Kristina Klara1, Jamie E Collins1, Ellen Gurary1, Scott A Elman1, Derek S Stenquist1, Elena Losina1, Jeffrey N Katz2. 1. From the Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, and the Division of Rheumatology, Immunology and Allergy, Brigham and Women's Hospital; the Department of Biostatistics, Boston University School of Public Health; and Harvard Medical School; Boston, Massachusetts, USA.K. Klara, BS, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital; J.E. Collins, PhD, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; E. Gurary, MS, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Department of Biostatistics, Boston University School of Public Health; S.A. Elman, BA, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; D.S. Stenquist, BA, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; E. Losina, PhD, Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, Brigham and Women's Hospital, and Department of Biostatistics, Boston University School of Public Health, and Harvard Medical School; J.N. Katz, MD, MS, Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, and Division of Rheumatology, Immunology and Allergy, Brigham and Women's Hospital, and Harvard Medical School. 2. From the Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, and the Division of Rheumatology, Immunology and Allergy, Brigham and Women's Hospital; the Department of Biostatistics, Boston University School of Public Health; and Harvard Medical School; Boston, Massachusetts, USA.K. Klara, BS, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital; J.E. Collins, PhD, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; E. Gurary, MS, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Department of Biostatistics, Boston University School of Public Health; S.A. Elman, BA, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; D.S. Stenquist, BA, Orthopedic and Arthritis Center for Outcomes Research, Brigham and Women's Hospital, and Harvard Medical School; E. Losina, PhD, Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, Brigham and Women's Hospital, and Department of Biostatistics, Boston University School of Public Health, and Harvard Medical School; J.N. Katz, MD, MS, Orthopedic and Arthritis Center for Outcomes Research, Department of Orthopedic Surgery, and Division of Rheumatology, Immunology and Allergy, Brigham and Women's Hospital, and Harvard Medical School. jnkatz@partners.org.
Abstract
OBJECTIVE: To dêtermine the reliability of radiographic assessment of knee osteoarthritis (OA) by nonclinician readers compared to an experienced radiologist. METHODS: The radiologist trained 3 nonclinicians to evaluate radiographic characteristics of knee OA. The radiologist and nonclinicians read preoperative films of 36 patients prior to total knee replacement. Intrareader and interreader reliability were measured using the weighted κ statistic and intraclass correlation coefficient (ICC). Scores κ < 0.20 indicated slight agreement, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 substantial, and 0.81-1.0 almost perfect agreement. RESULTS: Intrareader reliability among nonclinicians (κ) ranged from 0.40 to 1.0 for individual radiographic features and 0.72 to 1.0 for Kellgren-Lawrence (KL) grade. ICC ranged from 0.89 to 0.98 for the Osteoarthritis Research Society International (OARSI) summary score. Interreader agreement among nonclinicians ranged from κ of 0.45 to 0.94 for individual features, and 0.66 to 0.97 for KL grade. ICC ranged from 0.87 to 0.96 for the OARSI Summary Score. Interreader reliability between nonclinicians and the radiologist ranged from κ of 0.56 to 0.85 for KL grade. ICC ranged from 0.79 to 0.88 for the OARSI Summary Score. CONCLUSION: Intrareader and interreader agreement was variable for individual radiograph features but substantial for summary KL grade and OARSI Summary Score. Investigators face tradeoffs between cost and reader experience. These data suggest that in settings where costs are constrained, trained nonclinicians may be suitable readers of radiographic knee OA, particularly if a summary score (KL grade or OARSI Score) is used to determine radiographic severity.
OBJECTIVE: To dêtermine the reliability of radiographic assessment of knee osteoarthritis (OA) by nonclinician readers compared to an experienced radiologist. METHODS: The radiologist trained 3 nonclinicians to evaluate radiographic characteristics of knee OA. The radiologist and nonclinicians read preoperative films of 36 patients prior to total knee replacement. Intrareader and interreader reliability were measured using the weighted κ statistic and intraclass correlation coefficient (ICC). Scores κ < 0.20 indicated slight agreement, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 substantial, and 0.81-1.0 almost perfect agreement. RESULTS: Intrareader reliability among nonclinicians (κ) ranged from 0.40 to 1.0 for individual radiographic features and 0.72 to 1.0 for Kellgren-Lawrence (KL) grade. ICC ranged from 0.89 to 0.98 for the Osteoarthritis Research Society International (OARSI) summary score. Interreader agreement among nonclinicians ranged from κ of 0.45 to 0.94 for individual features, and 0.66 to 0.97 for KL grade. ICC ranged from 0.87 to 0.96 for the OARSI Summary Score. Interreader reliability between nonclinicians and the radiologist ranged from κ of 0.56 to 0.85 for KL grade. ICC ranged from 0.79 to 0.88 for the OARSI Summary Score. CONCLUSION: Intrareader and interreader agreement was variable for individual radiograph features but substantial for summary KL grade and OARSI Summary Score. Investigators face tradeoffs between cost and reader experience. These data suggest that in settings where costs are constrained, trained nonclinicians may be suitable readers of radiographic knee OA, particularly if a summary score (KL grade or OARSI Score) is used to determine radiographic severity.
Authors: L Gossec; J M Jordan; S A Mazzuca; M-A Lam; M E Suarez-Almazor; J B Renner; M A Lopez-Olivo; G Hawker; M Dougados; J F Maillefert Journal: Osteoarthritis Cartilage Date: 2008-04-15 Impact factor: 6.576
Authors: L Sheehy; E Culham; L McLean; J Niu; J Lynch; N A Segal; J A Singh; M Nevitt; T D V Cooke Journal: Osteoarthritis Cartilage Date: 2015-05-21 Impact factor: 6.576
Authors: J Damen; D Schiphof; S Ten Wolde; H A Cats; S M A Bierma-Zeinstra; E H G Oei Journal: Osteoarthritis Cartilage Date: 2014-05-21 Impact factor: 6.576
Authors: D T Felson; R C Lawrence; P A Dieppe; R Hirsch; C G Helmick; J M Jordan; R S Kington; N E Lane; M C Nevitt; Y Zhang; M Sowers; T McAlindon; T D Spector; A R Poole; S Z Yanovski; G Ateshian; L Sharma; J A Buckwalter; K D Brandt; J F Fries Journal: Ann Intern Med Date: 2000-10-17 Impact factor: 25.391
Authors: Elena Losina; Jamie E Collins; Meghan E Daigle; Laurel A Donnell-Fink; Julian J Z Prokopetz; Doris Strnad; Vladislav Lerner; Benjamin N Rome; Roya Ghazinouri; Debra J Skoniecki; Jeffrey N Katz; John Wright Journal: BMC Musculoskelet Disord Date: 2013-10-12 Impact factor: 2.362
Authors: William Alexander Cantrell; Ceylan Colak; Nancy A Obuchowski; Kurt P Spindler; Morgan H Jones; Naveen Subhas Journal: Knee Date: 2020-06-27 Impact factor: 2.199
Authors: Matthew D Li; Ken Chang; Ben Bearce; Connie Y Chang; Ambrose J Huang; J Peter Campbell; James M Brown; Praveer Singh; Katharina V Hoebel; Deniz Erdoğmuş; Stratis Ioannidis; William E Palmer; Michael F Chiang; Jayashree Kalpathy-Cramer Journal: NPJ Digit Med Date: 2020-03-26
Authors: Bhushan R Deshpande; Elena Losina; Savannah R Smith; Scott D Martin; R John Wright; Jeffrey N Katz Journal: BMC Musculoskelet Disord Date: 2016-04-11 Impact factor: 2.362