Low birth weight and neighbourhood of residence: a multi-level analysis

Abstract

Low birth weight (LBW) is most often measured at an individual level. However, increasingly, it has been shown that other factors, which are not directly tied to the individual, can influence low birth weight. Specifically, factors such as family income, education level, place of residence, and health care benefits may influence a child's health. Through the database available, I will test the hypothesis that low birth weight is an outcome that can be influenced by the macro-level environment. The objective of this study is to understand the relationship between neighbourhood level factors and LBW in a population of children. It is well known that individual level risk factors influence low birth weight. What is less known is the extent to which a potential determinant of LBW - such as neighbourhood of residence influences low birth weight in a birth cohort in a small prairie city. The main study question is as follows; Do factors related to neighbourhood of residence increase the risk of low birth weight children? The study was comprised of a birth cohort of 5,643 children born in 1992-1994, in Saskatoon, Saskatchewan, Canada. It was found through logistic regression models that the following variables contributed significantly to the prediction of low birth weight; sex (OR=1.75, 95% CI=1.27-2.41), financial assistance (OR=1.5, 95% CI=1.05-2.14), and gestational age (OR=85.8, 95% CI=54.02-136.35). There were also significant interactions between gestational age and parity and gestational age and stillborn births. Neighbourhood characteristics that were related to LBW (unadjusted) were; proportions of residents < grade 9 (p=0.056), dwellings owned (p=0.03), median income (p=0.018), park space (p=0.015), total number of person < $10,000/year (p