Randomized trial of a clinical decision support system: impact on the management of children with fever without apparent source

J Am Med Inform Assoc. Jan-Feb 2008;15(1):107-13. doi: 10.1197/jamia.M2164. Epub 2007 Oct 18.


Objective: To assess compliance with a clinical decision support system (CDSS) for diagnostic management of children with fever without apparent source and to study the effects of application of the CDSS on time spent in the emergency department (ED) and number of laboratory tests.

Design: The CDSS was used by ED nursing staff to register children presenting with fever. The CDSS identified children that met inclusion criteria (1-36 months and fever without apparent source (FWS)) and provided patient-specific diagnostic management advice. Children at high risk for serious bacterial infection were randomized for the 'intervention' (n = 74) or the 'control' (n = 90) group. In the intervention group, the CDSS provided the advice to immediately order laboratory tests and in the control group the ED physician first assessed the children and then decided on ordering laboratory tests.

Results: Compliance with registration of febrile children was 50% (683/1,399). Adherence to the advice to order laboratory tests was 82% (61/74). Children in the intervention group had a median (25(th)-75(th) percentile) length of stay at the ED of 138 (104-181) minutes. The median length of stay at the ED in the control group was 123 (83-179) minutes. Laboratory tests were significantly more frequently ordered in the intervention group (82%) than in the control group (44%, p < 0.001, chi(2) test).

Conclusion: Implementation of a CDSS for diagnostic management of young children with fever without apparent source was successful regarding compliance and adherence to CDSS recommendations, but had unexpected effects on patient outcome in terms of ED length of stay and number of laboratory tests. The use of the current CDSS was discontinued.

Publication types

  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Chi-Square Distribution
  • Child, Preschool
  • Clinical Laboratory Techniques / statistics & numerical data*
  • Decision Support Systems, Clinical*
  • Emergency Service, Hospital
  • Female
  • Fever of Unknown Origin / etiology*
  • Guideline Adherence
  • Humans
  • Infant
  • Inservice Training
  • Length of Stay
  • Male
  • Nursing Staff, Hospital
  • Patient Care Management*
  • Practice Guidelines as Topic
  • ROC Curve
  • Registries
  • Risk Assessment
  • User-Computer Interface