Table 3 Differences for stationary behaviour and physical activity levels in absolute time between the different software, CoLaus study, Lausanne, Switzerland, 2014–2017, on 2693 participants.

From: Comparison of different software for processing physical activity measurements with accelerometry

 

Stationary behaviour

Light physical activity

Moderate physical activity

Vigorous physical activity

GENEACTIV vs

 PAMPRO

− 499 [− 555; − 443]

− 66 [− 80; − 53]

74 [41; 119]

− 22 [− 29; − 17]

 GGIR-White

− 101 [− 157; − 44]

15 [− 9; 36]

141 [100; 192]

0 [0; 2]

 GGIR-MRC, v.1.5–9

15 [− 28; 57]

− 83 [− 116; − 57]

130 [91; 177]

0 [− 1; 1]

 GGIR-MRC, v.1.11–1

38 [1; 75]

− 99 [− 131; − 73]

128 [90; 174]

− 1 [− 1; 0]

PAMPRO vs

 GGIR-White

403 [352; 452]

78 [60; 96]

66 [55; 78]

24 [17; 31]

 GGIR-MRC, v.1.5–9

518 [478; 557]

− 20 [− 50; 7]

54 [44; 64]

23 [17; 30]

 GGIR-MRC, v.1.11–1

542 [500; 582]

− 36 [− 67; − 9]

52 [42; 61]

23 [17; 30]

GGIR-White vs

 GGIR-MRC, v.1.5–9

− 114 [− 138; − 89]

− 100 [− 122; − 78]

− 11 [− 16; − 7]

− 1 [− 1; 0]

 GGIR-MRC, v.1.11–1

138 [106; 170]

− 116 [− 140; − 93]

− 13 [− 20; − 8]

− 1 [− 2; 0]

GGIR-MRC, v.1.5–9 vs

 GGIR-MRC, v.1.11–1

23 [4; 42]

− 16 [− 24; − 7]

− 2 [− 5; 0]

0 [0; 0]

  1. SB, stationary behaviour; PA, physical activity. Results are expressed as median and [interquartile range]. Positive/negative values indicate that the software indicated in bold overestimates/underestimates relative to the other. Between-group comparisons performed using Wilcoxon signed-rank test; all differences are significant at p < 0.001.