Multi-centre modified Delphi exercise to identify candidate items for classifying early-stage symptomatic knee osteoarthritis

King, L. K.; Liew, J. W.; Mahmoudian, A.; Wang, Q.; Jansen, N. E.J.; Stanaitis, I.; Hung, V.; Berenbaum, F.; Das, S.; Ding, C.; Emery, C. A.; Filbay, S. R.; Hochberg, M. C.; Ishijima, M.; Kloppenburg, M.; Lane, N. E.; Losina, E.; Mobasheri, A.; Turkiewicz, A.; Runhaar, J.; Haugen, I. K.; Appleton, C. T.; Lohmander, L. S.; Englund, M.; Neogi, T.; Hawker, G. A.

Multi-centre modified Delphi exercise to identify candidate items for classifying early-stage symptomatic knee osteoarthritis

Mark

King, L. K. ; Liew, J. W. ; Mahmoudian, A. ^LU ; Wang, Q. ; Jansen, N. E.J. ; Stanaitis, I. ; Hung, V. ; Berenbaum, F. ; Das, S. and Ding, C. , et al. (2025) In Osteoarthritis and Cartilage 33(1). p.155-165

Abstract: Objective: To generate a list of candidate items potentially useful for discriminating individuals with Early-stage Symptomatic Knee Osteoarthritis (EsSKOA) from those with other conditions and from established osteoarthritis (OA), and to reduce this list based on expert consensus. Design: We conducted a three-round online international modified Delphi exercise with OA clinicians and researchers (“OA experts”). In Round 1, participants reviewed 84 candidate items and nominated additional item(s) potentially useful for EsSKOA classification; those nominated by ≥3 participants were added. In Round 2, participants rated perceived usefulness of 108 items (1 [not at all useful] to 9 [extremely useful]). In Round 3, participants could revise... (More); Objective: To generate a list of candidate items potentially useful for discriminating individuals with Early-stage Symptomatic Knee Osteoarthritis (EsSKOA) from those with other conditions and from established osteoarthritis (OA), and to reduce this list based on expert consensus. Design: We conducted a three-round online international modified Delphi exercise with OA clinicians and researchers (“OA experts”). In Round 1, participants reviewed 84 candidate items and nominated additional item(s) potentially useful for EsSKOA classification; those nominated by ≥3 participants were added. In Round 2, participants rated perceived usefulness of 108 items (1 [not at all useful] to 9 [extremely useful]). In Round 3, participants could revise their ratings after reviewing Round 2 group median and quartiles. Following Round 3, we retained items with a median usefulness score >5 and ≥33.3% of participants categorised the item as useful (7 to 9), overall and in subgroup analysis by clinician field. Results: There were 128 participants in Round 1 and 113 (88%) completed all rounds. We retained 77 items that spanned multiple domains (demographics, symptoms, physical exam, performance-based measures, imaging, laboratory investigations, and gross inspection/arthroscopy). Highly rated items included (median usefulness score): prior knee joint injury (8), diagnosis of OA in a different joint (7), and activity-related knee pain (7). The interquartile range was most often 3. Conclusion: We identified 77 items that OA experts consider potentially useful for EsSKOA classification. The results highlight experts’ uncertainty around item usefulness. Next, candidate items will be further assessed and reduced using data-driven and multicriteria decision analysis methods.
(Less)

Please use this url to cite or link to this publication: https://lup.lub.lu.se/record/aaa67b56-5d3e-4f54-aaed-64a7c864190e

author

King, L. K. ; Liew, J. W. ; Mahmoudian, A. ^LU ; Wang, Q. ; Jansen, N. E.J. ; Stanaitis, I. ; Hung, V. ; Berenbaum, F. ; Das, S. and Ding, C. , et al. (More)

King, L. K. ; Liew, J. W. ; Mahmoudian, A. ^LU ; Wang, Q. ; Jansen, N. E.J. ; Stanaitis, I. ; Hung, V. ; Berenbaum, F. ; Das, S. ; Ding, C. ; Emery, C. A. ; Filbay, S. R. ; Hochberg, M. C. ; Ishijima, M. ; Kloppenburg, M. ; Lane, N. E. ; Losina, E. ; Mobasheri, A. ; Turkiewicz, A. ^LU ; Runhaar, J. ; Haugen, I. K. ; Appleton, C. T. ; Lohmander, L. S. ^LU

; Englund, M. ^LU

; Neogi, T. and Hawker, G. A. (Less)

organization

publishing date

2025-01

type

Contribution to journal

publication status

published

subject

Orthopedics

keywords

Classification criteria, Delphi, Early-stage, Knee osteoarthritis, Symptoms

in

Osteoarthritis and Cartilage

volume

33

issue

1

pages

11 pages

publisher

Elsevier

external identifiers

pmid:39521366
scopus:85210075580

ISSN

1063-4584

DOI

10.1016/j.joca.2024.10.016

language

English

LU publication?

yes

id

aaa67b56-5d3e-4f54-aaed-64a7c864190e

date added to LUP

2025-01-15 14:44:38

date last changed

2025-01-29 16:12:06

@article{aaa67b56-5d3e-4f54-aaed-64a7c864190e,
  abstract     = {{<p>Objective: To generate a list of candidate items potentially useful for discriminating individuals with Early-stage Symptomatic Knee Osteoarthritis (EsSKOA) from those with other conditions and from established osteoarthritis (OA), and to reduce this list based on expert consensus. Design: We conducted a three-round online international modified Delphi exercise with OA clinicians and researchers (“OA experts”). In Round 1, participants reviewed 84 candidate items and nominated additional item(s) potentially useful for EsSKOA classification; those nominated by ≥3 participants were added. In Round 2, participants rated perceived usefulness of 108 items (1 [not at all useful] to 9 [extremely useful]). In Round 3, participants could revise their ratings after reviewing Round 2 group median and quartiles. Following Round 3, we retained items with a median usefulness score &gt;5 and ≥33.3% of participants categorised the item as useful (7 to 9), overall and in subgroup analysis by clinician field. Results: There were 128 participants in Round 1 and 113 (88%) completed all rounds. We retained 77 items that spanned multiple domains (demographics, symptoms, physical exam, performance-based measures, imaging, laboratory investigations, and gross inspection/arthroscopy). Highly rated items included (median usefulness score): prior knee joint injury (8), diagnosis of OA in a different joint (7), and activity-related knee pain (7). The interquartile range was most often 3. Conclusion: We identified 77 items that OA experts consider potentially useful for EsSKOA classification. The results highlight experts’ uncertainty around item usefulness. Next, candidate items will be further assessed and reduced using data-driven and multicriteria decision analysis methods.</p>}},
  author       = {{King, L. K. and Liew, J. W. and Mahmoudian, A. and Wang, Q. and Jansen, N. E.J. and Stanaitis, I. and Hung, V. and Berenbaum, F. and Das, S. and Ding, C. and Emery, C. A. and Filbay, S. R. and Hochberg, M. C. and Ishijima, M. and Kloppenburg, M. and Lane, N. E. and Losina, E. and Mobasheri, A. and Turkiewicz, A. and Runhaar, J. and Haugen, I. K. and Appleton, C. T. and Lohmander, L. S. and Englund, M. and Neogi, T. and Hawker, G. A.}},
  issn         = {{1063-4584}},
  keywords     = {{Classification criteria; Delphi; Early-stage; Knee osteoarthritis; Symptoms}},
  language     = {{eng}},
  number       = {{1}},
  pages        = {{155--165}},
  publisher    = {{Elsevier}},
  series       = {{Osteoarthritis and Cartilage}},
  title        = {{Multi-centre modified Delphi exercise to identify candidate items for classifying early-stage symptomatic knee osteoarthritis}},
  url          = {{http://dx.doi.org/10.1016/j.joca.2024.10.016}},
  doi          = {{10.1016/j.joca.2024.10.016}},
  volume       = {{33}},
  year         = {{2025}},
}

Lund University Publications

LUND UNIVERSITY LIBRARIES

Multi-centre modified Delphi exercise to identify candidate items for classifying early-stage symptomatic knee osteoarthritis