Leaks and there meaning

[proctor78]

Please help me understand this is what my leaks say
median 0 95th percentile 0 max 19.2

Thanks you guys are great

[BernieRay]

median = 0 means that your average leak was 0.
95% = 0 means that for 95% of the time you were using your S9, you had no leaks.
max = 19.2 means that for less than <5% of the total time, you had a sizable leak(s).

Resmed considers anything below 20 as manageable (meaning the S9 can compensate for it), but most of us feel better in the morning when we can keep the max value as low as possible.

[robysue]

Please help me understand this is what my leaks say
median 0 95th percentile 0 max 19.2

Thanks you guys are great

On the Resmed machines, the leak rate that is reported is the UNINTENTIONAL leak rate: In other words, the machine subtracts out an estimation for the expected, INTENTIONAL leak rate before reporting the numbers. (That estimation is based on the mask setting in the setup menu on the S9's.) For the S9 machines, the leak rates are reported in Liters/min (L/min). The Resmed engineers have decided that UNINTENTIONAL leak rates at or above 24 L/min are problematics --- if they last for a signficant amount of time.

The Median leak rate represents the leak rate number that is "half-way" through the leak rates. What I mean by this is that for 50% of the time the machine was running, your UNINTENTIONAL leak rate was AT or BELOW the median leak rate. And for 50% of the time the machine was running, the UNTENTIONAL leak rate was AT or ABOVE the median leak rate figure.

The 95th percentile leak rate represents the leak rate number that is 95% of the way through the leak rates. What I mean by this is that for 95% of the time the machine was running, your UNINTENTIONAL leak rate was AT or BELOW the 95% leak rat. And for 5% of the time the machine was running, the UNINTENTIONAL leak rate was AT or ABOVE the 95% leak rate figure. And for 5% of the time the machine was running, the UNINTENTIONAL leak rate was AT or ABOVE the 95% leak rate figure. To put this into perpsective: 5% of one hour is three minutes. And so you always know that you have a leak rate GREATER than the 95% leak rate figure for NO MORE than

(time machine ran in hours)*(three minutes per hour) = Time with Unintentioal leak rate greater than 95% leak rate.

The Maximum leak rate is simply the largest leak rate the machine happened to sample during the night. If the machine samples the leak rate while you are busy fiddling with the mask---pulling it off your face to reset it, for example, the maximum leak rate can be pretty high.

In the context of the data you provided: The machine could detect NO Unintentional leaks for 95% of the time the machine was running since the 95% leak rate was 0.0. But at some point there was a (short-lived) leak of 19.2 L/min recorded. Your unintentional leak rate was strictly greater than zero (and less than or equal to 19.2) for no more than 5% of the night.

These are GREAT leak figures, by the way.

[robysue]

median = 0 means that your average leak was 0.

Mathematically speaking, "average" and "median" are not the same thing.

95% = 0 means that for 95% of the time you were using your S9, you had no leaks.

Not quite, but close: The S9 estimates the value of the intentional leak rate for the mask and subtracts it off the leak data before reporting it. Very small leaks can (and do occur) that wind up being zeroed out by this process. For most people these small leaks aren't much of an issue. But if you're waking up with a very dry mouth every single morning, there might be a very small, slow leak through the lips that the S9 is simply not detecting as Unintentional leaking.

Resmed considers anything below 20 as manageable (meaning the S9 can compensate for it), but most of us feel better in the morning when we can keep the max value as low as possible.

Resmed draws the RedLine at 24L/min and Mr. Frowny Face only shows up if you are leaking at a rate of 24L/min for at least 25% of the time the machine is running. That's where Resmed really considers the leaks to become problematic.

You are right, most folks feel better when the leaks are much lower than the Red Line. However, the max leak rate is pretty much not the thing to watch: Any time you need to fiddle with the mask---including the very beginning of the night when you are doing the initial mask fitting---there's a good chance that the machine might sample the leak rate when the mask is not sealed against your face very well and artificially inflate that max leak rate. The 95% leak rate is a much better figure to try to keep low.

And while it's reasonable to aim for Median=0.0 and 95% = 0.0, getting the Max=0.0 may be darn near impossible.

[proctor78]

Thanks a bunch like I said you guys are great and now i am proud of myself....LOL

[BernieRay]

athematically speaking, "average" and "median" are not the same thing.

Practically speaking, do you really think the difference between the two has any real meaning in regards to leak rates, especially for a newcomer?

And while it's reasonable to aim for Median=0.0 and 95% = 0.0, getting the Max=0.0 may be darn near impossible.

Out of 221 days with a data-capable machine, I've had 32 days with max leak of zero - that's almost once a week. Only twice have I had consecutive days with no leaks, though. I've been on xPAP way too long to think max leak = 0 is an attainable goal on a daily basis. But it is attainable to some extent. For that reason, it's a worthwhile goal to me. In striving for it but not expecting it, I definitely have lower leaks.