Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Jul 1.
Published in final edited form as: Arch Gerontol Geriatr. 2019 Apr 29;83:161–168. doi: 10.1016/j.archger.2019.04.014

Does poor sleep impair cognition during aging? Longitudinal associations between changes in sleep duration and cognitive performance among older Mexican adults

Theresa E Gildner 1,*, Aarón Salinas-Rodríguez 2, Betty Manrique-Espinoza 2, Karla Moreno-Tamayo 3, Paul Kowal 4,5,6
PMCID: PMC6545158  NIHMSID: NIHMS1528579  PMID: 31059924

Abstract

Alterations in sleep patterns are common among older adults; further, short and long sleep durations have been linked with impaired cognitive performance in older individuals. Yet most research examining these relationships has been cross-sectional, limited to high-income nations, and has failed to consider how changes in sleep duration may impact cognitive decline. The present longitudinal study uses nationally-representative data to test whether changes in sleep length among “healthy” baseline sleepers are associated with reduced cognitive function in older Mexican adults (≥50 years old) at follow-up.

Data were drawn from the first and second waves of the World Health Organization’s Study on global AGEing and adult health. Self-report data captured sleep duration over two nights, and five cognitive tests (immediate and delayed verbal recall, forward and backward digit span, and verbal fluency) were used to measure various cognitive domains and create a composite z-score of cognitive performance. Linear regressions were performed to assess associations between sleep length changes and cognitive decline, controlling for relevant lifestyle and health factors.

Increased sleep durations at follow-up among individuals who reported intermediate sleep durations (6-9 hrs/night) at baseline were significantly associated with greater rates of decline in overall cognitive function. Longer sleepers also trended toward greater rates of decline for attention/working memory and executive function. This study suggests that long sleep durations are a risk factor for certain types of impaired cognition among older adults living in a middle-income country. These findings are clinically important given the growing rates of dementia and aging populations globally.

Keywords: sleep changes, cognitive function, longitudinal analysis, dementia

INTRODUCTION

Detrimental changes in sleep duration often occur as individuals age (Bombois et al., 2010; Descamps & Cespuglio, 2010), with older adults reporting higher rates of sleep disorders associated with disrupted and fragmented sleep patterns. These age-related sleep alterations are hypothesized to increase risk of cognitive decline in older adults, and both short and long sleep duration have been linked with impaired cognitive function (Chen et al., 2016; Crenshaw & Edinger, 1999; Devore et al., 2016; Faubel et al., 2009; Lo et al., 2016). Sleep deprivation dulls the senses, slows the reflexes, and impedes ability to recall information (Crenshaw & Edinger, 1999; Devore et al., 2016; Lo et al., 2016). Conversely, long sleep duration is thought to reflect poor sleep quality, physiological dysregulation, or disturbed sleep patterns – perhaps due to underlying illness – which may impair cognitive performance (Faubel et al., 2009; Devore et al., 2016; Lo et al., 2016; Tsapanou et al., 2017).

In addition, chronic problems related to sleep latency (time to sleep onset) and maintenance are frequently associated with adverse consequences, including drowsiness and reduced performance on cognitive tests (Nebes et al., 2009). Recurrent sleep problems which detrimentally affect sleep maintenance and duration can therefore impair daily functioning, increase risk of accidents, and negatively impact the physical and psychosocial status of older individuals. These patterns have been documented cross-culturally. For example, previous work conducted as part of the World Health Organization’s Study on global AGEing and adult health (SAGE) has demonstrated significant cross-sectional relationships between sleep duration and cognitive performance (Gildner et al., 2014). This study found that individuals with intermediate sleep durations (6-9 hrs/night) exhibited significantly higher cognitive scores than individuals with short (< 6 hrs/night) or long (> 9 hrs/night) sleep durations. This suggests that healthy sleep patterns are linked with improved cognition; however, this work was cross-sectional in nature.

Longitudinal work is needed to clarify how changes in sleep patterns might impact cognitive performance, especially in lower-income nations where comparatively little research on these topics have been conducted. Relatively few studies have tested how changes in sleep duration over time in association with cognitive function (Devore et al., 2016), although these few studies indicate that both decreased (Devore et al., 2014; Ferrie et al., 2011; Virta et al., 2013) and increased (Devore et al., 2014; Ferrie et al., 2011; Loerbroks et al., 2010; Virta et al., 2013) sleep durations over time are related to poor cognitive function at follow-up. Even less work has been done examining how changes in sleep duration are associated with differences in cognition between multiple time points; this research has produced non-significant or mixed results (Devore et al., 2014).

Additional work is therefore needed to clarify whether detrimental changes in sleep duration are linked with reduced cognitive function, as has been hypothesized. This information is critical since the ability to minimize age-related cognitive deficits are dependent on the identification of key modifiable factors. These clinical interventions have important implications toward addressing the global burden of dementia, which is projected to more than triple by the year 2050 (WHO, 2012), placing an increased burden on family and caregivers. This is especially true in middle-income countries like Mexico, which often lack the healthcare and social welfare institutions required to support a growing population of older adults (Aguila et al., 2012). Thus, improving behavioral and lifestyle factors (e.g., sleep patterns) during aging may represent an important strategy for maintaining cognitive function at older ages (Haimov & Shatil, 2013; Keage et al., 2012).

The present study examines links between changes in sleep duration and cognitive test performance in a middle-income country using longitudinal data drawn from SAGE (Kowal et al., 2012). Nationally representative data from Mexico were utilized to construct a more comprehensive picture of whether changes in sleep duration from baseline measures substantially impact cognitive performance among older adults at follow-up. The following hypothesis was tested: among older adults with intermediate baseline sleep durations (6-9 hrs/night), substantial changes in sleep duration at follow-up will be associated with reduced cognitive performance.

MATERIALS AND METHODS

Study Design and Participants

A nationally-representative sample of older Mexican adults (≥50 years old) was collected using a stratified multistage cluster sample design (Kowal et al., 2012; Naidoo, 2012). In-person interviews were used to collect baseline household and individual level data between 2009/10 (Wave 1), and follow-up data collected in 2014 (Wave 2). Specifically, three strata were defined by locality (metropolitan, urban, rural); within these strata, the Basic Geo-Statistical Areas defined by the National Institute of Statistics were used as primary sampling units. Households within these sampling units were randomly selected and constituted the secondary sampling units. Household weights were post-stratified by locality according to population census projections. Finally, individuals within households made up tertiary sampling units. Individual weights were post-stratified by sex and age-groups (18-34, 35-49, 50-59, 60-105) according to the census projections. Thus, household and individual weight calculations are based on the selection probability at each stage of selection (Kowal et al., 2012; Naidoo, 2012).

Of the 1,595 living individuals who participated in both Wave 1 and Wave 2 of data collection (Figure 1), those missing one or more of the variables of interest were excluded from regression analysis (n = 648), resulting in a final sample size of 947 participants. The missing adults were younger on average than those included in analyses, but otherwise no substantive differences were observed. To test the hypothesis, a subset of these participants was used during analysis; namely, those with intermediate sleep durations (6-9 hrs/night) at baseline (n = 782).

Figure 1.

Figure 1.

Open in a new tab

Flowchart of the SAGE Mexico design depicting participant retention between Waves 1 and 2.

Ethical Approval

SAGE was approved by the World Health Organization’s Ethical Review Committee. Additionally, the Research and Ethics Committees of Mexico’s National Institute of Public Health approved SAGE Mexico. Written informed consent was obtained from all study participants.

Cognition Variables

The present study used five standard cognitive tests – administered at both baseline and follow-up – to test associations between sleep duration and cognition. The five measures employed here are associated with different aspects of cognition, including: new verbal learning and recall (immediate and delayed verbal recall), attention/working memory (forward and backward digit span), and executive function (verbal fluency) (e.g., see Cullen et al., 2007 for review), providing a more complete picture of how sleep length may impact distinctive cognitive domains.

For the immediate verbal recall test, interviewers read a list of 10 words aloud and asked the participants to immediately recall as many words as they could in one minute. Three trials of this assessment were performed. Upon completing the third trial, the interviewer administered the other cognitive tests, after which delayed recall ability was determined by asking subjects to remember the list of words without the interviewer repeating the list. The digit span test required participants to repeat progressively longer series of numbers; the total score was recorded as the longest digit span repeated without error. This process was then performed with the respondent repeating new sets of increasingly longer digit spans in reverse. The verbal fluency test consisted of naming as many animals (without using proper nouns) as possible in one minute; the final score was correct responses minus errors.

In accordance with other cognitive studies, composite z-scores were calculated to facilitate the comparison of cognitive test performance between individuals (e.g., Devore et al., 2014; Gildner et al., 2014; Scarmeas et al., 2006; Tsapanou et al., 2017). This approach accounts for the fact that the cognitive tests were scaled differently, allowing the various test scores to be meaningfully compared and used to calculate an overall, composite cognitive performance score. Specifically, z-scores for each cognitive test were first computed (relative to the sample mean) and these five z-scores were then summed for each participant, resulting in a final composite z-score. Additionally, the change in participant composite cognition score between Waves 1 and 2 was calculated by subtracting the baseline composite cognitive z-score from the composite z-score at follow-up.

Sleep Duration

Average sleep durations at baseline (Wave 1) and follow-up (Wave 2) were calculated. Participants were asked their sleep duration on each of the preceding two nights; these measures did not include daytime sleep. In accordance with other sleep studies (Faubel et al., 2009; Patel et al., 2006), the duration values across two nights were averaged together to create a summary measure of sleep length so mean sleep duration could be compared. Following a standard approach (Faubel et al., 2009; Ohayon & Vecchierini, 2005; Tamakoshi & Ohno, 2004), sleep duration was categorized to identify individuals with “healthy”, intermediate sleep durations at baseline (6-9 hrs/night). These cutoff points were selected based on common medical definitions of short and long sleep (e.g., Mukheijee et al., 2015), while also remaining consistent with previous SAGE research (e.g., Gildner et al., 2014). However, varying definitions of “optimal” sleep duration exist (Cappuccio et al, 2010); a second set of sleep category cut-off points was therefore used during analysis, with individuals reporting 7-9 hrs/night categorized as “healthy” baseline sleepers. Both sets of sleep categories were tested during analysis. Finally, the change in participant sleep duration between Waves 1 and 2 was calculated by subtracting average baseline sleep duration from average sleep duration at follow-up.

Covariates

Variables demonstrated in past studies to impact cognitive performance were controlled for in all statistical analyses. Evidence suggests that participant sex and age influence cognitive decline (Beydoun et al., 2012; Mielke et al., 2014), these variables were therefore included in all analyses. In addition, education level was classified using the International Standard Classification of Education (UNESCO, 1997) and included during analysis. Given that ethnic background can influence health patterns (Alvarado-Esquivel et al., 2004), participants were also asked about their ethnicity and this response was included during analysis. History of unemployment has also been linked with cognitive performance (Alvarado-Esquivel et al., 2004), thus employment status was also included in the analyses. Moreover, previous research suggests that individuals living in rural settings exhibit a higher prevalence of dementia compared to urban areas (Jia et al., 2014). Participant household setting (urban vs. rural) was therefore controlled for during analysis.

Underlying medical conditions known to influence cognitive function were also added to all models (depression, prior diagnoses of stroke, hypertension, diabetes, and chronic lung disease) as a summed total of all diagnosed comorbidities (Iadecola, 2013; Kessler et al., 2010; Kling et al. 2013). Depression status was determined using symptom-based reporting and an algorithm to assign diagnosis based on the World Mental Health Composite International Diagnostic Interview (Kessler et al., 2010). It is worth noting that depression demonstrates a complex relationship with sleep patterns (Roberts et al., 2000), however mediation analyses indicated that depression diagnosis does not significantly mediate the relationship between sleep duration and cognitive performance in this sample (Z med = 0.009, | 0.009 | ≤ 1.96). Participants were also asked about medication use, including use of opioids, anxiolytics, sedatives, and stimulants. From this information a binary medication variable was created (not taking any medications vs. taking at least one) and included during analysis.

Health-related factors known to influence cognitive function (body mass index, physical activity levels, gait speed, tobacco use, and drinking patterns) were also included in all analyses (Anttila et al., 2004; Laurin et al., 2001; Ott et al., 1998). Specifically, participant height and weight were collected, and these values were used to calculate BMI (kg/m2). Physical activity level was measured based on responses to questions drawn from the Global Physical Activity Questionnaire (WHO, 2009). As has been described elsewhere (Rivas- Marino et al., 2015), three categories of physical activity (low, moderate, and high physical activity) were calculated from these questionnaire items, based on reported time spent in moderate or vigorous activities during work, recreational/leisure time, and transportation.

Previous research indicates that gait speed – a measure of time required to walk a set distance (in m/s) – represents an important indicator of physical function that may be linked with cognitive function (e.g., Buraccio et al., 2010; Watson et al., 2010). Participant gait speed was therefore included during analysis. Gait speed was assessed through timed walks over a 4-meter flat area. The lengths of time required to travel this distance at both the participant’s usual and rapid walking speed were recorded, and normal and rapid gait speed were calculated in meters per second (m/s). Height-standardized z-scores for usual and rapid gait speed were calculated and these values were used during analysis. Given the complex relationship proposed between gait speed and cognition (e.g., Buraccio et al., 2010; Watson et al., 2010), preliminary analyses were conducted testing whether gait speed mediated the relationship between sleep duration and cognitive performance. However, these mediation analyses indicated that neither normal nor rapid gait speed significantly mediated the relationship between sleep duration and any of the cognitive function measures in this sample (Z med = 0.167 – 1.67, all values ≤ 1.96). Normal and rapid gait speed were therefore included during analysis as covariates. Respondents were also asked if they had ever used tobacco or consumed alcohol and if participants answered affirmatively the frequency of use was recorded (see Table 1). Finally, following similar studies (Jelicic et al., 2002), baseline cognitive performance and length of the time interval between the first and second participant interview (i.e., time between baseline and follow-up measures) were included in the analyses.

Table 1.

Description of the study population (n = 782; intermediate baseline sleepers) at follow-up (weighted data), data were drawn from baseline SAGE Wave 1 (2009-10) and follow-up Wave 2 (2014).

Variables Mean SE
Age 62.12 1.06
Normal gait speed (in m/s) 0.801 0.026
Rapid gait speed (in m/s) 1.16 0.037
Follow-up interval (in days) 1727.82 2.15
Baseline composite cognitive z-score 0.122 0.246
Change in composite cognitive z-score −0.172 0.185
Change verbal recall z-score 0.056 0.061
Change in forward digit span z-score 0.374 0.090
Change in backward digit span z-score −0.019 0.083
Change in verbal fluency z-score −0.665 0.116
Change in delayed verbal recall z-score 0.082 0.086
Change in sleep duration −0.339 0.189
Proportion S.E.
Male sex (vs. female) 0.446 0.061
Live in urban area (vs. rural) 0.703 0.059
No formal education 0.156 0.046
Less than high school 0.698 0.050
High school or beyond 0.146 0.033
Positive depression diagnosis 0.131 0.025
Positive stroke diagnosis 0.025 0.010
Positive chronic lung disease diagnosis 0.029 0.011
Positive hypertension diagnosis 0.277 0.035
Positive diabetes diagnosis 0.263 0.054
Positive medication use 0.100 0.026
Low physical activity level 0.537 0.062
Moderate physical activity level 0.183 0.030
High physical activity level 0.280 0.057
Never used tobacco 0.763 0.049
Not currently using tobacco 0.115 0.030
Use tobacco occasionally 0.019 0.006
Use tobacco daily 0.102 0.042
Never consumed alcohol 0.559 0.053
Not currently drinking 0.141 0.023
Drink < 1/month 0.206 0.052
Drink 1-3 days/month 0.077 0.042
Drink 1-4 days/week 0.010 0.004
Drink 5+ days/week 0.006 0.003
Open in a new tab

Statistical Analyses

Tests for normality were performed, and no violations were observed. Multicollinearity was not detected between any of the variables. Parametric tests were subsequently conducted using Stata version 14 (StataCorp LLC, College Station, Texas, USA) to test the hypotheses. Sample weights were used in all analyses (Naidoo, 2012). A Bonferroni correction was applied to account for multiple comparisons (i.e., the six cognitive variables tested) and p-values were considered significant at less than 0.008. A series of linear regressions were conducted to test the hypotheses and evaluate the relative contribution of sleep duration to changes in cognitive score.

Initial analyses indicated that participant reported ethnicity, employment status, and BMI did not significantly contribute to any statistical model. These variables were consequently dropped from analysis. All subsequent analyses controlled for participant sex, age, education level, household setting, number of diagnosed comorbidities, medication use, physical activity level category, gait speed (both usual and rapid), tobacco and alcohol use, baseline cognitive score, and follow-up interval. Preliminary analyses examining the association between baseline sleep duration and changes in cognitive test performance between Waves 1 and 2 (controlling for all covariates) indicated that sleep duration category at baseline (short, intermediate, or long) was not significantly associated with change in cognitive test score for any of the six cognitive function measures; this was true for both calculated sleep duration categories (i.e., categorizing intermediate sleepers as individuals reporting 6-9 hrs/night or 7-9 hrs/night). However, the majority of participants reported an intermediate sleep duration at baseline (n = 782, 83% reported 6-9 hrs/night; n = 647, 68% reported 7-9 hrs/night). This lack of variation in sleep duration category may account for the absence of statistically significant relationships between baseline sleep duration and change in cognition. Of greater interest is how changes in sleep patterns among this large number of seemingly healthy baseline sleepers may be associated with detrimental alterations to cognitive performance over time.

Still, it should be noted that the relationship between cognitive decline and sleep patterns is complex and possibly bidirectional. For example, Alzheimer’s disease appears to cause circadian disruption and disturbed sleep patterns (Ju et al., 2014). To account for the potential influence of impaired cognition at baseline on subsequent sleep patterns at follow-up, sensitivity analyses were conducted. Based on previous methods (Vancampfort et al., 2017), SAGE participants exhibiting mild cognitive impairment at baseline (Wave 1) were identified. These individuals (n = 136) were then excluded from analysis during hypothesis testing; however, the exclusion of these participants did not substantively change the study results and they were subsequently included in the final analyses to increase sample size and statistical power.

Descriptive statistics

Average change in sleep duration or cognition (for each cognitive measure) were calculated. In addition, average values or frequencies for all other variables of included in the models were calculated.

Examination of the relative contribution of sleep duration change to variation in cognitive decline, among participants with intermediate baseline sleep durations

A series of linear regressions examined if changes in average sleep duration contributed to changes in cognitive test performance (the composite cognitive score and each of the individual cognitive performance tests) among participants with intermediate baseline sleep durations (6-9 hrs/night). Given the similar results evident between the two different classifications of intermediate sleep (6-9 hrs/night vs. 7-9 hrs/night), the classification of 6-9 hrs/night was retained to maximize sample size, increase statistical power, and remain consistent with previous SAGE research (e.g., Gildner et al., 2014). The change in sleep duration between Waves 1 and 2 was entered into these models.

RESULTS

Descriptive statistics are presented in Table 1. The sample was composed of slightly more women than men, lived in an urban area, had received some formal schooling (but lacked a high school degree), had not been diagnosed with one of the underlying medical conditions, did not report high rates of medicine use, reported low physical activity levels, and had never used tobacco or consumed alcohol. Average age at follow-up was roughly 62 years, mean normal gait speed was 0.801 m/s, mean rapid gait speed was 1.16 m/s and average follow-up interval was 4.7 years (between Waves 1 and 2).

As predicted, changes in sleep durations between baseline and follow-up were associated with greater cognitive decline among participants reporting intermediate sleep durations at baseline, such that increased sleep durations among this sample were associated with greater negative changes in certain aspects of cognition (Table 2; for full models see Appendix A). Specifically, a significant inverse relationship between change in sleep duration and magnitude of change in overall cognitive performance was evident (B = −0.291, p = 0.004). Likewise, the inverse association between changes in sleep duration and cognitive performance change in backward digit span trended toward significance (B = −0.096, p = 0.016), and a similar inverse relationship trended toward significance for cognitive performance change in verbal fluency (B = −0.103, p = 0.059). Overall, these findings suggest that increased sleep durations at follow-up are associated with greater cognitive impairment in select measures of cognitive performance.

Table 2.

Multiple regression models (using weighted data) for the prediction of changes in cognitive performance score from change in sleep duration between Waves 1 and 2 among participants with intermediate sleep duration (6-9 hrs/night) at Wave 1 (n = 782); B coefficients (SE). All analyses controlled for participant sex, age, education level, household setting, number of diagnosed comorbidities, medication use, physical activity level category, gait speed, tobacco and alcohol use, baseline cognitive score, and follow-up interval. See Appendix A for full models.

Composite Cognition Immediate Verbal Recall Forward Digit Spam Backward Digit Spam Verbal Fluency Delayed Verbal Recall
−0.291 (0.099)** −0.012 (0.027) −0.050 (0.042) −0.096 (0.039)* −0.103 (0.054)* −0.032 (0.042)
Open in a new tab

Comparisons are statistically significant (with Bonferroni adjustment) at:

*

= trend toward significance (p < 0.06)

**

= p < 0.008

***

= p < 0.001

DISCUSSION

This study provides a unique examination of longitudinal associations between sleep duration change and cognitive decline in older Mexican individuals. Previous studies examining similar patterns have been confined to high-income countries and have generally relied on data collected from cross-sectional and non-representative population samples. The present study found partial support for the hypothesis. Among participants with intermediate sleep durations (6-9 hrs/night) at baseline, increased sleep durations at follow-up were associated with greater cognitive decline on certain measures of cognitive function, as was predicted. Still, while the relationships between long sleep and reduced backward digit span and verbal fluency scores trended toward significance, longer sleep was only significantly associated with changes in overall composite cognition.

There are several possible neurobiological explanations for these results. First, evidence suggests that long sleep durations may accelerate the rate of frontotemporal gray matter atrophy among older adults (Spira et al., 2016), potentially impairing memory. It is also possible that long sleep durations may reflect circadian disruptions linked with sleep dysregulation and impaired cognition (Devore et al., 2014; Yoo & Eckel-Mahan, 2016). Mounting evidence indicates that irregular circadian rhythms are associated with short-term cognitive impairment and longer-term brain atrophy (Cho et al., 2000; Cho, 2001; Machi et al., 2012); although these relationships appear to be complex. Moreover, the hormone melatonin plays an important role in shaping sleep patterns and has also been linked with cognitive function (e.g., memory performance) and rate of neurocognitive decline (Daulatzai, 2016; Hardeland et al., 2006; Santoro et al., 2016; Tsapanou et al., 2017). Abnormal melatonin levels may therefore influence the relationship between sleep duration and cognitive function.

The results documented in the present study are consistent with previous findings in high-income nations, which have also produced results indicating that long sleep negatively impacts aspects cognitive performance. For example, work among older adults living in Spain and the United States indicates that participants reporting long sleep durations exhibited significantly lower scores on mental tests related to attention and working memory (Faubel et al., 2009; Patel et al., 2006; Schmutte et al., 2007; Xu et al., 2011), as the results here suggest. This may reflect cognitive deficits linked with underlying health conditions and sleep disorders thought to impair sleep. Unhealthy individuals may suffer from disrupted sleep patterns and therefore require more time in bed to feel rested, leading these individuals to report longer sleep durations. However, longer reported sleep durations may not entirely compensate for underlying health problems and cognitive decline may still occur. These kinds of disturbed sleep patterns have been linked with cognitive impairments (Blackwell et al., 2006), including tasks related to memory and attention.

However, systematic reviews of research examining these associations indicate that the relationship between long sleep and cognitive function is not consistent across studies (Devore et al., 2016; Lo et al., 2016). According to one systematic review of sleep and cognition literature, studies that have documented significant associations between long sleep duration and cognitive decline have all relied on self-reported sleep durations (Devore et al., 2016), such as the subjective measures used in the present study. Objective measures of sleep duration have failed to document a significant relationship between extreme sleep durations and cognitive decline (Devore et al., 2016). This suggests that it is the perception of sleep length, rather than actual time asleep, that is related to reduced cognitive performance. Additional research is therefore needed to clarify how subjective discernment of sleep patterns – perhaps shaped in part by participant health – might contribute to the development of disrupted sleep patterns and cognitive decline over time.

Future research directions

The present study provides novel data demonstrating significant relationships between increased sleep duration and decline in certain aspects of cognition. These findings could be expanded through the incorporation of underlying health conditions into future work examining relationships between sleep and cognition. For example, a next step would be to examine how various health conditions might occur jointly with detrimental changes in sleep duration (i.e., due to shared etiologies) and also serve as predictors of subsequent cognitive decline. In addition, future waves of SAGE data will help clarify the relationships between sleep length and cognitive function evident in the present study. Measuring changes in sleep patterns at additional time points may provide more insight into underlying health conditions which negatively impact sleep patterns (e.g., resulting in longer sleep durations), but only become apparent overtime and can only be captured by assessing sleep patterns and cognitive performance at several time points. Likewise, additional waves of data may better enable the detection of cognitive decline linked with sleep patterns (i.e., these changes in cognition may only become apparent after longer follow-up periods) (Devore et al., 2014).

Limitations and strengths

The present study has important limitations. First, the SAGE questionnaire did not collect information on participant naps or sleep disorders (including insomnia), which may impact sleep durations and cognitive function. Second, the sleep data used were reliant on participant reported sleep duration from only the two nights of sleep prior to the interview, which may not accurately reflect typical sleep patterns. An additional limitation related to the use of self-reported sleep data is that individuals may not be able to distinguish between time in bed and time asleep. Evidence indicates that when asked to report hours slept, participants often report time spent in bed; however, this value may exceed time spent asleep (Stenholm et al., 2011). Objective measures of sleep based on polysomnography or actigraphy would better capture individual sleep patterns, including sleep duration, transitions between sleep states, and length of night awakenings (Lockley et al., 2002).

However, the use of objective sleep duration measurements affords the opportunity to examine how individual impressions of sleep patterns are associated with cognitive function, an especially important measure when assessing the contribution of long sleep durations to cognitive decline (Devore et al., 2016). This study also provides an important test of the relationship between sleep duration and several diverse measures of cognition by evaluating these variables across multiple waves of data collection, thereby examining how longitudinal changes in sleep may be associated with various aspects of cognitive decline in a nationally-representative population.

Conclusion

In conclusion, this study documented a significant association between longer sleep durations and increased overall cognitive decline among older Mexican adults reporting intermediate sleep durations at baseline. These results support previous findings in high-income populations and suggest that perceptions of long sleep are related to health conditions which negatively affect cognitive performance over time. Subjective measures of long sleep duration in older adults may therefore represent an important indicator of dementia risk. Furthermore, promoting healthy sleep habits (e.g., avoiding excessive time in bed and establishing sleep hygiene practices that enhance sleep quality) could represent an important consideration in future clinical studies aimed at mitigating cognitive decline in older individuals; not only in wealthy countries, but in lower- and middle-income populations as well.

Highlights.

  • Links between sleep length and cognitive decline were tested among older Mexicans.

  • Increased sleep length in healthy baseline sleepers impaired aspects of cognition.

  • Attention and working memory were most strongly associated with sleep changes.

  • Increased sleep length may be a risk factor for some types of cognitive decline.

ACKNOWLEDGMENTS

We thank Nirmala Naidoo for help with data management, all the study respondents, and the study team in Mexico. WHO’s Study on global AGEing and adult health is supported by the respective governments in collaborating countries, WHO, and the US National Institute on Aging through Interagency Agreements and a research grant (NIH R01-AG034479).

CONFLICT OF INTEREST STATEMENT

Support for the research was provided by NIH NIA Interagency Agreement YA1323-08-CN-0020 with the World Health Organization and grant NIH R01-AG034479. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. All authors have indicated no conflicts of interest.

Appendix A. Multiple regression models for prediction of changes in cognitive performance from continuous sleep duration change between Waves 1 and 2, among healthy sleepers (6-9 hrs/night) at Wave 1, B (SE) (n =782)a,b.

Composite
Cognition		 Immediate
Verbal Recall		 Forward
Digit Spam		 Backward
Digit Spam		 Verbal
Fluency		 Delayed
Verbal Recall
Constant −8.86 (13.12) −2.57 (4.44) 4.46 (5.37) −6.49 (5.66) −15.78 (7.21)* 11.52 (8.23)
Sex 0.764 (0.306)* 0.236 (0.124)* 0.317 (0.156)* 0.034 (0.162) −0.070 (0.162) 0.246 (0.109)*
Age: 60-69 years old 0.113 (0.366) −0.099 (0.118) 0.156 (0.181) −0.140 (0.161) 0.235 (0.221) −0.039 (0.162)
70-79 years old −1.07 (0.367)** −0.475 (0.141)** 0.126 (0.169) −0.076 (0.203) −0.324 (0.195) −0.321 (0.179)
80-89 years old −1.99 (0.553)*** −0.534 (0.225)* −0.141 (0.260) 0.057 (0.204) −1.03 (0.304)** −0.348 (0.228)
90+ years old −2.06 (0.932)* −0.376 (0.283) −0.274 (0.357) −0.237 (0.333) −0.619 (0.487) −0.553 (0.561)
Household setting −0.787 (0.298)* 0.094 (0.130) −0.223 (0.126) −0.074 (0.145) −0.508 (0.264)* −0.075 (0.204)
Education: < high school 0.636 (0.334)* −0.082 (0.143) −0.239 (0.210) 0.626 (0.185)** 0.148 (0.188) 0.182 (0.232)
high school or beyond 2.21 (0.527)*** 0.338 (0.187) −0.148 (0.275) 1.01 (0.262)*** 0.535 (0.325) 0.476 (0.295)
Physical activity level: moderate −0.625 (0.314)* 0.177 (0.113) −0.186 (0.137) −0.184 (0.147) −0.359 (0.228) −0.073 (0.159)
high 0.399 (0.386) 0.036 (0.133) 0.109 (0.162) 0.083 (0.157) −0.095 (0.179) 0.265 (0.169)
Normal gait speed 2.58 (0.960)* 0.581 (0.462) 0.762 (0.361)* −0.483 (0.298) 1.00 (0.480)* −0.536 (0.419)
Rapid gait speed −0.356 (0.636) 0.232 (0.226) 0.271 (0.296) 0.766 (0.519) −0.733 (0.304)* 0.357 (0.221)
Number of comorbidities −0.133 (0.181) 0.005 (0.048) 0.075 (0.074) −0.032 (0.073) −0.121 (0.088) −0.061 (0.073)
Medication use 0.308 (0.649) −0.029 (0.169) 0.082 (0.179) 0.119(0.171) 0.181 (0.221) −0.045 (0.311)
Tobacco use: not currently using 0.113 (0.435) 0.063 (0.135) 0.199 (0.307) −0.242 (0.247) −0.030 (0.193) 0.123 (0.175)
not using daily −0.755 (0.893) 0.089 (0.229) −0.236 (0.343) −0.412 (0.205)* 0.014 (0.481) −0.209 (0.321)
using daily 0.015 (0.450) −0.257 (0.187) 0.030 (0.188) 0.656 (0.210)** 0.291 (0.213) −0.705 (0.261)**
Drinking frequency: do not drink −0.072 (0.348) 0.152 (0.132) −0.081 (0.166) 0.076 (0.140) −0.163 (0.164) −0.056 (0.148)
less than once per month −0.051 (0.315) −0.181 (0.166) −0.177 (0.174) 0.419 (0.180)* −0.025 (0.179) −0.087 (0.165)
1-3 days per month 0.235 (0.604) 0.361 (0.243) −0.165 (0.290) 0.010 (0.362) 0.292 (0.327) −0.263 (0.330)
1-4 days per week 0.325 (0.874) 0.484 (0.281) 0.845 (0.328)* 0.272 (0.239) −0.949 (0.303)** −0.327 (0.472)
5+ days per week 0.295 (0.889) −0.163 (0.237) 0.368 (0.218) −0.502 (0.353) −0.180 (0.454) 0.771 (0.367)*
Baseline composite cognition −0.527 (0.082)*** −0.100 (0.021)*** −0.001 (0.029) −0.032 (0.024) −0.250 (0.074)** −0.144 (0.021)***
Follow-up interval 0.005 (0.008) 0.001 (0.003) −0.002 (0.003) 0.003 (0.003) 0.009 (0.004)* −0.007 (0.005)
Sleep duration change −0.291 (0.099)** −0.011 (0.027) −0.050 (0.042) −0.096 (0.039)* −0.103 (0.054)* −0.032 (0.042)
Open in a new tab
a

Comparisons are statistically significant (with Bonferroni adjustment) at: *= trend toward significance (p < 0.06), **= p < 0.008, ***= p < 0.001

b

Reference groups used in the creation of dummy codes for each categorical variable:

(i) Sex = male

(ii) Age = 50 – 59 years

(iii) Household Setting = urban

(iv) Education Levels = no formal education

(v) Physical Activity Levels = low activity levels

(vi) Medication use = not taking any medication

(vii) Tobacco Use = never used

(viii) Drinking Frequency = never used

(ix) Sleep Category: intermediate sleep duration (6-9 hrs/night)

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

DATA STATEMENT

Data from the World Health Organization Study on global AGEing and adult health (SAGE) Waves 1 and 2 can be accessed online at: http://apps.who.int/healthinfo/systems/surveydata/index.php/catalog/sage/about

REFERENCES

  1. Aguila E, Diaz C, Fu MM, Kapteyn A and Pierson A Living longer in Mexico: Income security and health. RAND Health, 2012, 1(4): 1 10.1037/e525502012-001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alvarado-Esquivel C, Hernández-Alvarado AB, Tapia-Rodriguez RO, Guerrero-Iturbe Á, Rodriguez-Corral K and Martinez SE Prevalence of dementia and Alzheimer’s disease in elders of nursing homes and a senior center of Durango City, Mexico. BMC Psych, 2004, 4(1): 3 10.1186/1471-244x-4-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Anttila T, Helkala EL, Viitanen M, Kareholt I, Fratiglioni L, Winblad B, … and Kivipelto M Alcohol drinking in middle age and subsequent risk of mild cognitive impairment and dementia in old age: a prospective population based study. BMJ, 2004, 329(7465): 539 10.1136/bmj.38181.418958.be [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Beydoun MA, Boueiz A, Abougergi MS, Kitner-Triolo MH, Beydoun HA, Resnick SM, … and Zonderman AB Sex differences in the association of the apolipoprotein E epsilon 4 allele with incidence of dementia, cognitive impairment, and decline. Neurobiol. Aging, 2012, 33(4): 720–731. 10.1016/j.neurobiolaging.2010.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Blackwell T, Yaffe K, Ancoli-Israel S, Schneider JL, Cauley JA, Hillier TA, … and Stone KL Poor sleep is associated with impaired cognitive function in older women: the study of osteoporotic fractures. J. Gerontol. A Biol. Sci. Med. Sci, 2006, 61(4): 405–410. 10.1093/gerona/61.4.405 [DOI] [PubMed] [Google Scholar]
  6. Bombois S, Derambure P, Pasquier F, and Monaca C Sleep Disorders in Aging and Dementia. J. Nutr. Health Aging, 2010, 14: 212–217. 10.1007/s12603-010-0052-7 [DOI] [PubMed] [Google Scholar]
  7. Buracchio T, Dodge ΗH, Howieson D, Wasserman D, and Kaye J The trajectory of gait speed preceding mild cognitive impairment. Arch. Neurol, 2010, 67(8): 980–986. 10.1001/archneurol.2010.159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cappuccio FP, D’Elia L, Strazzullo P, & Miller MA Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep, 2010, 33(5): 585–592. 10.1093/sleep/33.5.585 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen JC, Espeland MA, Brunner RL, Lovato LC, Wallace RB, Leng X, … and Manson JE Sleep duration, cognitive decline, and dementia risk in older women. Alzheimers Dement, 2016, 12(1): 21–33. 10.1016/j.jalz.2015.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cho K, Ennaceur A, Cole JC, and Suh CK Chronic jet lag produces cognitive deficits. J. Neurosci, 2000, 20: RC66 10.1523/jneurosci.20-06-j0005.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cho K Chronic ‘jet lag’ produces temporal lobe atrophy and spatial cognitive deficits. Nat. Neurosci, 2001, 4: 567–568. 10.1038/88384 [DOI] [PubMed] [Google Scholar]
  12. Crenshaw M, and Edinger J Slow-wave sleep and waking cognitive performance among older adults with and without insomnia complaints. Physiol. Behav, 1999, 66: 485–492. 10.1016/s0031-9384(98)00316-3 [DOI] [PubMed] [Google Scholar]
  13. Cullen B, O’Neill B, Evans JJ, Coen RF, and Lawlor BA A review of screening tests for cognitive impairment. J. Neurol. Neurosurg. Psychiatry, 2007, 78(8): 790–799. 10.1136/jnnp.2006.095414 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Daulatzai MA Pharmacotherpy and Alzheimer’s Disease: The M-Drugs (Melatonin, Minocycline, Modafinil, and Memantine) approach. Curr. Pharm. Des, 2016, 22(16): 2411–2430. 10.2174/1381612822666160203142111 [DOI] [PubMed] [Google Scholar]
  15. Descamps A, and Cespuglio R Influence of aging on the sleep rebound induced by immobilization stress in the rat. Brain Res, 2010, 1335: 14–23. 10.1016/j.brainres.2010.03.087 [DOI] [PubMed] [Google Scholar]
  16. Devore EE, Grodstein F, Duffy JF, Stampfer MJ, Czeisler CA, and Schernhammer ES Sleep duration in midlife and later life in relation to cognition. J. Am. Geriatr. Soc, 2014, 62:1073–1081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Devore EE, Grodstein F and Schernhammer ES Sleep duration in relation to cognitive function among older adults: a systematic review of observational studies. Neuroepidemiol, 2016, 46(1): 57–78. 10.1159/000442418 [DOI] [PubMed] [Google Scholar]
  18. Faubel R, López- García E, Guallar- Castillón P, Graciani A, Banegas JR, and Rodriguez-Artalejo F Usual sleep duration and cognitive function in older adults in Spain. J. Sleep Res, 2009, 18: 427–435. 10.1111/j.l365-2869.2009.00759.x [DOI] [PubMed] [Google Scholar]
  19. Ferrie JE, Shipley MJ, Akbaraly TN, Marmot MG, Kivimäki M, and Singh-Manoux A Change in sleep duration and cognitive function: findings from the Whitehall II study. Sleep, 2011, 34: 565–573. 10.1093/sleep/34.5.565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gildner TE, Liebert MA, Kowal P, Chatteiji S, and Snodgrass JJ Associations between sleep duration, sleep quality, and cognitive test performance among older adults from six middle income countries: results from the Study on Global Ageing and Adult Health (SAGE). JCSM, 2014, 10(6): 613 10.5664/jcsm.3782 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Haimov I, and Shatil E Cognitive training improves sleep quality and cognitive function among older adults with insomnia. PLoS One, 2013, 8:e61390 10.1371/journal.pone.0061390 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hardeland R, Pandi-Perumal SR, and Cardinali DP Melatonin. Int. J. Biochem. Cell Biol., 2006, 38(3): 313–316. 10.1016/j.biocel.2005.08.020 [DOI] [PubMed] [Google Scholar]
  23. Iadecola C The pathobiology of vascular dementia. Neuron., 2013, 80(4): 844–866. 10.1016/j.neuron.2013.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jelicic M, Bosma H, Ponds RW, Van Boxtel MP, Houx PJ, and Jolles J Subjective sleep problems in later life as predictors of cognitive decline. Report from the Maastricht Ageing Study (MAAS). Int. J. Geriatr. Psychiatry, 2002, 17(1): 73–77. 10.1002/gps.529 [DOI] [PubMed] [Google Scholar]
  25. Jia J, Wang F, Wei C, Zhou A, Jia X, Li F, … and Cui Y The prevalence of dementia in urban and rural areas of China. Alzheimers Dement, 2014, 10(1): 1–9. 10.1016/jjalz.2013.01.012 [DOI] [PubMed] [Google Scholar]
  26. Ju YES, Lucey BP, and Holtzman DM Sleep and Alzheimer disease pathology—a bidirectional relationship. Nat. Rev. Neurol, 2014, 10(2): 115 10.1038/nrneurol.2013.269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Keage HA, Banks S, Yang KL, Morgan K, Brayne C, and Matthews FE What sleep characteristics predict cognitive decline in the elderly? Sleep Med, 2012, 13: 886–92. 10.1016/j.sleep.2012.02.003 [DOI] [PubMed] [Google Scholar]
  28. Kessler RC, Birnbaum HG, Shahly V, Bromet E, Hwang I, McLaughlin KA, … and Haro JM Age differences in the prevalence and co- morbidity of DSM- IV major depressive episodes: results from the WHO World Mental Health Survey Initiative. Depress. Anxiety, 2010, 27(4): 351–364. 10.1002/da.20634 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kling MA, Trojanowski JQ, Wolk DA, Lee VM, and Arnold SE Vascular disease and dementias: paradigm shifts to drive research in new directions. Alzheimers Dement, 2013, 9(1): 76–92. 10.1016/j.jalz.2012.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kowal P, Chatterji S, Naidoo N, Biritwum R, Fan W, Lopez Ridaura R, … and Boerma JT Data resource profile: The WHO Study on Global Ageing and Adult Health (SAGE). Int. J. Epidemiol, 2012, 41(6): 1639–1649. Available at: https://www.ncbi.nlm.nih.gov/pubmed/23283715 (Accessed March 20, 2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Laurin D, Verreault R, Lindsay J, MacPherson K, and Rockwood K Physical activity and risk of cognitive impairment and dementia in elderly persons. Arch. Neurol, 2001, 58(3): 498–504. 10.1001/archneur.58.3.498 [DOI] [PubMed] [Google Scholar]
  32. Lo JC, Groeger JA, Cheng GH, Dijk DJ, and Chee MW Self-reported sleep duration and cognitive performance in older adults: a systematic review and meta-analysis. Sleep Med, 2016, 17: 87–98. 10.1016/j.sleep.2015.08.021 [DOI] [PubMed] [Google Scholar]
  33. Lockley SW, Skene DJ, and Arendt J Comparison between subjective and actigraphic measurement of sleep and sleep rhythms. J. Sleep Res, 2002, 8: 175–183. 10.1046/j.1365-2869.1999.00155.x [DOI] [PubMed] [Google Scholar]
  34. Loerbroks A, Debling D, Amelang M, and Sturmer T Nocturnal sleep duration and cognitive imp airment in a population-based study of older adults. Int. J. Geriatr. Psychiatry, 2010, 25: 100–109. 10.1002/gps.2305 [DOI] [PubMed] [Google Scholar]
  35. Machi MS, Staum M, Callaway CW, Moore C, Jeong K, Suyama J, and Patterson PD, Hostler D The relationship between shift work, sleep, and cognition in career emergency physicians. Acad. Emerg. Med, 2012, 19: 85–91. doi.org/10.111l/j.1553-2712.2011.01254.x [DOI] [PubMed] [Google Scholar]
  36. Mielke MM, Vemuri P, and Rocca WA Clinical epidemiology of Alzheimer’s disease: assessing sex and gender differences. Clin. Epidemiol, 2014, 6: 37–48. 10.2147/clep.s37929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mukheijee S, Patel SR, Kales SN, Ayas NT, Strohl KP, Gozal D, and Malhotra A An official American Thoracic Society statement: the importance of healthy sleep. Recommendations and future priorities. Am. J. Respir. Crit. Care Med, 2015, 191(12): 1450–1458. 10.1164/rccm.201504-0767st [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Naidoo N WHO Study on global AGEing and adult health (SAGE) Waves 0 and 1 sampling information for China, Ghana, India, Mexico, Russia and South Africa. World Health Organization; [Internet, 2012], Available at: http://www.who.int/healthinfo/sage/SAGEWorkingPaper5_WavelSampling.pdf (Accessed March 23, 2018). [Google Scholar]
  39. Nebes RD, Buysse DJ, Halligan EM, Houck PR, and Monk TH Self-reported sleep quality predicts poor cognitive performance in healthy older adults. J. Gerontol, 2009, 64: 180–187. 10.1093/geronb/gbn037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ohayon MM, and Vecchierini M Normative sleep data, cognitive function and daily living activities in older adults in the community. Sleep, 2005, 28: 981–989. 10.1093/sleep/28.8.981 [DOI] [PubMed] [Google Scholar]
  41. Ott A, Slooter AJC, Hofiman A, van Harskamp F, Witteman JCM, Van Broeckhoven C, … and Breteler MMB Smoking and risk of dementia and Alzheimer’s disease in a population-based cohort study: the Rotterdam Study. Lancet, 1998, 351(9119): 1840–1843. 10.1016/sO140-6736(97)07541-7 [DOI] [PubMed] [Google Scholar]
  42. Patel SR, Malhotra ADJG, Gottlieb DJ, White DP, and Hu FB Correlates of long sleep duration. Sleep, 2006, 29: 881–889. 10.1093/sleep/29.7.881 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rivas- Marino G, Negin J, Salinas- Rodriguez A, Manrique- Espinoza B, Sterner KN, Snodgrass J, and Kowal P Prevalence of overweight and obesity in older Mexican adults and its association with physical activity and related factors: An analysis of the study on global ageing and adult health. Am. J. Hum. Biol, 2015, 27(3): 326–333. 10.1002/ajhb.22642 [DOI] [PubMed] [Google Scholar]
  44. Roberts RE, Shema SJ, Kaplan GA, and Strawbridge WJ Sleep complaints and depression in an aging cohort: a prospective perspective. Am. J. Psychiatry., 2000, 157(1): 81–88. 10.1176/ajp.157.E81 [DOI] [PubMed] [Google Scholar]
  45. Santoro SD, Giacheti CM, Rossi NF, Campos LMG, and Pinato L Correlations between behavior, memory, sleep-wake and melatonin in Williams-Beuren syndrome. Physiol. Behav, 2016, 159: 14–19. 10.1016/j.physbeh.2016.03.010 [DOI] [PubMed] [Google Scholar]
  46. Scarmeas N, Albert SM, Manly JJ, and Stem Y Education and rates of cognitive decline in incident Alzheimer’s disease. J. Neurol. Neurosurg. Psychiatry, 2006, 77: 308–331. 10.1016/sO197-4580(04)80374-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Schmutte T, Harris S, Levin R, Zweig R, Katz M, and Lipton R The relation between cognitive functioning and self-reported sleep complaints in nondemented older adults: results from the Bronx aging study. Behav. Sleep Med, 2007, 5(1): 39–56. 10.1207/sl5402010bsm0501_3 [DOI] [PubMed] [Google Scholar]
  48. Spira AP, Gonzalez CE, Venkatraman VK, Wu MN, Pacheco J, Simonsick EM, … and Resnick SM Sleep duration and subsequent cortical thinning in cognitively normal older adults. Sleep, 2016, 39(5): 1121–1128. 10.5665/sleep.5768 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Stenholm S, Kronholm E, Bandinelli S, Guralnik JM, andFerrucci L Self-Reported Sleep Duration and Time in Bed as Predictors of Physical Function Decline: Results from the InCHIANTI Study. Sleep, 2011, 34: 1583–1593. 10.5665/sleep.1402 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Tamakoshi A, and Ohno Y Self-reported sleep duration as a predictor of all-cause mortality: results from the JACC study, Japan. Sleep, 2004, 27: 51–54. 10.1093/sleep/27.1.51 [DOI] [PubMed] [Google Scholar]
  51. Tsapanou A, Gu Y, O’Shea DM, Yannakoulia M, Kosmidis M, Dardiotis E, … and Scarmeas N Sleep quality and duration in relation to memory in the elderly: initial results from the Hellenic Longitudinal Investigation of Aging and Diet. Neurobiol. Learn. Mem, 2017, 141: 217–225. 10.1016/j.nlm.2017.04.011 [DOI] [PubMed] [Google Scholar]
  52. UNESCO. International Standard Classification of Education 1997. [Internet, 1997], Available at: http://www.unesco.org/education/information/nfsunesco/doc/isced_1997.htm (Accessed March 16, 2018).
  53. Vancampfort D, Stubbs B, Lara E, Vandenbulcke M, Swinnen N, and Koyanagi A Mild cognitive impairment and physical activity in the general population: Findings from six low-and middle-income countries. Exp. Gerontol, 2017, 100: 100–105. 10.1016/j.exger.2017.10.028 [DOI] [PubMed] [Google Scholar]
  54. Virta JJ, Heikkilä K, Perola M, Koskenvuo M, Raiha I, Rinne JO, and Kaprio J Midlife sleep characteristics associated with late life cognitive function. Sleep, 2013, 36: 1533–1541. 10.5665/sleep.3052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Watson NL, Rosano C, Boudreau RM, Simonsick EM, Ferrucci L, Sutton-Tyrrell K, … and Harris TB Executive function, memory, and gait speed decline in well-functioning older adults. J. Gerontol. A Biol. Sci. Med. Sci, 2010, 65(10): 1093–1100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. World Health Organization (WHO). Global physical activity questionnaire. Geneva, Switzerland: [Internet, 2009], Available at: http://www.who.int/ncds/surveillance/steps/resources/GPAQ_Analysis_Guide.pdf (Accessed March 20, 2018). [Google Scholar]
  57. World Health Organization (WHO). Dementia. A public health priority. Geneva: Switzerland: [Internet, 2012], Available at: apps.who.int/iris/bitstream/10665/75263/1/9789241564458_eng.pdf (Accessed March 20, 2018). [Google Scholar]
  58. Xu L, Jiang CQ, Lam TH, Liu B, Jin YL, Zhu T, … and Thomas GN Short or long sleep duration is associated with memory impairment in older Chinese: the Guangzhou Biobank Cohort Study. Sleep, 2011, 34: 575–580. 10.1093/sleep/34.5.575 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Yoo SH, and Eckel-Mahan K Hippocampal PERI: A circadian sentinel controlling RSKy activity during memory formation. J. Neurochem, 2016, 138(5): 650–652. 10.1111/jnc.13727 [DOI] [PubMed] [Google Scholar]

RESOURCES