Injury and violence and the relationship to prematurity or low birth weight : a pilot study

This study examined the relationship between injury/violence and prematurity or low birth weight. This relationship was examined after controlling for confounding variables that have been linked with birth outcome. A survey was administered to women, 18 years or older, who had a single, live birth to an infant born preterm or LBW at Gundersen Lutheran in 2008. Two controls were matched on age, parity/gravid, date of delivery, and county of residence with each case. The distributions of the cases and controls were compared using chi-square and ANOVA. To evaluate the association of several potential risk factors with birth outcome, a matched analysis was carried out using conditional logistic regression. Analyses were carried out with SAS version 9.1.3. The response rate was 46%, resulting in 18 matches. In the univariate models, education, healthcare payment, WIC status, weight gain, any stress, health complication, and self harm was statistically different between cases and controls. The multivariate models indicated that the odds ratios of injury/violence were not statistically significant. Cases had a 1.8 times greater odds of injury than their matched controls