Diagnostic accuracy of surgeons and trainees in assessment of patients with acute abdominal pain.

Abstract

Diagnostic accuracy in the assessment of patients with acute abdominal pain in the emergency ward is not adequate. It has been argued that this is because the investigations are carried out predominantly by a trainee. Resource utilization could be lowered if surgeons had a higher initial diagnostic accuracy. Patients with acute abdominal pain were included in a prospective cohort study. A surgical trainee and a surgeon made independent assessments in the emergency department, recording the clinical diagnosis and proposed diagnostic investigations. A reference standard diagnosis was established by an expert panel, and the proportion of correct diagnoses was calculated. Diagnostic accuracy was expressed in terms of sensitivity, specificity, positive predictive value and negative predictive value. Interobserver agreement for the diagnosis and elements of history-taking and physical examination were expressed by means of Cohen's κ. Certainty of diagnosis was recorded using a visual analogue scale. A trainee and a surgeon independently assessed 126 patients. Trainees made a correct diagnosis in 44·4 per cent of patients and surgeons in 42·9 per cent (P = 0·839). Surgeons, however, recorded a higher level of diagnostic certainty. Diagnostic accuracy was comparable in distinguishing urgent from non-urgent diagnoses, and for the most common diseases. Interobserver agreement for the clinical diagnosis varied from fair to moderate (κ = 0·28-0·57). The diagnostic accuracy of the initial clinical assessment is not improved when a surgeon rather than a surgical trainee assesses a patient with abdominal pain in the emergency department.