Validity of 6 activity monitors during standard physical activities in COPD – Comparison with indirect calorimetry

Background: Reduced physical activity is an important feature of COPD. A range of monitors are available, but the validity of their various output parameters is not well established for patients with chronic diseases. The IMI PROactive project evaluated 6 activity monitors: SenseWear (SW); RT3; Actiwatch (AW); Actigraph (AG); Dynaport Minimod (MM) and Lifecorder Kenz (KZ). Methods: 40 patients (GOLD stage 1-4; Age 67±7, FEV1 56±18%pred, 6MWD 430±127) were recruited in 4 centres. Patients wore all activity monitors and a portable metabolic system and performed a 1 hour standard activity set (including walking, standing, stair climbing and upper limb tasks). Activity intensity (METs), assessed by indirect calorimetry, was compared to activity monitor outcomes. Average VO 2 during the protocol was also calculated (8.78±1.72 ml min -1 kg -1 ) and compared to activity monitor outputs; METs (SW, MM); activity counts or Vector Magnitude Units, VMU (AW, AG, RT3); or a general activity outcome (KZ). Results: Table 1 gives median (IQR) for Pearson correlation between VO 2 and activity monitor outcomes and that between mean METS over the study and mean activity monitor output. Table 1 SW MM RT3 AG AW KZ Min-by-min 0.72 0.82 0.74 0.73 0.53 0.57 (0.64–0.81) (0.73–0.85) (0.64–0.79) (0.60–0.83) (0.45–0.62) (0.39–0.65) Mean VO 2 0.72** 0.51** 0.19 –0.005 0.56** 0.40** **p<0.01 Conclusion: The MM, SW, RT3 and AG give acceptable estimates of minute-by-minute variation in metabolic activity during physical activity. SW, MM, AW and KZ show acceptable correlations between overall metabolic activity and monitor output. These findings could guide users in choosing valid activity monitors for research or clinical questions.