Nee Chen Khoo1, Martin Duffy. 1. University of New South Wales, St Vincent's Hospital, Sydney, New South Wales, Australia. khoonc@ozemail.com.au
Abstract
OBJECTIVE: To assess the accuracy of 'out of hours' (17.00-08.00 hours) non-contrast head computed tomography (NCHCT) scan interpretation by senior ED medical staff. METHODS: The study was undertaken in an inner-city tertiary hospital in Sydney. A sample of 315 consecutive 'out of hours' NCHCT scans was compiled from a log book in the Medical Imaging Department and randomized into 21 sets of 15 scans. The scans were reviewed by emergency physicians and registrars recruited for the study and their interpretations were compared with the formal radiologist report. RESULTS: Seven emergency physicians and 14 registrars took part in the study. Two hundred and eighty-seven ED interpretations were used in the 2 x 2 table analysis. The overall observed agreement was 0.67 (95% confidence interval [CI] 0.61-0.72), sensitivity 0.57 (95% CI 0.45-0.69), specificity 0.70 (95% CI 0.64-0.76) and kappa 0.24 (95% CI 0.13-0.36). There were 32 false negative interpretations including three small contre coup contusions and three subdural haemorrhages. CONCLUSION: 'Out of hours' NCHCT scan interpretation by senior ED medical staff is only correct two-thirds of the time. Further education for all senior ED staff is indicated to improve our accuracy. The safety of NCHCT scan interpretation by senior ED medical staff needs further study.
OBJECTIVE: To assess the accuracy of 'out of hours' (17.00-08.00 hours) non-contrast head computed tomography (NCHCT) scan interpretation by senior ED medical staff. METHODS: The study was undertaken in an inner-city tertiary hospital in Sydney. A sample of 315 consecutive 'out of hours' NCHCT scans was compiled from a log book in the Medical Imaging Department and randomized into 21 sets of 15 scans. The scans were reviewed by emergency physicians and registrars recruited for the study and their interpretations were compared with the formal radiologist report. RESULTS: Seven emergency physicians and 14 registrars took part in the study. Two hundred and eighty-seven ED interpretations were used in the 2 x 2 table analysis. The overall observed agreement was 0.67 (95% confidence interval [CI] 0.61-0.72), sensitivity 0.57 (95% CI 0.45-0.69), specificity 0.70 (95% CI 0.64-0.76) and kappa 0.24 (95% CI 0.13-0.36). There were 32 false negative interpretations including three small contre coup contusions and three subdural haemorrhages. CONCLUSION: 'Out of hours' NCHCT scan interpretation by senior ED medical staff is only correct two-thirds of the time. Further education for all senior ED staff is indicated to improve our accuracy. The safety of NCHCT scan interpretation by senior ED medical staff needs further study.
Authors: Zohair Al Aseri; Mohamed Al Aqeel; Badr Aldawood; Fahad Albadr; Rawan Ghandour; Abdulaziz Al Mulaik; Mohammed A Malabarey; Anas Khan Journal: Comput Intell Neurosci Date: 2022-09-26