Comparison of Four Actigraph Cutpoints for Physical Activity

Abstract

PURPOSE: The objective of this study was to examine four different equations for calculating moderate-intensity physical activity as well as to validate each against the 7 day Physical Activity Recall interview (PAR). Participants were healthy sedentary women (n=201). Characteristics of the sample include: mean age=47.5 (SD=10.5); 93.4% Caucasian; minutes of moderate intensity (and above) physical activity as reported on the PAR = 43.3 (SD=65.3); BMI=28.5 (SD=5.2). METHODS: Assessments were conducted prior to participants being enrolled in a 12-month intervention trial, with participants wearing the monitor for 3 days. A subsample of participants (n=83) were randomly selected to have a 3 day PAR conducted on the same days on which they wore the monitor. The 4 equations used were: Freedson et al. (1998), Hendelman et al. (2000), Swartz et al., (2000) and Nichols et al. (2000). RESULTS: Each equation produced the following amount of moderate intensity minutes (and above) of physical activity: Freedson: 9.15 minutes (SD=21.6); Hendelman: 313.3 (SD=210.3); Swartz: 62.5 (SD=73.1); Nichols: 9.05 (SD=21.2). The table below illustrates the correlations between each method. The correlations between the 3 day PAR and each of the four equations were: Freedson (r=0.24;p<.05); Hendelman (r=0.05;p=.67); Swartz (r=0.13;p=.25); Nichols (r=0.23;p<.05). CoNCLUsIoNs: Many of the equations established for the Actigraph were established using treadmill walking versus walking in free-living environments; therefore it is possible that there may be error introduced when accelerometers are used in free-living situations. Future directions, including the importance of establishing cut-points for free-living activity will be discussed.Table: Correlations Between Each Method.