A10 says AHI was 2.2, but didn't record them!

[cathyf]

According to the front panel of my machine, and according to MyAir, I had an AHI of 2.2 last night. I slept 5:48, so that's 13 events. But when I loaded the data into OSCAR, it only shows a single OA,

whole night

Screen Shot 2020-12-06 at 5.53.22 PM.png (579.81 KiB) Viewed 312 times

which I figured as sleep-wake junk, as it was 60 seconds after I hit start on the machine. But when I look at the graph that one doesn't look like an apnea at all!

OA but not an OA 60 seconds in

Screen Shot 2020-12-06 at 5.41.53 PM.png (1017.51 KiB) Viewed 312 times

I have user flags turned on as well -- UF1 is 20% & 10s, UF2 is 35% & 10s. That half-hour long cluster of orange events is 22 UF2s, and some of them look like unambiguous OAs! Maybe 10-12 to my eye...

These UF2s sure look like OAs to me!

Screen Shot 2020-12-06 at 5.48.39 PM.png (163.07 KiB) Viewed 312 times

What's also strange is that when I first woke up I thought it said 0.2 for the AHI, and then after I got out of the shower it said 2.2 and I figured I just mis-saw it. But 0.2 would be 0.17 rounded to one decimal point -- so maybe I did see "0.2" ?

I've been paging back through recent MyAir reports compared to OSCAR, and while nothing is that far off, I'm seeing some odd things where it looks like the numbers are being truncated not rounded. (Like for 1 event in 5.8 hours is 0.17, and MyAir would show that as 0.1 rather than 0.2). But nothing looks like it's off by a factor of 10, either!

Has anyone else seen this sort of thing happen?

[Miss Emerita]

There is rounding, and it doesn't work the way we were taught in school that it should, but the larger discrepancies are a mystery. Let us know if they persist.