• F T de Dombal, V Dallos, and W A McAdam.
• Clinical Information Science Unit, University of Leeds.
• BMJ. 1991 Jun 22;302(6791):1495-7.

ObjectiveTo compare three methods of support for inexperienced staff in their diagnosis and management of patients with acute abdominal pain--namely, with (a) structured data collection forms, (b) real time computer aided decision support, and (c) computer based teaching packages.DesignProspective assessment of effects of methods of support on groups of doctors in one urban hospital and one rural hospital.SettingAccident and emergency department at Whipps Cross Hospital, London, and surgical wards of Airedale General Hospital, West Yorkshire.PatientsConsecutive prospective series of all patients presenting to each hospital in specified time periods with acute abdominal pain; total patients in the various periods were 12,506.Main Outcome MeasuresDiagnostic accuracy of participating doctors, admission rates of patients with non-specific abdominal pain, perforation rates in patients with appendicitis, negative laparotomy rates.ResultsUse of any one modality resulted in improved diagnostic accuracy and decision making performance. Use of structured forms plus computer feedback resulted in better performance than use of forms alone. Use of structured forms plus a computer teaching package gave results at least as good as those with direct feedback by computer.ConclusionsThe results confirm earlier studies in suggesting that the use of computer aided decision support improves diagnostic and decision making performance when dealing with patients suffering from acute abdominal pain. That use of the computer for teaching gave results at least as good as with its use for direct feedback may be highly relevant for those who are apprehensive about the real time use of diagnostic computers in a clinical setting.

19 keywords...

hide…