Sleep watches, actigraphs and fitness trackers all have different names and different brands, but all claim to do the same thing — measure sleep. In 1973 (the year the very first cell phone call was made) Daniel Kripke, at the University of California San Diego, created the first actigraph or sleep watch — the first “wearable”. It was designed as a research tool.
The concept was that excessive wrist motion meant poor sleep. Proving this is much more complicated.
First, how do you measure motion on a wrist? This is not simple. Motion sensing technology has evolved over the years. Today 3 axis accelerometers are common and cheap (like the kind in most phones), but that has not always been the case. The early, expensive detectors prevented sleep watches from being cost effective and widely available, until now.
Second, once you have recorded the signals you have to figure out how they relate to sleep. In 1992 Daniel Kripke et al developed and validated what is now known as the Cole – Kripke algorithm that converts wrist motion into sleep/wake cycles. It can tell pretty accurately if you are awake or asleep. They made the algorithm open source and it is still widely used today. The problem is that it was validated using 2 axis accelerometers not the 3 axis ones used today. No wonder Daniel Kripke says that information coming from modern “sleep watches” or fitness trackers is as “informative as a fortune cookie”.
Medical-grade actigraphs are a different story. Largely used for research, medical-grade actigraphs are validated against polysomnography (PSG). This brain wave technology monitors you while you sleep in a hospital bed with electrodes on your head and body. The test shows how your brain waves change during sleep and if you are awake or not. It is used to diagnose sleep disorders. To determine if these actigraphs work researchers put them on the wrists of patients undergoing PSG. Researchers determined that medical-grade actigraphs were pretty accurate in figuring out if the patient was awake or asleep. In fact, these medical-grade actigraphs became so refined that in 2007 the American Academy of Sleep Medicine accepted validated actigraphs as an acceptable measure of total sleep time in some circumstances. That same year fitness trackers hit the market.
Piggy-backing on medical-grade actigraphs, fitness trackers sought to not only measure sleep, but other fitness goals as well such as steps taken. The general public was interested in their health and wanted some way to measure it. Manufacturers were happy to oblige and the consumer market exploded.
“The trouble with sleep watches” is that the manufacturers generally do not make their hardware, algorithms or validation studies public (if they even did a study). So how do we know if they work? Many now purport to measure sleep stages (e.g. REM sleep) as well as sleep wake cycles. At best, actigraphs can infer sleep. Currently, only brain wave technology can measure sleep stages. Remember that sleep is in the brain and our brains are not on our wrists. Inferring sleep from wrist motion is complicated and requires independent validation against PSG regardless of the hardware or algorithms.
The challenge for industry and sports is that even if the actigraphs being used are accurate and validated — what do you do with the information? Actigraphs are like weigh scales. They can tell you how much you weigh, but cannot tell you if that number is appropriate for you or what the causes(s) of any weight gain or loss are. The most sleep watches can do is estimate if you are awake or if you are asleep. They cannot tell you anything else about your sleep. The products that purport to display sleep stages have not published any validation studies. But even if they were accurate what would you do with the information? How can you go to bed and decide to get 20 more minutes of REM sleep? Good luck with that.
Sleep is simply a complex brain function. Poor sleep can be caused by sleep disorders, mental disorders, lifestyle choices, organic diseases and medications. The overwhelming use of sleep watches has been to let the user know that they are not sleeping well or at the right times. Like weigh scales, they cannot diagnose anything.
There are very legitimate uses for medical-grade validated actigraphs. Medically they can be used to infer sleep and even be a substitute for PSG total sleep time when PSG is not available. When the sleep issue is largely a lifestyle issue, actigraphs can assist in demonstrating the extent of the issue and track progress. Tracking sleep in this way can improve performance and health and safety. In addition, tracking sleep with validated actigraphs can provide teams and employers with aggregated data to improve scheduling. Lastly, validated actigraphs can be used for research which furthers our understanding about what we spend 1/3 of our lives doing.
However, there are many challenges for users. Sleep is a relatively new science. Most doctors receive an hour or two of sleep training in a 4 year program. Coaches, trainers and health and safety professionals are rarely taught anything about sleep. Where would they even learn this? While it is changing, many have learned from the marketing materials of the manufacturers of consumer-grade sleep watches. There is a great need for sleep science to be integrated into the education of health, safety, coach and trainer professionals.
No matter what technology you use or recommend it is always critical to ask for independent validation studies, not just marketing materials. That way you are in a better position to assess the usefulness of the data you collect. Of course, this applies to all technology not just sleep watches.