ResScan 3.11, averages of AHindices inconsistent

[locomotive_breath]

I try to compare two periods of time (several weeks) in which I used my S9 with different constant pressures.

April-June (7mmH20):
unknown: 0.0 / central: 2.1 / obstruct: 0.3 // Apnoe: 2.9 / hypopnoe: 0.3 /// AHI: 3.2

July-August (8mmH20):
unknown: 0.0 / central: 1.6 / obstruct: 0.2 // Apnoe: 2.0 / hypopnoe: 0.5 /// AHI: 2.5

As far as I understand, in both cases the Apnoe: (2.9&2.0) should be lower (2.4&1.8 ) or the
unknown: unequal 0.0 (0.5&0.2) ...

Am I missing the obvious?

[KatieW]

I'm not sure I understand your question. If you're asking why the numbers don't add up, my guess is that the Unknown, CAI, OAI numbers are the median, rather than the average, so they won't add up to your AI. But I'm not a mathematician.

[ruthven78]

I try to compare two periods of time (several weeks) in which I used my S9 with different constant pressures.

April-June (7mmH20):
unknown: 0.0 / central: 2.1 / obstruct: 0.3 // Apnoe: 2.9 / hypopnoe: 0.3 /// AHI: 3.2

July-August (8mmH20):
unknown: 0.0 / central: 1.6 / obstruct: 0.2 // Apnoe: 2.0 / hypopnoe: 0.5 /// AHI: 2.5

As far as I understand, in both cases the Apnoe: (2.9&2.0) should be lower (2.4&1.8 ) or the
unknown: unequal 0.0 (0.5&0.2) ...

Am I missing the obvious?

AHI = Apnea + hypopnea

Katie is correct

The central & obstructive breakdown is an index so it is basing it on total divided by the hours used. So they wont entirely match up. So you have an average of 2.1 centrals apneas per hour, and an average of 0.3 obstructive apneas, and and average of 3.2 combined apneas per hour. Since hypopneas are a creature all their own, they are figured out seperated and then added to the combined apneas to find your AHI. An AHI of less than 5 is what most companies target for therapy. (ie a person without sleep apnea is considered to have an AHI of less than 5)

[locomotive_breath]

Thank You for your quick responses! As you might guess, I'm fighting with the language too.
Yes, I am asking why the numbers don't add up to the expected results.

I have never seen 'unknown' being different from 0.0, thought it may be frozen at 0.0, so has anybody seen those creatures? (I'm glad the machine has no doubts about my apneae )

When I select only 1 day, the statistic is always consistent. But I want to use several/many days for conclusions ... I think this is reasonable.
So I select several days/weeks (drag the mouse) to get multipleday-averages for the indices, which are nighthour-averages themselves. KatieW, I hope it is clearer this time.

ruthven78: "So you have an average of 2.1 centrals apneas per hour, and an average of 0.3 obstructive apneas, and average of _2.9_ combined apneas per hour. Since hypopneas [...] figured out seperated [...] to find your AHI."
And inbetween I have missing 2.9 - 2.1 - 0.3 = 0.5 'unspecified' apneas. This is a lot for rounding errors, isn't it?
I don't know, how the program does it, but I could simply TRY to find the AVERAGE NUMBERS of the indices displayed on a DAILY BASE: I can write down the daily indexvalues, add them up and divide by the number of days, sitting in front of a modern high-end computer (I have hexcoded 8080, 8085, 8086, 80286, 80386s ... centuries ago) and the program has even more raw data than just the daily numbers.

I am insecure now, which pressure is better? 8cm resulting in AHI: 2.5 >>> 7cm resulting in AHI: 3.2?
(look on AI & AHI, without understanding the details, eg. number of centrals falling with raising pressure!?)
... I seek the MINpressure for Apap mode ...

Yes, I hope to get lower AHIs.