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

Area of Science:

• Rheumatology
• Orthopedics
• Clinical Epidemiology

Background:

• Early-stage Symptomatic Knee Osteoarthritis (EsSKOA) requires precise classification for effective management.
• Differentiating EsSKOA from established osteoarthritis (OA) and other knee conditions is clinically significant.
• A validated set of items for EsSKOA classification is currently lacking.

Purpose of the Study:

• To generate a comprehensive list of candidate items for EsSKOA classification.
• To reduce this list based on expert consensus and perceived usefulness.
• To establish a foundation for further data-driven assessment of classification items.

Main Methods:

• A three-round international modified Delphi exercise was conducted with 128 OA clinicians and researchers.
• Participants rated the usefulness of 108 candidate items on a 9-point scale.
• Items were retained if they had a median usefulness score >5 and were rated useful (7-9) by at least 33.3% of experts.

Main Results:

• 77 candidate items spanning demographics, symptoms, physical exam, performance measures, imaging, and lab investigations were retained.
• Highly rated items included prior knee joint injury, diagnosis of OA in another joint, and activity-related knee pain.
• The interquartile range for item usefulness ratings was most often 3, indicating some variability.

Conclusions:

• 77 items were identified as potentially useful for EsSKOA classification by OA experts.
• The study highlights existing uncertainty among experts regarding the usefulness of specific items.
• Further data-driven methods will be employed to refine and reduce the candidate item list.