APAP as good as PSG (avg. .5 AHI difference)!

[SleepyCPAP]

I was looking up recent published studies. This one from March 2013 was only the abstract online, so I can't tell what machine they were using for the APAP (but it references downloading from a Smart Card, so maybe M-Series? or an overseas brand?). As long as certain people were screened out (folks with the more complicated health issues), the AutoPAP machine was, on average, about .5 AHI different in its estimation compared with the PSG (with a range that was less than +/-5 AHI on a patient-by-patient basis).

Past studies had given a difference of one or more AHI, and some pretty extreme variations patient-by-patient (some of which could be explained by the fact that the sleep technician can tell if you are awake, and the machine didn't know and so dilutes the AHI with inflated "sleep" time of laying there but not really asleep all the time).

I'm guessing newest top-end machines are doing better. If it was M-Series I recall they were not picking up on things that PSR1 can spot.
http://www.ncbi.nlm.nih.gov/pubmed/22371206

and here is a very similar study which was with a smaller group at a VA setting in Buffalo (I don't think this was in a peer-review publication):
http://www.acsu.buffalo.edu/~grant/94.pdf
(That Buffalo study was using Respironics for the APAP)

For those folks who have trouble convincing their health care providers to pay attention to the data from the machine, I would hope that pointing out these studies might be useful to shift them from their reluctance.

--SleepyCPAP