Can Data Help Us Sleep Better?

I was recently honored with the opportunity to speak at the Sleep Technology Summit, a conference put on by the National Sleep Foundation exploring recent innovations and future projections in sleep technology.

While there, my conversations centered on two parallel themes, (1) can new technologies accurately measure sleep at a level worthy of displacing Polysomnography, the decades-old gold standard, and (2) why bother?

The first theme I was prepared for, we get asked this all the time and frankly, most of my job is working to enable us to truthfully answer “Yes.” However, given the audience of sleep professionals, the second major theme came as more of a surprise. Nonetheless, people were skeptical as to whether or not consumer sleep monitoring devices could offer their users value.

So, for those of you who weren’t there, this is a summary of some of the research and anecdotes I shared, mixed with some supplementary musings.

Navigating a “Medical” Problem with a Non-Medical Device

In one of my first conversations, the following question was posed: “As a non-FDA approved device, federal regulations prevent you from diagnosing sleep disorders. If you can’t do that, what good do you provide?”

As is tradition, I began my answer with “Great Question,” but the truth is, it was surprisingly parochial. There is tremendous room for non-medical sleep tech in healthcare that is fully kosher with the FDA as a case-finding and conversation-prompting tool. For example, imagine a doctor-patient conversation in which the patient expresses the following frustration: “I’m tired all the time, but get 8 hours of sleep per night.” As we reported earlier in a white paper examining the value of sleep stage tracking, 8 hours in bed usually only means about 7 hours of sleep, but the person-to-person variability in actual sleep efficiency is huge (standard deviation = 7.2%). This is important because people usually don’t know where they fall on this distribution, so two patients both honestly reporting that they believe they are getting 8 hours could actually be getting extremely different amounts of sleep. Furthermore, two people both getting similar amounts of total sleep could be losing sleep to different culprits – some people have trouble falling asleep while others have trouble staying asleep. But if our doctors knew more about how much sleep we were getting and where we were losing sleep, better-informed doctor/patient conversations could follow.

Just One More Hour

In a retrospective analysis conducted this spring, we found that athletes who joined WHOOP averaging fewer than 7.9 hours of sleep per night added 52 minutes in bed per night by their 129th day on the platform. Previous research has shown that adding an additional hour of sleep per night positively alters gene expression for over 500 genes, including genes implicated in diabetes, cancer, inflammation, and stress. Another study linked an hour of additional sleep to a reduction in blood pressure in the hypertensive. So in addition to improving free throw accuracy for basketball players, by prioritizing sleep, WHOOP athletes are also prioritizing health, and creating in their 20s and 30s sleep habits that will benefit them for the rest of their lives.

Getting Personal

At the conference, it seemed like a wide-spread criticism of the wearable sleep-tech industry was that while vast amounts of data were being generated, the data doesn’t actually benefit the user. I think this is a fair criticism of first-generation sleep tech which generated “data” alone but is significantly different from what WHOOP is doing now generating “actionable data.” Data tells us that people on the western edges of their time zones go to bed later than people on the eastern edges, but it doesn’t tell you what to do if you’re a sleep-deprived western-edge dweller. WHOOP’s business model from day 1 was about personalized, actionable feedback born from the recognition that the elite athletes we target are who they are because they are outliers, and therefore are less likely than “the average person” to have “average” physiological responses.

For example, common lore is that it takes one day per time zone traveled to adjust to jetlag; a decent rule of thumb that doesn’t actually work for everyone and vastly oversimplifies the complex dynamics of our circadian rhythm. Before the Olympic trials, WHOOP identified Connor Jaeger as someone who took longer than he realized to adjust to time zone changes and recommended that he travel to Omaha ahead of his team. He did, and the rest is history; he not only qualified to compete in Rio that trip but went on to win an Olympic silver medal in the 1500 meter freestyle race.

Starting Young

In the summit’s opening event, I learned that the average person who reports an interest in improving sleep is in their 40s or 50s, presumably because that’s when a lifetime of bad sleep habits are catching up to them. Because of our focus on the relationship between sleep and athletic performance, WHOOP is positioned to educate a younger population on the importance of sleep. This means that we have the opportunity to improve behavior before health consequences emerge, something we hope translates to fewer sleep-related issues down the road.

Because of the tremendous value of our data to both athletes and non-athletes, WHOOP is excited to be launching our first publicly available product this December. Preorder now to find out what truly understanding your sleep could do for you.