R J McClure1, J Burnside. 1. National Centre for Epidemiology and Population Health, Australian National University, Canberra.
Abstract
OBJECTIVE: To identify the representativeness and sensitivity of the data collected by the Australian Capital Territory Injury Surveillance and Prevention Project (ACTISPP) for ED injury cases seen in the Australian Capital Territory (ACT). METHODS: A retrospective descriptive study was conducted using a systematic sample of people injured in the ACT between January 20 and September 19, 1992, and who received their first treatment for this injury in an ACT ED. The ACTISPP surveillance system is dependent on ED patient or family completion and return of a comprehensive injury survey. The following population variables in the ACTISPP database were compared with those of the injured ED population, as determined by ED and hospital record review: age, gender, postal code of residence, nature of injury, body part injured, and whether the patient had been admitted to hospital. RESULTS: The sensitivity of the surveillance system for injuries seen in the ED was 31% (95% CI = 30-32%). The distribution of the ACTISPP sample did not differ from the expected distribution for sex or socioeconomic status. The distributions differed by age only in the 5-14-year age group (p < 0.001). With respect to the nature of injury and body part injured, the difference was significant (p < 0.001) but was no greater than 7% for any of the categories. Two percent of the surveillance sample were admitted to hospital, compared with an estimated 4% of the target population (p < 0.001). CONCLUSIONS: Approximately one-third of all incident injuries managed in the EDs of the ACT are included in the ACTISPP database. Although some sampling bias is introduced in the surveillance process, it does not appear to be of sufficient magnitude to compromise the project's goals.
OBJECTIVE: To identify the representativeness and sensitivity of the data collected by the Australian Capital Territory Injury Surveillance and Prevention Project (ACTISPP) for ED injury cases seen in the Australian Capital Territory (ACT). METHODS: A retrospective descriptive study was conducted using a systematic sample of people injured in the ACT between January 20 and September 19, 1992, and who received their first treatment for this injury in an ACT ED. The ACTISPP surveillance system is dependent on ED patient or family completion and return of a comprehensive injury survey. The following population variables in the ACTISPP database were compared with those of the injured ED population, as determined by ED and hospital record review: age, gender, postal code of residence, nature of injury, body part injured, and whether the patient had been admitted to hospital. RESULTS: The sensitivity of the surveillance system for injuries seen in the ED was 31% (95% CI = 30-32%). The distribution of the ACTISPP sample did not differ from the expected distribution for sex or socioeconomic status. The distributions differed by age only in the 5-14-year age group (p < 0.001). With respect to the nature of injury and body part injured, the difference was significant (p < 0.001) but was no greater than 7% for any of the categories. Two percent of the surveillance sample were admitted to hospital, compared with an estimated 4% of the target population (p < 0.001). CONCLUSIONS: Approximately one-third of all incident injuries managed in the EDs of the ACT are included in the ACTISPP database. Although some sampling bias is introduced in the surveillance process, it does not appear to be of sufficient magnitude to compromise the project's goals.