Objective: To investigate whether obstetric and perinatal outcomes in pregnancies differ after different frozen embryo transfer (FET) protocols. Design: Register-based cohort study. Setting: Not applicable. Patient(s): All singleton deliveries after assisted reproductive technology in Denmark from 2006 to 2014. Data consisted of 1,136 deliveries after frozen in vitro fertilization. Frozen embryo transfer cycles were grouped by type of FET protocol: programmed FET (n = 357); modified natural cycle FET (n = 611); and true natural cycle FET (n = 168). Intervention(s): None. Main Outcome Measure(s): Obstetric outcomes (hypertensive disorders in pregnancy, preterm prelabor rupture of membranes, placenta previa, placental abruption, induction of labor, postpartum hemorrhage, and cesarean section) and perinatal outcomes (post-term birth, preterm birth, birth weight, small for gestational age, large for gestational age). Result(s): The risk of hypertensive disorders in pregnancy, postpartum hemorrhage, and cesarean section was significantly higher after programmed FET compared with natural cycle FET (modified natural cycle FET and true natural cycle FET). A higher risk of birth weight > 4,500 g was observed in the programmed FET group compared with natural cycle FET. Conclusion(s): This study shows that obstetric and perinatal outcomes are adversely affected in programmed FET cycles. Hence, when possible, an endometrial preparation with the creation of a corpus luteum should be considered. Properly sized randomized controlled trials of FET in programmed cycle versus natural cycle including perinatal outcomes are warranted in the future. Clinical Trial Registration Number: ISRCTN11780826.