Real-World Measures of Cardiorespiratory Function Can Stratify Primary Sjogren's Syndrome Participants with...

Many individuals with various chronic diseases experience debilitating fatigue that substantially impacts their quality of life. Currently, assessments of fatigue rely on patient reported outcomes (PROs), which are subjective and prone to recall bias. Wearable devices, however, can provide valid and continuous estimates of human activity and physiology, which are essential components of health, and may provide objective evidence of fatigue. This study aims to stratify primary Sjogren's syndrome (PSS) patients with different fatigue levels using real-world measures of activity and cardiorespiratory function. 72 participants with PSS wore a VitalPatch sensor on the chest for two 7-day continuous periods. Concurrently, the participants completed PROs relating to fatigue up to 4 times a day. The mean, standard deviation, minimum, and maximum of the heart rate (HR) and respiratory rate (RR), both overall and during periods of walking, sitting, and standing were calculated, along with the difference in HR and RR between these activities, and the time spent in each activity. The Mann-Whitney U test and four machine learning classifiers were used to assess if the digital measures could separate the participants categorised as "persistent" or "non-persistent" fatigue. The categorization of these two groups were tested using 5 different thresholds.None of the activity-time measures were statistically different and very few of the RR measures were statistically different between the groups (p<0.05). However, 64% of HR measures differentiated persistent fatigue from non-persistent fatigue participants (p<0.05). Machine learning also found that HR measures could separate the fatigue persistency groups with accuracies up to 77%. Therefore, this analysis has shown that real-world measures from a digital wearable are able to stratify PSS participants with persistent and non-persistent fatigue. Thus, leading to an objective, single-device approach to identifying fatigue severity in an immune-mediated inflammatory disease.