Tried it: Sleep Restriction Therapy

Craig of that time period could curl up in the corner of Grace St (Richmond, VA) apartment on a hard floor with a bathtowel for a pillow and wake up 100% refreshed. Also pictured, at least three awesome people Craig of the current time period would love to see more often.

I've struggled with insomnia since my sophomore year in college. Organic Chemistry is a famously intense class, and many professors require students to memorize a large volume of reactions. I am terrible at rote memorization.

My roommate was also in the class, and he had a hard time too. Our study and coping methods did not mesh well. He stayed up late and hit the sack hard, developing a powerful snore. I dropped hobbies and stuck to a conventional diurnal schedule instead. That combo meant I got little sleep at night and little relaxation during the day.

That was 16 years ago (!) and I have suffered through insomnia since. I've tried various things, depression medication, anxiety medication, talk therapy, hypnosis, etc. They worked not at all, not for very long, not at all, and not consistently, respectively.

So now, I'm going to try something new, something with a lot of good press, something intense. I'm going to try sleep restriction therapy.

The idea is to increase the efficiency of your sleep first, then get the total quantity of sleep back up. Most of the instructions online are similar, but I'm going with the approach described in this video.

1) Calculate how much actual sleep you're getting. Thanks to my FitBit, I have some data to work with here already. It says I'm averaging 7 hours. But much of that is light sleep, and I know from being the actual person that a good chunk of that is me lying still hoping to get to sleep. So I'm going to peg my actual sleep average at 6 hours. Most instructions don't advise you set it at less than 5 hours.

2) Pick a wake-up time, then set a bedtime that gives you just the amount of time in bed that you calculated as your actual seep time in step 1. My commute requires me to be up at 6:55 am two days a week, and you have to use a consistent schedule in this therapy. So 6:55 am wake-up means 1 am bedtime.

3) For at least a week, don't go to bed before your bedtime, and don't stay in bed past your wake-up time. I'm told this will suck.

4) If, after a week, sleep efficiency has improved—which is to say that you're sleeping for most or all of the time you're in bed—move your bedtime back by half an hour. Make that check and shift once per week. If sleep efficiency hasn't improved, stick with the shorter sleep time. If it slips as you add time in bed, stop adding time in bed. If the slip stays put, remove a half hour of time in bed.

And you just do this till you're getting 8 hours of good sleep a night, or whatever it is your body needs (hint, for most people it's 8. A lot of people chronically get less and are chronically impaired).

So, we start tonight, 7/26/2018. Let's see how it goes.

• Day 1. The first night went fine. I watched a movie with my wife. The FitBit didn't get good data for some reason. It mistook the hour of me sitting around before going to bed as being in bed, and didn't get any detail about light vs deep vs REM sleep. But supposedly I was in bed 6:20 and asleep for 5:49, for a sleep efficiency of 349/380 = 92%.

• Day 2. Second night it was reeeaaaally hard not to fall asleep while putting my four-year-old to sleep. But I think I did it. I once again hung out with Whitney with the TV on for the last hour, which I know I shouldn't keep up. She shouldn't be sleep restricting with me, and late TV is bad. I accidentally slept in by 47 minutes, whoops. The FitBit says I was asleep for 5:58, with a time in bed of 6:34; 358/394 = 91%. The first solid hour of that was deep sleep and I got 1:37 of REM sleep. That's better than normal.

• Day 3 and 4. Next two nights went about the same. Asleep 5:28, in bed 6:30; 358/390 = 92%. Then asleep 4:59, in bed 5:32; 299/332 = 90%. This morning was the first time I managed to get up on time.

I want to pull a few numbers from before the experiment. So on July 25, I was in bed for 9 hours and asleep for 8:19; 499/540 = 92%. 7/12 it was 85%. Picking days at random because I can't find a good data export option and I'm too lazy to transcribe by hand, I have been rocking around 82% about a third of nights and around 90% for two thirds, according to the FitBit data.

If this is correct, I don't actually have a problem. But I think it overestimates my time asleep. FitBit measures sleep by movement and heart-rate variability. Are you still? Are the movements you are making ones that match with sleep? It's going to call you asleep. But I often will be lying very still and yet be awake and resenting it. Maybe I need to go into a lab and measure brain-waves to get an accurate count. But I'm going to focus more on how I feel and watch the FitBit data mostly for the delta.

One thing I've noticed is that I'm only a little bit more tired than I normally am. That suggests to me that around 6 hours of actual sleep isn't too far off normal. If that very soft math is right, my sleep efficiency is generally around 75%. But I'm a little afraid. What if 6 hours of sleep and 18 hours of feeling tired and being slightly above average is the best I can do? What if 8 hours of sleep and 16 hours of feeling good and being awesome isn't something I can achieve?

• Day 6. I'm so tired. It took immense willpower to stay up till 1am last night and getting up at 7am this morning was almost as hard. My work is slipping a little, and I'm having a bit of aphasia, which is normal for me when I'm sleep deprived but super inconvenient because I'm training a new hire right now. I have been sleeping deeply and staying asleep, though last night I was tossing and turning a bit. The combination of an uncomfortable bed and eating too much too late at night, I think. 

Current mood
By Martin Falbisoner - Own work, CC BY-SA 4.0, Link

... And that's about when I didn't have enough willpower to do my job, do this experiment, and keep blogging about it. But it was at about day 8, when I started to add sleep back in, that the whole thing fell apart.

I had been seeing the density of sleep I was aiming for, so I started step four from above. But as soon as I started to let myself go to sleep earlier, I slipped and slept in, too. I tried to claw it back, but I couldn't. So it was time to reflect and determine what gains this had yielded.

First: I got a lot done staying up to 1am. Of course I had the extra time, but no one else was awake and I knew the time was coming. I read books, watched movies, did some paid freelance work. And for some reason, late at night I was able to resist wasting time on social media. It'd be good to recapture some of that during the daytime.

Second: Honestly, it's been six months since I started the experiment. I kinda lost track of this post. But where before I was sleeping badly almost every night, which is to say waking up tired in the morning after waking up repeatedly at night, now it's down to "often." And after the sleep restriction therapy is when I started to sleep better. So maybe my brain and body did get a little taste of what really being asleep is supposed to feel like.

Third: Doing this experiment and talking about it forced me to observe and talk about exactly what is different with my experience of sleep. I wake up too much. I toss and turn. I'm told that for many people, most nights, they go to sleep and then pop it's morning and they weren't perceiving the passage of time all night long—sleep is almost never a time machine to breakfast for me. I wake up and need to go to the restroom. These are specific symptoms that I could research and discuss with an expert.

I think that if I'd held to it, I would have seen greater improvements. And I think it's worth trying again.

#goals
By FOTO:FORTEPAN / Magyar Bálint, CC BY-SA 3.0, Link