OBJECTIVE: There is little evidence that useful electronic data could be collected at Australian small rural emergency services. If in future their funding model changed to the Activity-Based Funding model, then they would need to collect and submit more data. We determine whether it is possible to collect episode-level data at six small rural emergency services and quantify the accuracy of eight fields. DESIGN: A prospective cross-sectional study. SETTING: South-West Victoria, Australia. PARTICIPANTS: Six small rural emergency services. INTERVENTION: We collected and audited episode-level emergency data from participating services between 1 February 2011 and 31 January 2012. A random sample of these data were audited monthly. Research assistants located at each service supported data entry and audited data accuracy for four hours per week. MAIN OUTCOME MEASURES: Rates for data completeness, accuracy and total accuracy were calculated using audit data. RESULTS: Episode-level data were collected for 20 224 presentations across six facilities. The audit dataset consisted of 8.5% (1504/17 627) of presentations from five facilities. For all fields audited, the accuracy of entered data was high (>93%).Triage category was deemed appropriate for 95.9% (95% confidence interval (CI): 94.9-96.9%) of the patient records reviewed. Some procedures were missing (28.7%, 95%CI: 27.2-30.3%). No significant improvement in data accuracy over 12 months was observed. CONCLUSION: All six services collected useful episode-level data for 12-months with four hours per week of assistance. Data entry accuracy was high for all fields audited, and data entry completeness was low for procedures.
OBJECTIVE: There is little evidence that useful electronic data could be collected at Australian small rural emergency services. If in future their funding model changed to the Activity-Based Funding model, then they would need to collect and submit more data. We determine whether it is possible to collect episode-level data at six small rural emergency services and quantify the accuracy of eight fields. DESIGN: A prospective cross-sectional study. SETTING: South-West Victoria, Australia. PARTICIPANTS: Six small rural emergency services. INTERVENTION: We collected and audited episode-level emergency data from participating services between 1 February 2011 and 31 January 2012. A random sample of these data were audited monthly. Research assistants located at each service supported data entry and audited data accuracy for four hours per week. MAIN OUTCOME MEASURES: Rates for data completeness, accuracy and total accuracy were calculated using audit data. RESULTS: Episode-level data were collected for 20 224 presentations across six facilities. The audit dataset consisted of 8.5% (1504/17 627) of presentations from five facilities. For all fields audited, the accuracy of entered data was high (>93%).Triage category was deemed appropriate for 95.9% (95% confidence interval (CI): 94.9-96.9%) of the patient records reviewed. Some procedures were missing (28.7%, 95%CI: 27.2-30.3%). No significant improvement in data accuracy over 12 months was observed. CONCLUSION: All six services collected useful episode-level data for 12-months with four hours per week of assistance. Data entry accuracy was high for all fields audited, and data entry completeness was low for procedures.