Background: Young graduates manning the emergency rooms in public hospitals often need guidance in diagnosing critically ill patients due to their limited clinical experience. Textbooks, manuals and several websites are of limited assistance, as they do not generate patient-specific advice. ISABEL diagnostic tool, an Internet-delivered pediatric diagnosis support system that provides such information has not been evaluated in developing countries.
Aim: To study the sensitivity of the ISABEL diagnostic tool.
Material and method: Records of patients admitted in the pediatric intensive care unit in a metropolitan hospital in India during January 2000-July 2002 were retrieved. Resident medical officers wrote key clinical and laboratory findings on the basis of admission notes and results of investigations carried out within 30 min of admission. The list of diagnoses generated by the diagnostic tool at the ISABEL site after submission of these terms was entered in a performa. The presence of final diagnosis in the list generated by the ISABEL was the outcome measure studied.
Results: Records of 200 subjects (boys 111, girls 89, aged 28 days-12 years) were analyzed. Congenital heart disease, respiratory tract infections, meningitis, tetanus and septicemia were the most frequently encountered diagnoses. The diagnostic tool missed 27 diagnoses (such as septicemia, tuberculosis and seizures) in 39 subjects providing a sensitivity of 80.5%.
Conclusion: Even without any training offered to the users, ISABEL provided a reasonable sensitivity of 80.5%. The tool holds promise of being useful in the developing countries.