Evaluating the Influence of Colorism in the Risk of Pregnancy Loss Among Black, Hispanic, and White American Women

Abstract

Prenatal loss is a common pregnancy outcome, affecting approximately 20% of all pregnancies. While there is a vast literature on the biomedical risk factors of pregnancy loss, potential social determinants have been less extensively or systematically studied. In recent years, an important expansion of research exploring race as a social determinant of health has been a consideration of the health consequences of colorism, defined as skin tone discrimination or stratification. However, explorations of colorism and women’s reproductive health are few, and possible colorism effects in pregnancy loss have not been explored to date. In this study, I begin to fill in this gap using data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), which includes information on skin tone and pregnancy experiences of American women born 1974-1983. This study also explores possible racial differences in skin tone effects on pregnancy loss between non-Hispanic Black, non-Hispanic White, and Hispanic women. An analysis of pregnancies to Add Health participants prior to 2001-02 identifies a “light privilege” pattern in pregnancy loss, in which mothers with medium or dark brown skin were at significantly higher risks of loss than those with light brown skin. Black mothers were at a significantly higher risk of loss than White, though there is insufficient evidence to distinguish between Hispanic and White mothers. This racial effect in pregnancy loss is weakened when skin tone differences are considered. However, race does not seem to modify skin tone effects in pregnancy loss. A supplementary analysis of pregnancies narrowing focus to non-Hispanic Black and Hispanic mothers prior to 2006-08 provides less support for either skin tone or racial disparities in pregnancy loss, though it does identify maternal age as a risk factor for pregnancy loss, consistent with the biomedical literature. This analysis does not support the hypotheses that such age effects are moderated by either race or skin tone, though further research with considerably larger sample sizes is desired to evaluate the possibility of such moderation more conclusively.