Technology assessment in diagnostic imaging. A proposal for a phased approach to evaluating radiology research

The authors propose an objective basis for critical evaluation of research trends and define and analyze a sample of radiology studies according to research phase.

A random sample of 146 original diagnostic studies from two radiology journals was categorized according to phase, modality, and design by three physician reviewers, collated with a microcomputer database, and analyzed using an SAS program.

Phase 1 studies (technical evaluation) constituted 18.5% of publications: phase 2 (standardization and tissue characterization), 10.3%; phase 3 (spectrum of appearances), 40.4%; phase 4 (diagnostic efficacy), 21.2%; and phase 5 (clinical evaluation), 9.6%. Of 48 diagnostic efficacy studies, 42% were prospective (versus 35% for the total sample), 38% were controlled (median sample size, 53 [versus 30 for the total sample]). Only 27% of the 48 diagnostic efficacy studies were externally funded. Research in magnetic resonance imaging (MRI), which comprised 45% of all publications, was oriented toward phase 1 (32%) rather than phase 5 studies (0%). Phase 5 studies were the focus of 18% and 8% of ultrasound (US) and computed tomography (CT) studies, respectively. There were more prospective, controlled efficacy studies in US than in MRI or CT.

Analyses of research trends will be facilitated by use of a standard taxonomy which adopts a modality-based, phased approach.