Previous studies of the relationship between maternal eating disorders and adverse perinatal outcomes have failed to control for familial transmission of perinatal phenotypes, which may confound the reported association. In a unique design afforded by the Norwegian Mother and Child Cohort Study and Medical Birth Registry of Norway, we linked three generations through birth register records and maternal-reported survey data to investigate whether maternal eating disorders increase risk after parsing out the contribution of familial transmission of perinatal phenotypes. The samples were 70,881 pregnancies in grandmother-mother-child triads for analyses concerning eating disorder exposure during pregnancy and 52,348 for analyses concerning lifetime maternal eating disorder exposure. As hypothesized, eating disorders predicted a higher incidence of perinatal complications even after adjusting for grandmaternal perinatal phenotypes. For example, anorexia nervosa immediately prior to pregnancy was associated with smaller birth length (relative risk = 1.62; 95% CI [1.20, 2.14]), bulimia nervosa with induced labor (relative risk = 1.21; 95% CI [1.07, 1.36]), and binge-eating disorder with several delivery complications, larger birth length (relative risk = 1.25; 95% CI [1.17, 1.34]), and large-for-gestational-age (relative risk = 1.04; 95% CI [1.01, 1.06]). Maternal pregravid body mass index and gestational weight mediated most associations. Our results support that exposure to eating disorders increases the risk for negative health outcomes in pregnant women and their babies.