Development and validation of a simple nomogram for predicting knee osteoarthritis using movement evoked pain in a community setting

Knee osteoarthritis (KOA) is a clinical disease with a high incidence rate. Early identification and treatment of KOA are of great significance. This study aims to establish a predictive model using the movement-evoked pain (MEP) test for the early diagnosis of KOA. From May to December 2018, we conducted a cross-sectional survey among 3374 residents from 12 communities in Hangzhou City, Zhejiang Province. Data collection included general demographic information, the MEP test and treatment history. The data set was randomly divided into training set and validation set at a ratio of 7:3 by computer randomization. We analyzed the prevalence of KOA based on imaging and determined the influencing factors using logistic regression. Based on these factors, we constructed a nomogram and conducted validation. Among the 6748 knees analyzed, 78.4% were diagnosed with KOA based on imaging (KL grade ≥ 2). From 13 initial variables, we identified 9 independent predictors for the nomogram: age, exercise habits, pain during squatting, stair climbing, and housework, maximum pain, and history of oral NSAIDs, physical therapy, or intra-articular injections. A nomogram was developed based on these variables. The Area Under the Curve of the training set and validation set in the model were 0.889 (95% CI: 0.878-0.902) and 0.878 (95% CI: 0.859-0.898), respectively. The Brier score of the calibration curve was 0.127 and 0.131, respectively. The decision curve showed that it could increase the net clinical benefit within the risk threshold range of 20-80%. The MEP test enables imaging-independent KOA risk stratification, offering a feasible decision-support tool for primary care.