{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T18:00:16Z","timestamp":1781632816179,"version":"3.54.5"},"update-to":[{"DOI":"10.1371\/journal.pmed.1005122","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000}}],"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T00:00:00Z","timestamp":1780963200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000865","name":"Bill and Melinda Gates Foundation","doi-asserted-by":"publisher","award":["OPP1139557"],"award-info":[{"award-number":["OPP1139557"]}],"id":[{"id":"10.13039\/100000865","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001602","name":"Science Foundation Ireland","doi-asserted-by":"publisher","award":["SFI\/21\/FIP\/SDG\/9948"],"award-info":[{"award-number":["SFI\/21\/FIP\/SDG\/9948"]}],"id":[{"id":"10.13039\/501100001602","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosmedicine.org"],"crossmark-restriction":false},"short-container-title":["PLoS Med"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Globally, pneumonia remains the single biggest cause of mortality in children under 5 years of age. This study sought to train and test a prediction model for hospitalisation within 7 days after initial presentation in 2- to 59-month-old Malawian children with WHO-defined pneumonia in primary care and compare its performance to existing risk prediction models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods and findings<\/jats:title>\n                    <jats:p>BIOTOPE is a cohort study of children with pneumonia in a primary healthcare setting in Malawi. The training cohort involved nine primary care centres and the testing cohort involved two primary care centres in Northern Malawi. The training cohort was recruited between December 2022 and April 2023 while the testing cohort was recruited in 2016. Participants were consecutive children aged 2\u201359 months presenting with cough and\/or difficulty breathing and who were diagnosed as WHO-defined pneumonia in primary care of any severity. The training cohort was used to train and validate a machine learning model with a prespecified primary outcome defined as hospitalisation and\/or death within 7 days as the outcome. This model was then further evaluated in the testing cohort.<\/jats:p>\n                    <jats:p>Median age was 15 months (interquartile range 8\u221227) in the training and 17 months (interquartile range 9\u221229) in the external testing cohort (52.1% and 54.4% male, respectively). Hospitalisation occurred in 14.3% (294) of the training cohort and 12.1% (55) of the testing cohort. There was one death in the training cohort only. WHO danger signs were present in 17.6% (360) and 15.9% (70) of children in the training and testing cohorts, respectively. The optimal machine learning model achieved an area under the receiver operating characteristic and precision recall curves of 0.87 and 0.57, respectively, in the testing cohort outperforming existing risk prediction models; furthermore, this model produced an expected calibration error of 0.16 (a logistic regression model using severity status as the response variable and the log odds of the machine learning model\u2019s calibrated probabilities produced an intercept estimate of \u22120.32 and a slope estimate of 1.13). Key limitations include the use of hospitalisation and\/or death as a severity outcome, which may reflect health system factors rather than true disease severity, that mortality-based comparisons were not possible due to low mortality in these primary care cohorts, and that comparator tools were developed for hospital populations rather than primary care populations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This machine learning score outperformed traditional pneumonia risk scores in predicting hospitalisation within 7 days in Malawian children presenting to primary care. Traditional pneumonia risk scores diminish in performance when externally applied to new datasets suggesting they may not generalise well beyond their original derivation settings. Mortality-related findings are not applicable as there was only one death in this cohort. Overall these findings support the potential of machine learning to meaningfully improve early identification of children at risk of severe pneumonia in low-resource primary care settings. Further external validation and clinical impact studies are needed to confirm these results.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pmed.1005122","type":"journal-article","created":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T17:40:47Z","timestamp":1781026847000},"page":"e1005122","update-policy":"https:\/\/doi.org\/10.1371\/journal.pmed.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Prediction of hospitalisation in young children with pneumonia in Malawi: A machine learning-based approach"],"prefix":"10.1371","volume":"23","author":[{"given":"Patrick","family":"Staunton","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7864-7320","authenticated-orcid":true,"given":"Mohammad Adib","family":"Makrooni","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8301-6184","authenticated-orcid":true,"given":"Master","family":"Chisale","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Billy","family":"Nyambolo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6155-9340","authenticated-orcid":true,"given":"Joseph","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-2632-0332","authenticated-orcid":true,"given":"Damien","family":"McCarthy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7391-2949","authenticated-orcid":true,"given":"Mark","family":"Ledwidge","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9720-5699","authenticated-orcid":true,"given":"Yasir","family":"Bin Nisar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chris","family":"Watson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4705-9064","authenticated-orcid":true,"given":"Balwani","family":"Mbakaya","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0496-4029","authenticated-orcid":true,"given":"Cathal","family":"Seoighe","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5564-2890","authenticated-orcid":true,"given":"Joe","family":"Gallagher","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,9]]},"reference":[{"issue":"10303","key":"pmed.1005122.ref001","doi-asserted-by":"crossref","first-page":"870","DOI":"10.1016\/S0140-6736(21)01207-1","article-title":"Global, regional, and national progress towards Sustainable Development Goal 3.2 for neonatal and child health: all-cause and cause-specific mortality findings from the Global Burden of Disease Study 2019","volume":"398","author":"GBD 2019 Under-5 Mortality Collaborators","year":"2021","journal-title":"Lancet"},{"key":"pmed.1005122.ref002","unstructured":"WHO, Department of Child and Adolescent Health and Development, UNICEF. Handbook IMCI integrated management of childhood illness [Internet]. 2014. Available from: https:\/\/www.who.int\/publications\/m\/item\/integrated-management-of-childhood-illness---chart-booklet-(march-2014)"},{"key":"pmed.1005122.ref003","unstructured":"World Health Organization. Integrated management of childhood illness: management of the sick young infant aged up to 2 months. IMCI chart booklet. 2019."},{"key":"pmed.1005122.ref004","unstructured":"World Health Organisation. Caring for the sick child in the community; 2011."},{"issue":"1","key":"pmed.1005122.ref005","article-title":"Appropriateness of clinical severity classification of new WHO childhood pneumonia guidance: a multi-hospital, retrospective, cohort study","volume":"6","author":"Clinical Information Network Author Group","year":"2018","journal-title":"Lancet Glob Health"},{"issue":"1","key":"pmed.1005122.ref006","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1001\/jama.284.1.79","article-title":"Users\u2019 guides to the medical literature: XXII: how to use articles about clinical decision rules","volume":"284","author":"TG McGinn","year":"2000","journal-title":"JAMA"},{"key":"pmed.1005122.ref007","article-title":"Predicting escalation of care for childhood pneumonia using machine learning: retrospective analysis and model development","volume":"6","author":"O Serin","year":"2025","journal-title":"JMIRx Med"},{"issue":"5","key":"pmed.1005122.ref008","doi-asserted-by":"crossref","first-page":"1256","DOI":"10.1002\/ppul.26897","article-title":"Machine learning models to evaluate mortality in pediatric patients with pneumonia in the intensive care unit","volume":"59","author":"S-R Lin","year":"2024","journal-title":"Pediatr Pulmonol"},{"issue":"1","key":"pmed.1005122.ref009","article-title":"Evaluation of the need for intensive care in children with pneumonia: machine learning approach","volume":"10","author":"Y-C Liu","year":"2022","journal-title":"JMIR Med Inform"},{"issue":"7","key":"pmed.1005122.ref010","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2020-046633","article-title":"Aetiology and severity of childhood pneumonia in primary care in Malawi: a cohort study","volume":"11","author":"J Gallagher","year":"2021","journal-title":"BMJ Open"},{"key":"pmed.1005122.ref011","volume-title":"Revised WHO classification and treatment of pneumonia in children at health facilities: evidence summaries","author":"World Health Organisation","year":"2014"},{"issue":"1","key":"pmed.1005122.ref012","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0027793","article-title":"Development of the Respiratory Index of Severity in Children (RISC) score among young children with respiratory infections in South Africa","volume":"7","author":"C Reed","year":"2012","journal-title":"PLoS One"},{"key":"pmed.1005122.ref013","article-title":"Causes of severe pneumonia requiring hospital admission in children without HIV infection from Africa and Asia: the PERCH multi-country case-control study","author":"Pneumonia Etiology Research for Child Health Study G","year":"2019","journal-title":"Lancet"},{"issue":"4","key":"pmed.1005122.ref014","article-title":"Derivation and validation of a novel risk assessment tool to identify children aged 2-59 months at risk of hospitalised pneumonia-related mortality in 20 countries","volume":"7","author":"CA Rees","year":"2022","journal-title":"BMJ Glob Health"},{"key":"pmed.1005122.ref015","article-title":"TRIPOD AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","author":"GS Collins","year":"2024","journal-title":"BMJ"},{"key":"pmed.1005122.ref016","first-page":"1","article-title":"Community of practice in practice: successful implementation of integrated community health information systems.","author":"C Kanjo","year":"2022"},{"issue":"7","key":"pmed.1005122.ref017","first-page":"1","article-title":"Imputation with the R package VIM","volume":"74","author":"A Kowarik","year":"2016","journal-title":"Comput Stat"},{"key":"pmed.1005122.ref018","unstructured":"Kaplan J. fastDummies: Fast Creation of Dummy (Binary) Columns and Rows from Categorical Variables. 2020."},{"key":"pmed.1005122.ref019","first-page":"785","article-title":"XGBoost: a scalable tree boosting system.","author":"T Chen","year":"2016"},{"key":"pmed.1005122.ref020","volume-title":"xgboost: Extreme Gradient Boosting version 3.1.0.1","author":"T Chen","year":"2025"},{"key":"pmed.1005122.ref021","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pmed.1005122.ref022","volume-title":"Caret: classification and regression training","author":"M Kuhn","year":"2022"},{"issue":"11","key":"pmed.1005122.ref023","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v036.i11","article-title":"Feature selection with the boruta package","volume":"36","author":"M Kursa","year":"2010","journal-title":"J Stat Softw"},{"issue":"1","key":"pmed.1005122.ref024","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1093\/bioinformatics\/btw570","article-title":"Precrec: fast and accurate precision-recall and ROC curve calculations in R","volume":"33","author":"T Saito","year":"2017","journal-title":"Bioinformatics"},{"key":"pmed.1005122.ref025","doi-asserted-by":"crossref","first-page":"180","DOI":"10.1186\/1471-2334-11-180","article-title":"Pneumonia and poverty: a prospective population-based study among children in Brazil","volume":"11","author":"LKAM Th\u00f6rn","year":"2011","journal-title":"BMC Infect Dis"},{"key":"pmed.1005122.ref026","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1186\/2049-9957-3-14","article-title":"Impact of the socioeconomic status on the severity and outcome of community-acquired pneumonia among Egyptian children: a cohort study","volume":"3","author":"SFAH Azab","year":"2014","journal-title":"Infect Dis Poverty"},{"key":"pmed.1005122.ref027","doi-asserted-by":"crossref","DOI":"10.1101\/2021.08.06.21261706","article-title":"Clinical pneumonia in the hospitalised child in Malawi in the post-pneumococcal conjugate vaccine era: a prospective hospital-based observational study","author":"PY Iroh Tam","year":"2021","journal-title":"Pediatrics"},{"key":"pmed.1005122.ref028","doi-asserted-by":"crossref","first-page":"04062","DOI":"10.7189\/jogh.11.04062","article-title":"External validation of the RISC, RISC-Malawi, and PERCH clinical prediction rules to identify risk of death in children hospitalized with pneumonia","volume":"11","author":"CA Rees","year":"2021","journal-title":"J Glob Health"},{"issue":"3","key":"pmed.1005122.ref029","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1056\/NEJMc2104626","article-title":"The clinician and dataset shift in artificial intelligence","volume":"385","author":"SG Finlayson","year":"2021","journal-title":"N Engl J Med"},{"key":"pmed.1005122.ref030","volume-title":"Stakeholder consultative meeting on prevention and management of childhood pneumonia and diarrhoea: report","author":"WHO","year":"2021"},{"issue":"7","key":"pmed.1005122.ref031","doi-asserted-by":"crossref","first-page":"623","DOI":"10.1001\/jama.2021.24935","article-title":"Proactive vs reactive machine learning in health care: lessons from the COVID-19 pandemic","volume":"327","author":"Y Luo","year":"2022","journal-title":"JAMA"},{"key":"pmed.1005122.ref032","doi-asserted-by":"crossref","first-page":"04004","DOI":"10.7189\/jogh.16.04004","article-title":"Outcome and management of 2-59-month-old Nigerian children with chest indrawing pneumonia at primary-level healthcare facilities: a prospective cohort study","volume":"16","author":"AG Falade","year":"2026","journal-title":"J Glob Health"},{"key":"pmed.1005122.ref033","doi-asserted-by":"crossref","first-page":"04089","DOI":"10.7189\/jogh.15.04089","article-title":"Outcomes of children aged 2-59 months with chest indrawing pneumonia managed on an outpatient basis in selected primary health facilities in Zambia","volume":"15","author":"C Jacobs","year":"2025","journal-title":"J Glob Health"},{"key":"pmed.1005122.ref034","doi-asserted-by":"crossref","first-page":"04217","DOI":"10.7189\/jogh.15.04217","article-title":"Management and outcomes of chest-indrawing pneumonia among children aged 2-59 months in a programme setting in Ethiopia: a prospective observational study","volume":"15","author":"Z Tigabu","year":"2025","journal-title":"J Glob Health"},{"key":"pmed.1005122.ref035","doi-asserted-by":"crossref","first-page":"04096","DOI":"10.7189\/jogh.15.04096","article-title":"Outcome and management of children with chest indrawing pneumonia at primary health care settings in Pakistan: an observational cohort study","volume":"15","author":"ZH Suhag","year":"2025","journal-title":"J Glob Health"},{"key":"pmed.1005122.ref036","doi-asserted-by":"crossref","first-page":"178","DOI":"10.12688\/gatesopenres.13208.2","article-title":"Care-seeking patterns amongst suspected paediatric pneumonia deaths in rural Malawi","volume":"4","author":"C King","year":"2021","journal-title":"Gates Open Res"},{"key":"pmed.1005122.ref037","doi-asserted-by":"crossref","first-page":"04169","DOI":"10.7189\/jogh.15.04169","article-title":"Management of chest indrawing pneumonia in children aged 2-59 months by community-level workers compared to standard care on clinical outcomes: systematic review and meta-analysis","volume":"15","author":"BG Pathak","year":"2025","journal-title":"J Glob Health"}],"updated-by":[{"DOI":"10.1371\/journal.pmed.1005122","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000}}],"container-title":["PLOS Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1005122","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T17:43:31Z","timestamp":1781631811000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1005122"}},"subtitle":[],"editor":[{"given":"Aaloke","family":"Mody","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,9]]},"references-count":37,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,9]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pmed.1005122","relation":{},"ISSN":["1549-1676"],"issn-type":[{"value":"1549-1676","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,9]]}}}