Maternal sleep quality during early pregnancy,risk factors and its impact on pregnancy outcomes: a prospective cohort study

ObjectivesWe aimed to assess the prevalence of poor sleep quality during early pregnancy and its risk factors, and to explore the association between sleep quality and adverse pregnancy outcomes.MethodsThis was a prospective birth cohort study that included 4352 pregnant women. Sleep quality were assessed using the Pittsburgh Sleep Quality Index (PSQI). The risk factors for poor sleep quality were analyzed by a logistic regression model. Log-binomial regression models were used to analyze the association between sleep quality and pregnancy outcomes.ResultsThe prevalence of maternal poor sleep quality during early pregnancy was 34.14%. The multivariate logistic model showed that stillbirth history (OR = 2.45; 95% CI: 1.34, 4.47), history of induced abortion (OR = 1.26; 95% CI: 1.07, 1.49), general health-related quality of life (OR = 3.98; 95% CI: 2.97, 5.34), insufficient physical activity (OR = 1.18; 95% CI: 1.03, 1.36), smoking (OR = 1.59; 95% CI: 1.18, 2.15), and vegetarian (OR = 2.18; 95% CI: 1.54, 3.08) were risk factors for poor sleep quality, while taking folic acid consistently before pregnancy (OR = 0.83; 95% CI: 0.72,0.97) was the protective factor. After controlling for all the confounders, poor sleep quality during early pregnancy increased the risk of premature rupture of membranes by 12% (95% CI: 1.00, 1.25).ConclusionPregnant women with a history of stillbirth and induced abortion, general health-related quality of life, insufficient physical activity, smoking, and a vegetarian diet tended to have poor sleep quality. More attention should be paid to healthy lifestyle of pregnant women to improve sleep quality and better pregnancy outcomes.     

Association between sleep duration during pregnancy and gestational diabetes mellitus: a meta-analysis

ObjectiveSleep is critical for glucose metabolism. Pregnant women often have sleep disturbances and extreme sleep duration. Investigations of the relationship between sleep duration during pregnancy and gestational diabetes mellitus (GDM) have reported inconsistent results. The present study aimed to meta-analyze the relationship between sleep duration during pregnancy and GDM risk.MethodsWe performed a systematic search of the PubMed, ISI Web of Science, and PsycINFO databases for studies that were published up to October 2017, that reported associations between sleep duration during pregnancy and GDM risk. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated as the effect sizes for all studies. Heterogeneity and potential publication biases were assessed.ResultsA total of 4366 papers were retrieved, among which seven studies assessed the relationship between sleep duration during pregnancy and GDM development. The seven articles included 18,203 subjects at baseline and 1294 GDM cases during follow-up. Compared to normal sleep duration, extreme sleep duration during early and middle pregnant stages had a close relationship with GDM based upon pooled data from prospective and cross-sectional studies. Prospective results showed that long sleep duration during pregnancy was a risk factor for GDM, but not short sleep duration. Publication biases were found when analyzing the relationship between extreme sleep duration and GDM.ConclusionsExtreme sleep duration during pregnancy is closely associated with GDM. Moreover, long but not short sleep duration can predict the risk of developing GDM. These findings remind us of the importance of sleep duration control during pregnancy and help optimize early strategies for the prevention of GDM.     

Chronic sleep loss during pregnancy as a determinant of stress: impact on pregnancy outcome

Short sleep duration, poor sleep quality, and insomnia frequently characterize sleep in pregnancy during all three trimesters. We aimed: (i) to review the clinical evidence of the association between conditions of sleep loss during pregnancy and adverse pregnancy outcomes; and (ii) to discuss the potential pathophysiological mechanisms that may be involved. A systematic search of cross-sectional, longitudinal studies using Medline, Embase, and PsychINFO, and MeSH headings and key words for conditions of sleep loss such as ‘insomnia’, ‘poor sleep quality’, ‘short sleep duration’, and ‘pregnancy outcome’ was made for papers published between January 1, 1960 and July 2013. Twenty studies met inclusion criteria for sleep loss and pregnancy outcome: seven studies on prenatal depression, three on gestational diabetes, three on hypertension, pre-eclampsia/eclampsia, six on length of labor/type of delivery, eight on preterm birth, and three on birth grow/birth weight. Two main results emerged: (i) conditions of chronic sleep loss are related to adverse pregnancy outcomes; and (ii) chronic sleep loss yields a stress-related hypothalamic–pituitary–adrenal axis and abnormal immune/inflammatory, reaction, which, in turn, influences pregnancy outcome negatively. Chronic sleep loss frequently characterizes sleep throughout the course of pregnancy and may contribute to adverse pregnancy outcomes. Common pathophysiological mechanisms emerged as being related to stress system activation. We propose that in accordance to the allostatic load hypothesis, chronic sleep loss during pregnancy may also be regarded as both a result of stress and a physiological stressor per se, leading to stress ‘overload’. It may account for adverse pregnancy outcomes and somatic and mental disorders in pregnancy.     

Associations between specific technologies and adolescent sleep quantity,sleep quality,and parasomnias

ObjectiveWe tested the hypothesis that weekday bedtime use of six technologies would be significantly associated with eight sleep parameters studied relating to sleep quantity, sleep quality, and parasomnias.MethodsIn our cross-sectional study, we previously administered validated age-appropriate questionnaires (School Sleep Habits Survey, Technology Use Questionnaire). Participating adolescents (n = 738; 54.5% boys) were aged 11–13 years and were from the Midlands region of the United Kingdom in 2010.ResultsFrequent use of all technology types was significantly inversely associated with weekday sleep duration (hours). Frequent music listeners and video gamers had significantly prolonged sleep onset (β = 7.03 [standard error {SE}, 2.66]; P < .01 and β = 6.17 [SE, 2.42]; P < .05, respectively). Frequent early awakening was significantly associated with frequent use of all technology types. The greatest effect was observed in frequent television viewers (odds ratio [OR], 4.05 [95% confidence interval {CI}, 2.06–7.98]). Difficulty falling asleep was significantly associated with frequent mobile telephone use, video gaming, and social networking, with music listeners demonstrating the greatest effect (OR, 2.85 [95%CI, 1.58–5.13]). Music listeners were at increased risk for frequent nightmares (OR, 2.02 [95%CI, 1.22–3.45]). Frequent use of all technologies except for music and mobile telephones was significantly associated with greater cognitive difficulty in shutting off. Frequent television viewers were almost four times more likely to report higher sleepwalking frequency (OR, 3.70 [95% CI, 1.89–7.27]).ConclusionsFrequent weekday technology use at bedtime was associated with significant adverse effects on multiple sleep parameters. If confirmed in other samples and longitudinally, improving sleep hygiene through better management of technology could enhance the health and well-being of adolescent populations.     

Comparison of longitudinal diary and actigraphy-assessed sleep in pregnant women

Study objectivesVarious methods are employed to assess sleep in pregnant women, including self-report, sleep diary, and actigraphy. Unfortunately, the data are often contradictory, and interpretations are often inconsistent. The current aims are to compare subjective and objective sleep data in pregnant women collected longitudinally in early pregnancy.MethodsIn this secondary analysis of 104 pregnant women, sleep was collected via diary and actigraphy for 14 days during three separate occasions (10–12 weeks; 14–16 weeks; and 18–20 weeks). Sleep variables included wake after sleep onset (WASO), sleep efficiency (SE), bedtime/lights out, sleep onset latency (SL), and total sleep duration (TST). Repeated measures ANOVAs compared each sleep variable across Time and by Method of data collection, while controlling for parity and daytime naps.ResultsSignificant differences were noted for only the method of data collection for sleep component studied for WASO, F (1, 98) = 147.20, p < 0.001; SE, F (1, 98) = 129.41, p < 0.001); bedtime/lights out, F (1, 103) = 5.33, p < 0.05); and sleep duration, F (1, 104) = 182.75, p < 0.001). Significant variation was not seen in any variable across time-period. Conclusions: There are substantial discrepancies between diary- and actigraphy-assessed sleep measures in pregnant women which is in alignment with previous literature. These data highlight that these methodologies assess different constructs. We contend that these data may be useful as a reference to compare high-risk women or those with sleep disorders. Using a multi-modal approach to identify sleep disturbance in pregnancy is likely a more clinically useful option.     

Polysomnographically determined sleep and body mass index in patients with insomnia

We assessed associations between polysomnographically determined sleep, especially the amount of slow-wave sleep (SWS), and body mass index (BMI) in patients with insomnia. One hundred and forty-one insomniacs and 55 healthy volunteers completed overnight polysomnographic recordings, and we measured height and body weight. No significant correlations were obtained between total sleep time and BMI among insomniacs. Compared with normal volunteers, insomnia patients exhibited longer sleep latency and shorter total sleep duration. While the two groups had no significant differences in BMI, insomniacs presented with more N1 but less time spend in SWS and rapid eye movement sleep (REMS). Based on their SWS time, we divided insomnia patients into three groups: short (26.99±13.88), intermediate (59.24±8.12), and long (102.21±26.17) SWS groups. The short-SWS group had significantly greater BMI than the long-SWS group. Further analyses with multiple linear regression showed a significant negative correlation between the amount of SWS and BMI scores in insomniacs, whereas no such correlation was found in healthy volunteers after controlling for potential confounds (e.g., age, sex and AHI). Our study suggests that low amounts of SWS may be associated with higher BMI in patients with insomnia.     

Cross-sectional comparison of various sleep disturbances among sex- and age-matched HIV-infected versus HIV-uninfected individuals in China

Objective/BackgroundThis study assessed the prevalence and correlates of various sleep disturbances in HIV-infected patients compared to sex- and age-frequency-matched HIV-uninfected controls in China.MethodsThis cross-sectional analysis included 1469 HIV-infected cases and 2938 HIV-uninfected controls. Insomnia symptoms, poor sleep quality (Pittsburgh Sleep Quality Index [PSQI] >5) as well as their specific domains, were assessed.ResultsPrevalence of insomnia symptoms, poor sleep quality, and long sleep duration were higher in HIV-infected vs uninfected participants (23.7% vs 19.8%, 24.1% vs 19.9%, and16.1% vs 8.7%, respectively; all p < 0.05), and remained significant after adjusting for age, sex and education. An Age-stratified analysis showed that such differences were significant only at ages 18–29 and 30–44 years for insomnia symptoms and poor sleep quality long sleep duration was significant across all age groups. Among HIV-infected patients, multivariate analysis indicated that older age, depressive symptoms and frailty score were the most consistent variables associated with sleep disorders (ie, insomnia symptoms, poor sleep quality, short and long sleep durations), as well as all associations (if significant) were positive, excluding the negative associations of older age and depressive symptoms with short sleep duration. Regarding HIV-specific factors, only current CD4 cell count ≥500 cells/μL was negatively associated with insomnia symptoms.ConclusionsThe impact of HIV infection on sleep disturbances may differ across age groups and are more pronounced among young adults. Additionally, the phenomenon of prolonged sleep duration among HIV-infected patients should be noted, and its link to poor physical health warrants further investigation.     

Reciprocal dynamics between self-rated sleep and symptoms of depression and anxiety in young adult women: a 14-day diary study

ObjectivesThe objective of this study was to characterize the day-to-day associations among sleep disturbance, depression, and anxiety in a sample of young adult women.MethodsOne hundred and seventy-one women (20.1 ± 3.3 years) completed in-laboratory baseline assessment followed by daily online surveys across a two-week period. Daily measures included the Mood and Anxiety Symptom Questionnaire-Short Form to assess shared and disorder-specific symptoms of depression and anxiety (general distress, anhedonic depression, and anxious arousal), as well as self-reported total sleep time (TST), sleep-onset latency (SOL), and sleep quality (SQ).ResultsFindings supported bidirectional day-to-day relationships between sleep and affective symptoms. When women felt greater general distress (shared features of anxiety and depression), they experienced longer SOL and worse SQ at night. Specificity among depression, anxiety, and sleep disturbance was observed such that higher levels of depression-specific anhedonia presaged longer SOL, shorter TST, and poorer SQ. In the other direction, when women had poor-quality sleep, they later experienced greater anhedonic depression and anxious arousal. The influence of TST on anhedonia was complex such that a single night of short sleep led to less anhedonic depression the next day, whereas women who obtained shorter sleep across the two-week period reported greater anhedonia.ConclusionsReciprocal dynamics between nightly sleep disturbance and daily experiences of depression and anxiety may serve as a process by which insomnia, depression, and anxiety develop into comorbid clinical states over time in women. The associations of anhedonic depression with nightly sleep disturbance and chronic short sleep were especially toxic, offering insight into daily mechanisms driving the most prevalent phenotype of comorbid insomnia.     

Insomnia symptoms increase during pregnancy,but no increase in sleepiness - Associations with symptoms of depression and anxiety

ObjectiveTo evaluate alteration in insomnia and sleepiness symptoms during pregnancy and assess early pregnancy risk factors for these symptoms, especially depressive and anxiety symptoms.MethodsA cohort of 1858 women was enrolled from the FinnBrain Birth Cohort Study. Insomnia and sleepiness symptoms were measured in early, mid- and late pregnancy with the Basic Nordic Sleep Questionnaire. Depressive symptoms were measured using the Edinburgh Postnatal Depression Scale and anxiety symptoms with the Symptom Checklist-90/Anxiety Scale. General linear models for repeated measures were conducted.ResultsGeneral sleep quality decreased (p < 0.001) and all insomnia types (p < 0.001) and sleep latencies (p < 0.001) increased as pregnancy proceeded. Snoring increased, but witnessed apneas remained rare. Nevertheless, morning (p = 0.019) and daytime (p < 0.001) sleepiness decreased from early to both mid-pregnancy and late pregnancy (p = 0.006 and p = 0.039). Women took more naps in early and late pregnancy compared to mid-pregnancy (both p < 0.001). Women with higher baseline anxiety symptoms had greater increase in sleep latency. At each pregnancy point, higher depressive and anxiety symptoms were associated with higher insomnia (p < 0.001) and sleepiness scores (p < 0.001) and higher depressive symptoms with longer sleep latencies (p < 0.001).ConclusionWe found a marked increase in insomnia symptoms throughout pregnancy. However, sleepiness symptoms did not increase correspondingly. Both depressive and anxiety symptoms in early pregnancy were associated with higher insomnia and sleepiness symptoms in later stages of pregnancy which emphasizes the importance of their assessment in early pregnancy.     

Trajectories of sleep problems from adolescence to adulthood. Linking two population-based studies from Norway

Study objectiveThe aim of the study is to assess stability and predictors of insomnia and sleep duration from adolescence to early adulthood.MethodsA longitudinal sample of 1257 individuals from three age cohorts were assessed from linked population-based studies, the youth@hordaland study from 2012 (age 16–18) and the SHoT2018 study (age 22–25). Identical measures of insomnia symptoms and sleep duration were analysed.ResultsThe stability of insomnia was high from adolescence to young adulthood, 50% of those with insomnia symptoms in adolescence still had insomnia symptoms six years later (adjusted IRR = 2.01; (CI 95%; 1.5–2.44)). Short sleep duration was also stable, with 67.8% of the adolescents in the lowest sleep duration quartile still remaining in the lowest quartile six years later. The overall rate of insomnia symptoms, long wake after sleep onset (WASO), and oversleeping increased from adolescence to young adulthood. Also, we observed a reduction in sleep efficiency and later rise times. There was no significant change in sleep onset latency (SOL).ConclusionInsomnia symptoms and short sleep duration are prevalent during both adolescence and young adulthood. Considerable individual stability and a rising rate of insomnia symptoms were observed over time. These findings underscore the importance of early identification and timely interventions to prevent chronic sleep problems.     

Sleep and cognitive performance: cross-sectional associations in the UK Biobank

ObjectiveThe relationship between insomnia symptoms and cognitive performance is unclear, particularly at the population level. We conducted the largest examination of this association to date through analysis of the UK Biobank, a large population-based sample of adults aged 40–69 years. We also sought to determine associations between cognitive performance and self-reported chronotype, sleep medication use and sleep duration.MethodsThis cross-sectional, population-based study involved 477,529 participants, comprising 133,314 patients with frequent insomnia symptoms (age: 57.4 ± 7.7 years; 62.1% female) and 344,215 controls without insomnia symptoms (age: 56.1 ± 8.2 years; 52.0% female). Cognitive performance was assessed by a touchscreen test battery probing reasoning, basic reaction time, numeric memory, visual memory, and prospective memory. Adjusted models included relevant demographic, clinical, and sleep variables.ResultsFrequent insomnia symptoms were associated with cognitive impairment in unadjusted models; however, these effects were reversed after full adjustment, leaving those with frequent insomnia symptoms showing statistically better cognitive performance over those without. Relative to intermediate chronotype, evening chronotype was associated with superior task performance, while morning chronotype was associated with the poorest performance. Sleep medication use and both long (>9 h) and short (<7 h) sleep durations were associated with impaired performance.ConclusionsOur results suggest that after adjustment for potential confounding variables, frequent insomnia symptoms may be associated with a small statistical advantage, which is unlikely to be clinically meaningful, on simple neurocognitive tasks. Further work is required to examine the mechanistic underpinnings of an apparent evening chronotype advantage in cognitive performance and the impairment associated with morning chronotype, sleep medication use, and sleep duration extremes.     

Objectively measured sleep duration and hyperglycemia in pregnancy

ObjectiveOur primary purpose was to assess the impact of objectively measured nighttime sleep duration on gestational glucose tolerance. We additionally examined associations of objectively measured daytime sleep duration and nap frequency on maternal glycemic control.MethodsSixty-three urban, low-income, pregnant women wore wrist actigraphs for an average of 6 full days in mid-pregnancy prior to screening for hyperglycemia using the 1-h oral glucose tolerance test (OGTT). Correlations of nighttime and daytime sleep durations with 1-h OGTT values were analyzed. Multivariable logistic regression was used to evaluate independent associations between sleep parameters and hyperglycemia, defined as 1-h OGTT values ⩾130 mg/dL.ResultsMean nighttime sleep duration was 6.9 ± 0.9 h which was inversely correlated with 1-h OGTT values (r = −0.28, P = .03). Shorter nighttime sleep was associated with hyperglycemia, even after controlling for age and body mass index (adjusted odds ratio [OR], 0.2 [95% confidence interval {CI}, 0.1–0.8]). There were no associations of daytime sleep duration and nap frequency with 1-h OGTT values or hyperglycemia.ConclusionsUsing objective measures of maternal sleep time, we found that women with shorter nighttime sleep durations had an increased risk for gestational hyperglycemia. Larger prospective studies are needed to confirm our negative daytime sleep findings.     

Sleep habits and sleep characteristics at age one year in the ELFE birth cohort study

ObjectiveInfant sleep plays a critical role in normal development. Sleep problems, including sleep onset difficulties (SODs) and night waking (NW), range from 20% to 30% in infants and young children and can be persistent over time up to adulthood. Young French children seem to have longer sleep durations and less sleep troubles than their counterparts worldwide. Here, we aimed at describing infant sleep characteristics (total sleep time (TST)/24 h, NW, and SODs) and associated sleep habits in infants at age one year from the French nationwide birth cohort Etude Longitudinale Française depuis l’Enfance (ELFE).MethodsThis study included 11,783 infants with information on both sleep characteristics and sleep habits (parental presence when falling asleep, eating to fall asleep, sucking a pacifier or finger to sleep and sleep arrangement and location). Associations were studied by multinomial logistic regression analyses adjusted for familial and infant characteristics.ResultsMean TST was 13 h36 min including 2 h54 min of naps; 20% of the infants had TST ≤12 h/24 h. About 46% did not present SOD or NW, 16% had frequent SODs and 22% had NW > 1 night in 2. Parental presence, feeding to fall asleep and infant sleep arrangements were frequent in infants with short sleep duration (≤12 h/24 h), NW and SODs. Non-nutritive sucking was associated with risk of NW, SOD and TST >14 h/24 h. Parental room sharing was associated with NW.ConclusionThis work provides new information on infant sleep arrangements and non-nutritive sucking that should be accounted for when considering sleep behaviors. In addition, most identified sleep habits associated with poor sleep characteristics may be amenable to change.     

Association of weekday-to-weekend sleep differences and stress response among a Japanese working population: a cross-sectional study

ObjectivesTo investigate the relationship between differences in weekday-to-weekend sleep habits and stress responses in a working population.MethodsThis cross-sectional study used data from university workers on sleep habits, differences in sleep duration between weekdays and weekends, and each midpoint of the sleep phase on weekdays and weekends. Social jetlag was defined as the difference in the midpoint of the sleep phase between weekdays and weekends. In addition, the Brief Job Stress Questionnaire assessed stress responses and stress-related factors. To examine sleep-related factors affecting stress responses, regression analysis was performed with adjustments for age, sex, and stress-related factors.ResultsAnalyzed were 2,739 participants. Sleep duration differences obtained by subtracting sleep duration on weekdays from that on weekends, social jetlag, and weekday sleep duration were significantly associated with an increased risk of stress responses in a univariate linear regression model. Adjusting for age, sex, job stressors, and stressor buffering factors did not change this trend. However, when additionally adjusting for all sleep parameters, only sleep duration differences and weekday sleep duration were significantly associated with stress responses (β 0.67 [95% CI 0.24, 1.10], p = 0.002), (−0.66 [-1.20, −0.13], p = 0.015).ConclusionsThis study provided further evidence that weekday sleep duration and weekday-to-weekend sleep duration differences were independently associated with stress responses even when considering stress-related factors. However, social jetlag was not clearly associated with stress responses. Our findings highlighted the necessity of securing sufficient sleep for stress management and mental health promotion in a working population.     

Short sleep and risk of proteinuria in 3-year-olds in Japan: a population-based cohort study

ObjectiveTo examine whether short sleep at 18 months of age is associated with risk of proteinuria at 3 years of age.MethodsA population-based cohort study using municipal health check-up data from 62,623 children born between March 1, 2004, and April 30, 2011, and followed to 3 years via questionnaire in Kobe City, Japan. Sleep duration was calculated from wake time and bedtime at the 18-month health check-up by standardized parent-reported questionnaires. Proteinuria, defined as dipstick urinary protein of ≥1+, was measured by dipstick urinalysis at 3 years. Multivariable logistic regression was used to generate adjusted olds ratios.ResultsIn total, 2253 (3.6%) children had short nighttime sleep (≤8 h) at 18 months. At 3 years, 765 (1.2%) children had proteinuria. The adjusted odds ratio for proteinuria among children with 8 h of sleep or less versus those with normal sleep of 10 h per night was 1.50 (95% confidence interval, 1.07–2.11; P = 0.02). Odds ratios of the same direction but weaker magnitude were observed for children with bedtime at 22:00 or later, and those with wake time at 6:00 or earlier.ConclusionsShort sleep at 18 months was associated with risk of proteinuria at 3 years. Future research is warranted to examine whether sleep programs targeting early childhood may have additional potential health benefits for kidney development.     

Epidemiology of sleep and sleep disorders in The Netherlands

ObjectiveThere is a surging public interest in The Netherlands concerning sleep, sleep disorders and associated health. For a proper perspective, it is necessary to have reliable information on the prevalence of sleep characteristics at the national level. This study set out to assess prevalence rates and key characteristics of sleep and sleep disorders in The Netherlands.MethodsIn 2012, a nationally representative sample of 2089 individuals, aged 18–70 years, responded to a set of 48 questions, including the Holland Sleep Disorders Questionnaire, a validated questionnaire based on the International Classification of Sleep Disorders.ResultsPrevalence rates were: 32.1% for a general sleep disturbance (GSD), 43.2% for insufficient sleep, 8.2 for insomnia, 5.3% for circadian rhythm sleep disorder, 6.1% for parasomnia, 5.9% for hypersomnolence, 12.5% for restless legs disorder and limb movements during sleep, 7.1% for sleep related breathing disorder, and 12.2% for the presence of comorbidity, ie, the presence of two or more concurrent sleep disorders. In addition, sleep onset time as well as sleep duration showed U-shaped relationships with GSD prevalence rates, with respectively the 22:00–24:00 period and seven to 8 h as optimal associates.ConclusionsSleep disorders and insufficient sleep have a high prevalence. As matter of concern, female adolescents reached the highest prevalence rates for most sleep disorders, insufficient sleep and daytime malfunctioning.     

Short sleep is associated with more depressive symptoms in a multi-ethnic cohort of older adults

ObjectivesTo evaluate cross-sectional and prospective associations between short and long sleep duration and depressive symptoms in older adults (aged >65 years).MethodsThe data from a subsample of the racially/ethnically diverse Northern Manhattan Study were analyzed. Depressive symptoms were assessed twice with the Center for Epidemiologic Studies Depression Scale (CES-D), approximately 5 years apart. The presence of depressive symptoms was defined as a CES-D score ≥16 or use of antidepressants. Self-reports of short (<6 h), intermediate (6–8 h) and long (≥9 h) sleep were assessed prior to the initial CES-D. Logistic regression was used to evaluate the cross-sectional associations between short and long sleep durations with depressive symptoms, using intermediate sleep as the reference. The prospective association between sleep duration and depression in a sample of participants without depressive symptoms at first CES-D was also analyzed. All models were adjusted for demographic, behavioral, and vascular risk factors.ResultsThe initial sample consisted of 1110 participants: 62% women, 69% Hispanic, 17% black, 14% white. Short sleep was reported by 25%, intermediate sleep by 65%, and long sleep by 9%. Depressive symptoms were described in 25% of the initial sample. Short sleep, but not long sleep, was associated with depressive symptoms at baseline (adjusted OR 1.8, 95% CI 1.3–2.6), and at follow-up (adjusted OR 1.9, 95% CI 1.1–3.5; median follow-up = 5.1 years).ConclusionShort sleep duration had a cross-sectional and prospective association with depressive symptoms in an urban multi-ethnic cohort of older adults.     

Sleep disturbances and dementia risk: A multicenter study

Introduction

Few longitudinal studies assessed whether sleep disturbances are associated with dementia risk.