Statistical Modelling for the Assessment of Low Birth Weight in Tertiary Care Settings (2019-2024)

Abstract

LBW is a major public health concern worldwide, particularly in developing countries, and is defined as a birth weight of less than 2,500 grams. It is essential to properly evaluate and manage LBW infants because this practice minimizes newborn health complications. This analysis reviews the application of statistical models (2019-2024) which evaluate risk components and treatment results for LBW cases found in tertiary medical facilities. Research uses logistic regression along with machine learning models in accord to survival analysis to discover maternal indicators alongside clinical indicators & socioeconomic indicators that predict LBW. Multiple risk factors are successfully integrated through advanced learning approaches starting from classical regression methods as the review demonstrates. Findings suggest that ensemble methods and deep learning models demonstrate superior predictive performance compared to conventional statistical approaches. The studies indicate that integrating machine learning methods with traditional biostatistics offers a more nuanced understanding of LBW risk. However, the need for interpretable models in clinical settings remains paramount.