Chief principal scientist from the US smart bed manufacturer, Sleep Number, talked to Health Tech World about the potential of smart technology and sleep science for improving health and wellbeing and predicting the onset of respiratory diseases like COVID-19
Sleep plays an integral role in health and wellbeing, yet many struggle with sleep problems. According to the NHS, one in three people in the UK suffer from poor sleep. Some of these issues are linked with clinical conditions like sleep apnoea and insomnia, but for most, poor sleep is often a result of bad sleeping habits. Stress, excessive screen time and working from home are common factors that can contribute to a rough night’s sleep.
Tech is becoming increasingly popular for improving sleep quality, with a wide range of devices and gadgets on the market that claims to support healthy sleep patterns.
Gary Garcia-Molina is chief principal scientist at Sleep Number, a US-based manufacturer of smart beds that gather data during the night and adjust in real-time to improve quality of sleep.
Garcia-Molina, who has been working in sleep science for the past 16 years, said there are several key factors when it comes to healthy sleep. These include regularity, duration, timing and efficiency.
The company’s 360 smart bed and its operating system, SleepIQ technology, deliver individualised sleep health evaluations by automatically sensing and responding to the needs of sleepers. The bed adjusts throughout the night, digitally sensing the sleeper’s movements and automatically adjusting the firmness of the bed to keep them comfortable.
According to Sleep Number, its SleepIQ technology conducts one of the largest real-world sleep studies every night and has learned from more than 10 billion hours of sleep data from 1.3 billion sleep sessions.
The bed tracks three types of movement; large body movements, respiratory patterns and cardiac activity without the need for a wearable device.
“SleepIQ has the capability to measure tiny movements that are produced by the body and particularly by the blood vessels. Those tiny movements reveal cardiac activity,” Garcia-Molina explained. “It’s quite impressive to be able to measure the instantaneous heart rate using the sensors that we have.
“[Our] smart bed is able to measure those signals, and from there, if it is possible to measure physiological activity during sleep and to estimate the sleep quality and sleep stage.
“And what is crucially important is to be able to communicate this information to the user in a way that facilitates the user to adopt behaviours and enhance the quality of their sleep.”
The app that connects with the smart bed provides a score for the user’s quality of sleep after every sleep session and provides feedback to help them adapt their behaviour and settle into a healthy sleep pattern.
Sleep Number presented data from two new studies using its 360 smart bed at SLEEP 2021, the 35th annual meeting of the Associated Professional Sleep Societies earlier this month. Data presented at the meeting included a predictive model of COVID-19 infection based on sleep metrics and results from a large study analyzing overnight heart rate variability (HRV).
The data presented at the annual meeting came in two parts. The first looked at COVID-19, which, according to Sleep Number, was the first study to evaluate real-world, longitudinal data collected unobtrusively and non-invasively during sleep, using a smart bed platform.
The company says that sleep metrics measured with the smart bed are a unique source of long-term health data and have the potential to predict and track the development of symptoms associated with COVID-19 and other respiratory diseases. Sleep Number is working on expanding these capabilities to detect symptoms for illnesses such as the common cold, Influenza and SARS.
“Immune function and sleep are very related,” Garcia Molina said. “When a challenge to our immunity occurs, that releases something that are called cytokines and these have an influence on duration and sleep quality.
Sleep has also been found to be a moderating factor for the pathology of COVID-19 with some recent papers revealing that poor sleep can make people more vulnerable to the disease.
Garcia-Molina added: “A few years ago, there was a seminal paper where investigators looked at the effectiveness of the flu vaccine in relation to sleep, and it was clearly demonstrated that for people that sleep six hours or less, the build-up of antibodies as a consequence of receiving the flu vaccine was delayed.
“It [shows] that sleep and the cardiorespiratory functions that occur during sleep has the potential to reveal important information for tracking the pathology of COVID-19 and potentially also to detect when the symptoms manifest, which we believe can be clinically important to mitigate the spread of the disease.”
An analysis of real-world data collected from 122 COVID-19 positive and 1,603 negative smart bed sleepers was conducted to build a predictive COVID-19 model based on unobtrusive sleep metrics. Sleep duration, sleep quality, restful sleep duration, time to fall asleep, respiration rate, heart rate and motion level obtained from ballistocardiography data from January 2019 to October 2020 were measured in the analysis.
In the COVID-19 positive group, worsening of symptoms was associated with an increase in sleep duration, average breathing rate, average heart rate and a decrease in sleep quality. For those in the COVID-19 negative group, no significant disruptions in sleep and cardiorespiratory metrics were observed.
“I would say one of the most important results of the research we did is that, by looking at the history of sleep, we could predict the onset of symptoms associated with COVID-19.
“When we thought about the symptoms of COVID-19, we quickly realised that similar symptoms occur in other pathologies. Influenza also causes changes in sleep [patterns], heart rate and respiratory rate. So, this approach could possibly be applied to other pathologies.”
The company also presented data on heart rate variability gathered from 18.2 million sleep sessions from 379,225 smart bed sleepers. The aim was to better understand changes in heart rate variability, a measurement that is commonly used to assess the activity of the autonomic nervous system (ANS). While there is broad interest in researching HRV, few studies to date have established normative overnight HRV values for a large population.
The ANS unconsciously regulates various essential bodily functions including breathing, heart rate and blood pressure. Changes to ANS function can result from factors including lifestyle, ageing, cardiorespiratory illnesses, sleep state and physiological stress.
Two important components of the ANS are the sympathetic nervous system and the parasympathetic nervous system. The sympathetic nervous system is activated during a fight or flight response and the parasympathetic nervous system is activated when we are relaxed. Generally speaking, the higher the heart rate variability is, the higher the parasympathetic activity.
This means that higher HRV generally correlates with better health and cardiac response to stress, and lower HRV numbers are an indicator of unhealthy cardiac activity. Results of the analysis found significant cross-sectional associations between overnight HRV and age, gender and day of the week.
“Heart rate variability is a good proxy to look at the cardiac activity and parasympathetic activity during sleep, so we looked at that in our consumer base and then we looked at gender dependencies and also the influence of age on heart rate variability,” Garcia-Molina said.
HRV was found to be highest during the weekend and lowest at mid-week for men and women under the age of 50. For women, values followed a U-shaped pattern, starting high at the beginning of the week, dipping mid-week, then increasing through the weekend, whereas values for men followed an L-shaped pattern, starting high at the beginning of the week, but quickly fell and stayed low through the week. The study also found that heart rate variability decreases with age then after the age of 50, the rate of change starts to level off.
Sleep Number says that these results show that measuring overnight HRV data using smart beds may be a useful, ecologically valid device for evaluating population health models.
For Garcia-Molina, the “limit is the imagination” when it comes to the potential of sleep science and he predicts that in the future, as sleep data from different technologies is aggregated, it will play an important role in the era of personalised medicine.
“We have the modelling capabilities, we have the data available, [so] the imagination is the roof. Day after day you’re going to see in the press and scientific journals more discoveries and associations with sleep; sleep and emotional health, sleep and ageing, sleep and immunity, sleep and the gastrointestinal system. It’s amazing the level of impact that sleep has on our health.
“What I see happening is all of this data being aggregated, then the scientific community can develop models and algorithms that can provide steps into the forecast of health outcomes.
“I think we are all wishing this to happen as soon as possible because that’s how these therapies and approaches and interventions can be most beneficial when it becomes personalised. And having technologies such as our smart bed will help; it’s one step closer to that goal.”