Objectives: To validate death coding in a statewide hospital database, and to determine the rate of agreement between hospital ICD-9 codes and vital records "cause of death" among injured children. Methods: Utah hospital discharge records for January 1992 to November 1997 were probabilistically linked to Utah death certificates for the years 1992 -1997. Injured pediatric patients aged 1-17 years who died within one month of discharge were analyzed. Results: There were 16,249 injured pediatric patients in the hospital discharge database of which 197 were in-hospital deaths. Death certificates were matched to 18 (91%) of the patients with an in-hospital death as well as 2 patients who died after discharge. The mortality rate was slightly higher for hospital records at 1212.4 per 100,000 injured patients as compared to the death certificate rate of 1120.1 per 100,000 injured patients (p=0.06). While injury was coded as cause of death for 95% of the injured pediatric patients, only a smaller percentage (31.3%) had complete agreement on specific ICD-9 injury codes between the hospital and death files. Rates of agreement differed by cause of injury (p<0.001), with submersion and suffocations having the highest agreement rate (67.7%) and homicides having the lowest (28.6%). Conclusion: The coding of death on hospital records appears to be an accurate measure for mortality rates, but the hospital ICD-9 codes may not be reliable surrogate for cause of death.
Learning Objectives: 1. To validate injury death coding in a statewide hospital database for pediatric patients. 2. To determine the rate of agreement between hospital ICD-9 codes and vital records cause of death codes for injured children. 3. To evaluate the rate of agreement between hospital and death certificate for cause of injury
Keywords: Mortality, Reporting
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.
The 128th Annual Meeting of APHA