A novel matching approach using 911 call time and patient characteristics can effectively identify duplicate EMS responses, enabling more accurate patient-level inferences from response-based datasets with 100% sensitivity and up to 98.6% specificity.