The primary aim of this study was to examine whether incarceration during pregnancy is associated with infant birthweight. Our second objective was to illustrate the sensitivity of the relationship between infant birthweight and exposure to prison during pregnancy to the method used to measure and model this exposure. The data consisted of delivery records of 360 infants born between January 1, 2002 and December 31, 2004 to pregnant women incarcerated in Texas state prisons. Weighted linear regression, adjusting for potential confounders, was used to model infant birth weight as a function of: (1) the number of weeks of pregnancy spent incarcerated (Method A) and (2) the gestational age at admission to prison (Method B), respectively. These two exposure measures were modeled as continuous variables with and without linear spline transformation. The association between incarceration during pregnancy and infant birthweight appears strongest among infants born to women incarcerated during the first trimester and very weak to non-existent among infants born to women incarcerated after the first trimester. With Method A, but not Method B, linear spline transformation had a distinct effect on the shape of the relationship between exposure and outcome. The association between exposure to prison during pregnancy and infant birth weight appears to be positive only among women incarcerated during the first trimester of pregnancy and the relation is sensitive to the method used to measure and model exposure to prison during pregnancy.