Screening for hip and knee osteoarthritis in the general population: predictive value of a questionnaire and prevalence estimates

Objective: To study the feasibility and validity of a two-step telephone screening procedure for symptomatic knee and hip osteoarthritis (OA) in the general population. Method: The screening questionnaire was based on signs and symptoms, previous diagnosis of OA and validated OA criteria. A random sample of telephone numbers was obtained and, at each number, one person aged 40-75 years was included. A physical examination and knee or hip radiographs were offered when the screen was positive. A sample of subjects with negative screens was also examined. The diagnosis of hip/knee OA was based on the American College of Rheumatology criteria for signs and symptoms and Kellgren-Lawrence radiographic stage 2 or greater. Prevalence rates were estimated with correction for the performance of the screening procedure. Results: Of 1380 subjects, 479 had positive screens, among whom 109 were evaluated; symptomatic radiographic OA was found in 50 subjects, at the knee (n = 35) or hip (n = 20). Corrected prevalence estimates of symptomatic OA were 7.6% (6.4%-8.8%) for the knee and 5% (3.9%-6.1%) for the hip. The screening procedure had 87% (95% Cl 79% to 95%) sensitivity and 92% (95% Cl 91% to 93%) specificity for detecting knee OA and respectively 93% (95% Cl 86% to 100%) and 93% (95% Cl 92% to 94%) for hip OA. Conclusion: This study establishes the feasibility of telephone screening for symptomatic knee/hip OA, which could be used for a nationwide prevalence study. Pain and previous OA diagnosis were the best items for detecting symptomatic OA.