SEATTLE—Although subjective sleepiness appears to increase the likelihood of depression, objective sleepiness may decrease the likelihood of depression, according to research presented at the 29th Annual Meeting of the Associated Professional Sleep Societies. “The data raise the question of whether people with depression misperceive their sleepiness, or whether common objective measures used to quantify sleepiness are insufficient in people with depression,” said David T. Plante, MD, Assistant Professor of Psychiatry at the University of Wisconsin School of Medicine and Public Health in Madison.
David T. Plante, MD
The literature indicates that hypersomnolence, which Dr. Plante and colleagues define as excessive daytime sleepiness or excessive habitual sleep duration, increases the risk of incident depression and depressive relapse. The highly treatment-resistant symptom also may increase the risk of suicidal ideation. The lack of an accurate and objective measure to quantify excessive daytime sleepiness in mood disorders, however, is an obstacle to research focused on hypersomnolence and depression, said Dr. Plante.
The Wisconsin Sleep Cohort Study
He and his colleagues examined subjective and objective measures of hypersomnolence among participants in the Wisconsin Sleep Cohort study. This ongoing study began in 1988 and follows Wisconsin state employees who undergo polysomnography every four years. Over the course of the study, many participants also have undergone multiple sleep latency test (MSLT) protocols to quantify sleep propensity and completed the Epworth Sleepiness Scale (ESS) to quantify subjective sleepiness, as well as other self-report measures of somnolence and sleep duration.
For their analysis, Dr. Plante and colleagues defined subjective sleepiness as an ESS score greater than 10. They also examined participants’ reports of whether and how often they experienced excessive daytime sleepiness. As another subjective measure, the investigators examined participants’ reports about their habitual sleep duration. Excessive habitual sleep duration was defined as an average sleep time of nine or more hours.
The researchers used the MSLT as their objective measure of sleepiness. Dr. Plante and colleagues defined excessive daytime sleepiness as a mean sleep latency over four naps of less than eight minutes on MSLT. About 40% of the data was collected using a standard clinical MSLT protocol that allowed patients to sleep for 15 minutes after sleep onset to look for sleep-onset REM periods. The remaining data were collected using a research protocol in which participants were awakened once sleep onset had been established.
The study’s main depression measure was the 20-item Zung self-rating scale. The researchers removed sleep-related items from the scale and rescaled the score to eliminate bias due to built-in associations. Depression was defined as a Zung score of greater than or equal to 50 or the use of antidepressant medication.
Dr. Plante’s group performed repeated measures logistic regression to analyze the data. They controlled for several covariates such as age, sex, BMI, smoking status, alcohol use, caffeine use, chronic medical conditions, sedative-hypnotic use, sleep disordered breathing, and insomnia.
Subjective Sleepiness Increased the Likelihood of Depression
The analysis included subjective measures of hypersomnolence from approximately 1,300 people and objective measures of hypersomnia for approximately 1,200 people. Participants’ ages ranged between 30 and 80, and most participants were middle-aged. The sample included slightly more men than women.
In fully adjusted models, subjective sleepiness increased the odds of depression by a factor of 1.6. Similarly, excessive habitual sleep time was associated with twofold increased odds of depression. However, objectively measured sleepiness using the MSLT was associated with an approximately 25% reduction in the odds of depression, which was statistically significant. “There is a divergent association between subjective and objective measures of hypersomnolence in this group,” said Dr. Plante. Secondary outcome measures, including questions about subjective sleepiness and cut-points used to define excessive sleep duration or sleep propensity on the MSLT, all supported the primary results.
The only hypersomnolence symptom that the researchers found to have a significant interaction with insomnia was habitual sleep time. In this case, excessive sleep duration predicted depression in patients with insomnia, but did not significantly predict depression in people without insomnia.
“Further research should be directed toward understanding the divergent associations that subjective and objective measures of hypersomnolence have with depression,” said Dr. Plante. His group plans to conduct several additional analyses, including an examination of longitudinal data.
—Erik Greb