Keywords:
Musculoskeletal joint, Musculoskeletal system
Authors:
L. Braga1, J. Shah2, A. Phadtare3, M. Vaghasia3, S. Patil3, R. Pietrobon4; 1Omaha, NE/US, 2Singapore/SG, 3Dombivli/IN, 4Durham, NC/US
DOI:
10.1594/ecr2010/C-2208
Methods and Materials
This is a prospective multi-institutional survey conducted by University of Nebraska Medical Center (UNMC), University of North Carolina at Chapel Hill (UNC) and Center for Excellence in Surgical Outcomes (CESO), Duke University (DUHS) conducted from July 2008 through July 2009. Institutional review board from UNMC, UNC and DUHS institutions approved this project. Consent was waived given the anonymous survey.
Design
Potential subjects had been sent an email requesting their participation in an anonymous survey. We had invited radiologists from 50 countries for their participation, but we are unaware of how many countries actually responded to our survey due to anonymous nature of the project.
One evaluation from each reader was executed in the form of an Internet survey to access the inter-reader agreement [2]. Observers were not allowed to review other observer's ratings or to review their own after they have been completed the survey. In the invitation email, before the initiation of the survey, participants were presented with a direct link to the original article from Kellgren and Lawrence. At total, 28 participants were included in this survey, but two were excluded because of incomplete responses.
Data Storage & Confidentiality
For recruitment purposes, the only identifiable data collected on Web survey respondents were their name and email addresses. All identifiable information of survey respondents were stored electronically on password-encrypted computers belonging to the principal investigators (PI's). Only the PI's and personnel directly involved with the study have had access to this information. Once recruitment emails have been sent to potential survey respondents, all identifiable data was erased from the storage sites. Furthermore, no IP addresses will be captured when respondents enter the Web application to rate posted images. This code were be kept encrypted within the Web server and will not be released in any format to researchers or any personnel.
Imaging
Knee x-rays were provided by the Johnston County Osteoarthritis Project based on UNC Chapel Hill. The Johnston County Osteoarthritis Project is an ongoing population-based prospective study of knee and hip OA since 1991 [3]. Five cases were randomly blinded selected by a third party to be included in this project (Figures 1-5).
K/L grade was classified as 0= no OA; 1= questionable OA; 2= mild OA; 3= moderate OA; and 4= severe OA based on the presence and severity of osteophytes and joint space narrowing.
Statistical Analysis
Fleiss kappa coefficient method were performed to analyze the inter-observer agreement as follows [4]: 0.81 to 1.00 (almost perfect); 0.61 to 0.80 (substantial); 0.41 to 0.60 (moderate); 0.21 to 0.40 (fair); 0.00 to 0.20 (slight); and less than 0.00 (poor).