Development of and gathering validity evidence for a theoretical test in thoracic ultrasound
Autor: | Pia Iben Pietersen, Ole Graumann, Nick A Maskell, Kristian Rørbæk Madsen, Lars Konge, Najib M. Rahman, Christian B. Laursen, Morten Bendixen |
---|---|
Rok vydání: | 2019 |
Předmět: |
Adult
Male Pulmonary and Respiratory Medicine medicine.medical_specialty Students Medical Point-of-Care Systems Delphi method Diagnostic accuracy Item discrimination Education Young Adult 03 medical and health sciences 0302 clinical medicine Cronbach's alpha Physicians Internal consistency medicine Humans Medical physics 030212 general & internal medicine Competence (human resources) Ultrasonography business.industry Internship and Residency Reproducibility of Results Medical training Middle Aged Thorax Thoracic ultrasound Ultrasound technique 030228 respiratory system Thoracic imaging Female Clinical Competence Educational Measurement business Health Physics |
Zdroj: | Pietersen, P I, Konge, L, Madsen, K R, Bendixen, M, Maskell, N A, Rahman, N, Graumann, O & Laursen, C B 2019, ' Development of and Gathering Validity Evidence for a Theoretical Test in Thoracic Ultrasound ', Respiration, vol. 98, no. 3, pp. 221-229 . https://doi.org/10.1159/000500146 |
ISSN: | 1423-0356 0025-7931 |
Popis: | Background: Thoracic ultrasound (TUS) has a high diagnostic accuracy for many common pulmonary diseases, but theoretic knowledge in sonographic physics, thoracic anatomy and physiology, and sonopathologic patterns is required to develop competence. Objectives: The aims of the study were to develop and gather validity evidence for a theoretical test in TUS and to establish a pass/fail standard. Methods: Content was provided based on expert interviews, leading to the creation of 113 initial multiple-choice question (MCQ) items. Consensus was reached on 92 proceeding items through a Delphi process, and items were presented to physicians with different knowledge and experience in TUS. Answers were used for item statistics in order to select the items with the most optimal item discrimination and difficulty (i.e., level I items) to be included in the final test. Mean scores of the novice, intermediate and experienced groups were compared, and a pass/fail score was established using the contrasting groups’ standard setting method. Results:Item statistics revealed 38 level I items, of which 30 were selected to be included in the final test. The internal consistency was high (Cronbach’s alpha = 0.88). Differences in mean scores were 8.6 points (p < 0.001), 6.3 points (p = 0.01), and 14.9 points (p < 0.001) between novices and intermediates, intermediates and experienced, and novices and experienced, respectively. A pass/fail standard of 20 points was established. Conclusion: The established MCQ test can distinguish between physicians with different levels of competence in TUS and enables an objective, evidence-based approach for assessing the theoretical knowledge of trainees undergoing an educational programme in TUS. |
Databáze: | OpenAIRE |
Externí odkaz: |