article · Dutse Journal of Pure and Applied Sciences
Neonatal mortality remains a critical public health challenge in developing regions, requiring robust statistical frameworks to identify its underlying factors. Standard Poisson regression relies on the assumption of equidispersion, but actual clinical datasets often exhibit overdispersion and excess zeros. To evaluate alternatives, standard Poisson regression, zero-inflated Poisson regression, and double Poisson regression were evaluated on neonatal mortality counts alongside five independent variables representing patient diagnoses. While zero-inflated Poisson regression accounts specifically for an excess of zeros, double Poisson regression provides the flexibility to accommodate both overdispersed and underdispersed data. Assessing model fit through the Akaike Information Criterion and Bayesian Information Criterion revealed that double Poisson regression attained the lowest values, demonstrating superior performance over the alternative approaches when modelling complex neonatal mortality records.
Accurately identifying the clinical determinants of neonatal mortality requires appropriate statistical techniques. Conventional counting models often break down when medical records contain disproportionate numbers of zeros and uneven variation. Establishing that double Poisson regression provides superior accuracy allows health researchers and demographers to select more reliable models when analysing high-stakes infant survival data.
The methodology could be integrated into epidemiological software, health informatics platforms, or public health planning tools to analyse mortality risks tied to clinical diagnoses. Primary end-users include health data analysts, biostatisticians, and public health agencies. The research sits at an early, analytical stage, establishing mathematical model validity rather than delivering a commercialised software product or clinical workflow application.
AI-generated from the published abstract. Always read the original work before citing.
Neonatal mortality remains a major public health issue, particularly in developing countries, requiring robust statistical approaches to identify its determinants. This study compares Poisson Regression (PR), Zero-Inflated Poisson Regression (ZIP), and Double Poisson Regression (DPR) in modeling neonatal mortality data. The dependent variable measures neonatal deaths, while five independent variables capture patient diagnoses.Although PR assumes equidispersion, diagnostics revealed overdispersion and excess zeros, necessitating ZIP and DPR. ZIP accounts for zero inflation, while DPR flexibly handles both overdispersion and underdispersion. Model fit was evaluated using AIC and BIC, with DPR outperforming the other models by achieving the lowest values, indicating superior performance.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.4314/dujopas.v11i3d.26
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
