P2B108: Identifying Neonates at Risk for Post-discharge Mortality in Dar Es Salaam, Tanzania and Monrovia, Liberia: Derivation and Internal Validation of a Novel Risk Assessment Tool

Background: Neonatal mortality rates are as high as 20/1,000 live births in Tanzania in East Africa and 30/1,000 live births in the West African nation of Liberia. The time after discharge from a neonatal ward is increasingly recognized as a vulnerable time that carries a large burden of mortality in sub-Saharan Africa. Our objective was to derive and internally validate a risk assessment tool to identify neonates at risk for all-cause mortality within 60 days of discharge from the neonatal wards of two national referral hospitals in sub-Saharan Africa.

Methods: We conducted a prospective observational cohort study of neonates who were discharged from the neonatal wards at Muhimbili National Hospital in Dar es Salaam, Tanzania and John F. Kennedy Medical Center in Monrovia, Liberia. Research staff made phone calls to caregivers of all enrolled neonates to ascertain their vital status at 7, 14, 30, 45, and 60 days after discharge. Socioeconomic, demographic, clinical, and anthropometric data were collected during hospital admission. We included candidate variables with P values < 0.20 in bivariate analyses in a multivariable logistic regression model with best subset selection to identify risk factors associated with all-cause, 60-day post-discharge mortality. We used the model with the lowest Akaike Information Criterion and Bayesian Information Criterion. We calculated adjusted log coefficients for each candidate variable and assigned weighted points to derive a risk assessment tool to identify infants and children at risk for post-discharge mortality. We used bootstrap validation with 500 repetitions to internally validate our risk assessment tool.

Results: There were 2,344 neonates discharged from the neonatal wards of the two included hospitals, 2,310 (98.5%) enrolled, and 2,294 (99.3%) had 60-day outcomes available. The median age (interquartile range) at discharge was 8 (4, 15) days and 1,238 (54%) were male. In total, 71 (3.1%) died within 60 days of discharge. While leaving against medical advice and meconium aspiration were most associated with 60-day mortality, nine variables contributed to the risk assessment tool for a total possible score of 63 (Table). A score of ≥6 demonstrated the most optimal cut point to identify neonates at risk for post-discharge mortality. This risk assessment tool demonstrated good discriminatory value (optimism corrected area under the receiver operating characteristic curve 0.77; Figure) and excellent calibration.

Conclusion: Our risk assessment tool accurately predicted all-cause, 60-day post-discharge mortality among neonates discharged from neonatal wards of referral hospitals in Tanzania and Liberia. After external validation, this risk assessment tool has the potential to facilitate clinical decision making focused on eligibility for discharge and the direction of resources to follow up high risk neonates with non-modifiable risk factors.