Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach
<p><strong>Background</strong> A better understanding of which children are likely to die during acute illness will help clinicians and policy makers target resources at the most vulnerable children. We used machine learning to characterise mortality in the 30-days following admiss...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Elsevier
2023
|
| _version_ | 1797110269984374784 |
|---|---|
| author | Diallo, AH Sayeem Bin Shahid, ASM Khan, AF Saleem, AF Singa, BO Gnoumou, BS Tigoi, C Otieno, CA Bourdon, C Oduol, CO Lancioni, CL Manyasi, C McGrath, CJ Maronga, C Lwanga, C Brals, D Ahmed, D Mondal, D Denno, DM Mangale, DI Chimezi, E Mbale, E Mupere, E Mamun, GMS Ouedraogo, I Githinji, G Berkley, JA Njirammadzi, J Mukisa, J Thitiri, J Haggstrom, J Carreon, JD Walson, JL Jemutai, J Tickell, KD Shahrin, L Mallewa, M Hossain, MI Chisti, MJ Timbwa, M Mburu, M Ngari, MM Ngao, N Aber, P Harawa, PP Sukhtankar, P Bandsma, RHJ Bamouni, RM Molyneux, S Feldman, S |
| author_facet | Diallo, AH Sayeem Bin Shahid, ASM Khan, AF Saleem, AF Singa, BO Gnoumou, BS Tigoi, C Otieno, CA Bourdon, C Oduol, CO Lancioni, CL Manyasi, C McGrath, CJ Maronga, C Lwanga, C Brals, D Ahmed, D Mondal, D Denno, DM Mangale, DI Chimezi, E Mbale, E Mupere, E Mamun, GMS Ouedraogo, I Githinji, G Berkley, JA Njirammadzi, J Mukisa, J Thitiri, J Haggstrom, J Carreon, JD Walson, JL Jemutai, J Tickell, KD Shahrin, L Mallewa, M Hossain, MI Chisti, MJ Timbwa, M Mburu, M Ngari, MM Ngao, N Aber, P Harawa, PP Sukhtankar, P Bandsma, RHJ Bamouni, RM Molyneux, S Feldman, S |
| author_sort | Diallo, AH |
| collection | OXFORD |
| description | <p><strong>Background</strong> A better understanding of which children are likely to die during acute illness will help clinicians and policy makers target resources at the most vulnerable children. We used machine learning to characterise mortality in the 30-days following admission and the 180-days after discharge from nine hospitals in low and middle-income countries (LMIC).</p>
<p><strong>Methods</strong> A cohort of 3101 children aged 2–24 months were recruited at admission to hospital for any acute illness in Bangladesh (Dhaka and Matlab Hospitals), Pakistan (Civil Hospital Karachi), Kenya (Kilifi, Mbagathi, and Migori Hospitals), Uganda (Mulago Hospital), Malawi (Queen Elizabeth Central Hospital), and Burkina Faso (Banfora Hospital) from November 2016 to January 2019. To record mortality, children were observed during their hospitalisation and for 180 days post-discharge. Extreme gradient boosted models of death within 30 days of admission and mortality in the 180 days following discharge were built. Clusters of mortality sharing similar characteristics were identified from the models using Shapley additive values with spectral clustering.</p>
<p><strong>Findings</strong> Anthropometric and laboratory parameters were the most influential predictors of both 30-day and post-discharge mortality. No WHO/IMCI syndromes were among the 25 most influential mortality predictors of mortality. For 30-day mortality, two lower-risk clusters (N = 1915, 61%) included children with higher-than-average anthropometry (1% died, 95% CI: 0–2), and children without signs of severe illness (3% died, 95% CI: 2–4%). The two highest risk 30-day mortality clusters (N = 118, 4%) were characterised by high urea and creatinine (70% died, 95% CI: 62–82%); and nutritional oedema with low platelets and reduced consciousness (97% died, 95% CI: 92–100%). For post-discharge mortality risk, two low-risk clusters (N = 1753, 61%) were defined by higher-than-average anthropometry (0% died, 95% CI: 0–1%), and gastroenteritis with lower-than-average anthropometry and without major laboratory abnormalities (0% died, 95% CI: 0–1%). Two highest risk post-discharge clusters (N = 267, 9%) included children leaving against medical advice (30% died, 95% CI: 25–37%), and severely-low anthropometry with signs of illness at discharge (46% died, 95% CI: 34–62%).</p>
<p><strong>Interpretation</strong> WHO clinical syndromes are not sufficient at predicting risk. Integrating basic laboratory features such as urea, creatinine, red blood cell, lymphocyte and platelet counts into guidelines may strengthen efforts to identify high-risk children during paediatric hospitalisations.</p>
<p><strong>Funding</strong> Bill & Melinda Gates Foundation OPP1131320.</p> |
| first_indexed | 2024-03-07T07:52:34Z |
| format | Journal article |
| id | oxford-uuid:286d8b3e-a52a-4de5-8655-54be4664d8dc |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:52:34Z |
| publishDate | 2023 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:286d8b3e-a52a-4de5-8655-54be4664d8dc2023-07-31T14:53:01ZCharacterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approachJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:286d8b3e-a52a-4de5-8655-54be4664d8dcEnglishSymplectic ElementsElsevier2023Diallo, AHSayeem Bin Shahid, ASMKhan, AFSaleem, AFSinga, BOGnoumou, BSTigoi, COtieno, CABourdon, COduol, COLancioni, CLManyasi, CMcGrath, CJMaronga, CLwanga, CBrals, DAhmed, DMondal, DDenno, DMMangale, DIChimezi, EMbale, EMupere, EMamun, GMSOuedraogo, IGithinji, GBerkley, JANjirammadzi, JMukisa, JThitiri, JHaggstrom, JCarreon, JDWalson, JLJemutai, JTickell, KDShahrin, LMallewa, MHossain, MIChisti, MJTimbwa, MMburu, MNgari, MMNgao, NAber, PHarawa, PPSukhtankar, PBandsma, RHJBamouni, RMMolyneux, SFeldman, S<p><strong>Background</strong> A better understanding of which children are likely to die during acute illness will help clinicians and policy makers target resources at the most vulnerable children. We used machine learning to characterise mortality in the 30-days following admission and the 180-days after discharge from nine hospitals in low and middle-income countries (LMIC).</p> <p><strong>Methods</strong> A cohort of 3101 children aged 2–24 months were recruited at admission to hospital for any acute illness in Bangladesh (Dhaka and Matlab Hospitals), Pakistan (Civil Hospital Karachi), Kenya (Kilifi, Mbagathi, and Migori Hospitals), Uganda (Mulago Hospital), Malawi (Queen Elizabeth Central Hospital), and Burkina Faso (Banfora Hospital) from November 2016 to January 2019. To record mortality, children were observed during their hospitalisation and for 180 days post-discharge. Extreme gradient boosted models of death within 30 days of admission and mortality in the 180 days following discharge were built. Clusters of mortality sharing similar characteristics were identified from the models using Shapley additive values with spectral clustering.</p> <p><strong>Findings</strong> Anthropometric and laboratory parameters were the most influential predictors of both 30-day and post-discharge mortality. No WHO/IMCI syndromes were among the 25 most influential mortality predictors of mortality. For 30-day mortality, two lower-risk clusters (N = 1915, 61%) included children with higher-than-average anthropometry (1% died, 95% CI: 0–2), and children without signs of severe illness (3% died, 95% CI: 2–4%). The two highest risk 30-day mortality clusters (N = 118, 4%) were characterised by high urea and creatinine (70% died, 95% CI: 62–82%); and nutritional oedema with low platelets and reduced consciousness (97% died, 95% CI: 92–100%). For post-discharge mortality risk, two low-risk clusters (N = 1753, 61%) were defined by higher-than-average anthropometry (0% died, 95% CI: 0–1%), and gastroenteritis with lower-than-average anthropometry and without major laboratory abnormalities (0% died, 95% CI: 0–1%). Two highest risk post-discharge clusters (N = 267, 9%) included children leaving against medical advice (30% died, 95% CI: 25–37%), and severely-low anthropometry with signs of illness at discharge (46% died, 95% CI: 34–62%).</p> <p><strong>Interpretation</strong> WHO clinical syndromes are not sufficient at predicting risk. Integrating basic laboratory features such as urea, creatinine, red blood cell, lymphocyte and platelet counts into guidelines may strengthen efforts to identify high-risk children during paediatric hospitalisations.</p> <p><strong>Funding</strong> Bill & Melinda Gates Foundation OPP1131320.</p> |
| spellingShingle | Diallo, AH Sayeem Bin Shahid, ASM Khan, AF Saleem, AF Singa, BO Gnoumou, BS Tigoi, C Otieno, CA Bourdon, C Oduol, CO Lancioni, CL Manyasi, C McGrath, CJ Maronga, C Lwanga, C Brals, D Ahmed, D Mondal, D Denno, DM Mangale, DI Chimezi, E Mbale, E Mupere, E Mamun, GMS Ouedraogo, I Githinji, G Berkley, JA Njirammadzi, J Mukisa, J Thitiri, J Haggstrom, J Carreon, JD Walson, JL Jemutai, J Tickell, KD Shahrin, L Mallewa, M Hossain, MI Chisti, MJ Timbwa, M Mburu, M Ngari, MM Ngao, N Aber, P Harawa, PP Sukhtankar, P Bandsma, RHJ Bamouni, RM Molyneux, S Feldman, S Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title | Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title_full | Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title_fullStr | Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title_full_unstemmed | Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title_short | Characterising paediatric mortality during and after acute illness in Sub-Saharan Africa and South Asia: a secondary analysis of the CHAIN cohort using a machine learning approach |
| title_sort | characterising paediatric mortality during and after acute illness in sub saharan africa and south asia a secondary analysis of the chain cohort using a machine learning approach |
| work_keys_str_mv | AT dialloah characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT sayeembinshahidasm characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT khanaf characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT saleemaf characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT singabo characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT gnoumoubs characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT tigoic characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT otienoca characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT bourdonc characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT oduolco characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT lancionicl characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT manyasic characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mcgrathcj characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT marongac characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT lwangac characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT bralsd characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT ahmedd characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mondald characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT dennodm characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mangaledi characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT chimezie characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mbalee characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT muperee characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mamungms characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT ouedraogoi characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT githinjig characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT berkleyja characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT njirammadzij characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mukisaj characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT thitirij characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT haggstromj characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT carreonjd characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT walsonjl characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT jemutaij characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT tickellkd characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT shahrinl characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mallewam characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT hossainmi characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT chistimj characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT timbwam characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT mburum characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT ngarimm characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT ngaon characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT aberp characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT harawapp characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT sukhtankarp characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT bandsmarhj characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT bamounirm characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT molyneuxs characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach AT feldmans characterisingpaediatricmortalityduringandafteracuteillnessinsubsaharanafricaandsouthasiaasecondaryanalysisofthechaincohortusingamachinelearningapproach |