Developing a Predictive Score for Chronic Arthritis among a Cohort of Children with Musculoskeletal Complaints—The Chronic Arthritis Score Study

Abstract

To explore if features obtained from a carefully taken medical history can be predictors of the final diagnosis in children with musculoskeletal complaints. We collected detailed clinical information on 178 children referred to our Pediatric Immunology and Rheumatology Unit by their primary care pediatrician for musculoskeletal complaints; a univariate logistic analysis was performed to identify variables correlated with the diagnosis of chronic arthritis. The variables identified were combined in a linear score that indicates the probability for a patient with musculoskeletal pain to receive the diagnosis of chronic arthritis. The joint swelling pattern (P < .0001), the precipitating factors of pain (P = .001), the duration of morning stiffness (P < .0001) and the frequency of pain (P < .0001), were found to be independently correlated with the diagnosis of chronic arthritis and were used to develop a diagnostic score. This score had a sensitivity of 90.9% and specificity of 95.3%. We developed a score that could be useful in the daily clinical routine to correctly direct the differential diagnosis in a child with musculoskeletal complaints, rationalizing time and resources necessary to reach a definitive diagnosis.