JMIR Publications

ABSTRACT

Background:

Regular physical activity is associated with improved quality of life in patients with Inflammatory Bowel Diseases (IBDs), although much of the existing research is based on self-reported data which are typically over-reported. Wearable devices provide objective mobile health data on many rich physical activity dimensions including steps, duration, distance, and intensity. Little is known about how patients with IBDs engage in these varying dimensions of exercise and how it may influence their symptom and disease-specific patient-reported outcomes (PROs).

Objective:

This study aimed to 1) cluster physical activity patterns from consumer-grade wearable devices and 2) assess the relationship between the clusters and PROs in patients with IBDs.

Methods:

Precision VISSTA uses data from the Crohn’s and Colitis Foundation Partners cohort of adults with IBDs. Participants access an online portal to contribute physical activity data through smartphone applications or wearable devices in a bring-your-own-device model. Participants also complete biannual PRO questionnaires from the Patient-Reported Outcomes Measurement Information System short forms in the domains of depression, anxiety, pain interference, sleep disturbance, satisfaction with social roles and activities, fatigue, and IBD-specific questionnaires. K-means cluster analysis was used to generate physical activity clusters using three key features: number of steps, duration of moderate-to-vigorous activity (minutes), and distance of activity (miles). Based on the clusters, we conducted a cross-sectional analysis to examine differences in mean questionnaire scores and sociodemographics using one-way analysis of variance (ANOVA) tests. We also conducted a longitudinal analysis to examine individual cluster transitions among participants who completed multiple questionnaires, and mean differences in questionnaire scores were compared using paired sample t-tests across 6-month periods.

Results:

Among 430 participants, we identified clusters of low (33.7%), moderate (46.0%), and high (20.3%) physical activity. Participants in the low activity cluster reported the worst scores across all PROs. Scores varied significantly across clusters for depression (P=.004), pain interference (P<.001), fatigue (P<.001), sleep disturbance (P<.001), social satisfaction (P<.001), and short Crohn’s Disease Activity Index (P<.001). Sociodemographic characteristics also differed across clusters, with those with low physical activity being older (P=.002), having higher body mass indices (P<.001), and longer disease durations (P=.02) compared to other clusters. Across 6-month periods, 67.8% of participants remained in the same cluster, and only 1.2% moved to/from the furthest clusters of low and high activity.

Conclusions:

For patients with IBDs, there were positive associations between physical activity and PROs related to disease activity and psychosocial domains. Patient physical activity patterns mostly did not fluctuate over time, suggesting little variation in exercise levels in the absence of an intervention. Use of real-world data to identify subgroups with similar lifestyle behaviors could be leveraged to develop and evaluate targeted interventions that provide support for psychosocial symptoms and physical activity for personalized IBD care.