Sleep quality can be predicted from the way one walks, study finds

A recent study utilized machine learning techniques to explore the relationship between walking patterns and sleep quality. The findings suggest that individuals with lower sleep quality exhibit less variation in their pelvic tilt, particularly noticeable when initiating walking after a turn. Additionally, these individuals often struggle to maintain a consistent walking speed. The paper was published in Sleep Science.

Sleep quality problems refer to issues that affect one’s ability to get enough restful and restorative sleep. Insufficient sleep leads to daytime fatigue, mood disturbances, and impaired cognitive function. Common sleep problems include difficulty falling asleep, frequent awakenings during the night, waking up too early, and non-restorative sleep.

Poor sleep quality can be caused by factors that include stress, anxiety, health problems, poor sleep environment, and lifestyle choices such as excessive caffeine or screen time before bed. Chronic sleep quality problems can lead to serious health consequences, including an increased risk of cardiovascular disease, obesity, diabetes, and mental health disorders.

Studies have indicated that sleep quality might be associated with the way one walks. This pattern of body movements a person makes in order to walk is called gait. Findings in older adults indicate that poor sleep quality might reduce gait speed, but also gait variability. These changes in walking patterns are associated with an increased risk of falling in the elderly. However, not many studies explored this in younger adults.

Study author Joel Martin and his colleagues wanted to examine the associations between sleep quality and gait characteristics in younger adults. They were particularly interested in identifying gait characteristics associated with poor sleep quality, characteristics that could be useful to screen individuals at a greater risk of injury.

Free daily newsletter

Study participants were 123 adults between 18 and 36 years of age. To be included in the study, participants were required to be able to stand and walk for two minutes without any assistive devices. Of these participants, 63% were female, with an average age of 24 years.

Participants reported their demographic data, answered a series of questions about their activities in the past 24 hours, and completed an assessment of sleep quality (the Pittsburgh Sleep Quality Index). After this, they completed a walking test during which they wore a system of inertial sensors used to assess gait during the test (APDM Wearable Technologies’ Mobility Lab). The walking test required participants to walk for two minutes at a pace that was comfortable to them (while wearing the sensors). Study authors used machine learning to find a way to predict sleep quality based on participants’ walking patterns as recorded by sensors.

Results showed that 59% of participants could be considered good sleepers, while 41% were poor sleepers. The most important gait feature for predicting sleep quality was the first step lumbar transverse range of motion i.e., the degree of rotational movement that occurs in the lower back region of the spine during the initial step of walking. The first step lumbar left frontal bending maximum was the second most important predictors of poor sleep. It refers to the maximum degree of bending to the left side in the frontal plane that occurs in the lower back region of spine during the initial step of walking.

Overall, poor sleepers tended to have decreased pelvic tilt angle changes, particularly during the initial step of walking and when starting to walk after turning. They also had more difficulty maintaining a constant walking speed.

The machine learning model these researchers developed was 62% accurate in predicting sleep quality, which is a bit lower than the accuracy reported in some other studies.

“The results of our post-hoc findings support that poor sleepers may display very subtle changes in gait normally associated with difficulty initiating and maintaining gait speed. Notably, these gait patterns are similar to individuals who are at a higher risk for lower extremity injuries or walking more slowly,” the study authors concluded.

The study sheds light on the links between walking patterns and sleep quality. However, it should be noted that the study did not control for participants’ moods which may also affect both gait properties and sleep patterns.

The paper, “Association between Self-reported Sleep Quality and Single-task Gait in Young Adults: A Study Using Machine Learning,” was authored by Joel Martin, Haikun Huang, Ronald Johnson, Lap-Fai Yu, Erica Jansen, Rebecca Martin, Chelsea Yager, and Ali Boolani.