Search tips
Search criteria 


Logo of annrheumdAnnals of the Rheumatic DiseasesVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Ann Rheum Dis. 2005 January; 64(1): 29–33.
Published online 2004 June 18. doi:  10.1136/ard.2004.022905
PMCID: PMC1755212

Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement


Background: In clinical trials, at the group level, results are usually reported as mean and standard deviation of the change in score, which is not meaningful for most readers.

Objective: To determine the minimal clinically important improvement (MCII) of pain, patient's global assessment of disease activity, and functional impairment in patients with knee and hip osteoarthritis (OA).

Methods: A prospective multicentre 4 week cohort study involving 1362 outpatients with knee or hip OA was carried out. Data on assessment of pain and patient's global assessment, measured on visual analogue scales, and functional impairment, measured on the Western Ontario McMaster Universities Osteoarthritis Index (WOMAC) function subscale, were collected at baseline and final visits. Patients assessed their response to treatment on a five point Likert scale at the final visit. An anchoring method based on the patient's opinion was used. The MCII was estimated in a subgroup of 814 patients (603 with knee OA, 211 with hip OA).

Results: For knee and hip OA, MCII for absolute (and relative) changes were, respectively, (a) –19.9 mm (–40.8%) and –15.3 mm (–32.0%) for pain; (b) –18.3 mm (–39.0%) and –15.2 mm (–32.6%) for patient's global assessment; (c) –9.1 (–26.0%) and –7.9 (–21.1%) for WOMAC function subscale score. The MCII is affected by the initial degree of severity of the symptoms but not by age, disease duration, or sex.

Conclusion: Using criteria such as MCII in clinical trials would provide meaningful information which would help in interpreting the results by expressing them as a proportion of improved patients.

Figure 1
 Aspects of the cumulative distribution function used to determine the MCII (changes in pain score in patients with knee OA; n = 265). Among patients considering their response to treatment as good on a five point Likert scale, 75% experienced ...
Figure 2
 Patients' assessment of their response to treatment.

Articles from Annals of the Rheumatic Diseases are provided here courtesy of BMJ Group