Data From 08/26/11 and 08/27/11: Redux

ThomasMcKean wrote:Doctor personally confirmed inability to read the charts.

He's probably just not used to trying to make sense of flow data without EEG, EKG, and EMG. FWIW, there's a fair amount of puzzlement in this thread about your charts too.

Pugsy wrote:Pro doesn't give us event time duration at all. One has to try to eyeball the waveform which we cannot zoom in on like you can with ResMed. So of course I can use SleepyHead to do that. I only mention the 8 second thing because if we are to believe 169 second event of some kind for Thomas then we have to believe 8 second events also. If your SH event OA duration correlates with the numbers in the little boxes in ResScan then that is enough for me. I don't need to dissect it further. It is obvious from looking at my flow line that the 8 sec scored events are much shorter than my 20 second events.

Given that Encore Pro doesn't report those numbers as event duration, I think we should take them with a little salt. They're kinda sorta like the little duration balloons in ResScan, but they might sometimes be misleading. It's better than squinting at a zoomed out chart, though.

I think there's some good "philosophical" points here.

The medical mafia has been dismissing the machine scored events for a long time because they're not as good as PSG data. I think this thread justifies that skepticism to some extent, but not the idea of just ignoring the data.

Thomas's machine logs some events that look serious in terms of the machine scored type of event and duration. However, if you look at the flow waveforms, it doesn't look that serious. He's still breathing, just at a lower rate. He doesn't appear to gasp for air after the event. It looks like the reason the machine triggered the event may be that he was breathing deeply before this shallow breathing event.

I still think it's wrong for the medical mafia to simply ignore CPAP data. If the CPAP machine doesn't show any events, there's probably a good chance there weren't any severe events. I think the CPAP machine scoring suffers mostly on borderline events. A lack of machine scored events is probably a good indication that therapy is going well. You can look at is as a test that has false positives, but few false negatives.

It's important to look at the waveforms. "169 second long obstructive apnea" sounds scary. The flow waveforms on Thomas's event don't look scary.

Even though machine scored events do not necessarily indicate a "real" problem, they are a clue. I suspect most of them DO indicate a real problem to some extent. If you have access to flow waveforms, I think you CAN make a definite statement something is wrong in many events. If Thomas's 169 second event had little or no airflow, I think you could safely say it's a problem.

Perhaps machine scored events do get some false positives. Plenty of commonly used medical tests have false positives, but are still used as a form of screening to be followed up with more definitive tests. It's wrong to simply ignore them because they patient had a PSG in a sleep lab. It's especially wrong to ignore them if you have waveform data that will often tell you whether the event is "real" or not. It's especially wrong to ignore test results that you get at a very low cost with little extra effort on the part of the patient.

There really needs to be software as good as SleepyHead easily available to all doctors selling CPAP machines. The Respironics software is nowhere near as useful as SH, costs money and is poorly maintained and clumsy to install. There really should be software that works with all CPAP machines, so doctors don't have to install several different programs and learn how to use each one.

Unfortunately, I can't see the medical mafia using any kind of "unoffical," and "unsupported" software like SleepyHead. It reminds me of my place of work. The computer mafia there is always talking about "unsupported and non-strategic" software, which translates to "embarrassingly superior to the expensive stuff the bureaucracy has chosen."