Polite inquiry about how to obtain the specific behavior pattern variables of accelerometry, such as the maximum daily sedentary bout length or the number of daily sedentary bouts?
Dear professors,
I currently get trouble in the study that requires specific behavioral pattern variables from the accelerometry data, such as the maximum daily sedentary bout length and the number of daily sedentary bouts. Unfortunately, these particular variables appear to be missing from the derived accelerometry dataset in the RAP.
I would greatly appreciate if you could spare some time to advise me on how I might access these variables.
Looking forward to your response, and thank you so much!
As you have found, UK Biobank does not have derived variables for maximum daily sedentary bout length or number of daily sedentary bouts.
Field 90004 could be used to generate variables like this. See https://biobank.ndph.ox.ac.uk/showcase/field.cgi?id=90004 . See the Notes tab, which says "These time-series files describe a participant's accelerometer measured physical activity intensity every five seconds. Each measurement represents the average vector magnitude across all samples recorded in each five second epoch/period."
Alternatively, Field 90001 contains the raw data from the Axivity monitors. See https://biobank.ndph.ox.ac.uk/showcase/field.cgi?id=90001 . The Notes tab says "These files contain the raw acceleration data, per participant, from a single use (over a 7-day period) of the physical activity monitor. Format is native cwa as returned by the device."
Comments
1 comment
As you have found, UK Biobank does not have derived variables for maximum daily sedentary bout length or number of daily sedentary bouts.
Field 90004 could be used to generate variables like this. See https://biobank.ndph.ox.ac.uk/showcase/field.cgi?id=90004 . See the Notes tab, which says "These time-series files describe a participant's accelerometer measured physical activity intensity every five seconds. Each measurement represents the average vector magnitude across all samples recorded in each five second epoch/period."
Alternatively, Field 90001 contains the raw data from the Axivity monitors. See https://biobank.ndph.ox.ac.uk/showcase/field.cgi?id=90001 . The Notes tab says "These files contain the raw acceleration data, per participant, from a single use (over a 7-day period) of the physical activity monitor. Format is native cwa as returned by the device."
To analyse these bulk files, some people use the GGIR package in R, and other researchers use an accelerometer virtual environment in Python. See https://cran.r-project.org/web/packages/GGIR/vignettes/GGIR.html and https://biobankaccanalysis.readthedocs.io/en/latest/methods.html
These pages might also be useful. https://biobankaccanalysis.readthedocs.io/en/latest/methods.html, and https://axivity.com/case-studies/biobank .
Please sign in to leave a comment.