Return to search

The diagnostic accuracy of reporting radiographer chest X-ray interpretations and their influence on clinicians' diagnostic decision-making : a comparison with consultant radiologists

Background: Diagnostic imaging plays an expanding and central role in patients' medical care. Radiographer clinical reporting is being increasingly used in patient focused services. There is a paucity of research that has examined radiographer chest X-ray reporting. Aim: To determine the diagnostic accuracy of reporting radiographer chest X-ray (CXR) reporting and the influence that CXR reports have on clinicians' diagnostic decision-making. Method: A quasi-experimental study determined the diagnostic accuracy of a cohort of reporting radiographers in CXR interpretation, using a free-response methodology. The influence of CXR reports on clinicians' diagnostic decision-making was determined with a cohort study. A non-inferiority approach was used, in line with Royal College of Radiologists and College of Radiographers guidance that reporting radiographers must be comparable to consultant radiologists. Results: The diagnostic accuracy of reporting radiographers (RR) was non-inferior to consultant radiologists (CR) for all measures, all p < 0.0001; unweighted JAFROC (RR Figure of Merit [FoM]=0.828, 95%CI 0.808-0.847; CR FoM=0.788, 95%CI 0.766-0.811), weighted JAFROC (RR FoM=0.830, 95%CI 0.811-0.849; CR FoM=0.786, 95%CI 0.764-0.808) and inferred ROC (RR Area Under the Curve [AUC]=0.909, 95%CI 0.887-0.931; CR AUC=0.903, 95%CI 0.882-0.924). No difference was found in the number of CXR reports that produced a correct most likely and/or most serious diagnosis (RR 876 of 1337 cases; CR 810 of 1368; p=0.103). Uncorrected most likely diagnostic confidence (RR 72.5 to 80.2; CR 71.0 to 80.4) and uncorrected most serious diagnostic confidence (RR 34.0 to 41.9; CR 33.5 to 39.2) of reporting radiographer CXR reports was non-inferior to consultant radiologists (p < 0.001). Corrected most likely diagnostic confidence, calculated using the Tsushima methodology, was lower (RR 4.61; CR 5.02) with no apparent difference, but noninferiority was not confirmed (p > 0.05). Conclusion: With appropriate postgraduate education, reporting radiographers are able to interpret chest X-rays at a level comparable to consultant radiologists.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:715970
Date January 2016
CreatorsWoznitza, Nick
PublisherCanterbury Christ Church University
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://create.canterbury.ac.uk/15847/

Page generated in 0.0101 seconds