Siddarth et al. (Reference Siddarth2020)’s paper added new findings to the evidence on relationship between sleep and cognitive function. In their paper, poor sustained attention was strongly associated with self-reported sleep quality measured with the Pittsburgh Sleep Quality Index (PSQI; Buysse et al., Reference Buysse, Reynolds, Monk, Berman and Kupfer1989) among middle-aged and older adults with mild subjective memory complaints. Their work enhances our understanding of subjective cognitive function across multiple cognition domains among adults with existing memory complaints, diagnosed using transitional staging models from normal cognition to Alzheimer’s disease or other types of dementia. Although Siddarth and colleagues did not focus on examining the mechanisms linking sleep and cognitive aging (which are still poorly understood), their findings suggest the importance of subjective sleep quality in relation to subjective and objective cognition.
Prior studies have shown strong associations between subjective and/or objective sleep and cognitive function among healthy adults both in cross-sectional and longitudinal study designs. Both objectively measured poor sleep (e.g. low sleep efficiency, increased sleep latency, wake after sleep onset, fragmentation; Campbell et al., Reference Campbell2020; Yaffe et al., Reference Yaffe, Falvey and Hoang2014) and self-reported sleep complaints (Campbell et al., Reference Campbell2020; Tardy et al., Reference Tardy, Gonthier, Barthelemy, Roche and Crawford-Achour2015; Yaffe et al., Reference Yaffe, Falvey and Hoang2014) are significantly associated with worse cognitive function in older adults; however, subjective sleep quality as measured by the PSQI and objective sleep are not consistently correlated with each other in middle-aged and older adults (Landry et al., Reference Landry, Best and Liu-Ambrose2015). Some sleep indices (e.g. snoring, nightmares) are difficult for people to self-assess and report; thus, objectively-measured sleep parameters (e.g. sleep stages, electroencephalogram power values, sleep fragmentation) using polysomnography or actigraphy may help to further elucidate underlying mechanism of sleep and cognitive function, particularly among those who are already experiencing subjective and/or objective memory decline.
Siddarth and colleagues focused on subjective sleep quality and did not measure objective parameters of sleep. Nevertheless, we agree with their argument addressing that subjective sleep complaints could predict future cognitive impairment and dementia even in the absence of objective cognitive impairment. In other words, perceived poor sleep may play a role as an early symptomatic marker in the spectrum of developing Alzheimer’s disease or other types of neurodegeneration. Pending further evidence of contribution of subjective sleep on cognitive decline, monitoring self-reported symptoms of poor sleep may be served as a more cost-effective and easier method to identify at-risk group in combination with biomarkers, neuroimaging, and other clinical features. Additionally, patients’ own assessment of their sleep is important as their own view of sleep may motivate them to engage in cognitive behavioral therapies (e.g. Cognitive Behavioral Therapy for Insomnia [CBT-I]) that have been shown to improve sleep quality (Qaseem et al., Reference Qaseem, Kansagara, Forciea, Cooke and Denberg2016).
Siddarth et al. (Reference Siddarth2020)’s paper extended the evidence on the importance of role of sleep on cognition to more broad groups of older adults from those who already had clinically significant cognitive impairments. The special issue (Volume 32, Issue 7) published in July 2020 in the International Psychogeriatrics discussed sleep and sleep disorders among different groups of older adults. Papers in the issue confirm significant relationship between sleep and cognitive function or other health outcomes such as depression among older adults with neurodegenerative disorders associated with cognitive impairment (e.g. Parkinson’s disease, nursing home residents with dementia). Overall, the clinical message from these findings was sleep as a potential target to prevent further cognitive deterioration and/or psychological well-being. On the other hand, Siddarth et al. (Reference Siddarth2020)’s paper examined older adults who may be at-risk for cognitive decline and reported subjective memory complaints without major cognitive impairment (i.e. dementia). Taken together, monitoring sleep patterns and screening chronic sleep problems such as insomnia may be one way to prevent or delay worsening cognitive impairments and related health issues among older adults with all stages of cognitive function.
Indeed, Siddarth et al. proposed a future sleep intervention study to test whether improved sleep could improve memory consolidation. A few intervention studies have found that improving sleep with CBT-I is associated with improved objective cognitive function; however, other studies have not found any improvements in objective cognitive function in older adults with behavioral treatment for insomnia (Cassidy-Eagle et al., Reference Cassidy-Eagle, Siebern, Unti, Glassman and O’Hara2018; Herbert et al., Reference Herbert, Kyle and Pratt2018). The current literature is limited as there have only been a handful of studies that have examined the dynamic relationship between sleep and cognition. These studies have found different results, and even studies that have found an effect do not consistently observe improvements in the same cognitive domain (e.g. some found improvement in memory whereas others observed improvements in executive function; Herbert et al., Reference Herbert, Kyle and Pratt2018). Additionally, it is not yet clear if these interventions have long-term effects and are effective in lowering the risk of future cognitive impairment.
Ultimately, we agree with Siddarth et al. in that interventions to improve sleep quality should be further studied as these interventions may be an efficacious way to improve subjective and/or objective cognition as well as quality of life in middle-aged and older adults with memory complaints. Although a behavioral approach such as CBT-I is recommended as the first-line therapy for persons with chronic sleep problems (Qaseem et al., Reference Qaseem, Kansagara, Forciea, Cooke and Denberg2016), evidence on effectiveness of the CBT-I for those with mild cognitive impairment is relatively unknown. While limited, an initial study (Cassidy-Eagle et al., Reference Cassidy-Eagle, Siebern, Unti, Glassman and O’Hara2018) found that CBT-I was successful in both improving sleep and cognition among older adults with mild cognitive impairment. Behavioral sleep interventions tailored to levels of cognitive function among persons with mild memory impairment are certainly needed. Some adaptations such as incorporating compensation strategies like alarms/calendars and more novel artificial intelligence-assisted compensation strategy technology (e.g. smart speakers such as Amazon’s Alexa, Google Home) may be particularly useful for this group.
One limitation of Siddarth et al. (Reference Siddarth2020)’s paper was that that they did not measure information about daytime naps and oversleeping (e.g., >9 hours per day) in the paper which have been known as significant risk factors of cognitive decline. However, one considerable strength is that the authors used a questionnaire to screen obstructive sleep apnea in the study (i.e., the STOP-Bang; Chung et al., Reference Chung, Abdullah and Liao2016). While a questionnaire cannot replace a clinical assessment of obstructive sleep apnea, many studies examining sleep and cognition do not include an assessment of obstructive sleep apnea even though obstructive sleep apnea has been shown to be a risk factor for cognitive decline. Nevertheless, being at high risk of obstructive sleep apnea did not change relationship between subjective sleep and cognitive function among study participants. Given that sleep apnea is often undiagnosed and significantly associated with daytime sleepiness, future studies should also consider assessing obstructive sleep apnea as well as daytime naps and sleeping too much in order to understand their role in cognitive function among middle-aged and older adults with memory complaints.
Another limitation in the current literature is that the majority of sleep and cognition research has focused on the relationship between average sleep and cognition measured at one timepoint. Until recently, it has been difficult to measure objective cognition on a daily basis, but one newer tool to help us to obtain rich data of objective cognitive function is smartphone-based ecological momentary cognitive tests. Studies utilizing mobile cognitive testing have shown feasibility in a number of populations including older adults as well as high levels of between- and within-person reliability and validity (Moore et al., Reference Moore, Swendsen and Depp2017). Additionally, these mobile cognitive tests can be paired with ecological momentary assessment questions to more frequently assess subjective sleep and subjective cognitive functioning. This technology can aid in understanding how and if daily fluctuations in subjective sleep impact subjective and objective cognition on an individual level. It may also help lay the groundwork for individualized interventions that could promote better sleep and, by extension, may also promote better cognitive functioning.
Another area to further explore is to monitor intra-individual cognitive variability (IIV) in relationship with sleep–wake cycle. Studies have shown that IIV in sleep is intertwined with sleep–wake cycles (Slavish et al., Reference Slavish, Taylor and Lichstein2019), and higher IIV in cognition has been linked to more widespread medial temporal lobe atrophy in persons with mild cognitive impairment relative to cognitively normal adults and may be a sensitive marker of early neurodegenerative changes (Bangen et al., Reference Bangen2019). A recent study using passively and unobtrusively obtained keystroke metadata (i.e. how you type, not what you type) from real-world daily smartphone keyboard interactions found an age effect and diurnal pattern with the keystroke data (Vesel et al., Reference Vesel2020). Results of the sample of 250 individuals demonstrated a nonlinear diurnal pattern for typing speed, wherein user’s typing speed was fastest midday (12:00 pm to 3:00 pm) and their slowest typing was at night. Older users (aged 70+) exhibited slower typing when compared to younger users, and there was an interaction between age and time of day such that older adults had significantly more pronounced slowing in their typing speed towards the end of the day. This may suggest that abnormal typing dynamics of slower speed and higher variability in speed may represent earliest changes in how a person with cognitive impairments interacts with technology, and potentially elucidate chances in circadian rhythms and IIV, which can be leveraged to differentiate between mild cognitive impairment and healthy controls (Chen et al., Reference Chen2019; Stringer et al., Reference Stringer2018).
The Siddarth et al. (Reference Siddarth2020) paper also implies the critical need of special attentions to vulnerable groups, particularly during the ongoing COVID-19 pandemic. Currently, people are encouraged or required by local ordinances to socially isolate and not to leave their homes as much as possible, or when they do leave their homes to practice social distancing and always wear masks. Many nonessential businesses are closed and access to health care is increasingly challenging as a result. This situation may result in increased isolation, a lack of physical activity, decreased social engagement, and a suspension of purposeful activity (Brown et al., Reference Brown, Kumar, Rajji, Pollock and Mulsant2020). These disrupted rhythms of daily life may further affect sleep among vulnerable populations, including persons with memory impairment and informal caregivers. Persons with memory problems may be further at risk of worsening cognitive function if their sleep quality is poor. A newly published report in JAMA Open Network stated depression rates have tripled in US adults during the COVID-19 pandemic (Ettman et al., Reference Ettman, Abdalla, Cohen, Sampson, Vivier and Galea2020). The burden and stress of caring for a loved one with dementia may put caregivers at even higher risk for mental health issues (e.g. depression, anxiety, burnout) and physical health (e.g. increased cardiovascular disease risk; Mausbach et al., Reference Mausbach, Patterson, Rabinowitz, Grant and Schulz2007) during this period. The negative impact of the COVID-19 pandemic on family caregivers has likely worsened as they now have less access to formal and informal supports that provide respite and/or assistance with daily activities that were available prior to the pandemic. Furthermore, prior studies (before the pandemic) have demonstrated that caregivers have higher rates of cognitive decline compared to non-caregivers, suggesting need of assessment of both sleep and cognitive function among caregivers (de Vugt et al., Reference de Vugt2006; Vitaliano et al., Reference Vitaliano, Echeverria, Yi, Phillips, Young and Siegler2005).
In this commentary, we have highlighted some active and passive mobile health technologies that can be used to support the use of digital biomarkers in future studies of sleep and cognition, which could improve real-world assessment and enhance precision measurement of predictors and outcomes to improve sleep interventions (Lenze et al., Reference Lenze, Rodebaugh and Nicol2020). Ultimately, continued research, such as Siddarth et al.’s examination of the relationships between sleep and subjective memory complaints, coupled with adequate assessment tools can continue to pave the way for providers to administer precision medicine in behavioral healthcare.
You have
Access
Siddarth et al. (Reference Siddarth2020)’s paper added new findings to the evidence on relationship between sleep and cognitive function. In their paper, poor sustained attention was strongly associated with self-reported sleep quality measured with the Pittsburgh Sleep Quality Index (PSQI; Buysse et al., Reference Buysse, Reynolds, Monk, Berman and Kupfer1989) among middle-aged and older adults with mild subjective memory complaints. Their work enhances our understanding of subjective cognitive function across multiple cognition domains among adults with existing memory complaints, diagnosed using transitional staging models from normal cognition to Alzheimer’s disease or other types of dementia. Although Siddarth and colleagues did not focus on examining the mechanisms linking sleep and cognitive aging (which are still poorly understood), their findings suggest the importance of subjective sleep quality in relation to subjective and objective cognition.
Prior studies have shown strong associations between subjective and/or objective sleep and cognitive function among healthy adults both in cross-sectional and longitudinal study designs. Both objectively measured poor sleep (e.g. low sleep efficiency, increased sleep latency, wake after sleep onset, fragmentation; Campbell et al., Reference Campbell2020; Yaffe et al., Reference Yaffe, Falvey and Hoang2014) and self-reported sleep complaints (Campbell et al., Reference Campbell2020; Tardy et al., Reference Tardy, Gonthier, Barthelemy, Roche and Crawford-Achour2015; Yaffe et al., Reference Yaffe, Falvey and Hoang2014) are significantly associated with worse cognitive function in older adults; however, subjective sleep quality as measured by the PSQI and objective sleep are not consistently correlated with each other in middle-aged and older adults (Landry et al., Reference Landry, Best and Liu-Ambrose2015). Some sleep indices (e.g. snoring, nightmares) are difficult for people to self-assess and report; thus, objectively-measured sleep parameters (e.g. sleep stages, electroencephalogram power values, sleep fragmentation) using polysomnography or actigraphy may help to further elucidate underlying mechanism of sleep and cognitive function, particularly among those who are already experiencing subjective and/or objective memory decline.
Siddarth and colleagues focused on subjective sleep quality and did not measure objective parameters of sleep. Nevertheless, we agree with their argument addressing that subjective sleep complaints could predict future cognitive impairment and dementia even in the absence of objective cognitive impairment. In other words, perceived poor sleep may play a role as an early symptomatic marker in the spectrum of developing Alzheimer’s disease or other types of neurodegeneration. Pending further evidence of contribution of subjective sleep on cognitive decline, monitoring self-reported symptoms of poor sleep may be served as a more cost-effective and easier method to identify at-risk group in combination with biomarkers, neuroimaging, and other clinical features. Additionally, patients’ own assessment of their sleep is important as their own view of sleep may motivate them to engage in cognitive behavioral therapies (e.g. Cognitive Behavioral Therapy for Insomnia [CBT-I]) that have been shown to improve sleep quality (Qaseem et al., Reference Qaseem, Kansagara, Forciea, Cooke and Denberg2016).
Siddarth et al. (Reference Siddarth2020)’s paper extended the evidence on the importance of role of sleep on cognition to more broad groups of older adults from those who already had clinically significant cognitive impairments. The special issue (Volume 32, Issue 7) published in July 2020 in the International Psychogeriatrics discussed sleep and sleep disorders among different groups of older adults. Papers in the issue confirm significant relationship between sleep and cognitive function or other health outcomes such as depression among older adults with neurodegenerative disorders associated with cognitive impairment (e.g. Parkinson’s disease, nursing home residents with dementia). Overall, the clinical message from these findings was sleep as a potential target to prevent further cognitive deterioration and/or psychological well-being. On the other hand, Siddarth et al. (Reference Siddarth2020)’s paper examined older adults who may be at-risk for cognitive decline and reported subjective memory complaints without major cognitive impairment (i.e. dementia). Taken together, monitoring sleep patterns and screening chronic sleep problems such as insomnia may be one way to prevent or delay worsening cognitive impairments and related health issues among older adults with all stages of cognitive function.
Indeed, Siddarth et al. proposed a future sleep intervention study to test whether improved sleep could improve memory consolidation. A few intervention studies have found that improving sleep with CBT-I is associated with improved objective cognitive function; however, other studies have not found any improvements in objective cognitive function in older adults with behavioral treatment for insomnia (Cassidy-Eagle et al., Reference Cassidy-Eagle, Siebern, Unti, Glassman and O’Hara2018; Herbert et al., Reference Herbert, Kyle and Pratt2018). The current literature is limited as there have only been a handful of studies that have examined the dynamic relationship between sleep and cognition. These studies have found different results, and even studies that have found an effect do not consistently observe improvements in the same cognitive domain (e.g. some found improvement in memory whereas others observed improvements in executive function; Herbert et al., Reference Herbert, Kyle and Pratt2018). Additionally, it is not yet clear if these interventions have long-term effects and are effective in lowering the risk of future cognitive impairment.
Ultimately, we agree with Siddarth et al. in that interventions to improve sleep quality should be further studied as these interventions may be an efficacious way to improve subjective and/or objective cognition as well as quality of life in middle-aged and older adults with memory complaints. Although a behavioral approach such as CBT-I is recommended as the first-line therapy for persons with chronic sleep problems (Qaseem et al., Reference Qaseem, Kansagara, Forciea, Cooke and Denberg2016), evidence on effectiveness of the CBT-I for those with mild cognitive impairment is relatively unknown. While limited, an initial study (Cassidy-Eagle et al., Reference Cassidy-Eagle, Siebern, Unti, Glassman and O’Hara2018) found that CBT-I was successful in both improving sleep and cognition among older adults with mild cognitive impairment. Behavioral sleep interventions tailored to levels of cognitive function among persons with mild memory impairment are certainly needed. Some adaptations such as incorporating compensation strategies like alarms/calendars and more novel artificial intelligence-assisted compensation strategy technology (e.g. smart speakers such as Amazon’s Alexa, Google Home) may be particularly useful for this group.
One limitation of Siddarth et al. (Reference Siddarth2020)’s paper was that that they did not measure information about daytime naps and oversleeping (e.g., >9 hours per day) in the paper which have been known as significant risk factors of cognitive decline. However, one considerable strength is that the authors used a questionnaire to screen obstructive sleep apnea in the study (i.e., the STOP-Bang; Chung et al., Reference Chung, Abdullah and Liao2016). While a questionnaire cannot replace a clinical assessment of obstructive sleep apnea, many studies examining sleep and cognition do not include an assessment of obstructive sleep apnea even though obstructive sleep apnea has been shown to be a risk factor for cognitive decline. Nevertheless, being at high risk of obstructive sleep apnea did not change relationship between subjective sleep and cognitive function among study participants. Given that sleep apnea is often undiagnosed and significantly associated with daytime sleepiness, future studies should also consider assessing obstructive sleep apnea as well as daytime naps and sleeping too much in order to understand their role in cognitive function among middle-aged and older adults with memory complaints.
Another limitation in the current literature is that the majority of sleep and cognition research has focused on the relationship between average sleep and cognition measured at one timepoint. Until recently, it has been difficult to measure objective cognition on a daily basis, but one newer tool to help us to obtain rich data of objective cognitive function is smartphone-based ecological momentary cognitive tests. Studies utilizing mobile cognitive testing have shown feasibility in a number of populations including older adults as well as high levels of between- and within-person reliability and validity (Moore et al., Reference Moore, Swendsen and Depp2017). Additionally, these mobile cognitive tests can be paired with ecological momentary assessment questions to more frequently assess subjective sleep and subjective cognitive functioning. This technology can aid in understanding how and if daily fluctuations in subjective sleep impact subjective and objective cognition on an individual level. It may also help lay the groundwork for individualized interventions that could promote better sleep and, by extension, may also promote better cognitive functioning.
Another area to further explore is to monitor intra-individual cognitive variability (IIV) in relationship with sleep–wake cycle. Studies have shown that IIV in sleep is intertwined with sleep–wake cycles (Slavish et al., Reference Slavish, Taylor and Lichstein2019), and higher IIV in cognition has been linked to more widespread medial temporal lobe atrophy in persons with mild cognitive impairment relative to cognitively normal adults and may be a sensitive marker of early neurodegenerative changes (Bangen et al., Reference Bangen2019). A recent study using passively and unobtrusively obtained keystroke metadata (i.e. how you type, not what you type) from real-world daily smartphone keyboard interactions found an age effect and diurnal pattern with the keystroke data (Vesel et al., Reference Vesel2020). Results of the sample of 250 individuals demonstrated a nonlinear diurnal pattern for typing speed, wherein user’s typing speed was fastest midday (12:00 pm to 3:00 pm) and their slowest typing was at night. Older users (aged 70+) exhibited slower typing when compared to younger users, and there was an interaction between age and time of day such that older adults had significantly more pronounced slowing in their typing speed towards the end of the day. This may suggest that abnormal typing dynamics of slower speed and higher variability in speed may represent earliest changes in how a person with cognitive impairments interacts with technology, and potentially elucidate chances in circadian rhythms and IIV, which can be leveraged to differentiate between mild cognitive impairment and healthy controls (Chen et al., Reference Chen2019; Stringer et al., Reference Stringer2018).
The Siddarth et al. (Reference Siddarth2020) paper also implies the critical need of special attentions to vulnerable groups, particularly during the ongoing COVID-19 pandemic. Currently, people are encouraged or required by local ordinances to socially isolate and not to leave their homes as much as possible, or when they do leave their homes to practice social distancing and always wear masks. Many nonessential businesses are closed and access to health care is increasingly challenging as a result. This situation may result in increased isolation, a lack of physical activity, decreased social engagement, and a suspension of purposeful activity (Brown et al., Reference Brown, Kumar, Rajji, Pollock and Mulsant2020). These disrupted rhythms of daily life may further affect sleep among vulnerable populations, including persons with memory impairment and informal caregivers. Persons with memory problems may be further at risk of worsening cognitive function if their sleep quality is poor. A newly published report in JAMA Open Network stated depression rates have tripled in US adults during the COVID-19 pandemic (Ettman et al., Reference Ettman, Abdalla, Cohen, Sampson, Vivier and Galea2020). The burden and stress of caring for a loved one with dementia may put caregivers at even higher risk for mental health issues (e.g. depression, anxiety, burnout) and physical health (e.g. increased cardiovascular disease risk; Mausbach et al., Reference Mausbach, Patterson, Rabinowitz, Grant and Schulz2007) during this period. The negative impact of the COVID-19 pandemic on family caregivers has likely worsened as they now have less access to formal and informal supports that provide respite and/or assistance with daily activities that were available prior to the pandemic. Furthermore, prior studies (before the pandemic) have demonstrated that caregivers have higher rates of cognitive decline compared to non-caregivers, suggesting need of assessment of both sleep and cognitive function among caregivers (de Vugt et al., Reference de Vugt2006; Vitaliano et al., Reference Vitaliano, Echeverria, Yi, Phillips, Young and Siegler2005).
In this commentary, we have highlighted some active and passive mobile health technologies that can be used to support the use of digital biomarkers in future studies of sleep and cognition, which could improve real-world assessment and enhance precision measurement of predictors and outcomes to improve sleep interventions (Lenze et al., Reference Lenze, Rodebaugh and Nicol2020). Ultimately, continued research, such as Siddarth et al.’s examination of the relationships between sleep and subjective memory complaints, coupled with adequate assessment tools can continue to pave the way for providers to administer precision medicine in behavioral healthcare.