For this study, data was used from the CHIEF-HF Trial of 425 patients with heart failure who wore provided wearable devices (Fitbit) and increased their step count over a 12-week period. The randomized controlled trial also had participants complete serial Kansas City Cardiomyopathy Questionnaires (KCCQ) relating to symptoms and physical limitations. The relationship between daily step count, floors climbed, KCCQ symptoms, limitation scores, and patient outcomes as well as their changes were evaluated over 12 weeks.
KCCQ scores on a scale of 0-100 with higher scores indicating better health, and changes in scores of 5+ are considered clinically significant being previously shown to be associated with heart failure outcomes. After two weeks the mean KCCQ-physical limitation score was 55.7 and the KCCQ-total symptom score was 62.7; physical limitation scores increased by 4 points on average through 12 weeks and total symptom scores increased by 2.5 points.
Patients climbed a mean of 2.7 floors per day, but daily floor climb was not significantly different across different KCCQ score ranges. Higher daily step counts were seen with increased KCCQ scores for both physical limitation scores and total symptom scores. Those with total symptom scores of 0-24 averaged 2,473 steps per day, and those with scores of 75-100 averaged 5,351 steps per day.
A daily step count of 1,000-5,000 was significantly associated with symptoms and physical limitations based on KCCQ scores, and little association was seen when step counts were higher than 5,000 steps per day. Those who took 1,000 steps per day had total symptom scores 3.11 points lower than those who took 2,000 steps. Those who took 3,000 steps per day had KCCQ total symptom scores 2.89 points higher than those who took 2,000 steps.
Changes in step count over time were associated with changes in scores, suggesting that step count data such as those for a wearable device may be leveraged to inform clinical care or clinical trial endpoints. According to the researchers, those who increased step counts by 2,000 a day experienced a 5.2 point increase in the KCCQ total symptom score and a 5.33 point increase in KCCQ physical limitation score compared to those with no step count changes. Those with a step count decline had declines in KCCQ scores that were not statistically significant when compared to those with no step count change.
“What does this mean at the end of the day? If providers see improvements in step counts, then that is a good thing and reflects that patients’ health status is likely improving,” Golbus said. “However, seeing a decrease in step counts does not necessarily mean the converse and would not necessarily require an intervention. It might mean following up with a patient though.”
“These data are part of a large body of necessary and incremental work that will be required for actigraphy to attempt to achieve its potential as a patient-centered and efficient measure of functional status,” said Mitchell Psotka, MD, Ph.D., section chief of heart failure and transplant at Inova Schar Heart and Vascular Institute in Falls Church, Virginia, in a related editorial comment. “The authors have thankfully moved our understanding of actigraphy forward, though it is still the new kid on the block and will require substantial further testing and validation prior to widespread reliable clinical and research use.”
