'dur_day_total_MOD_min' + 'dur_day_total_VIG_min' VS file summary report

[Laura Maenhout]

Dear,

When I compare the outcomes of physical activity in minutes/day from the variables 'dur_day_total_MOD_min' + 'dur_day_total_VIG_min' with the information provided in the file summary report ('Time spent in moderate or vigorous activity'), the values differ significantly. The values from 'dur_day_total_MOD_min' + 'dur_day_total_VIG_min' are much higher. What could be the reason for the values of 'dur_day_total_MOD_min' and 'dur_day_total_VIG_min' being so much higher than what is reported in the file summary report? Am I doing something wrong?

Best regards,
Laura

[Jairo Hidalgo Migueles]

Hi Laura, 

The visual report is not the perfect tool for quality checks, but only an orientative visualization. As for your case, in the visual report GGIR uses the information from the longest bout defined for MVPA, by default it would be taking 10-min bouts of MVPA and that's the reason it is lower than the total minutes classified as MOD + VIG in the day.

Best,
Jairo

[Laura Maenhout]

Hi Jairo, 

Thanks for the fast response and explanation. That makes sense. 

Perhaps I could ask a follow-up question. When examining the variables 'dur_day_total_MOD_min' + 'dur_day_total_VIG_min', I noticed that the values are reaching 120 minutes of MVPA per day, and sometimes even more, while the summary report indicates an average of only 14 minutes per day... Since I am studying a population that is generally not very active, I would have expected the values to be lower than 120 minutes of MVPA per day. Is there an explanation why these values are so high (in comparison to the values when considering bouts)? Could this be related to the sensitivity of the accelerometer? (I collected the data using an ActiGraph GT9X Link).

Best, 

Laura

Laura Maenhout

PhD Researcher

T: +32 9 264 63 12 

 

Department of Movement and Sports Sciences  - Physical Activity and Health

Ghent University - Faculty of Medicine and Health Sciences

Campus Dunant, Watersportlaan 2, B-9000 Gent, Belgium

www.ugent.be/ge/bsw

 

Facebook: Bewegings – en Sportwetenschappen – UGent

Instagram: bsw_ugent

Twitter: bsw_ugent; Physical Activity and Health

                      

 

---

Van: rpacka...@googlegroups.com <rpacka...@googlegroups.com> namens Jairo Hidalgo Migueles <jairo.hidal...@gmail.com>
Verzonden: maandag 6 mei 2024 16:29
Aan: R package GGIR <rpacka...@googlegroups.com>
Onderwerp: [GGIR] Re: 'dur_day_total_MOD_min' + 'dur_day_total_VIG_min' VS file summary report

 

--
You received this message because you are subscribed to a topic in the Google Groups "R package GGIR" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/RpackageGGIR/SKnpf0zAYkg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to RpackageGGIR...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/RpackageGGIR/27cab384-1a65-4d7e-b9c7-036599dfaf0fn%40googlegroups.com.

[Jairo Hidalgo Migueles]

Hi,

The number of minutes classified in the total time variables depends mainly on the cut-points used and some other parameters such as the epoch length or the body site where the accelerometer was attached. For example, for wrist-worn accelerometers, it is usual to get a higher total MOD and total VIG time than initially expected. From my experience, this mainly relates to gesticulation and sporadic wrist movement during the day that does not increase the energy expenditure considerably (therefore it should not be considered physical activity), but is increases the movement enough as to surpass the MOD and VIG thresholds. Usually, I find that utilizing either bouts of at least 1 min in MVPA or aggregating the data in 60-second epochs in part 5 (with parameter part5_agg2_60seconds) provides estimates more in line with traditional accelerometer-based estimations from hip accelerometers if that's what you're looking for.

Best,
Jairo