This cross-sectional study was conducted between November 22, 2015, and December 31, 2016 at Showa University Hospital in Tokyo, Japan. We recruited patients when they visited the hospital as outpatients. The eligibility criteria were as follows: 1) fulfillment of the American Rheumatism Association 1987 revised Criteria for RA(21) and 2) age of 20–65 years at consent acquisition. The excluded patients were judged ineligible by the research doctor for the following reasons: 1) bedridden/requiring a wheelchair, 2) dementia, and 3) limb deficiency. For evaluating construct validity, a study with 50 participants was designed on our preliminary research, and 41 patients agreed to wear an accelerometer. However, seven patients were excluded (Additional file 1). Finally, 34 patients were included in the study. Disease activity was measured using the Disease Activity Score 28-erythrocyte sedimentation rate (DAS28-ESR).(22) We classified patients with a DAS28-ESR of less and more than 3.2 into the low-disease-activity (LDA) and moderate/high-disease-activity (MHDA) groups, respectively.(23) Of the 34 patients, 20 were classified in the LDA group and 14 in the MHDA group.
The primary endpoint was PA evaluated using a triaxial accelerometer. The secondary endpoint was PA evaluated using the International Physical Activity Questionnaire (iPAQ).(24, 25)
PA was assessed using the triaxial accelerometer Active Style Pro HJA-750C (Omron Healthcare, Kyoto, Japan). We measured the wear time, time of vigorous-intensity PA (VPA), moderate-intensity PA (MPA), light-intensity PA (LPA), SB, and the number of steps per day using this device. MET-based cutoffs were used to define the intensity of each activity as follows: ≤1.5 METs for SB, 1.6–2.9 METs for LPA, and ≥ 3 METs for MVPA.(26, 27) The participants wore the accelerometer on their waist for seven consecutive days while they were awake. They did not wear the accelerometer when engaging in water-involving activities, such as swimming and showering. Records obtained when the accelerometer was worn for at least 10 h/day were considered valid, and data were considered as “nonwear” when acceleration signals were not observed continuously for more than 60 min.(28) The CSV data files of the accelerometer were downloaded using Omron health management software BI-LINK for PA Professional Edition ver. 1.0. The files were processed using custom software (i.e., a custom-written macro program for compiling data).
A short iPAQ form was used to determine PA during leisure time, domestic work, paid or unpaid work, and transportation.(24, 25, 29, 30) The patients were questioned regarding the following three specific types of PA, in which they participated at any time during their daily routine: walking, moderate-intensity activity, and vigorous-intensity activity. Scores for each type of activity were calculated by summing the scores for duration and frequency. Published guidelines for data processing and analysis of IPAQ data were used (available from: http://www.ipaq.ki.se). Comorbidities were evaluated using the Charlson comorbidity index.(31) At the outpatient clinic visit, the subjective physical function was assessed using the Modified Health Assessment Questionnaire (mHAQ).(32) The health-related quality of life was measured using the Medical Outcomes Study (MOS) 12-item Short-Form Health Survey (SF-12).(33) The Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F version 4) questionnaire was used to calculate the level of fatigue.(34, 35) The Center for Epidemiologic Studies Depression Scale (CES-D) was used to assess depressive symptoms.(36) Medical data (e.g., body mass index [BMI], disease duration, and medication history) of patients with RA were collected from medical charts. Radiological evaluation of the lower extremity involved six classes of the Larsen classification.(37)
Statistical analysis was performed using JMP® 13 software (SAS Institute Inc., Cary, NC, USA). Continuous data were expressed as means with standard deviations (SD) or medians with interquartile ranges (IQR). The normal distributions of each dataset were evaluated by performing the Shapiro–Wilk test. Spearman’s rank-sum test was used to assess the correlation. The variables obtained from an accelerometer and iPAQ for the LDA and MHDA groups were compared using multilinear regression analysis with adjustment for age and the FACIT-F score. For all analyses, statistical tests were two-sided, and significance was defined as p < .05.