To improve your sleep the first thing to do is to get a baseline of where your sleep is at.
To do this, there are a number of options that can generally be split into two categories: Objective Measures and Subjective Measures.
The gold standard for Objective Measures is called a polysomnography (PSG). This typically involves spending a night in a sleep laboratory or hospital, with many wires hooked up to record brain waves (EEGs), eye movement (EOGs), heart rate (ECGs) and limb movements (EMGs) as well as oxygen saturation and blood flow. The end result looks something like this guy bellow.
The measuring equipment used in a PSG is the most accurate way of recording sleep. It is the most reliable way to pick up and diagnose sleep disorders such as Obstructive Sleep Apnea, Restless Legs Syndrome, Periodic Limb Movement Disorder, and many parasomnias. Similar equipment would also be used the following day during a Multiple Sleep Latency Test, which makes it easier to diagnose Narcolepsy and Hypersomnia.
The problem with a hospital based PSG is just how “normal” is the sleep that is being recorded, especially with sensitive sleepers or those that suffer from insomnia? Sleeping in a foreign environment and a different bed with restricted movement, limited choice in sleeping positions and sleep technicians coming into the room during the night to replace any wires that fall off can be a challenge. I normally don’t have difficulty in falling asleep anymore, but I did the night I spent in the Austin Hospital back in 2014 when I did my own PSG.
Fortunately there are other options.
There are now home PSG kits that can typically be hired at a cost from a sleep physician for a night. The benefit is that you get to sleep in your own bed, which may feel more normal, and there aren’t any interruptions during the night if any wires fall off. The negative is that you still have to wear all of the wires, which does limit position and comfort, and if something does go wrong or wires fall off then you won’t get as much or all of the required data.
A Sleep Profiler is potentially even better at getting a picture of what your “normal” sleep is like, even if it is slightly less accurate, as it records your sleep at home over multiple nights, and doesn’t have wires all over the place:
Once again, this can generally be borrowed for a fee from a sleep physician or specialist sleep clinic, who will then be able to score up the data for you and give you accurate, easy to understand feedback on what is happening with your sleep and what can be done to improve it.
During the clinical trial that I ran as part of my Doctoral research, we used a wrist activity monitor called an ActiWatch 2 by BMedical.
The ActiWatches were a good choice to record participants sleep as they were quite unobtrusive by comparison. They were simply worn on the participant’s non-dominant wrist like a watch for the duration of the study, and recorded the participants sleep, activity levels and light exposure over 14 consecutive days and nights at a time. This was convenient as I saw participants for a treatment session every two weeks, and could easily compare this objective data with their subjective data from the sleep diaries that they were completing to track their perceived quality and duration of sleep.
Wrist activity monitors are surprisingly accurate, especially the Actiwatches by BMedical, but they can come with a hefty price tag, with the ActiWatch 2 retailing for $1485 at the time of this writing. They can also be borrowed or rented out for less from some places if you want to just get a 2 week baseline period of how your sleep is.
The last objective measures that we can use to monitor our sleep are all of the new trackers and mobile apps that are being developed by companies recently for mass consumption by the general public. Some of these work by placing a phone or apparatus under your sheet or pillow, which is fine if you sleep alone, but not so good if you share the bed with a partner, as it is likely to pick up their movement too.
The three activity trackers that I have tried over the last few years are the Jawbone UP3, the Fitbit Charge and the Misfit Ray.
I found the Fitbit to be the most appealing initially, as it had a visual display which showed me the time and my progress throughout the day, versus the Jawbone and Misfit, which had no display and needed to be synced up to my phone so that I could view their data through their respective mobile apps.
The UP3 was definitely the most accurate for recording my sleep, and with the added benefit of the heart rate monitor gave me a reading of light sleep, deep sleep and REM sleep, whilst the other two only gave me light vs deep sleep. The Jawbone was the most expensive however, and the last time I checked the stores in Melbourne, I couldn’t find Jawbone products anywhere.
The big benefit of the Misfit Ray, and the reason why this is the activity tracker that I continue to use is that it uses batteries that only need to be replaced every 4 months versus the 5 day charge of the Fitbit and 7 day charge of the Jawbone. The Misfit Ray can also be worn in the shower and in the pool when I go swimming, so I don’t have to keep taking it off every time I am around water. It even has a bike and a swim feature, which helps to track a wider variety of activities more accurately across my week.
There are plenty of other companies that are getting into this market, and many clients will ask me if it is helpful to get these products. I do believe that they are helpful, providing that they are internally consistent. What I mean by this is that if you have a good night’s sleep and the data from the tracker typically shows a better quality of sleep, and then you have a bad night’s sleep and the data shows a poorer quality of sleep, it is useful to keep monitoring your sleep using this device. If the data that is being produced shows absolutely no relationship to how your sleep feels, it’s probably not going to be very helpful to keep monitoring your sleep in this way.
If you have tried the over-the-counter sleep tracking devices and still have no idea of what your sleep is actually like, a referral to a sleep physician from your general practitioner would be advisable, especially if you are concerned about your sleep quality, sleep duration, or the impact that your sleeping difficulties are having on your level of daytime functioning.
Subjective measures include questionnaires and sleep diaries. If you want to get a quick idea of your sleep difficulties, a validated questionnaire can help. The ones that I used during my research were the Insomnia Severity Index (ISI), The Pittsburgh Sleep Quality Index, The Epworth Sleepiness Scale, the Fatigue Severity Scale, the Sleep Hygiene Index, and the Dysfunctional Beliefs About Sleep Scale (DBAS-16). I continue to regularly use the ISI and the DBAS-16 to assess and monitor the severity of people’s insomnia in my clinical work that I do at the Melbourne Sleep Disorders Centre.
A more thorough way to get an understanding of someone’s perceived sleep is by using a sleep diary. There are many different sleep diary templates that can be downloaded from the internet and used for free. The one that I like the best is the American Academy of Sleep Medicine’s two week sleep diary. By getting client’s to graph their sleep visually after getting out of bed each morning, it makes it easy to see clear patterns and potential problems in their two weeks of data.
For my research I modified this diary slightly by adding in four boxes on the right side of the diary for more specific recording of the participant’s bedtime, how long it took them to fall asleep at the beginning of the night, how long they were awake for during the night, and the time that they got out of bed in the morning. Using just these four variables alongside the visual diary, I was then able to calculate an individual’s average time to bed (TTB), their sleep onset latency (SOL), their number of awakenings (NOA), their wake after sleep onset (WASO), their rise time (RT), their time in bed (TIB), their total sleep time (TST) and their sleep efficiency (SE), which is their TST/TIB. If filled out thoroughly by a client or participant, it is also possible to look at the impact of medication, caffeine, alcohol, exercise and work on sleep, as well as the timing of each of these variables.
Here’s my two week sleep diary from the start of 2017:
Now hopefully you can read the instructions as well as the data, but if not I will break down the important information to be gleaned from this chart:
- Even though I was not trying to do anything special in relation to my sleep, I was pretty good at sticking to my eight easy steps for good sleep.
- I went to bed and woke up at similar times, with the only exception being the Sunday night where I crashed early after playing a beach volleyball tournament all day. For my wake times, 6:50am was the earliest and 8:00am was the latest, which is well within the maximum 2 hour window that I recommend to clients.
- I only had caffeine once after 1pm, and not at all after 6pm.
- I waited until I felt sleepy before going to bed, and as a result I always managed to fall asleep within 10 minutes.
- I woke up on average once per night, but generally had no difficulty in returning to sleep. The most times that I recalled waking up was three times, but that was an extremely hot night.
- I still have a minor delay in my circadian rhythm, which means that I am a little tired upon awakening, but this typically wears off once I shower and have breakfast.
- I do some form of exercise everyday, but sometimes do it too late, such as when I have a 10:10pm basketball game. I need to stay up a bit later to wind down and wait until I feel sleepy on these nights (see 12/1). For this reason I sometimes don’t get as much sleep as I would like to on weeknights.
- In general, I am getting enough sleep with 7hours and 9minutes per night on average.
- I am definitely not spending too much time in bed, with an average of 7hours and 25minutes per night.
- Based on my excellent sleep efficiency of 96.4%, I would suggest that I could actually benefit by spending 15-30minutes more in bed each night.
If a client or participant came to me with a sleep diary that was identical to this, I would recommend for them to aim for a consistent sleep schedule between 11:45pm and 7:30am over the next two weeks. If they wanted to bring their bedtime and rise time forward earlier than that (say 11:15pm-7:00am), I would encourage them to still wait until they feel sleepy before going to bed and to also focus on minimising late night exercise, getting 20-30 minutes of sunlight exposure when possible after rising in the morning, staying away from bright screens in the last two hours before bed and actively winding down and relaxing before bed to lower their arousal level and bring on sleepiness earlier.
Objective vs Subjective
Here is the objective data from my Misfit Ray for 4 nights of the above sleeping period. In order to compare the Misfit to the sleep diary data, look at Jan 10 on the Misfit and 9/1/17 on the sleep diary, as the Misfit shows the date of the next morning and the sleep diary the date that the night begins on:
The main discrepancy is the 37 minutes awake shown on the Jan 13 Misfit data, which is the night that I had a 10:10pm basketball game (12/1/17). I don’t recall being awake for that long (or at all) once I fell asleep that night, but I could have potentially been more restless at the start of that night due to the late exercise. My poor quality of sleep on that night relative to the other nights of data does back up this theory.
The Jan 15 data also shows an exceptional night of sleep, with almost double the amount of restful to light sleep. Considering that this was a Saturday with no work, no caffeine after 10am, plenty of relaxing and a nice swim between 4 and 5pm, it does make sense that it was objectively a really good sleep, even if the awakening wasn’t picked up by the Misfit.
Apart from a few minor differences, the consistency between the subjective and objective data is actually quite impressive, and something that I hope will continue as I begin my journey of manipulating one variable every 2 weeks this year to look at the impact that it has on my sleep. The first episode that will be released using both video footage and a blog post on caffeine.
Thanks for reading and please check out my sleep detective website for more interesting sleep experiments!