Cavallo wrote:... Now, if I'm basing a setting decision on this range of numbers, it would seem prudent to fix a minimum pressure just shy of the median value (particularly since it agrees so well with my lab value,) and a maximum just above the 90th percentile, right? Or was the DME correct in saying that it would be just fine to run things with a fixed pressure at the median?
Well, first of all, all I can offer is my opinion based on my experience.
A fixed pressure works fine if everthing remains the same with you each and every night. It might work "ok" on other nights and be almost close enough.
However, the last time I checked a single pressure does not always work for me. If I am on my side, the pressure I need is different than if I am on my back. If I had a drink (rare, but occassionally I do have one) before bed, then I need a different (higher) pressure than when I've not had anything to drink. If I have more stress during the day, I tend to obstruct more (I also have more problems with central apneas, but you probably don't have that). Oh, and what happens if you gain or lose weight. That can and does impact your pressure requirement.
So - again this is just my own opinion based on the 18 year of my own personal experience - a tight range would seem to offer the best chance to adapt to the changes in your need for differing pressures. So, if you start at slightly (2 to 3cm H2O) below the mean, and set the maximum value slightly above the 90th percentile (again 2 to 3cm H2O), you will cover the full range of pressure you will most likely require. A fixed pressure would miss the mark most of the time.
By the way, it's a statistical nit, but I wanted to help clarify the meaning of a few words. The mean is the average. The median is just the center number in a sorted list. Those can be quite different.