Agreement between the physical activity intensity level obtained by different outcome measures in people with COPD

Abstract

Introduction: Intensity of physical activity (PA) must be measured to confirm that people with chronic obstructive pulmonary disease (COPD) meet PA recommendations and ensure participants’ safety. Quantifying the oxygen consumption (VO2) is the gold standard out-come measure to assess single free-living PA intensity. Nevertheless, several other outcome measures that are more economic, simpler and user-friendly than VO2 have also been used to assess a single-free living PA-related intensity, namely heart rate (HR) or dyspnoea Borg score. Different methodologies will probably yield different PA intensity levels, but this is yet unknown. This systematic review aimed to explore the agreement between the intensity level obtained by different outcome measures assessing the same single free-living PA. Methods: A systematic search was conducted in May 2020 on PubMed, Scopus, Web of Science, Cochrane Library and EBSCO. We included original studies on COPD, assessing single free-living PAs-related intensity (individual types of PAs, which were performed by participants at their own pace within a restricted period of time, and pertained to leisure, occupation, home or transport PAs) and reporting on at least two of the following outcome measures: %VO2peak, %VO2reserve, %HRpeak, %HRreserve, metabolic equivalent task [METs], dyspnoea, perceived exertion or fatigue scores on the Borg 0-10 or 6-20 scales, or walking speed. Each single free-living PA had its intensity categorised as light, moderate or vigorous, following the cut-offs proposed by the American College of Sports Medicine and World Health Organization (table). Agreement was calculated as: number of agreements between two measures [same intensity level]/number of comparisons using both measures*100. In case of no agreement, we determined which outcome measure yielded the highest intensity using this formula: number of comparisons where the outcome measure had the highest intensity/number of comparisons where there was no agreement*100.Results: Nineteen studies, enrolling 574 people with COPD (65 years, 61% men, 53% FEV1pp) were included. Percentages of agreement varied between 0 to 100% (table). %VO2 peak and %VO2 re-serve consistently yielded the highest intensity level. Therefore, we can infer that PA-related intensity assessed with Borg scores, %HR reserve and METs was underestimated. Nevertheless, these results should be interpreted with caution, as 8 of the 18 comparisons were performed using only one study. Conclusions: There is inconsistency among the PA intensity levels elicited by different outcome measures. Cut-offs points regularly used in healthy people to categorise PA intensity may not be suit-able for people with COPD, hence, future studies developing specific cut-offs for COPD and formal guidelines on how to accurately measure single free-living PAs-related intensity in people with COPD are urgently required.