Background: We hypothesized that trauma patients could be discharged safely from the emergency department (ED) before the availability of official readings for their radiologic examinations. We also sought to determine whether trauma patients were more prone to alterations of preliminary interpretations than other ED patients.
Methods: Alterations of preliminary readings (PR) for patients discharged from the ED were reviewed. If the official readings conflicted with the PR used for the patient’s disposition, attempts were made to contact the patient and provide the appropriate follow-up. Data recorded included the type of radiographic examination, the presence of a missed injury, and the follow-up. By using institutional data, the incidence of inaccurate PR were compared for trauma patients and other ED patients (χ2 test, Fisher exact test, p < 0.05).
Results: Between January of 1998 and December of 1998, 102 of 38,260 discharged ED patients had official readings differing from PR. Forty-three of the changed readings involved 42 of the 1,073 discharged trauma patients, who were more likely to harbor inaccurate PR (<0.0001) than other discharged ED patients. Twenty-eight altered readings involved plain films and 15 involved computed tomographic scans. The most common altered readings involved computed tomographic scans of the head and face (n = 13). Twelve missed injuries were detected, most commonly related to a missed injury of the extremity (7 cases). Nine other cases involved the detection of incidental pathologic conditions. Eight patients required repeat ED visits for clinical and radiographic evaluation, and one patient required subsequent hospital admission.
Conclusion: Discharged trauma patients are more likely to harbor alterations of preliminary interpretations than other ED patients. Although the official readings for these trauma patients will occasionally reveal previously undetected pathologic conditions, the majority of such cases can be managed with outpatient follow-up.
Higher hospital operating costs and decreased reimbursements have affected the care of hospitalized patients in recent years. To maintain financial viability, all hospital services have scrutinized ways to decrease operating costs. Perhaps the most common method of realizing these aims has been to decrease the number of total hospital beds and maximize the capacity of the remaining beds. In addition, hospitals have also promoted an increase in outpatient facilities. Consequently, the total number of hospital admissions has decreased and many patients who might have been admitted previously for observation or work-up now are being discharged from the emergency department (ED).
One group of patients whose admissions in many centers have decreased has been trauma patients. These decreasing admissions have resulted from decreased crime in certain areas, changing population demographics, 1 and fewer admissions strictly for observation. These trauma patients may receive an initial ED work-up consisting of an abbreviated history and physical along with selective diagnostic tests consisting of radiographic and laboratory examinations. The laboratory tests have variable turnover times and the radiology films are usually read by a resident or fellow before review by an attending radiologist. If the patients seem to harbor no serious injuries and the diagnostic tests are negative, they are often discharged from the ED. Frequently, these discharged patients actually leave the ED before the availability of official readings by an attending radiologist. We hypothesized that these patients could be discharged safely from the ED without having documented final readings, but solely the preliminary interpretation of a nonattending radiologist. We sought to confirm this hypothesis by a review of the data of our medical center. We also sought to determine whether incorrect preliminary readings of radiographic studies were more likely to occur in trauma patients compared with other ED patients.
From the Departments of Surgery (S.R.E., E.F., L.J.H., P.S.B.), and Emergency Medicine (N.F., C.S.), Weill Medical College of Cornell University, New York, New York.
Submitted for publication November 2, 1999.
Accepted for publication January 5, 2000.
Poster presentation at the 59th Annual Meeting of the American Association for the Surgery of Trauma, September 16–18, 1999, Boston, Massachusetts. Address for reprints: Soumitra R. Eachempati, MD, 525 E. 68th Street, Room P 711A, New York Presbyterian Hospital, New York, NY 10021.