My AHI math doesn't add up. Why?

I'm a CPAP newbie -- but a damned happy one considering I'm now sleeping normally for the first time in at least 40 years. (I had a sleep study and the news that I have severe OSA with an AHI of 44.) I've adjusted to my CPAP life with pleasure and now my AHI scores are typically lower than 1. Great, but I'm a bit confused how the readings on my machine work.

For example, last night I slept 7:13 hours and I had an AHI of 0.2. I believe that means that I had 1.44 events in total for the night. (My scores aren't always this good, BTW.) But why the .44? How can anyone have a fraction of an event? I'm wondering if someone will explain this. There must be a logical answer.

AHI is a PER HOUR average number over the course of the sleep session.
Averages are often fractions.

Of course AHI is per hour. If you'll read my comment you will see that is understood. And that there is no explanation for a fractional event.

guswah wrote: ↑

Fri Oct 15, 2021 3:31 pm

Of course AHI is per hour. If you'll read my comment you will see that is understood. And that there is no explanation for a fractional event.

pugsy said-
"Averages are often fractions."

Why do I even try??

If you understood averages then why the question?

Here you go. From a real life math professor.
http://adventures-in-hosehead-land.blog ... de-to.html

Have a nice day.

That link, at least at first glance, seems to be just what I was looking for. With thanks. I did say that there must be a logical answer.

Average number of own children per family U.S.

https://www.statista.com/statistics/718 ... er-family/

The typical American picture of a family with 2.5 kids might not be as relevant as it once was: In 2020, there was an average of 1.93 children under 18 per family in the United States.

Goodness gracious! How can you have a .93 child, much less a .5 child!

It would take genius to figure that one out.

JPB

guswah wrote: ↑

Fri Oct 15, 2021 3:31 pm

Of course AHI is per hour. If you'll read my comment you will see that is understood. And that there is no explanation for a fractional event.

Why don't you look at the events list and see how many there were, and then perhaps you can work out the math, don't forget to round.

guswah wrote: ↑

Fri Oct 15, 2021 12:30 pm

For example, last night I slept 7:13 hours and I had an AHI of 0.2. I believe that means that I had 1.44 events in total for the night. (My scores aren't always this good, BTW.) But why the .44? How can anyone have a fraction of an event? I'm wondering if someone will explain this. There must be a logical answer.

Are you referring to AHI reported by the machine? The answer is in the way the CPAP machine rounds session lengths and calculates. I guarantee you had an integral number of events during the night -- there is no such thing as a fractional event. If you have OSCAR, look at the list of events, or flow graph, or events graph, and see what AHI OSCAR reported.

For example, we often see a difference in sleep duration between ResMed and OSCAR. This is because ResMed rounds each session duration to the nearest minute, then adds those numbers together, so the result is always an even number of minutes. OSCAR, on the other hand, adds all the session durations together, in milliseconds, then converts to total elapsed time in seconds. This calculation difference alone can shift the calculated AHI significantly. You would have to look at your session details to replicate the result that the machine showed you.

Only Aldi sells hot dogs and buns in packages of exactly eight each.
Engineering for the arithmetically challenged.