Reaping the deep sleep

Thought it was about time for an update on my sleep therapy. I’ve got an appointment at the sleep clinic later this month, but really the only reason I’m keeping it is because I don’t want to go all the way to the back of the queue if problems arise again.

First of all, the truth: I don’t think I’m ever going to get much more than 6 hours of sleep while I’m working. While I’m on holiday – turns out – I can squeeze in 8 hours or so around all the interruptions and roll out of bed somewhere around the time the (official) working day would be starting in term time. But even then, I’m waking up multiple times in the night, and sometimes for an extended period.

On work days, I’m getting up at 6 a.m. Doesn’t quite have to be that early, but I’d rather be at work early and get in an hour or more before the first bell than having to stay after school and do things then. Going home early has always been my jam.

Things that CBT encouraged me to try:

  • Stopped reading in bed. This was the big one, because it is/was a lifelong habit, and it was frustrating and weird to stop it. So now, when I read, it’s downstairs, and there’s no reading in bed. I get into bed, turn out the light, and roll over.
  • Got a new mattress. Not strictly part of CBT, but certainly about addressing the environment and making it encourage sleep as long as possible. A New mattress was long overdue. I got an Eve, it’s okay. It won’t last long, I don’t think, but we’ll see.
  • Cut down on naps. In many ways, the hardest thing to do. The number of times I get in from work and feel like dropping off on the couch for 40 minutes or so! But the truth is, not lately. Getting better sleep at night has made me less tired in the day and I do not need to nap. I did on holiday, but I was also getting better sleep at night.
  • Sleep restriction. The most important and most effective part of the therapy. Don’t go to bed early. Starting with midnight, and gradually inching my way back to 11 p.m. means that the time I’m spending in bed is mostly asleep. Probably my optimum time, if I’m honest, is around 11:30 p.m., at which point I sleep through to the alarm at 6. I still wake up multiple times a night, but I don’t remember them. Last night, I woke up about 20 times (according to Fitbit data) for a total of 38 minutes, but because I don’t remember, it doesn’t bother me. It’s not the same as waking up at 4:30 and knowing you won’t get back to sleep.

So here we are. I still sometimes feel that 6 hours isn’t quite enough, but I don’t feel broken and exhausted all day; I have a more positive attitude to sleep, and it doesn’t worry me as much as it did (for years on end, until recently). So CBT, and Sleepio, works for me. Ask your doctor about it.

How do you sleep?

Just finished week 5 of the Sleepio course, and things are looking up. My sleep restrictions have been lifted by 15 minutes for each of the past two weeks, and I can now go to bed at an entirely reasonable 11:30pm. My sleep pattern also looks fairly mainstream, with a few good stretches of deep sleep early in the night, and lighter or REM sleep later on.

When I think back to the first week of this, when I was only allowed in bed between midnight and 6 a.m., I find it hard to credit just how hard it was, and how reasonable it seems now. I described that first week of restriction as “brutal” and so it was, but I’ve become adept at it, and the quality of my sleep has been improving. In one sense, I’ve made rapid progress; on the other, the desired 90% efficiency is just beyond me (I’m averaging 89% – so very close), and my Fitbit still reports that I wake up multiple times every night, for a total of 40-odd minutes.

Sleepio gives you a long set of rules to following, and strategies to control your thoughts, but there are a few core things that I think are the most effective. Some of them are blindingly obvious, but if you’ve been having sleep problems, your behaviour gets irrational.

  • No caffeine for 6 hours before bed. If you’re going to bed at 11:30 it means, no caffeine after 5:30 p.m., which is disruptive of my weekend bourbon and coke schedule.
  • No alcohol for 4 hours before bed. Less of a problem for me as I tend to have a cocktail hour and leave it at that. The chances of me drinking after 7:30 this side of Christmas are remote.
  • Bed is for sleep and not reading. So the Kindle is relegated to the downstairs room and my life-long habit of reading just before lights out is banished. I do feel I’ve been neglecting my reading but given the choice, I’d rather sleep.
  • No naps. This is hard, because there’s always a crash moment early in the evening, when I struggle to hold my eyes open. This is true whether I drink or not, and it’s currently around 7:00 p.m. There is a maximum 15 minute nap rule, and what this means is that when I feel myself drooping, I quickly set a 15-minute timer on my phone and let nature take its course. Sometimes the setting of the timer is enough to snap me awake. If I’m still awake 15 minutes later, I repeat the timer. If I do doze, it ends up being for 5-10 minutes and then the alarm goes off. And five minutes after that, I’m fine.
  • And no early nights. As I said before, I’d already surmised that going to bed early was doing me no favours. I’m accepting that even after sleep restriction, I’d be a fool to be in bed before 11:00 p.m. It may even be that my new 11:30 p.m. bedtime is it.

So here we are. The one rule I have been breaking is the one that says you shouldn’t read for 90 minutes before bed. That last hour and a half, when the house is put to bed and the dishwasher loaded etc., is so deadly dull. I’m supposed to watch telly or something, but I’m actually usually done with TV by about now, shortly after 9:00, and there’s not that much good stuff on. And if I put on music or a podcast, I’d certainly fall asleep on the couch. So I do allow myself to read – nothing too stimulating: no modernism or stream of consciousness, no weird science fiction; just a nice relaxing thriller. I read no more than an hour, and stop 15 minutes before bed, and it seems to be fine. Well, 89% fine.

And so to brood

I have to say, I could see this moment coming. If you have a modicum of intelligence and an analytical bent, you can’t keep a sleep diary for very long without realising a few things. It was simply a matter of waiting for the shoe to drop. Obvious things like caffeine and alcohol and the  bedroom environment come first. But then, with three weeks of accumulated evidence, it’s obvious: if you’re getting about six hours of sleep but you’re in bed for eight hours, you’re spending too long in bed awake. Which we already knew. But the beginning of a solution is not to somehow, magically, get more sleep, but to spend less time in bed.

Inevitably, as I’ve struggled with sleep, I’ve been tireder earlier in the evening and I’ve been getting in bed earlier. And it’s only when you keep the sleep diary and look at what you’re doing to yourself that you realise how wrongheaded that is. Getting into bed at 9:45 in the evening isn’t going to net you an extra couple of hours sleep. So Sleepio is now telling me that I can get my six hours by going to bed at midnight. The theory is that you will get a better quality six hours, with fewer interruptions, and much less time spent awake. So the offer at the moment from Sleepio is not to increase my sleep but to improve it. Again, this much I had already surmised.

There are other parts of the programme that are less palatable. No reading in bed. And getting up when you are awake for more than 15 minutes, so that you’re not spending too much time in bed awake. Forcing yourself out of a warm bed in the middle of the night is no fun. And there’s no napping allowed, and even though you’re staying up later, you’re not supposed to do anything too active or stimulating. So while I might not normally mind staying up, if it meant I was reading something good or even creating something, I don’t think those activities fit the template of passive relaxation.

The first night, I started yawning and feeling really tired shortly after 9 p.m.. I put this down to a psychological effect. Anyway, I did it. And I got 5 hours and 39 minutes of sleep for my 6 hours in bed. Not bad, but it felt bad, because I was in deep sleep when the alarm went off at 6 a.m., and I did wake up at about 4:30 for about 10 minutes.

The second night was brutal. I stayed up watching Tinker, Tailor, Soldier, Spy on Britbox, made the pizza dough, watched some music videos on YouTube, watched more Tinker, Tailor, Soldier, Spy, and went to bed at midnight. This time, though I got less sleep: just 5 hours and 12 minutes, and I was fitfully awake from 5 a.m. Still, I was only lightly asleep when the alarm went, so it didn’t feel quite so hard to get out of bed.

I tried to read a bit, but I’m really not in the mood to read first thing in the morning, so I don’t know when I’m going to fit my reading in. 

Anyway, phase three is going to be the toughest, but I’ll try to stick it out. I am in blood stepped in so far that should I wade no more, returning were as tedious as go o’er.

The quality of drowsy is not trained

Having completed my second on-line Sleepio CBT session, and my first week of sleep diary, it’s clear that this process is primarily designed to retrain your thinking about sleep. It’s also clear that this process can only work if you’re willing to give it a go and have a positive and humble attitude.

Reading some of the reviews of the (not very good) Sleepio iOS app, it’s very obvious that a lot of people have felt fobbed off with this course, and that anger has stopped them from getting anything out of it. As I’d already given up hope with regard to the NHS, it would be hard to disappoint me further. For example, to see a doctor on 23 October about my ongoing sleep problem, I made an appointment at the beginning of September. In other words, at a point where I felt desperate enough about insomnia to see a doctor, I was told to wait 5-6 weeks.

So it’s no surprise to be handed a card and told to log on to a web site. But! I have an open mind and I was already primed to think about CBT, so I have given the programme my full cooperation. Note that the laptop/browser Sleepio experience is what I’ve been doing; I have the app but just for reference. There have been a couple of flaky moments with the online version (a freeze at a certain point in session 2, forcing me to use the Chrome browser instead of Safari to get over that particular hump), but it doesn’t crash like the iOS app does.

Session two discussed my sleeping environment, and made some obvious points about noise/distractions, light levels, temperature: nothing I wasn’t already aware of. I would never have a television in my bedroom anyway, and I know I sleep better when it’s dark. The double whammy of British Summer Time and lighter mornings can have a devastating effect. All that said, I do think I probably need to invest in a new/better mattress, and I’m going to ban clothes drying from the bedroom. I might also write to the school about the bloody milkman.

Another aspect of session 2 was to begin to challenge my negative thinking about sleep. Again, I can see how this particular moment might upset some people who might already feel fobbed off. We already know that facts are fairly useless in terms of challenging all kinds of attitudes. For example, the great nation of SUV drivers who are killing life on Earth with their unnecessary and antisocial cars don’t really respond to statistics about their emissions. Still: the Fitbit does tell a story, and I had already realised that it was sleep quality that I needed to work on, not sleep quantity.

Some common sense steps were suggested. Avoid caffeine 6 hours before bed. Avoid alcohol 4 hours before. Cocktail hour is fine, but drinking into the evening not so much. Which we knew, right? Add to this a kind of relaxation tape, coaching you with an hypnotically slow voice to clench and unclench your muscles, and you might get a better night. 

I’ve been on holiday anyway, and would normally sleep better, but for the hacking cough that is persisting well into its second week. But I approached bed last night in a more positive frame of mind, and the result was a much better night and a Sleepio score of 7.5/10. I still woke up an inordinate number of times, and the red dots around 4 o’clock still tell a story about the persistence of a bad habit. But that lovely trench of deep sleep around 5 a.m. gives me hope, it really does.

It’s still too early to say if this is working. The night before was 2.5, after all, and the one before that was 5.0. There is no trend line here, and I have yet to deal with sleeping in the home bed with the need to get up and go to work. But I’m giving Sleepio a cautious up-thumb at the end of the first week.

Sleep less with the rattle

Day 6. I woke up for that last time this morning, telling myself it was probably 4 a.m. Pitch dark, of course, and quiet, as it usually is around here. No milkman to wake me up in Auxelles Bas, though there is the chance of a rooster or a donkey. 

The day before after the night before, I’d recorded the diary in Sleepio, which awarded my night a score of 2.5/10 for sleep quality. One thing this diary has taught me is that, although I feel dissatisfied with the amount of sleep I’m getting, it is in fact the quality of my sleep that is the bigger problem. Not enough deep sleep, perhaps, but also too many wakes, and when I do wake, too many times when I don’t just roll over and go to sleep.

The rattling cough doesn’t help. It’s hanging around my chest and my ribs and head both ache from the violence of the coughing. So the roll-over-in-bed move is always accompanied by a helpless, strangled hack. Also, too many trips to the loo. I tried to knock myself out a bit last night with an extra glass or two of whiskey, plus some kind of tea drink supplied by our herbalist of a neighbour. But more trips to the bathroom seem to be part of this virus in any event.

I also made myself stay awake for longer the night before, because going to bed too early is a rookie move, really. 

So it was that I woke several times in that silent dark, and used my phone screen to light my way across the hazards of the bedroom floor to the bathroom, where I barked and hawked and peed. One of the wakes was at half-past midnight; the next two hours later. So it really could have been four in the morning, but it was, oh joy, five to six.

Five to six!

And Sleepio  awards me a doubled score of 5/10 for sleep quality. The triumph for me is that, although I did in fact wake (three times) at around four in the morning, it was never for long enough to be awake awake. But those three thick red dots at that time do show how hard my body tries to wake me up at four o’clock. They are the adversary: the three-eyed face of my enemy.

Slepys Diary: andso tobed

So here we are then, four days after the doctor’s visit.

I found a fix for the lying-awake-thinking-about-the-diary problem by buying a cheap* Fitbit Versa. It lacks many of the activity features of the more expensive Fitbits, but it does track your sleep, which is more than an Apple Watch can manage (out of the box). Also, its battery life is much better: I’m on 66% after wearing it for two days and nights without putting it on the charger. It also tells the time, though I suspect its screen waking in response to an arm lift is a lot slower and less reliable than an Apple Watch. Usefully, Fitbit syncs with Sleepio, which is convenient, and provides you with a useful graph of your night. It’s a shame the Sleepio phone app isn’t very good, however, as it crashes more often than not when I try to edit the sleep diary on it.

While (it says here) it’s “quite normal” to wake up 20-30 times during the night (☑️), it is less normal to be awake for 40 minutes for two of those periods. I’m also a bit more relaxed about the amount of “deep sleep” I’m getting, which seems to be slightly below average, but not terribly so.

No, the major problem is still the trigger pulled at 2:30 or 4:00, or whenever, or both, and the being wide awake after that, even though I don’t want to be.

Yes, I know about Second Sleep, and I completely buy the idea, but unfortunately our nation’s great* employers are not so sympathetic. If I could indeed potter abut for an hour or so and then go back to bed for Second Sleep, that would be great. But school timetables do not work on flexitime.

Day two gave me a Fitbit sleep score of 56 (rated “Poor”), although the amount of time I was asleep was fairly decent. I do think the Fitbit confuses lying still reading with being asleep, however.

Day three was always going to be a shitshow, because I was driving overnight to France. The Channel Tunnel was quiet at midnight, and we crossed over without delays and without seeing too many sharp-elbowed SUV drivers. But! Junctions 7-9 of the M20 were closed and the diversion was long and traumatic. I believe this closure was part of the preparation for leaving the EU, which made the diversion longer and even more traumatic. I drove for the first three hours, slept a tiny bit in the tunnel, then drove another hour or so. We then swapped drivers and I dozed a bit in the passenger seat. But I do jerk awake a lot (my other half is not a smooth driver) and I had the jimmy legs.

I then drove for the last 3+ hours and we arrived at 10 a.m. local time. Between 11 and noon I slept on the couch: 58 minutes, so that was my night’s sleep. That hour allowed me to do the round of visits until early evening, when I was wrecked, so an early night was called for.

Day four: A “fair” Fitbit score of 78, although it ended my night early, counting nothing after 4 a.m. as sleep, even though I think I did sleep a tiny bit more after that. But it was always going to be a weird night: in bed super-early and getting a reasonable 6 hours-ish, even if that did mean being awake too early. Also I’ve had a heavy cold and spent a lot of the night coughing or trying not to. But it’s the holidays, so I could just keep turning over in bed until 7 (local time).

The experiment continues. Today will involve some mowing and I may bury my parmesan in the garden. No cycling though, because of sore lungs.

*These things are, of course, relative