To determine the adequacy of hospital discharge data for describing the nature and severity of multiple trauma, injury data coded from full hospital records were compared with injury data coded from discharge summaries for 83 plane crash survivors admitted to 14 different hospitals.
Full hospital records indicated that 33% sustained intracranial, 39% thoracic, and 27% abdominal/pelvic injuries. Thirty-five percent had spinal fractures, 40% upper limb fractures, and 70% lower limb fractures. Discharge summary data missed 52% of the intracranial injuries, 28% of the thoracic injuries, and 30% of the spinal fractures, yet only missed 5% of the lower limb and 15% of the skull, pelvic, and upper limb fractures.
These differences point to concerns regarding the use of discharge data for determining the nature, severity, and sequelae of injuries. The observed underreporting of neurological injury suggests a need to re-evaluate discharge data systems for measuring brain and spinal cord injury patterns, particularly given the increasing reliance on these systems for health policy and programmatic decision making.
From the Brookhaven National Laboratory, Upton, New York (C.F.K., H.C.T, C.J.K., J.I.B.), School of Public Health, SUNY Albany, New York (C.F.K., J.I.B.), Health Sciences Center, SUNY Stony Brook, New York (H.C.T), and SUNY Stony Brook, New York (J.I.B.).
Part of this work was conducted by the senior author in fulfillment of her Master's degree at SUNY Stony Brook, School of Allied Health Professions, and was awarded a Top Student Prize Paper for the 1992 competition sponsored by the Southern California Injury Prevention Research Center. This work also was presented in part at the 121st Annual APHA meeting, Oct. 27, 1993, San Francisco, California.
Address for reprints: Caroline F. Kramer, RN, MS, Epidemiology Program, Medical Department, Brookhaven National Laboratory, Upton, NY 11973.