niescierenko
Julia Kamara
Abraham Samma
Evance Godfrey
Chris A. Rees
Readon C. Ideh
Ye-Jeung G. Coleman-Nekar
Hussein K. Manji
Christopher R. Sudfeld
Adrianna L. Westbrook
Claudia R. Morris
Karim P. Manji
Christopher P. Duggan
Rishikesan Kamaleswaran
author_facet Rodrick Kisenge
Todd A. Florin
Cynthia G. Whitney
Michelle Niescierenko
Julia Kamara
Abraham Samma
Evance Godfrey
Chris A. Rees
Readon C. Ideh
Ye-Jeung G. Coleman-Nekar
Hussein K. Manji
Christopher R. Sudfeld
Adrianna L. Westbrook
Claudia R. Morris
Karim P. Manji
Christopher P. Duggan
Rishikesan Kamaleswaran
author_sort Rodrick Kisenge
collection DOAJ
description Background The time after hospital discharge carries high rates of mortality in neonates and young children in sub-Saharan Africa. Previous work using logistic regression to develop risk assessment tools to identify those at risk for postdischarge mortality has yielded fair discriminatory value. Our objective was to determine if machine learning models would have greater discriminatory value to identify neonates and young children at risk for postdischarge mortality.Methods We conducted a planned secondary analysis of a prospective observational cohort at Muhimbili National Hospital in Dar es Salaam, Tanzania and John F. Kennedy Medical Center in Monrovia, Liberia. We enrolled neonates and young children near the time of discharge. The outcome was 60-day postdischarge mortality. We collected socioeconomic, demographic, clinical, and anthropometric data during hospital admission and used machine learning (ie, eXtreme Gradient Boosting (XGBoost), Hist-Gradient Boost, Support Vector Machine, Neural Network, and Random Forest) to develop risk assessment tools to identify: (1) neonates and (2) young children at risk for postdischarge mortality.Results A total of 2310 neonates and 1933 young children enrolled. Of these, 71 (3.1%) neonates and 67 (3.5%) young children died after hospital discharge. XGBoost, Hist Gradient Boost, and Neural Network models yielded the greatest discriminatory value (area under the receiver operating characteristic curves range: 0.94–0.99) and fewest features, which included six features for neonates and five for young children. Discharge against medical advice, low birth weight, and supplemental oxygen requirement during hospitalisation were predictive of postdischarge mortality in neonates. For young children, discharge against medical advice, pallor, and chronic medical problems were predictive of postdischarge mortality.Conclusions Our parsimonious machine learning-based models had excellent discriminatory value to predict postdischarge mortality among neonates and young children. External validation of these tools is warranted to assist in the design of interventions to reduce postdischarge mortality in these vulnerable populations.
format Article
id doaj-art-c9cfbd3b5e7a4b43812afe4ad7f8d5fa
institution OA Journals
issn 2399-9772
language English
publishDate 2025-06-01
publisher BMJ Publishing Group
record_format Article
series BMJ Paediatrics Open
spelling doaj-art-c9cfbd3b5e7a4b43812afe4ad7f8d5fa2025-08-20T02:37:09ZengBMJ Publishing GroupBMJ Paediatrics Open2399-97722025-06-019110.1136/bmjpo-2025-003547Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, LiberiaRodrick Kisenge0Todd A. Florin1Cynthia G. Whitney2Michelle Niescierenko3Julia Kamara4Abraham Samma5Evance Godfrey6Chris A. Rees7Readon C. Ideh8Ye-Jeung G. Coleman-Nekar9Hussein K. Manji10Christopher R. Sudfeld11Adrianna L. Westbrook12Claudia R. Morris13Karim P. Manji14Christopher P. Duggan15Rishikesan Kamaleswaran16Department of Paediatrics and Child Health, Muhimbili University of Health and Allied Sciences, Dar es Salaam, United Republic of TanzaniaDivision of Emergency Medicine, Ann and Robert H. Lurie Childrens Hospital of Chicago, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USAFrom the Department of Medicine (D.M.T., S.M.R., T.A.J.), Emory University School of Medicine; The Office of Health Promotion and Disease Prevention (F.J.M., T.A.J.), Grady Health System; The Atlanta Veterans Affairs Medical Center (J.S.D., G.O.) and the Georgia Emerging Infections Program (D.M.T., S.M.R.,J.S.D., G.O.); Centers for Disease Control and Prevention (C.G.W.), Atlanta, GA.Harvard Humanitarian Initiative, Cambridge, Massachusetts, USADepartment of Pediatrics, John F. Kennedy Medical Center, Monrovia, LiberiaDepartment of Paediatrics and Child Health, Muhimbili University of Health and Allied Sciences, Dar es Salaam, United Republic of TanzaniaDepartment of Paediatrics and Child Health, Muhimbili University of Health and Allied Sciences, Dar es Salaam, United Republic of TanzaniaDivision of Pediatric Emergency Medicine, Emory University School of Medicine, Atlanta, Georgia, USADepartment of Pediatrics, John F. Kennedy Medical Center, Monrovia, LiberiaDepartment of Pediatrics, John F. Kennedy Medical Center, Monrovia, LiberiaAccident and Emergency Department, Aga Khan Health Services, Dar es Salaam, TanzaniaDepartments of Nutrition and Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USAPediatrics Biostatistics Core, Department of Pediatrics, Emory University, Atlanta, Georgia, USADivision of Pediatric Emergency Medicine, Emory University School of Medicine, Atlanta, Georgia, USADepartment of Pediatrics and Child Health, Muhimbili University of Health and Allied Sciences, Dar es Salaam, United Republic of TanzaniaDepartments of Nutrition and Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USADivision of Translational Biomedical Informatics, Department of Biostatistics and Bioinformatics, Duke University, Durham, North Carolina, USABackground The time after hospital discharge carries high rates of mortality in neonates and young children in sub-Saharan Africa. Previous work using logistic regression to develop risk assessment tools to identify those at risk for postdischarge mortality has yielded fair discriminatory value. Our objective was to determine if machine learning models would have greater discriminatory value to identify neonates and young children at risk for postdischarge mortality.Methods We conducted a planned secondary analysis of a prospective observational cohort at Muhimbili National Hospital in Dar es Salaam, Tanzania and John F. Kennedy Medical Center in Monrovia, Liberia. We enrolled neonates and young children near the time of discharge. The outcome was 60-day postdischarge mortality. We collected socioeconomic, demographic, clinical, and anthropometric data during hospital admission and used machine learning (ie, eXtreme Gradient Boosting (XGBoost), Hist-Gradient Boost, Support Vector Machine, Neural Network, and Random Forest) to develop risk assessment tools to identify: (1) neonates and (2) young children at risk for postdischarge mortality.Results A total of 2310 neonates and 1933 young children enrolled. Of these, 71 (3.1%) neonates and 67 (3.5%) young children died after hospital discharge. XGBoost, Hist Gradient Boost, and Neural Network models yielded the greatest discriminatory value (area under the receiver operating characteristic curves range: 0.94–0.99) and fewest features, which included six features for neonates and five for young children. Discharge against medical advice, low birth weight, and supplemental oxygen requirement during hospitalisation were predictive of postdischarge mortality in neonates. For young children, discharge against medical advice, pallor, and chronic medical problems were predictive of postdischarge mortality.Conclusions Our parsimonious machine learning-based models had excellent discriminatory value to predict postdischarge mortality among neonates and young children. External validation of these tools is warranted to assist in the design of interventions to reduce postdischarge mortality in these vulnerable populations.https://bmjpaedsopen.bmj.com/content/9/1/e003547.full
spellingShingle Rodrick Kisenge
Todd A. Florin
Cynthia G. Whitney
Michelle Niescierenko
Julia Kamara
Abraham Samma
Evance Godfrey
Chris A. Rees
Readon C. Ideh
Ye-Jeung G. Coleman-Nekar
Hussein K. Manji
Christopher R. Sudfeld
Adrianna L. Westbrook
Claudia R. Morris
Karim P. Manji
Christopher P. Duggan
Rishikesan Kamaleswaran
Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
BMJ Paediatrics Open
title Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
title_full Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
title_fullStr Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
title_full_unstemmed Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
title_short Machine learning approaches to identify neonates and young children at risk for postdischarge mortality in Dar es Salaam, Tanzania and Monrovia, Liberia
title_sort machine learning approaches to identify neonates and young children at risk for postdischarge mortality in dar es salaam tanzania and monrovia liberia
url https://bmjpaedsopen.bmj.com/content/9/1/e003547.full
work_keys_str_mv AT rodrickkisenge machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT toddaflorin machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT cynthiagwhitney machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT michelleniescierenko machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT juliakamara machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT abrahamsamma machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT evancegodfrey machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT chrisarees machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT readoncideh machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT yejeunggcolemannekar machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT husseinkmanji machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT christopherrsudfeld machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT adriannalwestbrook machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT claudiarmorris machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT karimpmanji machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT christopherpduggan machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia
AT rishikesankamaleswaran machinelearningapproachestoidentifyneonatesandyoungchildrenatriskforpostdischargemortalityindaressalaamtanzaniaandmonrovialiberia