How to shift your bedtime/waking time to earlier

When I was using light therapy alone, I found that my 25 hour pattern shifted to 24 hours beautifully but that sooner or later, I would stay up too late and would end up stuck on falling asleep at 4 am or so again.  DSPS is a tough nut to crack.  This was in the days before I found darkness therapy, which has made my sleep/wake pattern rock solid, so here’s how I treated it at the time.

My preferred sleeping tablet when I need one is 20mg temazepam; many people find that 10mg is enough, but I need a higher dose.  While my GP advised me that I should be able to take it for up to a week, I found that taking it for a week straight caused rebound insomnia when I stopped, presumably because the ME/CFIDS makes me over-sensitive to medication.  I find that taking it for three consecutive nights is fine, so I work with that.  It’s entirely possible that I’d be fine with five nights, but I’ve never needed to try.

I would always recommend strongly that you discuss this with your GP.  They should be made aware of your sleeping problems anyway, even if you’re trying to solve them yourself, and sleeping tablets are fairly serious things, even the over-the-counter ones.  Discuss which sleeping tablets will best meet your needs, and if you’ve had a sleep problem for a while you’ve most likely tried several by now anyway.  Do tell them that you will only be using the tablets for a few nights in order to shift your sleep pattern, as otherwise they will be a lot more reluctant to let you have any meds.  I’ve managed to convinced a GP who’d never met me before to let me have four sleeping tablets for this purpose when it was an emergency (missed flight, unexpected night flight), and I doubt that he’d have let me have anything at all if I’d just wandered in to say that I was a bad sleeper and wanted some drugs please.

I’ve always been able to move my sleep back enough over three days using this method, but if I had needed to control a larger shift I would probably have done three days, waited a week, and then repeated the process.  My general lightbox treatment time is 45 minutes, but for this I would sometimes use a longer treatment time to help reinforce the circadian shift.  As far as I can recall, I used an ordinary alarm clock to make sure I got my light therapy at the right time the next morning, but dawn simulation would probably be even better.

Let’s assume that my bedtime is usually 1 am but has shifted to 4 am.

Day 0 – bed at 4 am, wake the next day at 12 pm.
Day 1 – sleeping tablet at 2.30 am for 3 am bedtime, bright lightbox at 11 am.
Day 2 – sleeping tablet at 1.30 am, lightbox at 10 am.
Day 3 – sleeping tablet at 12.30 am, lightbox at 9 am.
Day 4 – no sleeping tablet, lightbox at 9 am.

After that I would relax and use the lightbox whenever I woke up, instead of setting an alarm clock.  This method worked for me every time, including when I had to go on a night flight and deal with a 3 hour time difference to boot.

If you are using darkness therapy, you may not need the sleeping tablets at all.  Just put on the tinted glasses/switch over to orange lighting an hour earlier every evening.  The usual recommendation is to commence darkness therapy three hours before your desired bedtime, though some people find that they get sleepy too early if they do this.  Obviously getting sleepy too early is not a deterrent for DSPS sufferers!

For the above problem, I’d suggest starting the darkness therapy three or four hours before your current bedtime to begin with, and seeing what happened.  It may take a few nights to kick in fully.  Use the bright lightbox when you wake up.  You can wait until you wake up naturally, depending on how effective the darkness therapy is for you, or you can set an alarm so that you make yourself use the lightbox an hour earlier every day.  Since you’re not using sleeping tablets and therefore don’t need to worry about getting the process completed in a hurry, you can try shifting your sleep more slowly, even by 15 min a night.  Keep a diary of what you’re doing so that you don’t lose track.

If you have Non-24 Sleep-Wake Disorder, I’d suggest waiting until your circadian clock has shifted around to your ideal bedtime and waking time before trying anything, then hitting it with light therapy, darkness therapy and/or sleeping tablets to stabilise it there.  The sleeping tablets are a short-term measure, but the light and darkness therapies can be continued full-time and indeed should if you have a circadian rhythm disorder.

If your problem is jet lag or shift work, rather than a misbehaving body clock, you’ll have to experiment to find out what suits you best, and you may only need to use light/darkness therapies occasionally.  Sleeping tablets are best reserved for occasional use, so if you are going to be moving your sleep pattern every week or so, I wouldn’t advise them, and I would certainly suggest that you discuss this with your doctor.  Some companies give their night shift workers yellow safety goggles to wear when they go home in the daytime, so that the light on their journey won’t keep them awake.

If your problem is instead Advanced Sleep Phase Syndrome, then use bright light therapy in the evenings instead to keep you awake for longer.  I have absolutely no idea how darkness therapy would factor in here, but if you’re using it for other purposes (e.g. sounder sleep), I would imagine you would want to be careful not to start it too early in the evening, and remember that darkness therapy alone is unlikely to shift your body clock in the desired fashion.  Light therapy will be the key here.

For any of these problems, dawn simulation alone is unlikely to be enough to shift your body clock, at least in my experience, but may be very helpful in sticking to a good pattern once you have one in place.  I recently went back to dawn simulation, and while I had already stabilised my body clock using light therapy and darkness therapy by now, I think it may be adding a little extra help, and my partner is certainly finding that the dawn simulation makes it much easier to get up in the mornings.

Dawn simulation

Dawn simulation works by gradually turning on a light, generally 40w or 60w, over a period of time, most typically 30 min, in order to simulate a sunrise.  The light goes through your eyelids and moves your sleep stage to the point where it should wake you up naturally, meaning that you wake feeling as refreshed as you’re going to get and that hormones such as cortisol do what they’re meant to.  I find that when it wakes me up, it feels like I blink and I’m awake, rather than being dragged kicking and screaming into consciousness.  They all have backup alarms you can set just in case.  Most people don’t need them, but some do, and can still find that while the light isn’t quite enough to wake them up on its own, they do feel much better when they wake up.

As well as making it easier to get up in the morning, some research has found dawn simulation to be as effective or nearly as effective as bright light therapy for treating SAD, and it may work well for non-seasonal depression too.  This is despite the fact that it doesn’t use such bright light, just an ordinary bedside lamp or equivalent.  It’s thought to be something to do with the gradual increase in light having the effect.  The advantage over bright light therapy is that the treatment is over by the time you get up, you don’t have to tether yourself to a light box in the morning, but the disadvantage is that it may not be as effective.  Companies selling both tend to recommend it for mild SAD and bright light boxes for more severe SAD, I’ve noticed.

Dawn simulators also generally offer dusk simulation as well, where the light fades down gradually to help you get to sleep.  I think there’s some research around showing that it helps a bit with insomnia.  I’ve not really used this setting much as it’s less convenient, though I find it relaxing when I do.  I use dawn simulation because it helps stabilise my sleep pattern even further, and because it helps both myself and my partner to get up in the morning.

One common problem with dawn simulators is that they tend to buzz when the light is ramping up or down.  The best solution is to get one of the new energy-saving halogen bulbs, also called halogen incandescent, which are the same shape as standard incandescent bulbs and can be dimmed.  Not only will they save you at 30% on energy (42w is equivalent to 60w and so on, though I and many other people think they actually seem to be brighter than that) and last longer, but the light is good quality and they don’t buzz.  You can’t use fluorescent bulbs in a dawn simulator (I never recommend them anyway as the light causes problems for so many people), you can’t use the usual halogen bulbs, you can’t use LED bulbs if you’ve managed to find one, you can just use them with standard incandescent or energy-saving halogen.  This is for where you put in the bulb yourself: there are a couple of types of dawn simulator around using fluorescent or LED lights already built in.  If you get the type of dawn simulator which plugs into a lamp, again it has to be the sort of lamp which takes incandescent bulbs, and it can’t be a lamp which already has any sort of dimmer fitted, such as a touch lamp.  Just use an ordinary table, bedside or desk lamp.

There’s a new type of dawn simulator which is combined with a lightbox, such as this one (do read reviews, there are some poor quality ones out there by other manufacturers).  They may be a good solution for people wanting to try both dawn simulation and bright light therapy, but bear a few things in mind.  Firstly, it comes up to full lightbox brightness rather than the brightness of a 40w or 60w bulb, so that you will most likely be awake long before it’s anywhere near fully bright.  One solution could be to set a 90 minute sunrise and assume that you’ll be awake after, say, 30 min.  You need your dawn simulation to be by your bed but are unlikely to want to be using your bright light box by the bed, though if you read in bed in the morning or have your laptop close to your bed you could be OK.  The other point that occurs to me is that since the light is much brighter than is needed for dawn simulation, you could place it further away from the bed (e.g. by your computer, if it’s in the bedroom) and just point it in the right direction.  If I was starting light therapy all over again, I’d probably have gone for one of these and hoped that I’d be able to work something out once I’d fiddled with it enough.

I started off years ago with an all-in-one Bodyclock by Lumie where the lamp is built in, but even though it was a 60w bulb I found that it wasn’t bright enough for a reading lamp, plus Lumie often get slated for poor product build quality.  I sold it on eBay and bought myself a Sunrise System which plugs into a lamp or lamps of your choice, and vastly prefer it.  Mine  eventually became faulty and I sent it in for repair, where they reported that it had been damaged by a power surge (at which point I promptly bought a surge protector) and sent me a replacement which is slightly different from the older model.  They can take a bit of getting used to, and I think the company needs to continue tweaking them (though at the very picky level), but generally they’re excellent.  They have lots of useful features, such as being able to set the time individually for each day of the week.  We have it set to finish at 9 am Mon-Sat and 10.30 on Sunday, when my partner starts work later.  Actually he starts quite a lot later on Sundays, but I want to keep myself in a good routine while having a bit of a lie-in.

One of the great things about dawn simulators of this sort is that you can plug them into more than one lamp as long as you don’t exceed the total wattage, which in the case of the Sunrise System is 200w in the US and 300w in Europe.  That’s enough to hook it up to lights all over the bedroom if you like, though most people like to hook it up to a light on each side of the bed.  We’ve found that I get woken up better than my partner does by light, that he doesn’t like waking up to full light and generally prefers it to be dimmer on his side, that he still needs his alarms as he’s trained himself to respond to those, and that I take longer to get up once I’m awake, so that by the time he’s had a shower and got ready for work it’s about the time that I’ll be getting up.

So on my side of the bed, the dawn simulator is hooked up to my bedside light, which is a 40w equivalent low-energy halogen bulb in an anglepoise lamp that faces the wall, and on my partner’s side there’s a little 25w spotlight hidden behind a vase which points into the corner and is enough for ambient lighting.  My partner’s bedside reading light is independent of the dawn simulator.  The dawn simulation starts at 8 am, which is when the first of his three alarms go off.  He gets up at 8.30 when the light is up to half brightness, and I get up at 9 when the light is at full brightness.  As there are alarms going off from 8, I can be anywhere from awake to dozing to having fallen asleep again while the light is increasing, but it doesn’t seem to prevent it from being effective.  Occasionally I sleep in later than 9, but it’s probably only once a week, which for me is fantastic.  This helps keep my sleep stabilised even if I went to bed too late the night before.  I’m getting up earlier than I used to (even when I was using the lightbox), and it’s great being able to get up earlier in the morning and go to bed at the same time as my partner.

What colour light should a lightbox produce? White, blue, full-spectrum?

There’s rather a kerfuffle going on amongst lightbox manufacturers as to what is the correct colour for bright light therapy.  Older models of lightboxes all used white fluorescent light, and I think that what is going on is mainly about the longer-established manufacturers trying to hang onto the market, while the new ones are trying to persuade customers that their exciting new product is superior.  So on one side you have old-fashioned white lightbox manufacturers baying that blue light will ruin your eyes, and on the other you have blue lightbox manufacturers claiming that blue light is all that matters.

From all the research I’ve read, it does seem that blue light is the most effective bandwidth.  Blue light is, of course, contained in white light, and even the brightest lightbox will not be as bright as outdoor sunlight.  If thirty minutes in front of a little lightbox was enough to wreck everyone’s vision, we’d all be blind from the sun by now, even with modern indoor living.  Psycheducation.org has written an excellent article on the purported dangers of blue light, concluding that unless you are at high risk of macular degeneration, you don’t need to worry, and to be sensible and not stare directly at your lightbox.

This spectrograph was produced by Apollo Health, the former manufacturers of the first blue lightbox, the GoLite, so it’s not from an unbiased source.  However, all the research I’ve read agrees that 470nm or thereabouts is the most effective bandwidth for affecting the circadian clock, and I’ve seen several spectrographs of fluorescent light (which is the  “standard full-spectrum” in this graph) that look the same.  Fluorescent light is spiky, which is apparently why it’s so harsh on the eyes, and even the full-spectrum versions don’t peak at 470nm.  Presumably this is why fluorescent lightboxes are a great deal larger than LED lightboxes and require longer treatment times.

Something interesting about LEDs is that the blue ones all seem to peak around 470nm naturally (warning: the page has lots of images and takes a long time to load).  This is why I suspect that using a standard blue LED bulb may not be as strong as a fancy lightbox, but it’s the right type of light and if used for longer, may have the same effect.  Typical white LEDs contain a broader bandwidth of light overall, but they turn out to peak around 470nm too (also a long page).  Here’s a typical example of a white LED bulb.

Some lightbox manufacturers claim that their LEDs are “blue-enriched white”, but frankly I think that they’re perfectly ordinary LEDs, the manufacturers are just cashing in on the fact that they happen to peak in the blue bandwidth.  (The advertising shenanigans of light therapy manufacturers can drive you demented at times.)

This means that white LEDs produce a cold light that most people don’t particularly like for domestic lighting, but it makes them ideal for bright lightboxes.  I’ve concluded that while blue light is the most effective, white LEDs are so close behind that you may as well choose your lightbox based on your personal preferences.  Some people find the blue light to be softer and less disruptive, other people find that they need white light so that they can look at colours properly (I once made the mistake of trying to do some embroidery with my GoLite on, and had to unpick it all afterwards as I’d picked up the wrong coloured threads), and a few people, such as my partner, just can’t get on with coloured light.  Choose according to your visual comfort and the other factors involved in choosing a lightbox, such as price, overall product quality and the various features that a lightbox can have.

If you prefer fluorescent lightboxes, there is the question of whether to go for normal or full-spectrum.  As far as I can tell, it doesn’t make any difference therapeutically, so go for whichever light you find to be most comfortable for your eyes, if you can even tell the difference.

***

Images courtesy of the LED Museum.

Bright light therapy

This is best known for treating Seasonal Affective Disorder, where it has been found to be as effective as anti-depressants, although the most effective treatment of all is to combine light therapy with anti-depressants.  What is less well-known is that bright light therapy is almost as effective in non-seasonal depression and is extremely useful for circadian rhythm disorders.

The main myth about bright light therapy is that it’s about full-spectrum lighting. It isn’t.  There are quite a few myths about light therapy which have been set up by rivalrous product manufacturers, and this one is an attempt to sell certain fluorescent lightboxes by claiming that they are in some way better than the others, which has somehow turned into the myth that all you need is a standard full-spectrum bulb.  The most effective wavelength for resetting the circadian clock is 470nm, blue light, so you need either blue light or white light which has plenty of blue in it.  (A couple of manufacturers favour green light for rather complicated reasons which you can read about here, where it’s concluded that they’re not worth trying unless you are at high risk of maculuar degeneration.)  The other important thing is the intensity of the light.  When a traditional fluorescent lightbox is used, 10,000 lux is preferred, though there are a few models around which use 5,000 or even 2,500 and require proportionately longer usage times.  Merely installing a full-spectrum/natural daylight bulb into your domestic light fittings will not make a blind bit of difference, as the light isn’t anywhere near strong enough.  Full-spectrum bright lightboxes were probably introduced because so many people find fluorescent light to be visually unpleasant, and there is a demand for a better colour temperature.  If you are going for a fluorescent lightbox, you may want to see if you can view a normal and a full-spectrum version to find out if you do prefer one or the other, but to be honest I think very few people are bothered about this, and that you’re better off using an LED lightbox if you don’t like fluorescent light.

Lightboxes are used for periods ranging from 15 min to 2 hours, depending on the type and the patient’s needs.  Unless you have Advanced Sleep Phase Syndrome, where you fall asleep too early in the day, the usual time of treatment is first thing in the morning, just after you get up (which for some people isn’t actually morning!), whether this is for sleep disorders or depression.  If you have DSPS or Non-24 Sleep-Wake Cycle, using bright light therapy in the morning has a high chance of stabilising your circadian clock and can even move it backwards so that you are falling asleep and waking up earlier.  I found that using a bright light box just after awakening immediately stabilised my sleep pattern at 24 hours instead of the 25 it had been on for years, and I started this four and a half years ago.  However, if my bedtime and waking time were too late, I needed to combine bright light therapy with a few days of sleeping tablets, taking the tablets an hour earlier each evening.  After that, the morning light therapy would serve to keep my waking time where it should be.  This has worked pretty well for me, although my sleep pattern wandered out of synch a few times a year and needed to be chased back.  Adding darkness therapy into the mix seems to have stabilised it completely.

There are two main types of lightbox, compact fluorescent and LED.  Fluorescent are the traditional ones and have been around for longer.  They are larger, require a longer treatment time, produce white light, and may cause medical problems for people sensitive to fluorescent light, such as migraine, visual problems, dizziness and so on.  The manufacturers try to dodge here and say that they have better ballast, but in my experience they’re just as bad as strip lighting and can cause nasty migraines. Perhaps they are better constructed, but since the light is so much brighter and it’s right by your face, it’s still more than enough to cause problems if you have difficulty with fluorescent light.  On the other hand, because they’re larger and the light is not as directional as LED light, they permit a little more freedom of movement when you are in front of them, though you still can’t move far away.  The light is also more likely to end up in the optimal position above your eyes, though since fluorescent lightboxes take much longer to use than LED lightboxes despite this, the advantage is probably cancelled out.  When looking at the stats for a fluorescent lightbox, most will say that they produce 10,000 lux, but you should also check at what distance that measurement is accurate.  If it’s 10,000 lux at 20 cm, that’s far too close to the light to be comfortable.  60 cm is more sensible, or alternatively just realise that you will need to use the lightbox for longer.  I think that this is how manufacturers are getting around the problem that everyone is told to look for 10,000 lux, but no one is quite sure what it really means and don’t realise that it’s only the measurement of the light intensity at a certain distance.

LED lightboxes have been around for a few years and may produce either white or blue light, which you can read more about here.  They are smaller and require a shorter treatment time.  Not everyone gets on with coloured light, and as the LED panel is composed of lots of little LEDs that look like dots, some people report getting spotting in front of their eyes, although I suspect that they were using the lightbox incorrectly.  You’re not meant to stare into it, you’re meant to position it at the side, or even better above your eyes, so that it hits your peripheral vision. LED lightboxes do have the disadvantage of needing to be placed more exactly and requiring you to stay in the same position, as the light produced by LEDs has a very narrow beam angle.  I use mine by my laptop or while sitting at my sewing desk, and while sometimes I need to prop up the lightbox to get it at the right angle, after that it’s fine.

I have a GoLite, previously made by Apollo and now made by Philips, which is a blue LED lightbox and very highly thought-of.  I also have a Lite-Pad, which is a cheap white LED lightbox I picked up on eBay for a tenner.  I use this one by my sewing table to save messing around with cables when I want to quilt during my morning lightbox stint, and also so that it doesn’t affect my colour judgement.  The GoLite does have a lot more bells and whistles, such as a clock and the ability to set both the light intensity and the length of time for up to three different preset programmes, but you’re paying a lot for them.  The Zadro looks like the best of the cheap lightboxes if you’re in the US, and indeed looks like a very good lightbox in its own right.  You can also buy combination lightbox/dawn simulators, of which more later.

There’s another type of bright light therapy around, known as the light visor, where the unit is placed in a sort of cap worn on the head, and the light is shone into the eyes from above.  Lumie makes a few, and there’s one which produces blue-green light around.  Their one advantage is that you don’t have to be tethered to your lightbox, you can move around.  Psycheducation.org is doubtful about them, and I agree.

Since you can’t tell which lightbox you will get on with in advance, or even if it will work for you, and since these things are very expensive, find a company who will offer either hire-purchase (e.g. the National Light Hire Company) or a free trial (e.g. Lumie).  You may need to spend a while working out the right amount of time to use the lightbox for.

Of course, an entirely free alternative is simply to make sure you get outside for at least one hour every morning.  It doesn’t matter if it’s cloudy.  I’ve run into a woman online who managed to get the same effect by using a 150w fluorescent light (that’s actually 150w, not the equivalent to 150w incandescent), such as this grow light, in her overhead light during the whole day, not just for an hour.  This is far, far brighter than normal domestic lighting, and while it won’t save you energy and many people (especially with ME) will find it uncomfortable on the eyes and/or likely to provoke migraine, for some people it’s a good solution.

Another possible solution is to buy an LED bulb in white or blue and sit with it at a level just above your eyes for a couple of hours a day.  The blue bulb will be exactly the right wavelength, and while the white won’t have as much blue, white LEDs peak at the right wavelength so it will have a lot of it.  It won’t be as strong as a therapeutic lightbox but if you use it for long enough and keep it close to the top of your eyes, you may be able to get a decent result out of it.  If anyone tries this, do let me know how you get on.  I’ve got one of the blue bulbs which I’ve used for experimenting with moonlight simulation (this is popularly known as Lunaception and is meant to improve menstrual cycles, although for me it just wrecked my sleep), and I’ve noticed that I tend to feel wired after looking at it, and that using it for half the day was enough stimulation that I had a great deal of trouble sleeping that night.  So I reckon there’s quite a bit of potential with those little blue bulbs, and they’re better filtered than lightboxes so you don’t get spotting even if you look straight at them.

Light and darkness: an overview

Arguably the biggest factors in sleep pattern regulation are light and darkness.  Humans evolved outdoors, getting plenty of strong daylight during the day and complete darkness at night, and averaging 12 hours of each.  It’s this light/dark signal that keeps the body on a 24 hour schedule: people who are completely blind almost all have sleep disorders, as the natural body clock runs on a 25 hour schedule for some bizarre reason and they don’t have the light/dark signals to keep it at 24 hours.  Now we sleep indoors, we mostly work indoors where the lighting is nowhere near as strong as sunlight, many of us barely get any   sunlight (and those of us with ME, or housebound due to other medical conditions, may not get any), and instead of following the natural pattern of darkness, we are in darkness only for the time we sleep and that may not even be complete darkness, and we will be under artificial light right up until bedtime.  This chart shows the relative light level from various outdoor and indoor conditions.  Even a well-lit office is still only 10% as bright as an overcast sky, and nighttime road lighting is 50 times as bright as a night with a clear full moon.  Our light/dark signals are all mixed up, and this is showing in the  high prevalence today of not only sleep disorders, but medical conditions which are affected by light/dark.

The very basic version is that bright light stimulates serotonin, and a lack of it can cause low serotonin levels and thus depression, as well as daytime sleepiness.  The main antidepressants used today are SSRIs, selective serotonin reuptake inhibitors, and there is a form of depression which is directly caused by low light levels during the winter, SAD (Seasonal Affective Disorder).  Cortisol is another hormone affected by light levels.  Melatonin is the hormone which makes us feel sleepy, along with a host of other roles in the body, and melatonin is produced when we are in darkness, which should average out to 50% of our time over the year but is now nothing of the sort.  The healthy pattern is to start producing melatonin a few hours before going to bed.  By using artificial lighting until right up to bedtime, melatonin production is inhibited, thus ensuring that we are less likely to feel sleepy when we go to bed, and also that we get less melatonin overall than we should.  All the research I’ve read agrees that we need to have melatonin coursing through our bodies for a certain number of hours per day, and that getting insufficient melatonin impacts on various areas of health, such as the immune system, as well as sleep.