Performance and Evaluation for CT Colonography Screening

Overview

Colorectal cancer (CRC) is an important United Kingdom healthcare issue affecting 1 in 20 individuals, half of whom will die from the disease. Late presentation of CRC has a poor prognosis, whereas excellent cure rates (\>95%) are seen in those who present early. Accurate and early diagnosis of CRC is therefore crucial. In most patients this is achieved via colonoscopy, a camera test which is widely available and allows tissue samples to be taken of any abnormalities seen during the procedure. However, a non-invasive alternative is Computed Tomography Colonography (CTC) which uses X-rays to produce images of the large bowel which are then interpreted by Radiologists. CTC has high sensitivity for the diagnosis of CRC (comparable to colonoscopy) and the cancer precursor - adenomatous polyps. Unlike colonoscopy, however, there is no accreditation process for CTC and there is no infrastructure to ensure that all reporting Radiologists are able to do so adequately and, as a result, there is a wide range of diagnostic accuracy. There are no universally-accepted standards to monitor quality or assess diagnostic performance, partly because we do not know what the quality markers are and there is currently no system to quantify them. Overall, this contributes to low cancer detection rates, missed cancers and inequity for patients across the National Health Service (NHS). This study aims to assess the impact of a structured training programme with assessment and feedback on NHS radiologist performance. If the impact is positive and results in significantly improved performance, then such a scheme could be adopted into an accreditation programme for CTC in the English Bowel Cancer Screening Programme (BCSP).

CT Colonography is a test which can can detect early cancer accurately, however, unlike colonoscopy (or mammography in breast cancer screening) there is currently no formalised program to help ensure Radiologists interpret CTC accurately. Colonoscopists are highly regulated, must be accredited and have ongoing quality assurance metrics, based on evidence linking these to clinical outcomes. Conversely, for CTC in the United Kingdom, there is no accreditation process, little regulatory oversight and no universally-accepted metrics to monitor quality or diagnostic performance. Radiologists who interpret CTC images are often inexperienced; one-third have interpreted \<300 cases in total and 20% report \<100 cases per annum (the minimum recommendation). Futhermore, in clinical practice across the NHS Bowel Cancer Screening Programme (NHS BCSP) CTC may have up to a 50% lower cancer and polyp detection rate than colonoscopy. Missed abnormalities at CTC are likely to underpin this difference, as centres with highly experienced radiologists (\>1000 cases) using 3-dimensional interpretation have significantly higher detection rates. There is a Quality Assurance (QA) framework for the NHS BCSP (with an overseeing QA Committee for Radiology) but recommendations are largely based on opinion rather than evidence and any radiologist can report CTC, without accreditation; a fact which our patient representatives find extremely disappointing. Prior to wide-scale implementation of a possible accreditation programme and to justify both financial and human resource, Public Health England and the British Society of Gastrointestinal and Abdominal Radiologists have mandated that such a program is scientifically proven to improve Radiologist performance. Consequently, this project will investigate training and assessment methods to improve diagnostic accuracy and elucidate the factors associated with higher Radiologist sensitivity.