Getting a proper night’s sleep is important for several health reasons, including supporting brain health, immune function, and muscle recovery. Still, falling asleep is not always easy for many individuals, and the National Institute of Health estimates that between 20% and 30% of adults in the United States have experienced insomnia at some point in their lives. In fact, researchers have shown that insomnia and the use of sleep aids are increasing among adults.
Despite the importance of sleep to our health, prior studies on sleep may not be the most reliable, since many have largely relied on subjective measurements. These include surveys asking participants to recall how well they slept, which can be prone to inaccurate recollections of past experiences, known as recall bias. While subjective studies have reported many health risks associated with poor sleep, do objective measures of disrupted sleep show the same impacts on health?
Recently, researchers from Peking University and Army Medical University in China sought to answer this question by examining data from the United Kingdom Biobank, a large-scale repository in the UK that includes health information from over 500,000 volunteers who were enrolled between 2006 and 2010. From this dataset, the researchers identified 88,461 adult participants who wore wrist-worn electromechanical sensors, known as accelerometers, that provide data, such as sleep duration, and included them in the study.
From the accelerometer data, the researchers examined 3 previously identified areas of interest related to the participants’ sleep, which they referred to as sleep dimensions. These included how long and when the participants went to sleep each night, how deeply the participants slept, or their sleep amplitude, the daily variation in their sleep, and how fragmented the participants’ sleep was each night in terms of waking numbers.
Next, the researchers analyzed the data from the UK Biobank to identify specific diseases that occurred in participants for whom they had accelerometer sleep data. They estimated whether certain diseases were more likely to occur in individuals with poorer sleep behavior within these areas of interest. The team identified a total of 172 diseases that were statistically more likely than not to be impacted by sleep length or quality. The researchers then assessed the risk of disease by comparing the top 25% versus the bottom 25% in a specific sleep dimension. For example, the 25% of participants with the shortest sleeping times might be more likely to be diagnosed with a disease compared to the 25% with the longest sleeping times.
The researchers quantified the difference in risk between the top 25% and bottom 25% in each area of interest, finding that several diseases were associated with an increased risk in individuals with poor sleep. These included Parkinson’s disease, with 2.8 times increased risk, age-related frailty, with 3.36 times increased risk, liver fibrosis or cirrhosis, with 2.57 times increased risk, and type 2 diabetes mellitus, with 1.6 times increased risk. They also found that roughly 20% of all participants who self-reported as being long-sleepers actually slept less than 6 hours a night, supporting the idea that studies relying on subjective self-report measures of sleep can have recollection errors.
While this study has the strength of utilizing objective sleep data, the researchers identified several limitations. For one, the participants in the UK Biobank were mainly middle-aged or elderly, which is not representative of the general population and may limit the generalizability of their study’s results. Next, accelerometer measurements were taken from a single time point in the participants’ lives, and individuals’ sleep patterns or sleep behaviors can change over time.
In addition, because this was an association study, the researchers couldn’t infer causality. For example, participants with certain diseases may be more likely to have poorer sleep, rather than poorer sleep causing their diseases. Nonetheless, the team’s study adds to existing literature suggesting sleep is important to overall health. They suggested that further researchers investigate how sleep patterns impact diseases in samples more representative of the general population by using accelerometer data that tracks participants’ sleep over longer periods.
