The objective of this study was to determine if ED triage nurses could appropriately interpret the Ottawa Ankle Rules (OAR). We conducted a prospective, observational trial of a clinical decision rule in a suburban ED on a convenience sample of ED patients, aged >17 years with acute ankle injuries. Nurses and EPs were trained in the appropriate use of the OAR. Nurses and physicians recorded their initial blinded patient assessments on standardized data collection instruments that included the OAR. X-rays were ordered without specific discretion to OAR by nurses or physicians. Sensitivity, specificity, negative predictive value, and positive predictive value were calculated as appropriate; kappa (k) values were calculated to assess interobserver agreement (IOA). One hundred three patients enrolled: mean age 37 +/- 16 years; 67% female; 27 had fractures. IOA between nurses and physicians was moderate for overall interpretation of OAR (kappa = 0.44). IOA (kappa) for each criterion varied from (1) moderate for fifth metatarsal pain (0.56), posterior malleolar pain (0.44), medial malleolar pain (0.40), and weight bearing with foot pain (0.48); to (2) fair for weight bearing with ankle pain (0.32) and navicular pain (0.21). Sensitivity of the nurse's interpretation of OAR for fracture was 92%, specificity 36%, negative predictive value 90%, and positive predictive value 32%. Sensitivity of the EP's utilization of the OAR for fracture was 92%, specificity 47% with a negative predictive value 94%, and a positive predictive value 38%. Nurses showed only a moderate ability to interpret the overall OAR for ordering of x-rays. Nurses' understanding of the individual criterion were variable.