(Another) Respironics Encore Data Failure

I have been using an M-Series Auto w/AFlex since March 20, 2009. I used a "loaner" for 12 days, and then received my own unit on April 3, 2009 (399.6 machine therapy hours to date). I am retired with a technical background and am using Encore Pro, and EPA to help me analyze my sleep periods so I can improve my sleep time. I purchased six SmartCards (SC) and have used as many as five in one night to separately record each of my sleep periods. Before going to bed, I put SC1 in my unit (my unit is always on - I only turn it off when I change/clean the filters). I put my mask on, turn on the fan or A/C, lay down, hook up the hose, start the blower, and, hopefully, have a restful sleep period. On my 1st awakening, I turn off the blower, detach the hose from my mask, get up, remove SC1 and replace it with SC2. The main reason for my awakening seems to be my need to use the bathroom. I repeat this same procedure throughout the night, using SC3, SC4, and SC5 if necessary. Each morning, I download each SmartCard to Encore Pro (and obtain the Therapy Hours and leak rate), print the Night Statistics report from EPA to get the # of Apneas, Hipopneas, Flow Limitations, etc., and enter this information into my Excel Spreadsheet along with info from notes that I have taken during the night on external factors such as sleep position, mask being used, laying awake time, etc. Once all the SCs are downloaded, I save and reformat them using the MyCPAP card utility and they are good to go for the following night. In two months of following this procedure, I have never lost any data from one of my sleep periods. During the past two+ months, I have lowered my avg weekly AHI from 8.8 to 0.9, weekly "time in Apnea / total therapy hours" from 2.36% to 0.12%, and my nightly awakenings (trips to the bathroom) from an avg of 4+ to less than 3. I have been able to make this progress from reading and following the wisdom of all the caring and knowledgeable experts (Rested Gal, SWS, Muffy, and many others) on CPAPTalk in combination with my nightly analyses. Kenny's recent "blog" thread was especially helpful. I plan to continue my detailed nightly analyses for another months or two.

cinco777 wrote: On my 1st awakening, I turn off the blower, detach the hose from my mask, get up, remove SC1 and replace it with SC2. The main reason for my awakening seems to be my need to use the bathroom.

Many thanks. I can empathize with the bathroom routine!

A key piece of information your post seems to offer to this conversation is your experience that a data card can be removed almost immediately after turning the blower off, without loss of data.

Question for you (just my curiosity): when you load each of the (up to) 5 cards, how does EncorePro handle the (up to) 5 separate data segments? Does it put all (up to) 5 together on a single nightly graph, or do you get (up to) 5 separate nightly graphs showing a few hours each (which would be my guess)? I would assume that it combines all of the sessions for the night on the final data summary page.

I am also assuming (in contrast to klockemy's hypothesis that "so it probably keeps the data in some sort of RAM then commits it to the card after the 1 hr time period has expired" that you only get each sleep segment's own data from each smartcard, i.e. that there is no "downloading" from "machine memory" of a prior session onto a later session's card. Are you able to confirm that view? (Whether or not you can tell probably depends on whether, how and when you look at each segment's data after reading the card into your computer.)

Many thanks for your contribution. Velbor

Velbor wrote

Many thanks. I can empathize with the bathroom routine!

One of my improved sleep goals is to reduce my "bathroom" awakenings/visits to 1/night. I am making good progress in reaching this goal as last Friday night (first time ever!) I made just one visit after sleeping for over 4 hours for my first sleep period of the night.

Yes, I can definitely confirm that my machine does not lose any data when, on getting out of bed, I immediately remove & replace the SmartCards (Each one is labeled and I keep them in sequence next to the machine). Early on, I recorded on paper, for each sleep period, the time that I started the blower and the time that I stopped it. I quit doing this step as the start and stop times on the Encore Pro Sleep Therapy Daily Details Report (Page 4) always showed the same start/stop times within a minute of my hand written ones.

More on my morning SmartCard procedure. I download SC1 to Encore Pro, save the generated report, open the Sleep Therapy Daily Details report, and print Page 4. Page 4 of this report provides me with, for example, views of the sleep events and leak profile for the sleep period. Both these views give me a clear picture of how I did re getting a restful sleep for this sleep period. (After downloading all my SmartCards, I will enter the start/stop TODs, avg leak, and AHI from the Page 4s into my spreadsheet). Next, I start EPA and print (and save) the night statistics for this first (SC1) sleep period. The night statistics show me, for this sleep period (SC1), how many of the various events occurred and, very importantly, the total time spent in Apnea. (I will enter this information into my spreadsheet later). I repeat this process for each of the SmartCards (SC2, SC3, ...). After completing all the downloads, I have multiple Page 4 printouts and multiple Nightly Statistics printouts, one set for each sleep period. Since the Nightly Statistics are cumulative, I subtract out the counts for the previous sleep period to get the counts for the follow-on sleep period. I will continue my response to your questions in my next posting.

Velbor wrote

I am also assuming (in contrast to klockemy's hypothesis that "so it probably keeps the data in some sort of RAM then commits it to the card after the 1 hr time period has expired" that you only get each sleep segment's own data from each smartcard, i.e. that there is no "downloading" from "machine memory" of a prior session onto a later session's card. Are you able to confirm that view? (Whether or not you can tell probably depends on whether, how and when you look at each segment's data after reading the card into your computer.)

I only get each sleep period's data on its SmartCard, nothing from anything prior that "might" be in machine memory. Each morning, I look at the Page 4 (Sleep Therapy Daily Details) and Nightly Statistics page for each sleep period. I enter specific data (see prior post) from each of these sets (two pages total for each sleep period) into my spreadsheet - one row per sleep period. Certain sums (total Apneas, etc.), and averages (Therapy Hours/night, etc.), are shown for each week. One in particular, the "Total Time in Apnea for the week / Total Therapy Hours for the week" is gaining importance in my continuing efforts to improve my sleep performance with CPAP. My first sleep study in late 2005 (no CPAP) showed me at 12% for "% of Sleep Time with events (Apneas and Hipopneas)" and I felt like a Zombie in the morning! The sleep lab had me sleeping on my back (which I never did and won't ever do since I now understand from CPAPTalk that it is the worst sleeping position for OSA sufferers) plus they had me take an Ambien (10mg) before the overnight sleep study commenced. With me "sleeping" on my back for the night, they had no problem qualifying me for Sleep Apnea insurance coverage!

You guys are amazing! My hat's off to you for your persistence

I found the missing data annoying at first but since it didn't occur more than about once a month (and seemed to decrease in frequency once I started sleeping regular hours), I ultimately decided to ignore it. It didn't seem to negatively affect my monitoring and tweaking but I must admit that now that I have my Sandman and don't lose data anymore, I do appreciate it.

I do find that doing something that's a chore (like starting cpap) is much easier for me if I feel I'm controlling my therapy instead of it controlling me

Mindy

cinco777 wrote:I am retired with a technical background .....

We seem to have much in common. I was a "data diver" in my previous life, and the habit is hard to break. I now go to what might be considered excessive lengths to collect and analyze my sleep data. But your process of multiple cards has me beat!

To be sure I understand: you use the multiple card process in part to insure that, if you lose data at all, you lose no more than one sleep period. If you could be confident of data integrity, one card for the entire night would give you the same full-night information, though at the expense of the exacting statistical detail for each sleep period which you want. If my summary is accurate, I'm not aware of anyone else on this board who goes to the lengths you do for that degree of precision. I join Mindy (who I believe has a more formal statistical background than I do) in taking my hat off to you, sir.

You've provided all of us with a level of understanding of the Respironics data management system which I doubt could have easily been put together in any other way. Hopefully, this wil eventually assist someone in making a definitive determination as to whether the reports part-night and full-night data loss are random and arbitrary glitches, or definable (and hopefully preventable) "features." Again, my thanks. Velbor

Velbor wrote

To be sure I understand: you use the multiple card process in part to insure that, if you lose data at all, you lose no more than one sleep period.

Not at all. I didn't learn about the possible data loss until more than a month after I implemented my multiple SmartCard procedure.

I use multiple cards for two purposes: 1) to achieve better granularity/quantification of sleep events (Apneas, Hipopneas, etc.) and mask leakage in a shorter time interval - a sleep period not the entire night, and 2) to gather more observations/results (4X) of "changes/experiments" since I can treat/utilize each sleep period as a "mini-night" test.

Re improved granularity/quantification, the Encore Pro Analyzer provides accurate counts of the sleep events and Apnea time per sleep interval - Encore Pro reports/views do not. Encore Pro's graphical view (Sleep Therapy Daily Details, Page 4) shows sleep events as tick marks and does not use a Time-Of-Day (TOD) axis. Early on, when my AHI was > 5, I found it impossible to ascertain if a tick mark represented one, or more than one, event. I also found it time consuming to associate a sleep event (tick mark) to the TOD that it occurred when I wanted to try to match it to an external event such as my switching sleep positions. When I have time, I plan to use the Encore Pro SQL event tables or parse the exported XML file and create a time-stamped sleep event TOD timeline that provides much finer event granularity.

Experiencing as many as five sleep periods/night has allowed me to try/test MORE things (different masks, mask fittings, mouth taping, masks with or without REMZzzs, different sleeping/face/pillow positions, and a host of other "experiments" that I have read about on CPAPTalk and want to see if they work for me. I only do pressure or CPAP modality changes once a week or every two weeks. I have already tried Auto CPAP from 7 - 10, straight CPAP @ 8, and Auto CPAP @ 8 and 7.5 (pressure that I have used for the past two weeks). Auto CPAP at a fixed pressure seems to work best for me (lower #s).

I

If I understand you correctly, you get up 5 times during the night to put in a new smart card and maybe change masks, settings, etc.

Doesn't that play havoc with your sleep quality? If I got up 5 times a night and did that, I would be exhausted the next day.

SmartCard "hot swap":
By design, Smartcards are meant to be inserted and removed with power applied. We use them extensively as "CAC" cards (Common Access Card). We gain access to controlled buildings/rooms by inserting them into wall-mounted readers (sometimes along with entering a PIN); and access to computer systems via a slot in the keyboard. Plug it in--you are logged on...pull it out--logged off.

And, by all accounts here, powering off makes no difference relative to the data loss issue. (But it WILL wear out your switch and/or finger more quickly...)

Hawthorne said

If I understand you correctly, you get up 5 times during the night to put in a new smart card and maybe change masks, settings, etc.

No. Before starting CPAP, and in my first weeks on CPAP, I was getting up as many as five times a night to go to the bathroom. After starting CPAP, when I got up for a bathroom "visit", I put a new SmartCard in my machine to record my next sleep period (plus often made a change such as sleeping in a new position, trying a different mask/fitting, etc.). After 2+ months of using CPAP, my number of bathroom "visits" has dropped from an average of 4+/night to less than 3/night, and I had (first time in recent memory) only one "visit" last Friday night. A Very Important sleep improvement goal for me is to get my nightly bathroom "visits" down to 1/night. I am making good progress. Once I am down to 1/night, I will only need two SmartCards because I will only have two sleep periods/night.

From reading this forum and from other reading/research, I learned that CPAP therapy lowers nightly blood pressure, and lower nightly blood pressure (and possibly other effects from sleep therapy improvements) leads to fewer "visits". My experience so far suggests that this is true for me.

Cinco, just curious, during which part of the night do you seem to get the best reports?

Doric asked

Cinco, just curious, during which part of the night do you seem to get the best reports?

Good question.

After nine weeks of CPAP therapy with data capability, a number of findings are already apparent. My best numbers - lowest AHI, lowest % of time in Apnea, lowest leak rate, and longest sleep interval w/o waking - almost always occur for the first sleep period of the night. My second sleep period of the night is a close 2nd. This finding makes sense to me as I am most "sleepy" when I go to bed. Until I stopped taking late afternoon naps (I quit after seeing Muffy's recommendations), my late nap sleep periods always had the lowest numbers. Please be aware that my nightly sleep numbers are now much lower than they were in late March when I started CPAP therapy using my new M-Series Auto with data capability. Almost every sleep period during the night is now a Good One (< 1.0 AHI). My analysis week is from Wednesday - Tuesday and last week's AHI was 0.7 with a 0.06% Time in Apnea. My Therapy Hours for the week were 50.2 (avg 7.2/night).

You may have already seen Muffy's list of recommendations on "how to get a restful night's sleep" in a recent post to a thread started by Bleeping Beauty but, if not, I highly recommend you have a look. His list and supporting rationale is the best that I have come across in all my reading and research. I have posted those that apply to me, such as no more late afternoon naps, on the mirror in my bathroom so that I have a constant reminder and don't cheat.

If I can provide any more information on my CPAP therapy analyses and "discoveries", please let me know. This forum has been the source of almost all my learning/education. What I have learned here has allowed me to make great progress in adapting to and benefiting from my CPAP sleep therapy, and I want to help others if I can.

cinco777 wrote: Experiencing as many as five sleep periods/night has allowed me to try/test MORE things (different masks, mask fittings, mouth taping, masks with or without REMZzzs, different sleeping/face/pillow positions, and a host of other "experiments" that I have read about on CPAPTalk and want to see if they work for me.

Thank you for the several clarifications you have provided. Again, we have much in common. I too am given to performing such "experiments".

The one caveat I would point out regarding your protocol is that the underlying "architecture" of sleep may be quite different for each "segment" as your sleep progresses through the night. For example, there may be more or less REM sleep in the early, middle, and later "segments", which itself may affect your breathing patterns. Unless you multiply repeat each experiment on different nights and "randomize" WHEN during the night each variable is altered, your data reliability may suffer. (This is an inherent problem in "split-night" sleep studies, in which, in an attempt to minimize both expense and inconvenience, both baseline and treatment data are collected sequentially during one night, rather than doing each on a separate night, which would provide more rigorous information.)

This is perhaps why (in addition to the discipline required), as I had commented, your approach appears to be rather unique. Much more common, and logistically easier, is to do full-night data collection, altering just one variable each night, and analyzing data on a nightly rather than on a "segmental" basis. Multiple repetitions and, if not "randomization", at least interlacing of variables, are still required to compensate as best as possible to normal night-to-night variability, as well as potential longer-term temporal trends. Since your shorter "segmental" periods each collect a relatively short duration of data, that much more repetition would presumably be needed to move toward statistical (as well as clinical) significance, with the result that just as many "nights" may be required before there is adequate data upon which to draw conclusions.

Be that as it may, you unique approach is a very welcome addition to the "shared knowledge and experience" of the community here which is interested in these sorts of things!

Velbor wrote

The one caveat I would point out regarding your protocol is that the underlying "architecture" of sleep may be quite different for each "segment" as your sleep progresses through the night.

Thanks for your insightful comments re my "experiments" and sleep period analyses. I agree with all of them and am trying to compensate for not having all the data (I wish, for example, that I could collect and analyze data on sleep stages as was done in my two sleep studies). My undergraduate degrees (so long ago!) were in Math and Physics so I understand the difficulties that I face and which you make clear in your comments. Changes that show promise are repeated with some randomness re selecting a sleep period for subsequent testing. I intend to collect enough data/results from my observations to ensure that I can reach statistically reliable conclusions. Of course, once I have reached that level of "sleeping like a baby", I intend to cut back on all my data collection and analysis and switch into monitoring only mode.

I'm sure you've read here that any changes should be made one at a time and give each change a few nights to settle in. I learned that the hard way at the beginning when I made too many changes at one time and couldn't figure out what was working and what was not.