Evidation partnered with Sanofi to conduct a prospective study assessing the utility of person-generated health data as a marker of constipation symptom severity. It was found that on irregular or constipation days, there was a statistically significant association with several activity and sleep metrics. Results also suggest the likelihood of medication use increased with increasing symptom severity. Through our platform of 4 million individuals, we were able to engage virtually with individuals living with constipation. Our work with Sanofi was presented at the UEGW 2020 Conference showcasing the importance of understanding the everyday life of individuals living with constipation, specifically symptoms, behavior and medication to improve patient outcomes. This is another example of how PGHD may be a useful tool for evaluating real world patient centered outcomes for individuals with constipation.
Background: Understanding day-to-day variations in symptoms, behavior, and medication management can be important for improving patient centered outcomes for people with constipation. Patient Generated Health Data (PGHD) from digital devices is a potential solution, and the objective of this study is to assess the utility of PGHD as a marker of constipation symptom severity.
Methods: A virtual, 16-week prospective study of people with frequent constipation from an online platform connecting mobile digital devices including wearable monitors capable of collecting steps, sleep and heart rate data. Participants wore a Fitbit device for the study duration and were administered daily and monthly surveys assessing constipation symptom severity and medication usage. 38 predetermined day-level behavioral activity metrics were computed from minute-level data streams for steps, sleep and heart rate. Mixed effects regression models compared activity metrics between constipation status (irregular or constipated vs regular day) and medication use (medication day vs none, medication on irregular day vs medication or irregular day only, medication on constipation day vs medication or constipation day only) strategies as well as to model likelihood to treat with constipation medications based on daily self-reported symptom severity. Non-parametric ANOVA tests evaluated differences across treatment groups. Correction for multiple comparisons was performed with the Benjamini-Hochberg procedure for False Discovery Rate.
Results: 1,540 enrolled participants with completed daily surveys (mean age 36.6 sd 10.0, 72.8% female, 88.8% Caucasian). 1,293 completed all monthly surveys and 756 had sufficient Fitbit data density for activity analysis. 22 of the 38 activity metrics were significantly associated with bowel movement (BM) or medication treatment patterns for constipation. Compared to baseline regular days with no medication use, participants had fewer steps, longer periods of inactivity and reduced total sleep time on irregular and constipated days (Table). Participants were 4.3% (95% CI 3.2, 5.3) more likely to treat with medication on constipated days versus regular. Daily likelihood to treat increased for each 1-point change in symptom severity of bloating (2.4%, 95% CI [2.0, 2.7]), inability to pass (2.2%, 95% CI [1.4, 3.0]) and incomplete BM (1.3%, 95% CI [0.9, 1.7]).
Conclusion: Constipation status irregular or constipated was associated with a number of activity metrics in steps and sleep. Given the small magnitude of effect, further research is needed to understand the clinical relevance of these results. Likelihood to treat with medication increased with increasing severity for a number of constipation symptoms. PGHD may be useful as a tool for evaluating real world patient centered outcomes for people with constipation.