Mobile Apps & Web

Garmin Connect Web Sleep interpretation over time

This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Sleep interpretation over time

Former Member over 10 years ago

Hi there

we've just bought vivofit for ourselves (myself and wife) and started using it, but even after only 3 days use it becomes obvious that the sleep interpretation and display over time is sorely lacking in what it can deliver.

I've searched the forum and haven't found the same feature request - apologies if my search fu is weak!

gamin connect shows sleep data in 3 methods;

pie chart of good, bad, average
line chart of total hours slept over days
line chart of sleep movement over single sleep instance

while all well and good, the first two, sleep mood and sleep total are pretty useless, I can't really envisage a time when I would want to look at a months sleep status and think 'hm I've had 40% bad sleep' and be happy with that as an isolated stat. By itself it doesn't mean anything, nor allow a drill down into reports to show more.

likewise, I don't see a real need from after a months use of thinking 'hm I really want to see a total graph of when I slept more or less' by itself it means nothing, so is not a useful stat.

the thing that both of these individual stats is missing is to make them relevant and make you aware of patterns - to allow you to contrast the mood / time against movement or vice versa.

if we had a correct display of movement over time where the movement graphs are tracked both individually over a week and month as well as options to display an average with time and mood tracked against them could show up important data and make the entire record meaningful... eg

on Wednesday I slept bad, got 7 hours of sleep but had 3 spikes of high movement at 2, 4 and 6 am.
the same repeats on Thursday and Friday
I slept well on Tuesday with 7 hours sleep and only one spike at 4am

if you took the above example and break it down in current stats you'd get this:

4x 7 hours sleep
75% bad sleep
25% good sleep
a single sleep wave with 3 spikes

by themselves they are pretty useless, woot I slept bad for 3 days, which three days?, woot I got 28 hours sleep, and?

but include the 'movement over time' and it highlights the descriptive text in the example... I had a bad nights sleep on 3 of the nights due to high movement, at specific times. having only 1 period on high movement is not enough to cause a bad nights sleep, it also shows over time that 4am is a vulnerable time for high movement - probably because kitty is currently sitting on my head to wake me up because the suns up and that obviously means breakfast time!

so you see, if all of that data can pull together and be presented together, with reports being able to drill down over time it becomes useful. at the moment its just a couple of pretty graphs with only a partial subset of data.

0 Former Member over 10 years ago

Good Morning,

I would like to add that from the sleep chart, I can not tell what is normal and what is not. it simply shows movements on a scale of low to high. What is Normal? middle of the graph?or bottom of the chart? if it is on scale of 1 to 10, then we could understand that below a number, within a range, is a good sleep pattern.

Who can help me interpret this sleep chart?
Nabil

0 moosetender over 10 years ago

I would like to see weekly, monthly and yearly sleep averages. Something similar to steps. Being able to know how much I sleep each night, on average, would be very useful.

0 Former Member over 10 years ago

I agree - it would be good to have more detailed feedback to be able to drill down.
I am not using an actual sleep tracker (unfortunately), and am simply entering in my sleep times - for me I would like to see the line graph of sleep hours mapped against my mood.

0 Former Member over 10 years ago

Purchased a FR15 a couple weeks ago, have been immersing myself in all that has been posted about interpreting the sleep data. I know at the heart of all this, what is recorded is a reference to movement that may or may not correlate well with quality of rest. Seems to me, leveraging an analysis tool used in drug development, if we had "area under the curve" (AUC) we'd have a metric that could be used to track over time. Think of it as the surface area under the line. Most nights, I sleep 7 to 8 hours so seeing that graphed over time ... I know that, the 3 smiley faces are about equally as useful (useless).

I anticipate use of an AUC may not be comparable from one person to the next, but to see it for an individual over time, when adjusting our daily behaviors (alcohol, caffeine, physical exertion, mattress, turning notifications off on smartphones, not reading on backlit devices before bedtime if you subscribe that that theory messing with our sleep, time from last meal, etc), we might have a better chance at making a correlation between wake-time behaviors and quality of sleep.

Other's thoughts?

0 Former Member over 10 years ago

Hi there

we've just bought vivofit for ourselves (myself and wife) and started using it, but even after only 3 days use it becomes obvious that the sleep interpretation and display over time is sorely lacking in what it can deliver.

I've searched the forum and haven't found the same feature request - apologies if my search fu is weak!

gamin connect shows sleep data in 3 methods;

pie chart of good, bad, average
line chart of total hours slept over days
line chart of sleep movement over single sleep instance

while all well and good, the first two, sleep mood and sleep total are pretty useless, I can't really envisage a time when I would want to look at a months sleep status and think 'hm I've had 40% bad sleep' and be happy with that as an isolated stat. By itself it doesn't mean anything, nor allow a drill down into reports to show more.

likewise, I don't see a real need from after a months use of thinking 'hm I really want to see a total graph of when I slept more or less' by itself it means nothing, so is not a useful stat.

the thing that both of these individual stats is missing is to make them relevant and make you aware of patterns - to allow you to contrast the mood / time against movement or vice versa.

if we had a correct display of movement over time where the movement graphs are tracked both individually over a week and month as well as options to display an average with time and mood tracked against them could show up important data and make the entire record meaningful... eg

on Wednesday I slept bad, got 7 hours of sleep but had 3 spikes of high movement at 2, 4 and 6 am.
the same repeats on Thursday and Friday
I slept well on Tuesday with 7 hours sleep and only one spike at 4am

if you took the above example and break it down in current stats you'd get this:

4x 7 hours sleep
75% bad sleep
25% good sleep
a single sleep wave with 3 spikes

by themselves they are pretty useless, woot I slept bad for 3 days, which three days?, woot I got 28 hours sleep, and?

but include the 'movement over time' and it highlights the descriptive text in the example... I had a bad nights sleep on 3 of the nights due to high movement, at specific times. having only 1 period on high movement is not enough to cause a bad nights sleep, it also shows over time that 4am is a vulnerable time for high movement - probably because kitty is currently sitting on my head to wake me up because the suns up and that obviously means breakfast time!

so you see, if all of that data can pull together and be presented together, with reports being able to drill down over time it becomes useful. at the moment its just a couple of pretty graphs with only a partial subset of data.

Yes, Yes, Yes! I just logged into this forum to ask these same questions! Makes no sense to me to track the data that is tracked but not the vitally important one of WHY sleep is not good! I also agree with the next post, that giving us an idea of what is normal would be helpful. Does everyone normally wake up several times a night? (I think I've heard that is true, but what does it look like on the chart?). I have sleep apnea so I am particularly concerned that my sleep patterns are normal. If not, I should probably get another extremely expensive sleep study done! :)

0 Former Member over 10 years ago

I concur with all the requests and add one. PLEASE make it possible for those of us that are up late to be able to indicate a went to bed time of after midnight without it telling us that we slept for 24 + 6 hours!!!

0 Former Member over 9 years ago

Sleep apnea

I agree with the previous posts but wonder if anyone else with untreated sleep apnea has seen any sort consistency in the graphs indicating if the apnea is worse one night or another. It is easy to tell when one gets up in the night to go urinate, the peak is usually very high, and when one is very still and apparently sleeping soundly, but what do all those other peaks relate to? Will sleep apnea show a particular pattern or is the sensor used for this graph just not sensitive enough or designed to pick up whatever sleep apnea movement would look like?
kitster

0 Former Member over 9 years ago

As already mentioned here, it would be good to see the sleep time over the last week / month / year, like with the steps, in connect. Just to be able to see charts for this and track average sleep over certain periods.