We established criteria for appropriate use of the prostate-specific antige
n? (PSA) assay and used them to evaluate PSA test utilization nr 1 tertiary
care institution. During a 6-month period 2,330 PSA results were reported
for outpatients and 95 for inpatients. We reviewed medical records for a ra
ndom sample of 338 outpatient results (14.51%) and all 95 inpatient results
, of which 21% (71/338) of outpatient and 17% (16/95) of inpatient results
were inappropriate according ro our rest utilization criteria. Among outpat
ients, 52% of rests were done for screening and 19% for monitoring for canc
er recurrence. For inpatients, workup for cancer (53/95 [56%]) was the most
frequent indication for testing and screening the second (24/95 [25%]). Of
tests failing the criteria, 66 (76%) of 87 resulted from excessively frequ
ent and age-inappropriate screening. We assessed the potential effect on cl
inical outcome if these tests were not performed. Of the 87 rests considere
d inappropriate, only I test result influenced clinical management for pati
ents younger than 75 years. By instituting simple limits on age and frequen
cy, we estimate that 74% (64/87) of the inappropriate tests could have been
eliminated.