Multimorbidity and Clinical Characteristics in Physiotherapy Utilization Among Osteoarthritis Patients: An Association Rules Mining Approach

Abstract

Osteoarthritis (OA) is a chronic degenerative joint disease frequently accompanied by multimorbidity, defined as the coexistence of two or more chronic conditions in an individual. Physiotherapy is recommended as a first-line conservative modality in OA management. This study aimed to identify multimorbidity patterns associated with physiotherapy utilization and to analyze sociodemographic and clinical factors associated with physiotherapy utilization among patients with OA. This retrospective cohort study was conducted using electronic medical records of patients with OA at Sultan Agung Islamic Hospital, Semarang, Indonesia, from 2020 to 2025. Total sampling was employed, resulting in 704 patients with OA who underwent physiotherapy. Descriptive analysis was performed to characterize the patients, while logistic regression was used to identify factors associated with physiotherapy utilization. Multimorbidity patterns were identified using the FP-Growth algorithm based on support, confidence, and lift values. Most patients were aged ≥60 years (494 patients), female (448 patients), diagnosed with knee OA (633 patients), and obese (429 patients). ARM identified 22 association rules, with the combination of diabetes mellitus, hyperlipidemia, and grade 2 OA showing the highest lift value (1.24). Factors associated with physiotherapy utilization included age ≥60 years (aOR 1.82), knee OA (aOR 1.48), grade 3–4 OA (aOR 1.34), and ≥2 comorbidities (aOR 1.71). In conclusion, specific multimorbidity patterns and clinical characteristics were associated with physiotherapy utilization among patients with OA, particularly older age, knee OA, higher disease severity, and multimorbidity.