Wearables show how sleep patterns change depending on health status

In a recent study published in Digital Medicine NPJresearchers used a large data set consisting of five million nights of sleep monitoring data from mobile devices to examine changes in an individual’s sleep phenotype over time and determine whether these changes in sleep patterns or phenotypes provide information about periods of acute illness, such as such as fever or coronavirus disease 2019 (COVID-19), etc.

Test: Five million nights: temporal dynamics in human sleep phenotypes. Photo credit: New Africa/Shutterstock.com

Background

Rapid advances in mobile device technology have made portable health monitoring devices readily available and affordable. Apart from various other health parameters, these devices are widely used to monitor sleep patterns and quality.

However, despite the abundance of sleep monitoring data, translating the conclusions drawn from this data into actionable changes is challenging due to variability in the combination of sleep parameters across individuals and over time.

Recommendations from the National Institutes of Health state that adults should get seven to nine hours of sleep each night. However, sleep research has shown that sleep patterns vary greatly in duration and quality, and these differences are related to lifestyle and health factors.

Studies that have used cluster analyzes of large-scale sleep data to quantify differences in sleep characteristics have been effective in characterizing sleep phenotypes, but have only used cross-sectional data and have not considered the conclusions that can be drawn about disease and health based on longitudinal sleep data.

About the study

In this study, researchers used a large data set of wearable sleep monitoring devices on over 33,000 people, totaling data on over 5 million nights, to determine changes in sleep phenotype over time. They also wanted to understand whether these changes informed health parameters or periods of acute illness.

Sleep periods in the large dataset were treated independently. Using cluster analysis, researchers obtained a set of sleep phenotypes, including an insomnia-like phenotype that consisted of segmented sleep of less than 6.5 hours per night and a recommended sleep phenotype of 8 hours of monophasic sleep.

The relevance of these phenotypes and changes in sleep phenotypes to disease and health was examined by examining whether patterns of transition probability in a chronically ill cohort differ significantly from those in a healthy cohort.

The researchers also examined whether patterns of transition probability differed before and after the disease in the same people.

Data was collected from self-reported survey responses and sleep-wake time series from 33,152 people over a ten-month period. Sleep monitoring data were also obtained from all participants using a portable smart ring device.

The data were divided into sleep periods, i.e. periods of nonoverlapping three to six consecutive nights, which were then used to determine sleep phenotypes through cluster analyses.

Characteristics typical of each sleep period were used to identify predefined groups of sleep phenotypes. Transition patterns and the distribution of sleep periods over time for each individual were used to determine changes in sleep phenotype.

Additionally, the study also examined the distribution of changes in sleep phenotypes among groups of people suffering from sleep apnea, influenza, diabetes, fever and COVID-19.

Results

The results revealed 13 sleep phenotypes associated with sleep quality and duration, and provided evidence of changes in an individual’s sleep phenotypes over time.

Moreover, patterns of change in sleep phenotypes showed significant differences between groups of people with and without chronic diseases or health problems, and within an individual over time.

The study found that not only are sleep phenotypes individual, but changes in sleep phenotypes provide information about health status.

Furthermore, assessment of the temporal dynamics of sleep patterns revealed that current sleep patterns are indicative of potential changes in sleep phenotypes. For example, shorter periods of deep sleep indicated a change in the insomnia-like sleep phenotype.

The temporal dynamics of sleep phenotype changes have also been found to indicate an individual’s chronic disease or health-related factors. The dynamic transition model was found to provide more information about an individual’s respiratory and cardiometabolic health factors than specific sleep phenotypes.

Conclusions

Overall, the study identified 13 sleep phenotypes associated with sleep duration and quality and found that these phenotypes varied across individuals based on health status and within an individual over time.

Temporal transition patterns in sleep phenotypes were also indicative of chronic diseases such as respiratory and cardiometabolic diseases.

These findings highlight the importance of longitudinal sleep analyzes and temporal dynamics assessments in drawing actionable conclusions from wearable device sleep monitoring data.