Objectives: The disease activities of rheumatoid arthritis (RA) tend to fluctuate between visits to doctors, and a self-assessment tool can help patients accommodate to their current status at home. The aim of the present study was to develop a novel modality to assess the disease activity of RA by a smartphone without the need to visit a doctor.
Subjects and methods: This study included 65 patients with RA, 63.1 ± 11.9 years of age. The 28-joint disease activity score (DAS28) was measured for all participants at each clinic visit. The patients assessed their status with the modified Health Assessment Questionnaire (mHAQ), a self-assessed tender joint count (sTJC), and a self-assessed swollen joint count (sSJC) in a smartphone application. The patients' trunk acceleration while walking was also measured with a smartphone application. The peak frequency, autocorrelation (AC) peak, and coefficient of variance of the acceleration peak intervals were calculated as the gait parameters.
Results: Univariate analyses showed that the DAS28 was associated with mHAQ, sTJC, sSJC, and AC (p<0.05). In a stepwise linear regression analysis, mHAQ (β = 0.264, p<0.05), sTJC (β = 0.581, p<0.001), and AC (β = -0.157, p<0.05) were significantly associated with DAS28 in the final model, and the predictive model explained 67% of the DAS28 variance.
Conclusions: The results suggest that noninvasive self-assessment of a combination of joint symptoms, limitations of daily activities, and walking ability can adequately predict disease activity of RA with a smartphone application.