DISCREPANCIES BETWEEN SCENARIO-BASED AND ACTUAL MANAGEMENT DECISIONS IN YOUNG FEBRILE INFANTS SA Finch, RC Wasserman, J Bernzweig, EJ Slora, RH Pantell. Pediatric Research in Office Settings (PROS), Center for Child Health Research, American Academy of Pediatrics, Elk Grove Village, IL; and General Pediatrics, University of CA at San Francisco, San Francisco, CA. Presented as a platform presentation at the 2000 Pediatric Academic Societies? Annual Meeting.

OBJECTIVE: To determine the degree to which office-based practitioners who participated in the PROS study on management of young febrile infants actually managed infants as they said they would in a pre-study scenario-based questionnaire.

DESIGN/METHODS: Prior to participating in the PROS Fever Study, practitioners completed a 1995 survey, including scenario-based questions about management of a febrile six-week-old female presenting with either minimal clear rhinorrhea (MCR), right otitis media (OM), or no apparent source of infection (NSI). Between 2/95-3/98, 3066 infants <=3 months old presenting with a temperature of >=38° C were enrolled by 573 practitioners from 219 practices. Of those enrolled, 166 infants (MCR=118, OM=29, NSI=19) seen by 143 practitioners matched those described in the survey in terms of age, gender, temperature, presence of family illness, diagnosis, and ill appearance. Management was studied with respect to diagnostic testing (ie, CBC, UA/urine culture, CXR, lumbar puncture (LP)). Agreement occurred whenever a practitioner indicated on the survey that a test would be performed and subsequently performed it on a comparable patient, or when s/he indicated that a test would not be performed and subsequently did not perform it.

RESULTS: With respect to diagnostic testing, the following agreement rates were generated:

	CBC	UA	CXR	LP
MCR	55%	57%	72%	85%
OM	71%	62%	76%	76%
NSI	90%	79%	63%	58%

In analysis of the scenario-versus-actual discrepancies among the MCR and OM infants, for 3 of the 4 tests (excepting LP for MCR or CBC for OM) the majority of practitioners indicated they wouldn't perform tests and subsequently performed them. Among NSI infant discrepancies, however, nearly all practitioners indicated for all four tests that they would perform the tests and subsequently did not.

CONCLUSION: A scenario-based questionnaire failed to predict actual practitioner decisions made when managing young febrile infants. The direction of the discrepancies between what practitioners said and what they actually did was not consistent across scenarios (eg, higher actual test use in some scenarios, lower test use in others). Although scenario-based questionnaires are useful, they cannot replace studies of actual practitioner behavior.