Rapid and rigorous technology evaluation is important for improving qu
ality and cost in health care, particularly for swiftly changing, high
ly technologic fields like radiology. Currently, however, evaluations
are generally seriously deficient in quality, and rigorous evaluations
typically require 4 years or more. Therefore, the authors developed a
n appropriate methodology. Its principal characteristics include study
of outcomes, clinical relevance, multiinstitutional design, intensive
communication, experienced data management and statistical centers, s
ophisticated analysis, careful attention to the protocol and reference
standard, on-site managers, and extensive pretesting and refinement.
The authors successfully tested the methodology in a seven-institution
study. Completed in 11/4 years, the study achieved active participati
on of treating physicians, which much enhanced the clinical relevance
of the end points studied. The data supported extensive analyses, whic
h included the effect of imaging on treatment plans, an important outc
ome measure. The authors report the (limited) difficulties encountered
and identify changes to ameliorate them. Thus revised, the methodolog
y can serve as a model for future technology assessments.