Development of a prediction model for infant hospitalisation and death using clinical features assessed by community health workers during routine postnatal home visits in Dhaka, Bangladesh - Scorecard - MDSpire
Advertisement
Development of a prediction model for infant hospitalisation and death using clinical features assessed by community health workers during routine postnatal home visits in Dhaka, Bangladesh
Clinical Scorecard: Creation of a predictive model for infant hospitalization and mortality based on clinical indicators evaluated by community health workers during standard postnatal home visits in Dhaka, Bangladesh
At a Glance
Category
Detail
Condition
Infant hospitalization and mortality
Key Mechanisms
Use of clinical indicators and predictive modeling
Target Population
Young infants (<2 months) in Dhaka, Bangladesh
Care Setting
Community health worker home visits
Key Highlights
9.2% of infants experienced hospitalization or death during the study.
The best-performing Cox model included baseline covariates and visit-specific features.
Aggregative predictors did not improve prediction compared to traditional models.
Adding four visit-specific clinical features to WHO danger signs improved predictive performance.
Guideline-Based Recommendations
Diagnosis
Assess WHO danger signs during home visits to identify infants needing referral.
Management
Incorporate visit-specific clinical features alongside WHO danger signs for better prediction.
Monitoring & Follow-up
Utilize cumulative assessments from multiple visits for monitoring infant health.
Risks
Relying solely on WHO danger signs may miss cases of severe illness.
Patient & Prescribing Data
Generally healthy infants assessed from birth to 60 days of age.
Routine home visits by CHWs can reduce neonatal mortality through early identification of illness.
Clinical Best Practices
Use aggregative time-varying predictors to reflect illness trajectory.
Conduct multiple home visits for comprehensive assessment of infant health.
by Alastair Fung, Marimuthu Sappani, Cole Heasley, Chun-Yuan Chen, Shaun K Morris, Peter J Gill, Diego G Bassani, Davidson H Hamer, Prakesh S Shah, S M Abdul Gaffar, Sultana Yeasmin, Shafiqul A Sarker, Shamima Sultana, Joseph Beyene, Daniel E Roth
