Objectives To evaluate national trends in electrocardiogram (EKG) monitoring in Veterans Affairs (VA) beneficiaries prescribed high-dose citalopram before and after the 2011-12 Food and Drug Administration (FDA) safety warnings. Design Interrupted time-series analyses. Setting National VA healthcare system data linked to Medicare data for veterans dually eligible for VA and Medicare services. Participants Adult VA outpatients prescribed citalopram or sertraline from February 2010 to September 2013 (N=1,068,816). Measurements EKG monitoring for VA outpatients prescribed high-dose citalopram (>40 mg/d aged ≤60, >20 mg/d aged >60) or sertraline. Results For individuals prescribed high-dose citalopram, EKG monitoring increased from 9.0% before the start of the first FDA warning to a peak of 12.6% for individuals aged 18 to 60 and from 14.0% to 19.4% for individuals aged 61 to 100. However, following the second FDA warning in 2012, EKG monitoring declined, returning to prewarning levels in both age groups. EKG monitoring did not increase in individuals with a history of previous cardiac risk factors prescribed high-dose citalopram in either age group. Conclusions EKG frequency did not significantly change in individuals with cardiac risk factors at greatest potential risk for QT prolongation. Lack of responsiveness to the FDA warnings may be due to many factors, including lack of clarity about which individuals should undergo EKG monitoring, provider substitution of alternative antidepressants for citalopram, conflicting evidence regarding risk for adverse cardiac events with high-dose citalopram use, and lack of provider knowledge regarding the warnings.