Activity monitoring in ME/CFS

The article was authored by Ingrid Gurvin Rekeland, Kari Sørland, Ove Bruland, Kristin Risa, Kine Alme, Olav Dahl, Karl Johan Tronstad, Olav Mella and Øystein Fluge, and you can read the full text here:   

The main purpose of this study was to explore the feasibility and usefulness of using activity armbands in a clinical trial to monitor changes in activity levels in clinical trial participants with ME/CFS, and to explore natural variation in symptoms and activity in a group of ME/CFS patients over time. We aim to apply this knowledge to future clinical trials, in order to distinguish between natural symptom variation and a clinical response to treatment. The participants in this study were thus not exposed to any form of intervention, but simply recorded their symptoms and activity levels during everyday life with the help of Fitbit armbands and questionnaires. 

The study included 27 patients with ME/CFS, of whom five had mild and six had mild-moderate disease, ten had moderate disease, four had moderate-severe and two had severe disease. For the purpose of statistical analysis, the mild and mild-moderate patients were lumped together into a “mild” group, and the moderate-severe and severe patients were classed as one group, “severe”.

In the article, we conclude that it is feasible to use activity armbands to record steps per 24 hours and resting heart rate in clinical trial participants, but that patients with ME/CFS cannot be evaluated exclusively on measures of physical activity. Due to the complexity of the disease, it is important to combine objective measures with patient-reported questionnaires to capture the patients’ own experience of symptoms and daily function. In the article, we describe in more detail how the activity parameters correspond to data from several different questionnaires on symptoms, daily function and quality of life. 

Another take-home message is that both activity levels and self-report measures fluctuate over time, and these changes were more evident in patients with milder symptoms. Many patients describe that their symptom burden varies from day to day or over longer periods of time, and this natural variation can be captured using activity monitoring combined with questionnaires. In a clinical trial, natural symptom variation can be wrongly interpreted as a clinical response to medication or therapy. We therefore suggest that clinical trials should include a run-in period where the individual patient’s pattern of function and activity variation is recorded before any kind of intervention is administered. 

The Fitbit study does not aim to investigate or report what is the typical or average number of steps for an ME/CFS patient. Patients with mild ME/CFS outnumbered the patients with severe ME/CFS. This was accepted, as we aimed to capture greater and smaller variations in function level. We also found that the number of steps measured were not always accurate. As with other wrist-worn monitors, Fitbit sometimes registers arm and hand movements as steps. The Fitbit watch systematically overestimated the number of steps compared to another (older) activity monitor by 30 % on average. However, we conclude that activity armband measurements are useful to monitor participants continually and record changes in the day-to-day activity in each individual over time – whether the changes are caused by natural variation or response to treatment. 

We intend to use the results from the Fitbit study to improve the design of our future clinical trials, and it is our hope that other researchers will find the publication useful in their efforts to design clinical studies in ME/CFS. 

We are at present performing a pilot medical intervention study («Daratumumab ved ME/CFS»), and have included both objective measures (steps and resting heart rate measured by Fitbit) patient-reported measures (questionnaires on symptoms, function and quality of life), as well as a 12-week observation period before start of treatment.​
​