Blinding practices during acute point-of-care ultrasound research: the BLIND-US meta-research study
- 11 November 2020
- journal article
- research article
- Published by BMJ in BMJ Evidence-Based Medicine
- Vol. 26 (3) , 110-111
Abstract
Objectives: Point-of-care ultrasound (POCUS) diagnostic accuracy research has significant variation in blinding practices. This study characterises the blinding practices during acute POCUS research to determine whether research methodology adequately reflects POCUS use in routine clinical practice.Design, settings and participants: A search for POCUS diagnostic accuracy studies published inEmergency Medicine,AnaesthesiaandCritical Carejournals from January 2016 to January 2020 was performed. Studies were included if they were primary diagnostic accuracy studies. The study year, journal impact factor, population, hospital area, body region, study design, blinding of the POCUS interpreter to clinical information, whether the person performing the POCUS scan was the same person interpreting the scan, and whether the study reported incremental diagnostic yield were extracted in duplicate by two authors. Descriptive statistics were provided and prespecified subgroup analysis was performed.Main outcome measures: The primary outcome was the number of studies that blinded the POCUS interpreter to at least some part of the clinical information. Secondary outcomes included whether the person performing the POCUS scan was the same person interpreting it and whether the study reported incremental diagnostic yield.Results: 520 abstracts were screened with 97 studies included. The POCUS interpreter was blinded to clinical information in 37 studies (38.1%), not blinded in 34 studies (35.1%) and not reported in 26 studies (26.8%). The POCUS interpreter was the same person obtaining the images in 72 studies (74.2%), different in 14 studies (14.4%) and not reported in 11 studies (11.3%). Only four studies (4.1%) reported incremental diagnostic yield for POCUS. Inter-rater reliability was moderate (k=0.64). Subgroup analysis based on impact factor, body region, hospital area, patient population and study design did not show significant differences after completing pairwise comparisons.Conclusions: Although blinding the POCUS interpreter to clinical information may be done in a perceived attempt to limit bias, this may result in accuracy estimates that do not reflect routine clinical practice. Similarly, having a different clinician perform and interpret the POCUS scan significantly limits generalisability to practice as it does not truly reflect ‘point-of-care’ ultrasound at all. Reporting incremental diagnostic yield from implementing POCUS into a diagnostic pathway better reflects the value of POCUS; however, this methodology was infrequently used.Trial registration number: The study protocol was registered on Open Science Framework (https://osf.io/h5fe7/).Keywords
This publication has 23 references indexed in Scilit:
- Reporting of imaging diagnostic accuracy studies with focus on MRI subgroup: Adherence to STARD 2015Journal of Magnetic Resonance Imaging, 2017
- Ultrasonography in the emergency departmentCritical Care, 2016
- STARD 2015: an updated list of essential items for reporting diagnostic accuracy studiesBMJ, 2015
- Point of Care Cardiac Ultrasound Applications in the Emergency Department and Intensive Care Unit - A ReviewCurrent Cardiology Reviews, 2012
- Review article: Use of ultrasound in the developing worldInternational Journal of Emergency Medicine, 2011
- Effect of Clinical Information on Diagnostic Performance in Breast SonographyJournal of Ultrasound in Medicine, 2009
- Accuracy of Diagnostic Tests Read With and Without Clinical InformationJAMA, 2004
- The Electrocardiographic Exercise Test in a Population with Reduced Workup Bias: Diagnostic Performance, Computerized Interpretation, and Multivariable PredictionAnnals of Internal Medicine, 1998
- The effect of clinical bias on the interpretation of myelography and spinal computed tomography.Radiology, 1982
- The Clinical History as a Factor in Roentgenogram InterpretationJAMA, 1963