Background: Agitated and violent patients may present a danger to themselves and emergency medical services (EMS) providers. Behavioral emergencies may also distract from diagnosing and treating critical medical and traumatic pathologies. In order to improve patient and provider safety as well as allow EMS providers to safely conduct an accurate and thorough medical assessment, severely agitated patients frequently require chemical restraints. Antipsychotics such as haloperidol, often in combination with benzodiazepines, have long been employed for chemical restraint in many EMS services. Recently, ketamine has been added as an option. Minimizing on-scene time to facilitate rapid transport to a receiving emergency department (ED) is a key quality metric in EMS. We aimed to evaluate whether the use of ketamine for chemical restraint was associated with an increased on-scene time compared to haloperidol. Methods: Following IRB approval, the electronic prehospital care report (E-PCR) database from an urban, firebased, single-tier EMS system over a 35-month period was retrospectively reviewed. The EPCR database was queried for “ketamine” and separately for “haloperidol” under the medication administered data field. Patient demographic information (age, gender, chief complaint), on-scene time (time from patient contact to scene departure), and co-administration of haloperidol with a benzodiazepine or diphenhydramine were abstracted. Patient demographic information was compared between groups using the Student t-test (age) and chisquared (gender; chief complaint). On-scene time was compared using an unadjusted Student t-test. Results: A total of 99 cases were identified during the study period (haloperidol=59; ketamine=40). Benzodiazepineswere coadministered with haloperidol in 47/59 (80%) cases while ketamine was given as monotherapy in all cases. There were no differences between haloperidol and ketamine treatment groups in terms of age (37.6 ¡¾ 17.8 vs. 36.3 ¡¾ 12.9; p = 0.70),% male (59.3% vs. 70.0%; p= 0.28) or% of patients with psychiatric chief complaint (44.1% vs. 55.0%; p = 0.29). There was no statistical difference in on-scene time for patients receiving haloperidol compared to ketamine (18.2 ¡¾ 9.3 vs. 15.5 ¡¾ 10.4; p = 0.19). Conclusions: In this urban EMS system, the use of prehospital ketamine for chemical restraint was not associated with an increased on-scene time compared to haloperidol.