The National Institute of Health and Care Excellence (NICE) criteria for osteoarthritis (OA) obviate the need for physical exam or imaging, and their use may improve timely diagnosis of OA. However, they have not been validated. Within a larger study of individuals with type 2 diabetes, participants with and without self-reported knee pain underwent assessment of the NICE criteria for knee OA by questionnaire (index test), and clinical evaluation for established or possible knee OA by a rheumatologist (reference standard). We calculated the sensitivity, specificity, likelihood ratio positive (LR+) and likelihood ratio negative (LR-) of the NICE criteria and modified NICE criteria without the stiffness criterion. Our study included 96 participants: mean age 65.4 (SD 8.3) years and 52% were women. Individuals who fulfilled NICE criteria for knee OA (55.2%) included a spectrum of pain severity on a 11-point pain numeric rating scale with a median score of 5 (range: 1-9). Rheumatologist assessment identified 56 (58.3%) participants with symptomatic knee OA. The sensitivity, specificity, LR+, and LR- of NICE criteria for symptomatic knee OA were 0.84 (95% CI 0.74, 0.94), 0.85 (95% CI 0.74, 0.96), 5.6 and 0.19, respectively. For modified NICE criteria, these were 0.89 (95% CI 0.82, 0.97), 0.85 (95% CI 0.74, 0.96), 5.93 and 0.13. The NICE criteria have high sensitivity and specificity for detecting symptomatic knee OA in a population with type 2 diabetes. We found that a modified version, omitting the stiffness criterion, performed similarly. These criteria should be validated in other settings and populations.
Read full abstract