Characteristics for Describing Patients With Knee OA
Characteristics for Describing Patients With Knee OA
Between 2007 and 2012, we identified 364 randomized trials involving patients with osteoarthritis of the knee. From 1149 authors, 76 were invited to participate in the survey. Of those, 32 agreed to participate, and 23 finally took part in both rounds. The mean age of the panel members was 54 years (range 39 to 75 years); 7 of the participants were female. Medical specialties of the panel members included rheumatologists (n = 13), physiotherapy (n = 6), rehabilitation medicine (n = 1), occupational therapy (n = 1), orthopedic biomechanics (n = 1), and musculoskeletal medicine (n = 1). Nine panel members were from North America (USA/Canada), eleven from Europe, two from Australia, and one from Asia.
Twenty-three of 32 experts returned the list with additional baseline characteristics. The additions comprised a total of 267 baseline characteristics in addition to the 30 characteristics nominated in the first list sent out. After deleting repeated characteristics and merging highly similar characteristics, 121 remained in the final list (Additional file 1).
All 23 participants from the first round completed the questionnaire in round 2. The medians of the expert ratings varied between 1 (e.g., click on knee motion) and 10 (e.g., global knee pain, WOMAC score).
To ensure the list of baseline characteristics remained within reasonable limits, we only included characteristics that had been rated a median of seven or above by the expert panel. A total of 39 characteristics fulfilled this criterion; 20 were related to a pain-reduction outcome, 15 were related to a functional improvement outcome, and 23 to a structural improvement outcome. The list of baseline characteristics with a median rating ≥7 is displayed in Table 1. Details of the median, 25–75% interquartile range, and the range of estimates are shown in Additional file 1. Six baseline characteristics were rated ≥7 in all three outcome categories (age, gender, BMI, global knee pain, function of knee, duration since onset of symptoms indicating knee osteoarthritis).
A consensus on the relevance of a baseline characteristic was arbitrarily defined as a calculated range of four (± 2) points or less around the median. Only one parameter, the global knee pain (e.g., VAS, WOMAC), fulfilled this criterion; all other characteristics listed in Table 1 displayed ranges greater than four points but are still rated as relevant baseline characteristics.
Results
Recruitment and Selection of Experts
Between 2007 and 2012, we identified 364 randomized trials involving patients with osteoarthritis of the knee. From 1149 authors, 76 were invited to participate in the survey. Of those, 32 agreed to participate, and 23 finally took part in both rounds. The mean age of the panel members was 54 years (range 39 to 75 years); 7 of the participants were female. Medical specialties of the panel members included rheumatologists (n = 13), physiotherapy (n = 6), rehabilitation medicine (n = 1), occupational therapy (n = 1), orthopedic biomechanics (n = 1), and musculoskeletal medicine (n = 1). Nine panel members were from North America (USA/Canada), eleven from Europe, two from Australia, and one from Asia.
Results of First Round
Twenty-three of 32 experts returned the list with additional baseline characteristics. The additions comprised a total of 267 baseline characteristics in addition to the 30 characteristics nominated in the first list sent out. After deleting repeated characteristics and merging highly similar characteristics, 121 remained in the final list (Additional file 1).
Results of Second Round
All 23 participants from the first round completed the questionnaire in round 2. The medians of the expert ratings varied between 1 (e.g., click on knee motion) and 10 (e.g., global knee pain, WOMAC score).
To ensure the list of baseline characteristics remained within reasonable limits, we only included characteristics that had been rated a median of seven or above by the expert panel. A total of 39 characteristics fulfilled this criterion; 20 were related to a pain-reduction outcome, 15 were related to a functional improvement outcome, and 23 to a structural improvement outcome. The list of baseline characteristics with a median rating ≥7 is displayed in Table 1. Details of the median, 25–75% interquartile range, and the range of estimates are shown in Additional file 1. Six baseline characteristics were rated ≥7 in all three outcome categories (age, gender, BMI, global knee pain, function of knee, duration since onset of symptoms indicating knee osteoarthritis).
A consensus on the relevance of a baseline characteristic was arbitrarily defined as a calculated range of four (± 2) points or less around the median. Only one parameter, the global knee pain (e.g., VAS, WOMAC), fulfilled this criterion; all other characteristics listed in Table 1 displayed ranges greater than four points but are still rated as relevant baseline characteristics.