The American journal of emergency medicine
-
The objective of this study was to determine if a neural network model can identify victims of intimate partner violence (IPV). A custom neural network model was constructed and trained using the 1995 ED databases at Truman Medical Center of all female visits. The input vector developed was an array of 100 binary elements containing, in coded form, the patient's age, day of week, primary diagnosis (excluding 995.81), disposition, race, time, and E-code. ⋯ The neural network identified 231 of 297 known IPV victims (sensitivity 78%) in the 1996 database. It also categorized 2234 false-positive patients out of 19,533 IPV-negative patients (specificity 89%). A computer-based neural network model, when supplied with information commonly available in the ED medical record, can identify victims of IPV.
-
It has been proposed that the workup of suspected subarachnoid hemorrhage should begin with lumbar puncture (LP) rather than computed tomography (CT) scan. We investigated whether EPs would in fact advocate this strategy in an index hypothetical case and in variations of the index case. An eight-question survey was distributed to EM physicians attending national continuing medical education meetings. ⋯ The "LP first" strategy for workup of subarachnoid hemorrhage was rejected by most EM physicians except when the CT scan would entail delay. Compared with the index case, an added number of physicians would advocate this strategy if the patient was uninsured. The ethical implication of advocating a strategy because of financial concerns and that most physicians would not favor for themselves or their patients merits attention.
-
Urolithiasis is not a frequent diagnosis in school-aged children. The clinical presentation can lack many of the clinical clues such as flank pain that are seen in older patients. We present four cases demonstrating this potential diagnostic dilemma.
-
The objective of this study was to determine the rate of positive ED diagnostic evaluations and significant interventions during the hospitalization of infants after an apparent life-threatening event (ALTE). The study was performed at a single, tertiary care children's hospital. Patients under 6 months of age were identified for a potential ALTE from the ED chief complaint log. ⋯ Risk factors for significant medical interventions included prematurity, a positive medical history, and age >60 days. The overall rate of either positive ED diagnostic evaluations or significant medical interventions during hospitalizations of infants after an ALTE is low. A majority of these patients can be best managed with a limited ED diagnostic evaluation and a period of observation.
-
Letter Case Reports
Descending necrotizing mediastinitis from upper respiratory infection.