Factors Predicting Compliance With Passive and Active Monitoring in Inflammatory Bowel Disease Patients

BACKGROUND: Updated goals of management of IBD patients include monitoring disease activity, yet compliance with such strategies is poorly studied. We are performing a multi-year study of patients with active and quiescent IBD to assess the utility of a wearable, continuously active biosensor and validating disease activity with scheduled stool (calprotectin) and serum (CRP) testing. This study assessed compliance with these passive and active monitoring strategies.

METHODS: Patients with diagnosed IBD were recruited for a prospective study using a biosensor (Fitbit Alta HR, San Francisco, CA) and mobile application (Litmus Health, Austin, TX). Each patient was instructed to wear a biosensor for a one-year period with stool sample drop-offs and blood draws for the first 6 months. The biosensor collected step count, sleep time, and heart rate during the study. Compliance was calculated as the number of days of heart rate, step count, and sleep data received over the duration of the study measured in days. Compliance was defined as ?80% data collection in the time studied. A t-test was performed to determine the significant difference (P < 0.05) in compliance based on age, sex, disease activity, and body mass index (BMI); and between day time and night time wear.

RESULTS: A total of 40 patients were analyzed; 23 men (57.5%), 24 with ulcerative colitis (60%), and 20 in remission (50%). 28 patients were <50 years of age (70%). The average BMI was 26.5 kg/m². Overall, there was a 78% biosensor compliance rate among IBD patients. 45% of patients in clinical remission were compliant in stool drop-offs, while 60% of patients with clinically active disease were compliant in stool drop-offs (P > 0.05; 95% CI: -14.8 to 41.39). There were significant differences between biosensor compliance at night (57%) and day (98%) (P < 0.01; 95% CI: 0.31 to 0.5) and active disease (P < 0.01; 95% CI: -1.13 to -0.80). There was no significant difference in compliance and sex (P > 0.05; 95% CI: -0.53 to 0.11), age (P > 0.05; 95% CI: -11.14 to 7.84), and BMI (P > 0.05; 95% CI: -3.62 to 5.57). There were insufficient data to analyze compliance in answering questionnaires.

CONCLUSION(S): In this study, patients were more compliant wearing the biosensor during the day than during the night and were more compliant if they had clinically active disease. In future studies, it will be important to provide an incentive and intervention to promote compliance in both active and passive monitoring strategies.