Objective To assess trends in patent ductus arteriosus (PDA) management and examine concurrent changes in neonatal mortality and morbidities. Methods This retrospective observational study examined infants born at 23 to 32 weeks' gestational age with PDA and admitted to a neonatal unit during 2006 to 2012. Multivariable logistic regression assessed trends in yearly PDA treatment rates and compared a composite outcome of mortality or any severe morbidity (bronchopulmonary dysplasia, intraventricular hemorrhage, periventricular leukomalacia, retinopathy of prematurity, or necrotizing enterocolitis) between and within time periods and PDA treatments. Results Study subjects included 5,824 preterm neonates with clinical/echocardiographic PDA diagnosis. During 2006 to 2012, conservative management increased (14–38%), whereas pharmacotherapy-only (58–49%), surgical ligation-only (7.1–2.5%), and both pharmacotherapy and surgical ligation (21–10%) decreased (p-values <0.01). From 2006 to 2008 and 2009 to 2012, the composite outcome decreased for infants managed conservatively (AOR = 0.70, 95% CI 0.52–0.92), with no changes detected for pharmacotherapy and/or ligation. Lower composite outcome after conservative management versus pharmacotherapy-only during 2009 to 2012 (AOR = 0.61, 95% CI 0.51–0.74), but not during 2006 to 2008 reflect significant effect modification by time period. Conclusion In Canada, during 2006 to 2012, conservative PDA management increased while pharmacotherapy and/or surgical ligation decreased. Lower composite outcome was detected during later years after increases in conservative management; however, bias due to unmeasured confounders remains possible.