Subjective-Objective Sleep Discrepancy in a Predominately White and Educated Older Adult Population: Examining the Associations with Cognition and Insomnia

OBJECTIVES

This study examined associations between various cognitive domains and sleep discrepancy (self-reported versus objectively measured sleep), and evaluated interactive associations with insomnia status (non-insomnia versus insomnia).

METHOD

Older adults (N=65, Mage=68.72, SD=5.06, 43 insomnia/22 non-insomnia) aged 60+ reported subjective sleep (7-days of sleep diaries), objective sleep assessment (one-night polysomnography, PSG, via Sleep ProfilerTM during the 7-day period), and completed cognitive tasks (NIH Toolbox-Cognition Battery) measuring attention and processing speed, working memory, inhibitory control, cognitive flexibility, and episodic memory. The sleep diary variable corresponding to same one-night of PSG was used to calculate the sleep discrepancy (diary minus PSG parameter) variables for total sleep time, sleep onset latency, wake after sleep onset, and sleep efficiency. Regression analyses determined independent and interactive (with insomnia status) associations between cognition and sleep discrepancy, controlling for age, sex, apnea-hypopnea index, and sleep medication usage.

RESULTS

Working memory interacted with insomnia status in associations with sleep discrepancy related to total sleep time and sleep efficiency. In those with insomnia, worse working memory was associated with shorter self-reported total sleep time (p=.008) and lower sleep efficiency (p=.04) than PSG measured.

DISCUSSION

In older adults with insomnia, worse working memory may be a contributing factor to sleep discrepancy. Future investigations of underlying neurophysiological factors and consideration of other objective sleep measures (actigraphy) are warranted. Prospective findings may help determine whether sleep discrepancy is a potential marker of future cognitive decline.

Smartphone Addiction Proneness Is Associated With Subjective-Objective Sleep Discrepancy in Patients With Insomnia Disorder

OBJECTIVE

Subjective reports of patients with insomnia often show a discrepancy with their objective assessments of sleep. We aimed to assess subjective-objective sleep discrepancy in subjects with insomnia disorder as well as the psychological factors associated with the discrepancy.

METHODS

This study is a secondary analysis of the baseline data of a randomized controlled study on 110 adults aged 18 years to 59 years with insomnia disorder. Subjective reports on sleep and the objective measures acquired by an overnight polysomnography were used to measure the sleep discrepancy. Smartphone Addiction Proneness Scale (SAPS), Center for Epidemiologic Studies Depression Scale (CES-D), beck anxiety inventory (BAI), and Global Assessment of Recent Stress (GARS) were used to evaluate the psychological factors associated with the sleep discrepancy.

RESULTS

Mean total sleep time (TST) discrepancy of the participants was -81.65±97.41 minutes. Multivariable logistic regression analyses revealed that age (adjusted OR=1.07, 95% CI=1.01-1.13, p=0.027), years of education (adjusted OR=0.69, 95% CI=0.48-0.91, p=0.017), and smartphone addiction proneness (adjusted OR=1.14, 95% CI=1.04-1.27, p=0.008) were independent predictors of TST misperception. Mean sleep onset latency (SOL) discrepancy of the participants was 41.28±45.01 minutes. Only anxiety was an independent predictor of SOL misperception (adjusted OR=1.16, 95% CI=1.05-1.31, p=0.006).

CONCLUSION

The present study provides empirical evidence to increase our understanding of the various factors that are associated with subjective-objective sleep discrepancy. Screening insomnia patients with smartphone addiction proneness may help predict the potential discrepancy between the patients' subjective reports and objective measures of sleep duration.

The paradox of paradoxical insomnia: A theoretical review towards a unifying evidence-based definition

Paradoxical insomnia is one of the most intriguing yet challenging subtypes of insomnia. Despite being recognized for a long time by the international community, it is still unclear whether this entity really exists, which are its features and boundaries. Much of the debate is fuelled by the lack of a consensus on its precise definition. To help filling some of the existing gaps, a systematic review of the literature was conducted, through which 19 different quantitative definitions were obtained. These definitions were then applied to two distinct datasets. The first consisted of 200 chronic primary insomnia patients, diagnosed according to the DSM-IV-TR criteria. The second consisted of 200 age- and sex-matched healthy persons without insomnia. For each dataset, available data from the objective sleep parameters and their subjective estimation were imported and analysed in MATLAB. Depending on the definition used, the prevalence of paradoxical insomnia ranged from 8 to 66%, while agreement between different definitions ranged from -0.19 to 0.9 (using Cohen's kappa coefficient). Based on the results garnered, necessary features for a quantitative definition of paradoxical insomnia were identified. Several open questions remain, such as whether there is a minimum number of hours a patient should sleep to fulfill the criteria for a diagnosis of paradoxical insomnia, and whether sleep latency can be used in the definition along with total sleep time. We conclude by advocating continued study of paradoxical insomnia and sleep state misperception and by providing specific directions for future research. STATEMENT OF SIGNIFICANCE: The current understanding of paradoxical insomnia and, more broadly, of sleep state misperception, is greatly hampered by the lack of agreement on a quantitative and evidence-base measure of the discrepancy between subjective and objective sleep evaluation. The current study provides a critical analysis about the strength and the limitations of the available definitions, using both a data-driven and a theory-driven approach. The overarching goal is to motivate a rigorous discussion involving the main experts of the field, to build a consensus, and develop an evidence-based measure of sleep state misperception and/or of paradoxical insomnia.

Assessing sleep quality using self-report and actigraphy in PTSD

Sleep disturbance is commonly reported by participants with post-traumatic stress disorder, but objective evidence of poor sleep is often absent. Here we compared self-report and actigraphic evaluations of sleep between veterans with post-traumatic stress disorder and controls. Participants reported their sleep retrospectively for the month before the recording night and on the recording night. On the recording night, they wore an Actiwatch-64 and were instructed to press the marker button upon getting into bed, each time they awoke, and at their final awakening. The post-traumatic stress disorder group reported much worse sleep than controls on the Pittsburgh Sleep Quality Index for the previous month and somewhat poorer sleep on the recording night. However, on the recording night, neither diary nor actigraphic measures of number of awakenings, total time in bed, nor time lying awake after sleep onset differed between participants with and without post-traumatic stress disorder. Diary-reported number of awakenings was fewer than actigraphically captured awakenings. These results suggest a memory bias towards remembering worse sleep on the nights before the recording night.

Sleep misperception in persons with epilepsy

Being able to confidently ascertain the amount of sleep is critical to the clinical management of epilepsy. Sleep misperception is the phenomenon in which an individual underestimates the amount of time spent asleep. Little is known about sleep misperception in patients with epilepsy. We conducted retrospective chart reviews on individuals who self-identified as having epilepsy in a questionnaire database of patients undergoing polysomnography (PSG) at the Massachusetts General Hospital Sleep Laboratory. Our metric for sleep misperception was the difference between subjective and objective sleep latency (S-O SL) and subjective and objective total sleep time (S-O TST) with subjective values based on questionnaire and objective values based on PSG. We confirmed 64 patients with epilepsy. We then selected age- and sex-matched diagnostic PSG data for comparison from 50 patients with insomnia symptoms but no obstructive sleep apnea (OSA) and another 50 patients with OSA but no insomnia symptoms. In our cohort with epilepsy, the median SL overestimation was 20 min (p<0.05), and the median TST underestimation was 45 min (p<0.05). Sleep misperception was similar regardless of potential confounding factors such as categorical epilepsy refractoriness, cognitive impairment, or psychiatric comorbidity. Our findings suggest that sleep misperception occurs similarly in patients with epilepsy as in patients without epilepsy with insomnia. Our findings further support the potential clinical utility of objective PSG testing in patients with epilepsy, as this may not only identify occult OSA but also disclose sleep misperception.