Introduction: Finite resources limit the amount of time EMS agencies can dedicate to continuing education in pediatric emergencies. EMS instructors need effective, efficient, and affordable educational strategies for these high-risk, low frequency events. Objective: To compare the effectiveness of four training methods in management of pediatric emergencies for paramedics. Methods: A validated, performance-based, simulated clinical assessment module was used to provide a baseline measurement of paramedics’ resuscitation skills during three simulated pediatric emergencies. Educational modules were developed that targeted deficiencies identified by the baseline assessment, including advanced pediatric life support skills, airway management, use of the Broselow-Luten Tape®, pediatric drug dose calculations and drug delivery, seizure management, and trauma assessment. Paramedics from five EMS agencies in Michigan were randomized to four education intervention groups. The control group used an existing, online, continuing education course. Three experimental groups were exposed to the same content during five, one-hour sessions conducted over 2.5 years. Instruction was delivered using high-fidelity, simulated case-based training, low-fidelity simulation training, or lecture with procedural skills lab, based on group assignment. After the training, all groups were tested within 4–6 months using methods identical to baseline testing. Results: One hundred forty-seven subjects completed the study. There were no differences in baseline skill levels among the four groups. Only the low fidelity simulation training group demonstrated improvement of combined scenario scores (p = 0.0008). Scores for targeted skills improved in one scenario in the high-fidelity group, two in the low-fidelity group, one in the lecture/lab group, and none in the control group. Conclusions: Although improvements in those skills included in the training were found in three groups, two hours of training in pediatric emergencies per year was insufficient to produce a substantial improvement overall. Expensive, high-fidelity simulators were not necessary for teaching pediatric resuscitation skills to paramedics; instructive scenarios using low-fidelity manikins and debriefings appear to be adequate. The content delivered by an online refresher course did not provide any improvement in performance as measured by simulated, case-based assessments.