So what do you do in your bedroom at night?

Telling someone at a dinner party you are studying a PhD, has a similar result to telling them you are an accountant (professional apologies, but true). However, when I say I watch people sleeping in their bedroom at night, the normal buzz of dinner conversation drops as others tune in. Creepy. Voyeurism. Interesting. Depends on people's perspective. 

Given we sleep (or for some, attempt to sleep) approximately 1/3 of our lives, it is amazing that we don't know more about what individuals do in their own bedrooms ..... after shutting their eyes.

Now, most people that are crazy enough to embark upon an 8 year plan of study, have a personal burning motivation. It seems that I am weird in this space, because I have always enjoyed curling up in bed and pumping out the zzz's without little issue. Sure, I have the occasional nights where sleep is not ideal, but these are few and far between. What motivated me to peer into people's bedrooms and examine what they do is purely of professional interest. In particular, why is it that some people go to sleep with no spinal symptoms, yet wake with degrees of spinal stiffness and sometimes severe pain e.g. a lumbar shift or acute cervical torticollis?

Textbooks provided opinions about why this might occur, and that was generally all they were. A search of the literature at the time showed that the majority of research was based around sleeping pathologies - sleep apnea, restless legs syndrome, insomnia. Furthermore it was conducted in large metropolitan sleep based research laboratories, which meant access was limited for rural folk and expensive. There was nil information looking at sleep posture and a possible relationship it may or may not have on morning spinal symptoms. This was surprising to me, given the regularity of presentation once I started asking clients how they felt first thing in the morning. So my journey began.

Generally research classifies sleep posture into supine, prone and sidelying (also called lateral, fetal). Around 3 months of age, children first develop favorite sleep postures, as they are able to move around freely in bed. By age 7 they have developed definitive sleep postures and spend equal times in supine, sldelying and prone. As we grow older we start to habitulise our sleep routine, in teenage years we spent around 60% of the night in sidelying, 30% in supine and 10% in prone - rough figures. Generally past 60 yeas of age, more and more time is spent in sidelying, right more so than left. Sidelying in research is usually defined as everything not supine or prone, and that contains a large variety of intermediate sleep postures. On reviewing spinal biomechanics, it seem plausible that postures involving extension and torsion forces would most likely to be the ones that if sustained, could result in physiological creep of soft tissues and resultant pain. Applying this thought to the broad classification of sidelying, meant that postures like the 'recovery' posture for CPR, where the top hip is flexed and adducted with resultant spine rotation, needed to be differentiated from a fetal position, where the limbs are kept together with no resultant spinal rotation. Now we had four postures to define and differentiate, supine, supported sidelying, 3/4 sldelying and prone.

The first stage of my Masters was to look at developing a method of nocturnal sleep posture examination, that could differentiate between the four postures and secondly could be tested for reliability and validity. This was important, as I wanted to develop a protocol (hardware and software) that could be utilized by all clinicians in their own practice, if they wanted to offer a sleep assessment service for their clients. Furthermore, it needed to be portal, cheap and robust enough to be used in the client's habitual sleep environment.

Nicknamed OWLS - Observation Without Light System, it includes two infra red (IR) cameras (able to stream video in pitch black surrounds), stands, a video recorder, monitor and cables. The OWLS has gone through several periods of evolution and continues to be refined as technology advances forwards.

In the early days of testing, I situated the two cameras side by side, on a sturdy camera tripod.

Now, you have to appreciate that each camera has 30 - 50 small red IR LEDs that glow in the night. My wife basically freaked out on the first testing night, because when you woke, the first thing you saw were these two huge red, glowing eyes staring down at you from the foot end of the bed - bit like an unfriendly ET. It was back to the drawing board....

The benefit of having 2 cameras, was that the second camera provided a backup on data collection in case of failure of one, and that certainly happens. Placing the two cameras in different planes of view, also meant better 'illumination' and made visualisation of body parts easier. This essentially was the challenge. Create detailed and accurate enough images, to enable the determination of body positions, under sheet and duvet/blankets. It hadn't been done anywhere else in the world and I was staring to see why.

So the final setup for my Masters project was this. One camera was placed at the foot end of the bed, 1.8m high and the other one on an overhead support, directly over the middle of the average bed. It was built to be 2.3m high, enabling it to fit under a standard 2.4m ceiling. These were linked to a monitor, enabling checking of the camera alignment and a digital video recorder.

There was a small degree of fan noise from the recorder and with the proximity of the cameras, we also wanted to determine if there was any 'first night effect' due to our setup. The first night effect, is a phenomenon noted when participants undergo examination in sleep centers. Basically, the more foreign the environment, the longer it takes for a person to adapt and assume their normal sleep routine. In our situation, we wanted to be sure that the sleep postures we were observing, were reflective of a normal nights routine.

After initial testing and tweaking of the equipment and protocol on family, 15 participants were recruited for the Masters project. Each person was videoed for 2 consecutive nights and I got to sit down and watch people sleep - 30 nights in total. With clear sailing (viewing), that equated to 360 hours of watching people sleep. Unfortunately some of the video needed to be repeated several times. Each night is automatically recorded from 2000hrs to 0800hrs and I certainly came to appreciate those participants that went to bed after 2200hrs or later. I came to think of myself as a 'visual vampire'. I come out at night and absorb pictures of sleep, rather than pitchers of blood!

The results from this testing of the protocol are at the 'articles in press stage' with the open source Journal of Sleep Disorders: Treatment and Care Vol. 5 Iss. 2. Full text is available here.

A big thank you to Kathy and Michele for their guidance and support in keeping me on track.

On the writing front we are currently preparing a paper demonstrating the reliability and validity of the recording protocol. This will complete the Masters phase of the research.

On the PhD phase, I have been using the same recording protocol and am nearing the end of collecting data for our intervention study. We are looking at associations between sleep postures and morning symptoms, and whether people that are asleep can actually change their sleeping habit.

As always, look forward to hearing your thoughts in the Comments section.

- Doug