Psychosomatic Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow An erratum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dew, M. A.
Right arrow Articles by Reynolds, C. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dew, M. A.
Right arrow Articles by Reynolds, C. F., III
Related Collections
Right arrow Other Epidemiology
Right arrow Sleep and Biological Rhythms
Psychosomatic Medicine 65:63-73 (2003)
© 2003 American Psychosomatic Society

ORIGINAL ARTICLE

Healthy Older Adults’ Sleep Predicts All-Cause Mortality at 4 to 19 Years of Follow-Up

Mary Amanda Dew, PhD, Carolyn C. Hoch, PhD, Daniel J. Buysse, MD, Timothy H. Monk, PhD, Amy E. Begley, MA, Patricia R. Houck, MSH, Martica Hall, PhD, David J. Kupfer, MD and Charles F. Reynolds, III, MD

From the Departments of Psychiatry (M.A.D., C.C.H., D.J.B., T.H.M., A.E.B., P.R.H., M.H., D.J.K., C.F.R.), Epidemiology (M.A.D.), and Psychology (M.A.D.), University of Pittsburgh School of Medicine and Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania.

Address reprint requests to: Dr. Dew, Department of Psychiatry, University of Pittsburgh School of Medicine, 3811 O’Hara St., Pittsburgh, PA 15213. Email: DewMA{at}msx.upmc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: Evidence concerning whether sleep disturbances in older adults predict mortality is mixed. However, data are limited to self-reported sleep problems and may be confounded with other comorbidities. We examined whether electroencephalographic (EEG) sleep parameters predicted survival time independently of known predictors of all-cause mortality.

METHODS: A total of 185 healthy older adults, primarily in their 60s through 80s, with no history of mental illness, sleep complaints, or current cognitive impairment, were enrolled in one of eight research protocols between October 1981 and February 1997 that included EEG sleep assessments. At follow-up (mean [SD] = 12.8 [3.7] years after baseline, range = 4.1–19.5), 66 individuals were positively ascertained as deceased and 118 remained alive (total N = 184).

RESULTS: Controlling for age, gender, and baseline medical burden, individuals with baseline sleep latencies greater than 30 minutes were at 2.14 times greater risk of death (p = .005, 95% CI = 1.25–3.66). Those with sleep efficiency less than 80% were at 1.93 times greater risk (p = .014, CI = 1.14–3.25). Individuals with rapid eye movement (REM) sleep percentages in the lowest 15% or highest 15% of the total sample’s distribution (percentage of REM <16.1 or >25.7) were at 1.71 times greater risk (p = .045, CI = 1.01–2.91). Percentage of slow-wave sleep was associated with time to death at the bivariate level, but not after controlling for potential confounders.

CONCLUSIONS: Older adults with specific EEG sleep characteristics have an excess risk of dying beyond that associated with age, gender, or medical burden. The findings suggest that interventions to optimize and protect older adults’ sleep initiation, continuity, and quality may be warranted.

Key Words: sleep • mortality • aging • gerontology.

Abbreviations: AHI = apnea-hypopnea index; CIRS = Cumulative Illness Rating Scale; CNS = central nervous system; EEG = electroencephalographic; HRSD = Hamilton Rating Scale for Depression; ICD-9 = International Classification of Diseases, 9th edition; MMSE = Mini Mental Status Examination; NDI = National Death Index; REM = rapid eye movement; SSDI = Social Security Death Index.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 
Complaints of insomnia, fragmented sleep, and poor quality sleep are more prevalent in older adults (age 55+) than in any other age group (1–10). Moreover, the public health burden associated with sleep disturbances in the elderly is significant: Studies have repeatedly linked these problems to greater use of health services (1, 11–14), increased use of hypnotics (9, 14–16), and reduced functional capabilities and quality of life (2, 3, 6, 17). Especially in older adults, sleep disturbances are frequently comorbid with physical and mental illness (6, 10, 13, 14, 18–21) and often increase the likelihood of nursing home placement for ill individuals (22, 23).

Of particular concern is that beyond these concurrent associations, some epidemiologic and community-based studies have reported that sleep complaints and sleep duration predict future physical health decline and all-cause mortality (5, 22, 24–28). For example, all-cause mortality rates over periods as long as 9 to 12 years have been observed to be almost doubled among individuals reporting sleep difficulties or aberrant sleep patterns at baseline (eg, 25, 28). However, other investigations have failed to find any evidence of these effects (29–32). The discrepancy in findings may be partially explained by differences between studies in the level of control for preexisting health status and for other known correlates of physical morbidity and mortality, such as depression level and cognitive status. Thus, the possibility remains that any linkages between sleep and subsequent health outcomes in older adults may be spurious, that is, explained by concurrent physical and psychological morbidities, which may themselves worsen sleep and lead to continued health changes and death.

An additional element clouding the interpretation of data on the potential predictive role of sleep disturbance on health outcome pertains to difficult construct validity issues in these community-based studies: The measures of sleep disturbance typically employed—single self-report items or questionnaires—may be better indicators of psychological distress than indicators of abnormalities or actual change in objective sleep parameters (eg, 33). Indeed, subjective reports of sleep disturbance generally bear low correlations with electroencephalographic (EEG) sleep measurements (34–36).

EEG sleep evaluation has, however, been almost exclusively limited to samples of individuals either presenting with a sleep complaint or physical illness or, most commonly, individuals with preexisting or concurrent psychiatric illnesses (eg, 37–43; for reviews, see 44 and 45). In such samples it is generally not possible to determine the unique contribution of sleep abnormalities, beyond the impact of these other factors, in predicting subsequent health outcomes and mortality. Although physically and mentally healthy samples of older adults are often included as control groups in these studies, the control group samples are usually too small to have adequate statistical power for examining EEG sleep effects on long-term health outcomes. Moreover, the control groups are often studied only cross-sectionally to provide a baseline for the patient groups of interest.

Because our center has conducted a variety of studies of EEG sleep in older adult patient groups, in which healthy control groups were included, as well as studies focused specifically on the EEG sleep characteristics of healthy older adults, we had an unusual opportunity to begin to move beyond some of the methodologic constraints noted above. We were able to assemble a relatively large cohort of community-residing older adults in whom we could evaluate the impact of EEG sleep on mortality over an average of over 12 years of follow-up, disentangled from other physical health and psychological parameters. Our primary research question concerned whether EEG sleep parameters contributed to the prediction of survival time over and above the contribution of other known predictors of all-cause mortality in older adults.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 
Subjects
Subjects were healthy older adults drawn from the total of eight National Institutes of Health–funded protocols involving healthy elders conducted at our center from October 1981 to February 1997 (36, 46–53). A total of 185 subjects were eligible for the present study, based on criteria described below. Of these, 184 (99.5%) could be determined positively as living or deceased at the time of follow-up (April 2001) and could thus be included in our analyses. Deaths were ascertained through three sources. First, death certificates were located for 52 persons through the National Death Index (NDI) (Centers for Disease Control and Prevention, National Center for Health Statistics, Hyattsville, MD). The NDI is a central, computerized index of death record information submitted by the vital statistics offices of all 50 states. The NDI currently has data on all deaths between 1979 and 1999. Following review and approval of our request for access, the NDI provided data that allowed us to positively identify deceased persons in our sample (based on full name, social security number, and other sociodemographic data that we had available on each subject), as well as the date and cause of death. The second source of information for determining deaths was the Social Security Death Index (SSDI) (Social Security Administration, Baltimore, MD), which we used to ascertain deaths between January 2000 and April 2001 (the last date for which SSDI data had been compiled). Because the SSDI is a public database, it does not provide information on death certificate number or cause of death. Thus, we were unable to ascertain the cause of death for the 14 additional deceased persons identified through the SSDI. The third source of information on vital status for subjects in our cohort was personal contact: For subjects not determined through the NDI or the SSDI to be deceased, we ascertained that all but one individual were alive through contact with subjects themselves or contact with immediate family members.

In sum, we identified 66 deaths and determined that 118 individuals were alive at the time of follow-up. The status of only one subject in the original group of 185 persons could not be definitively determined because we could not relocate him or any family members, and, although his social security number did not match any deaths in the NDI or SSDI, his name was relatively common and thus generated numerous possible matches based on it alone (ie, our concern was that if we had incorrectly recorded his social security number originally, then one of the possible matches could have been the correct match). Thus, rather than include him as a possible "false-negative" (coded as alive in our analyses, but actually deceased), we omitted him from our primary analyses. (However, all analyses described below were repeated with this individual included first as alive and then as deceased, and the findings were not changed in any respect from those reported in the remainder of this report.) In the total of 184 persons included in our analyses reported below, the average time to follow-up was 12.8 years (SD = 3.7, median = 12.2, interquartile range = 10.6.–16.5, full range = 4.1–19.5).

Subjects were required to meet almost identical mental and physical health inclusion criteria across the original eight protocols. All were community-residing older adults without complaints of insomnia, excessive daytime sleepiness, or behavioral disturbances related to sleep. They had no evidence of current or past psychiatric disorder, as determined by the Schedule for Affective Disorders and Schizophrenia–Lifetime Version (54). They were required to have scores indicating normal cognitive function on the Mini Mental Status Examination (MMSE; Ref. 55) (scores >=28 in four protocols, >=27 in two protocols, and >=26 in two protocols). They had no significant symptoms of depression based on the Hamilton Rating Scale for Depression (HRSD; Ref. 56) (scores <=7 in four protocols, <=6 in one protocol, and <=5 in three protocols). Physical examinations, electrocardiograms, and laboratory studies were used to screen out persons who had serious or uncontrolled physical health problems or who were taking medications that affected sleep or mood. However, individuals with stable medical conditions under a physician’s care (eg, heart disease, hypertension, thyroid disease, or arthritis) were accepted into all protocols. These individuals’ health conditions had little or no effect on activities of daily living and were not associated with sleep complaints. All subjects were required not to be taking any medications known to affect sleep or the circadian system. In all protocols subjects were required to abstain from use of alcohol for 2 weeks before sleep studies, and they were required not to nap on any of the days of the sleep studies.

The age criterion varied across protocols. Individuals were eligible if they were aged 55+ in one protocol, 60+ in four protocols, 65+ in one protocol, 70+ in one protocol, and 80+ in one protocol. There was no upper age limit for any protocol. In addition, the protocols differed in their requirement that subjects have an apnea-hypopnea index (AHI) of less than 10 (in the two oldest-age group protocols) vs. stipulating no exclusions based on AHI. Aside from the differences in minimum age and exclusions based on AHI, and the relatively minor differences in cutoff points on the MMSE and the HRSD, there were no sizable or significant sociodemographic differences among subjects recruited across the eight protocols.

Baseline demographic and clinical characteristics for the entire sample of 184 subjects, as well as for the subgroups alive and deceased at follow-up, are shown in Table 1. Underlying causes of death are also summarized, as determined from NDI-identified death certificates and coded by the National Center for Health Statistics according to ICD-9 criteria (58). Subjects deceased at follow-up were significantly older at baseline and more likely to be male. Despite the fact that by protocol design Cumulative Illness Rating Scale (CIRS; Ref. 59) scores indicated low levels of medical burden (with a maximum score of 16 out of a possible 63, median = 5, interquartile range = 3–7), subjects deceased at follow-up were more likely to have had higher baseline medical burden. (The CIRS is a validated and reliable clinician-rated measure of an individual’s medical status across 13 organ systems and considers the presence of diseases and disorders [eg, heart disease, hypertension, and diabetes] as well as symptoms and general health status.) None of the other variables differed across groups.1 Because of the relationships of age, gender, and medical burden to status at follow-up, these three factors were included as critical covariates in the statistical analyses below.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Demographic and Clinical Characteristics at Study Enrollment
 
Procedures
In all eight protocols, a standard procedure was followed for subject recruitment and for the collection of psychosocial, clinical, and polysomnographic information at what constitutes the baseline assessment in the present study design. Subjects were recruited through advertisements, word of mouth, and community presentations. We have previously been unable to identify any demographic, sleep, or health-related differences between subjects according to specific method of recruitment (60). After providing written informed consent, subjects were initially asked to keep a daily sleep/wake log (61) for the 14 consecutive days before the laboratory assessment. They then had either 2 or 3 consecutive nights of polysomnography (two of the protocols required only 2 nights) in private bedrooms of the Sleep and Chronobiology Center, Western Psychiatric Institute and Clinic. Bedtime and waketime for each subject was at his or her habitual time as determined by the 2-week sleep/wake log. Polysomnographic technologists used standard criteria for scoring sleep (62) with periodic checks of interrater reliability (63).

Polysomnographic Measures
Measures of sleep continuity and architecture were derived from data collected during nights 2 and 3 when available or during night 2 only for the two protocols that required only 2 nights of sleep. Night 1 was considered an adaptation night. (The two protocols that had the oldest minimum age for study inclusion (70+, 80+) had available only 2 nights of naturalistic sleep before additional experimental procedures were implemented on night 3. Key measures for the present analyses represent averages of nights 2 and 3 (when available) and include 1) total time spent asleep (minutes), 2) sleep latency (minutes), 3) sleep efficiency (percentage of total recording period spent asleep), 4) rapid eye movement (REM) latency (minutes), 5) REM sleep (percentage of time asleep spent in REM sleep), and 6) slow-wave sleep (percentage of total recording period spent in delta sleep).

These measures were selected because they have previously been found to be related to aging and to older adults’ psychosocial and health status (26, 33, 37, 64, 65) and because, unlike the AHI, none had been fixed by the design of any of our protocols. The six measures were not highly intercorrelated (absolute values of Pearson correlations, mean = 0.15, median = 0.14, interquartile range = 0.04–0.22). These correlations indicate that the measures tap only mildly overlapping elements of EEG sleep. Only one correlation was of moderate size (sleep efficiency and sleep latency, r = -0.41) (66).

Analyses
The sleep variables were either skewed (sleep latency, sleep efficiency, percentage of slow-wave sleep) or can not be interpreted unidirectionally (in the cases of time spent asleep, REM sleep, and REM latency, in which extreme scores in either direction can indicate impairment). For these reasons and to maximize clinical utility, we initially di- or trichotomized the sleep variables. When possible cutoff points were based on a previous analysis (36) that indicated clinically sensitive cutoff points for most of the variables. Sleep latency of greater than 30 minutes, sleep efficiency less than 80%, and percentage of slow-wave sleep less than the median value for the sample (median value = 1% slow-wave sleep) were considered to represent relative impairment. REM sleep and REM latency scores in either the upper 15% or in the lower 15% of the distributions were considered to be extreme values and to represent impairment (see Ref. 36 for further examination of these cutoff points in relation to health outcomes in older adults). Although time spent asleep is usually considered as abnormal when it is unusually brief (<6 hours) or unusually extended (>9 hours) (26, 67), no subjects had recorded times of greater than 8.3 hours. Therefore, we created a dichotomous variable of sleep duration <6 hours vs. 6+ hours.

We initially compared subjects alive vs. deceased at follow-up on the proportions scoring in the impaired range on each baseline measure of sleep using {chi}2 tests. To examine survival time as a function of subjects’ status on each sleep variable, we began with Kaplan-Meier analysis with log-rank (Mantel-Cox) tests for differences in the survival functions. Thus, for each sleep variable, we examined the zero-order relationship with survival time. Then, to determine whether each sleep variable affected survival time independent of critical covariates, we fit a Cox proportional hazards model. Thus, a total of six Cox models were fit, one for each of the six sleep variables (with each model including the age, gender, and medical burden covariates). The proportionality assumption of the Cox model was evaluated before its use (68). Individual as well as interactive effects of the variables were considered.

Before multivariate analyses, we evaluated the potential for multicollinearity between covariates and sleep variables. Medical burden was moderately correlated with age (r = 0.40), and female gender was moderately related to higher percentage of delta sleep (r = 0.40), but the size of these interrelationships was not high enough to pose difficulties for the analyses (68, 69). The remainder of intercorrelations between covariates and sleep variables were small (absolute values of r values, mean = 0.09, median = 0.09, interquartile range = 0.04–0.12).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 
Contingency table analyses indicated that deceased subjects at follow-up were significantly more likely to have longer sleep latencies, poorer sleep efficiency, and lower percentages of slow-wave sleep at baseline compared with subjects alive at follow-up (Table 2). Deceased subjects were marginally more likely to have extreme values of percentage of REM sleep (p < .06). (For each REM variable we also examined whether extremely low values vs. midrange values vs. values in the upper tail of the distribution were differentially associated with subjects’ status at follow-up. There was no indication that this was the case for either REM latency ({chi}2(2) = 3.61, p = .164) or REM percentage ({chi}2(2) = 2.96, p = .228).)


View this table:
[in this window]
[in a new window]
  		TABLE 2. Proportions of Study Groups With Each EEG Sleep Characteristic at Baseline
 
Kaplan-Meier survival analysis for each of the six sleep variables indicated that time to death was significantly shorter for older adults with longer sleep latency at baseline, poorer sleep efficiency, and extremely low or high percentage of REM; the survival curves as a function of each of these sleep parameters (and associated log-rank tests of these effects) are shown in Figure 1. Percentage of slow-wave sleep was only marginally associated with survival time (log-rank test = 3.67, p = .06), and time spent asleep and REM latency were not associated with survival time (log-rank tests = 1.39 and 2.93, respectively; p values > .08).



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1. Kaplan-Meier survival curves. a, Survival as a function of sleep latency (log-rank test = 9.63, p = .002). b, Survival as a function of sleep efficiency (log-rank test = 11.60, p = .0007). c, Survival as a function of percentage of REM sleep (log-rank test = 6.58, p = .010). df = 1 for all tests. Mean survival times (in weeks) were 614 and 752 for greater and lesser sleep latency, respectively; 624 and 754 for lesser and greater sleep efficiency, respectively; and 608 and 751 for extreme percentage of REM sleep and less extreme percentage of REM sleep. + = censored observation.

 
Most important, as shown in Table 3, the Cox proportional hazards model fit separately for each sleep variable—with each model controlling for age, gender, and medical burden—indicated that the impact of sleep latency, sleep efficiency, and percentage of REM sleep remained statistically reliable. The sizes of these variables’ effects—the hazard rate or relative risk—are given in the first column of Table 3 along with confidence intervals. Thus, although each covariate exerted its own effect on survival time, longer sleep latency increased the relative risk of death by 2.14, sleep efficiency increased the relative risk by 1.93, and extreme values of percentage of REM sleep increased the relative risk by 1.71. The effects of time spent asleep, REM latency, and percentage of slow-wave sleep were small and nonsignificant after the three covariates were controlled. We found no evidence (in terms of either statistical reliability or effect size) of any interaction effects between any sleep variable and the three covariates (results available from M. A. Dew); variables’ effects were additive.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Independent Associations of Baseline Sleep Characteristics With Mortality, Controlling for the Covariates of Age, Gender, and Medical Burden: Cox Regressiona
 
Because the effects were additive and the putative risk factors were not highly intercorrelated, it was appropriate to construct a final "dose-response" analysis to quantify the total incremental impact of the six predictors of survival time that emerged from our analyses: the three background variables (older age, male gender, higher medical burden), sleep latency >30 minutes, sleep efficiency <80%, and extreme values of REM sleep percentage. We created a "dose" variable indicating whether, at baseline, each subject possessed 0 factors, 1 to 2, 3 to 4, or 5 to 6 factors. A Kaplan-Meier analysis stratifying by dose was performed. Results indicated that none of the 18 subjects with 0 risk factors died during the follow-up period. In contrast, 53% of subjects with 1 to 2 factors were alive at follow-up (mean survival time, 15.7 years), and 0% of individuals with 3 to 4 factors or 5 to 6 factors were alive at follow-up (mean survival times, 11.6 and 6.4 years, respectively). The set of survival curves differed significantly (log rank {chi}2(3) = 89.26, p < .0001); significant pairwise differences (p < .03) between all of the dose groups were obtained. Descriptively, it is noteworthy that the sleep factors were well represented in various dose groups: 43% of subjects with 1 to 2 risk factors, 88% of subjects with 3 to 4 factors, and (by definition) 100% with 5 to 6 factors possessed at least 1 sleep-related risk factor.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 
The present study’s findings suggest that older adults with specific EEG sleep characteristics have an excess risk of dying beyond that associated with age, gender, or the presence of physical or mental illness. Sleep latencies of >30 minutes more than doubled the risk of mortality in our cohort of initially healthy elders across an average of almost 13 years of follow-up. Poor sleep efficiency, which served as a second measure of sleep continuity (and was only moderately associated with sleep latency in our sample), was also associated with an almost doubling of the risk for mortality.2 An unusually high or low percentage of REM sleep independently increased risk as well. We found no evidence that any of these sleep measures’ effects interacted or varied on the basis of the respondents’ baseline health status, age, or gender. Instead, the effects of the three sleep variables and these other baseline characteristics were additive: Cumulative survival rates were significantly and dramatically reduced as the total number of risk factors possessed by subjects increased.

These findings suggest that the inconsistently observed relationships between reports of sleep disturbances and health outcomes, including mortality, found in epidemiologic studies may indeed reflect true, predictive associations between sleep and these variables (70) rather than spurious relationships that are more parsimoniously explained by preexisting, known physical or mental health deficits. Our focus on EEG sleep parameters also leads to the conclusion that it is in fact sleep per se, and not merely the existence of high levels of psychological distress (as may be reflected by self-reported sleep complaints), that predicts mortality. Finally, in contrast to previous efforts, which at best have been able to statistically control for confounding variables, our focus on a pooled sample of older adults who were generally very physically healthy, had no history of mental illness, were cognitively intact, and reported no sleep/wake complaints at baseline allowed us to carefully examine the association of EEG sleep measures with subsequent mortality relatively independent of health factors known to affect mortality rates. However, despite the limited range of medical burden in the sample, this variable continued to be an important predictor of time to death. Its continuing role shows that even relatively mild health deficits merit careful consideration when evaluating individual elders’ risk profiles for more serious health outcomes.

The present findings of a unique EEG sleep-mortality relationship support the validity and clinical utility of conceptualizing sleep, particularly sleep initiation and continuity and REM sleep, as sensitive psychobiological markers of successful aging and adaptation in late life (37, 71–73)—markers sufficiently sensitive that they may serve as early indicators of decline that cannot be or is not easily detectable with standard physical or psychiatric examinations, history taking, and laboratory assessments. Although previous studies have shown that age-related health declines, or the older adult’s movement from a pattern of "successful" aging to "pathological" aging and deterioration, can be predicted by elements of EEG sleep (36, 47, 72), the present investigation is the first to link eventual mortality to EEG sleep measures of individuals who, on the basis of numerous criteria, were examples of successfully aging older adults.

A variety of biological mechanisms may be implicated in the specific linkages that we observed between sleep initiation/continuity, REM sleep processes, and mortality. Our study was not designed to examine these mechanisms; indeed, at this point we urge that replication of our findings in other cohorts is crucial to determine their generalizability. At a broad level, however, EEG sleep abnormalities in our sample could have reflected nonspecific effects of systemic medical burden on central nervous system (CNS) function. Relevant to this point is the fact that, although our analyses revealed sleep-mortality effects even after controlling for (relatively low) chronic medical burden, the majority of our subjects died from systemic medical illnesses rather than established CNS diseases. A linkage between sleep and mortality from systemic medical illness is consistent with other correlational evidence that sleep disturbances modulate changes in immune system activity and disease processes in both humans and animals (63, 65, 74).

Alternatively, EEG sleep problems related to difficulty initiating and maintaining sleep and to either too much or too little REM sleep may reflect specific changes in brain function that are themselves ultimately relevant to mortality. Sleep efficiency and latency reflect ease of transition from one operating state of the CNS to another and can serve as markers of homeostatic sleep drive (45, 51). This drive weakens considerably in late life, perhaps indicating an increased fragility or brittleness of the CNS due to degenerative changes, even in apparently healthy, cognitively intact elders (45, 71). More specifically, sleep onset is related not only to reduced neuronal activity in most cortical and brainstem regions, but also to increased activity among a set of neurons in the ventrolateral preoptic area of the hypothalamus (75, 76). Once initiated, sleep is maintained by hyperpolarization and rhythmic oscillations in thalamocortical circuits (77). It is possible that prolonged sleep onset or disrupted sleep maintenance (resulting in poorer overall sleep efficiency) are subtle indicators of dysfunction in these brain regions that could subsequently be associated with more substantial brain dysfunction and decreased longevity.

Similarly, although the neurobiology of REM sleep abnormalities is not fully understood, there is strong evidence that too much or too little REM sleep is selectively related to illnesses that are themselves known to ultimately increase mortality risk (33, 72). Low percentages of REM sleep are common in neurodegenerative diseases such as dementia of the Alzheimer’s type, whereas a high REM percentage is often associated with mood disorders (33, 37, 39, 40, 50, 52, 72). The reduction in REM sleep associated with dementia may reflect loss of cholinergic neurons, and the increase in REM sleep associated with mood disorders may reflect a functional increase in brainstem cholinergic activity (50, 72). In the present sample, no subjects initially presented with evidence of either a dementing or mood-related illness. Yet their REM data may have been indicators of subtle and/or very early deficits in these areas. It is noteworthy that we found no evidence of differential effects of low vs. high REM percentage; what mattered was not the direction but the extremity of REM percentage in predicting mortality. The availability of positron emission tomography, which permits examination of brain function during different operating states, may be useful in further elucidating the relationship of REM and non-REM sleep changes to mortality (78, 79).

Several limitations of our study must be borne in mind. First, our total sample size and the number of deaths observed were both relatively small. Thus, although we had adequate power (>80%) to examine all-cause mortality (66), our sample was not large enough to examine sleep parameters in relation to specific causes of death. There do not seem to have been an undue number of unnatural deaths (accidents and/or suicides), although causes of death ascertained from death certificates are always limited in their accuracy and completeness. Second, our focus on a healthy volunteer sample, whose sleep was assessed in a laboratory rather than the home environment, potentially limits the generalizability (external validity) of our findings. Yet in the context of previous work in this field, as discussed earlier, the fact that we studied a very healthy volunteer sample is a unique and important asset from the internal validity standpoint of disentangling the impact of factors such as medical comorbidity from the impact of sleep on mortality. Moreover, the collection of carefully controlled, laboratory EEG sleep data also increases internal validity (and at the time of most of our baseline sleep studies in the 1980s and early 1990s, ambulatory, home-based EEG sleep monitoring was not adequately reliable). In one of the few reports comparing laboratory and home polysomnographic recordings of sleep in healthy elderly control subjects (80), it seems that sleep efficiency was slightly higher, slow-wave sleep lesser, and REM latency longer in the laboratory setting (although formal analyses were not conducted on these differences). Ultimately, therefore, our findings will require replication in other cohorts studied in more naturalistic settings. The ongoing Sleep Heart Health Study (81) may provide a valuable opportunity for replication since it is currently following morbidity and mortality outcomes in a large community-based sample of middle-aged and older adults whose EEG sleep was initially assessed with a home polysomnogram.

An additional limitation of the present study is the fact that the subjects were originally recruited under eight different protocols that, although highly similar in many respects, were not originally conducted expressly for the purpose of fully addressing the relationship of sleep parameters to mortality. For example, we could not examine the role of sleep-disordered breathing because of AHI exclusion criteria set only in the protocols enrolling the oldest subjects (leading to a biased estimate of any association of AHI with age and, by extension, with mortality in the pooled sample). Sleep-disordered breathing has been found to be directly or indirectly related to mortality (24, 27, 82, 83). However, it is not clearly known whether there are joint or interactive effects of this problem with other EEG sleep abnormalities in increasing an individual’s likelihood of death. In addition, we could not consider the role of napping because potential subjects were excluded if excessive daytime sleepiness was present and because of a requirement in all of our protocols that subjects not nap during the days of the sleep studies. Because napping influences other sleep parameters such as sleep latency, and because daytime sleep has been found to be associated with mortality (27), its role requires continued consideration in future work. Finally, we could only partially explore the potential relationship of nighttime sleep duration to mortality (cf, Refs. 26 and 67) because none of the subjects in our sample had extremely long (>9 hours) sleep durations.

Despite these limitations, and even in the absence of a clear understanding of mechanism, the fact that certain sleep parameters independently predicted cumulative probability of death may have practical implications for sleep-related interventions in older adults. In particular, one could hypothesize that interventions that optimize and/or protect sleep initiation and sleep quality in old age might not only add quality to life but prolong life as well. Given replication of our findings in other cohorts, the design and evaluation of interventions appropriate for both healthy and impaired older adults warrant careful consideration.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
REFERENCES
 
This research was supported in part by Grants MH52247, MH52266, MH30915, MH37869, MH43832, MH00295, AG15136, AG15138, AG00972, and AG13396 from the National Institute of Mental Health and the National Institute on Aging, National Institutes of Health.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
ACKNOWLEDGMENTS
 REFERENCES
 
1 In particular, it is noteworthy that subjects’ date of enrollment into their respective study protocols was unrelated to status at follow-up, and additional correlational analyses indicated that it was unrelated to any EEG sleep characteristic (median r = .05) and that it did not affect the strength of the associations between EEG sleep measures and mortality. This indicates that there is little likelihood that the relationships of central interest in the present analyses differed as a function of whether subjects entered protocols earlier vs. later during the 15-year window of recruitment. Back

2 Sleep efficiency is determined in part by sleep latency. We chose to examine both of these variables, however, because each has previously been implicated in relation to health outcomes and each is a very commonly reported measure of sleep. In the present data set, when we corrected sleep efficiency for sleep latency (ie, we computed a standard index of sleep maintenance), this variable too was significantly related to mortality in a Cox model controlling for the covariates (hazard rate = 1.78, CI = 1.02–3.11, p = .043). Although sleep maintenance is a purer measure of sleep continuity than is sleep efficiency, and although it is less correlated with sleep latency (r = -0.22 in the present sample), sleep efficiency, rather than pure sleep maintenance, is by far the more frequently used measure in both empirical and clinical applications. Back

Received for publication October 1, 2001.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Bixler EO, Kales A, Soldatos CR, Kales JD, Healey S. Prevalence of sleep disorders in the Los Angeles metropolitan area. Am J Psychiatry 1979; 136: 1257–62.[Abstract/Free Full Text]
  2. Foley DJ, Monjan AA, Brown SL, Simonsick EM, Wallace RB, Blazer DJ. Sleep complaints among elderly persons: an epidemiologic study of three communities. Sleep 1995; 18: 425–32.[Medline]
  3. Foley DJ, Monjan A, Simonsick EM, Wallace RB, Blazer DJ. Incidence and remission of insomnia among elderly adults: an epidemiologic study of 6,800 persons over three years. Sleep 1999; 22 (Suppl 2): S366–72.
  4. Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders: an opportunity for prevention? JAMA 1989; 262: 1479–84.[Abstract/Free Full Text]
  5. Ganguli M, Reynolds CF III, Gilby JE. Prevalence and persistence of sleep complaints in a rural older community sample: the MoVIES project. J Am Geriatr Soc 1996; 44: 778–84.[Medline]
  6. Henderson S, Jorm AF, Scott LR, Mackinnon AJ, Christensen H, Korten AE. Insomnia in the elderly: its prevalence and correlates in the general population. Med J Aust 1995; 162: 22–4.[Medline]
  7. Hohagen F, Käppler C, Schramm E, Rink K, Weyerer S, Reimon D, Berger M. Prevalence of insomnia in elderly general practice attenders and the current treatment modalities. Acta Psychiatr Scand 1994; 90: 102–8.[Medline]
  8. Kronholm E, Hyyppä MT. Age-related sleep habits and retirement. Ann Clin Res 1985; 17: 257–64.[Medline]
  9. Partinen M. Epidemiology of sleep disorders. In: Kryger M, Roth T, Dement WC, editors. Principles and practice of sleep medicine. 2nd ed. Philadelphia: WB Saunders; 1994. p. 437–59.
  10. Roberts RE, Shema SJ, Kaplan GA. Prospective data on sleep complaints and associated risk factors in an older cohort. Psychosom Med 1999; 61: 188–96.[Abstract/Free Full Text]
  11. Angst J, Vollrath M, Koch R, Dobler-Mikola A. The Zurich study. VII. Insomnia: symptoms, classification and prevalence. Eur Arch Psychiatry Neurol Sci 1989; 238: 285–93.[CrossRef][Medline]
  12. Ohayon MM, Caulet M, Lemoine P. Comorbidity of mental and insomnia disorders in the general population. Compr Psychiatry 1998; 39: 185–97.[CrossRef][Medline]
  13. Weissman MM, Greenwald S, Nino-Murcia G, Dement WC. The morbidity of insomnia uncomplicated by psychiatric disorders. Gen Hosp Psychiatry 1997; 19: 245–50.[CrossRef][Medline]
  14. Weyerer S, Dilling H. Prevalence and treatment of insomnia in the community: results from the Upper Bavarian Field Study. Sleep 1991; 14: 392–8.[Medline]
  15. Balter MB, Uhlenhuth EH. New epidemiologic findings about insomnia and its treatment. J Clin Psychiatry 1992; 53 (Suppl): 34–9.
  16. Hohagen F, Rink K. Käppler C, Schramm E, Reimann D, Weyerer S, Berger M. Prevalence and treatment of insomnia in general practice: a longitudinal study. Eur Arch Psychiatry Clin Neurosci 1993; 242: 329–36.[CrossRef][Medline]
  17. Ohayon MM, Caulet M, Philip P, Guillerminault C, Priest RG. How sleep and mental disorders are related to complaints of daytime sleepiness. Arch Intern Med 1997; 157: 2645–52.[Abstract/Free Full Text]
  18. Gislason T, Reynisdóttir H, Kristbjarnarson H, Benediklsdóttir B. Sleep habits and sleep disturbances among the elderly—an epidemiological survey. J Intern Med 1993; 234: 31–9.[Medline]
  19. Hyyppä M, Kronholm E. Quality of sleep and chronic illnesses. J Clin Epidemiol 1989; 42: 633–8.[CrossRef][Medline]
  20. Klink ME, Quan SF, Kaltenborn WT, Lebowitz MD. Risk factors associated with complaints of insomnia in a general adult population: influence of previous complaints of insomnia. Arch Intern Med 1992; 152: 1634–7.[Abstract/Free Full Text]
  21. Rodin J, McAvay G, Timko C. A longitudinal study of depressed mood and sleep disturbances in elderly adults. J Gerontol 1988; 43: 45–53.
  22. Pollak CP, Perlick D, Linsner JP, Wenston J, Hsieh F. Sleep problems in the community elderly as predictors of death and nursing home placement. J Community Health 1990; 15: 123–35.[CrossRef][Medline]
  23. Rabins PV, Mace NL, Lucas MJ. The impact of dementia on the family. JAMA 1982; 248: 333–5.[Abstract/Free Full Text]
  24. Hays JC, Blazer DG, Foley DJ. Risk of napping: excessive daytime sleepiness and mortality in an older community population. J Am Geriatr Soc 1996; 44: 693–8.[Medline]
  25. Kojima M, Wakai K, Kawamura T, Tamakoski A, Aoki R, Lin Y, Makayama T, Horibe H, Aoki N, Ohna Y. Sleep patterns and total mortality: a 12-year follow-up study in Japan. J Epidemiol 2000; 10: 87–93.[Medline]
  26. Kripke DF, Simons RN, Garfinkel L, Hammond EC. Short and long sleep and sleeping pills: is increased mortality associated? Arch Gen Psychiatry 1979; 36: 103–16.[Abstract/Free Full Text]
  27. Newman AB, Spiekerman CF, Enright P, Lefkowitz D, Manolio T, Reynolds CF III, Robbins J. Daytime sleepiness predicts mortality and cardiovascular disease in older adults. The Cardiovascular Health Study Research Group. J Am Geriatr Soc 2000; 48: 115–23.[Medline]
  28. Wingard DL, Berkman LF. Mortality risk associated with sleeping patterns among adults. Sleep 1983; 6: 102–7.[Medline]
  29. Althuis MD, Fredman L, Langenberg PW, Magaziner J. The relationship between insomnia and mortality among community-dwelling older women. J Am Geriatr Soc 1998; 46: 1270–3.[Medline]
  30. Brabbins CJ, Dewey ME, Copeland JRM, Davidson IA, McWilliam C, Saunders P, Sharma VK, Sullivan C. Insomnia in the elderly: prevalence, gender differences and relationships with morbidity and mortality. Int J Geriatr Psychiatry 1993; 8: 473–80.
  31. Foley DJ, Monjan AA, Masaki KH, Enright PL, Quon SF, White LR. Associations of symptoms of sleep apnea with cardiovascular disease, cognitive impairment, and mortality among older Japanese-American men. J Am Geriatr Soc 1999; 47: 524–8.[Medline]
  32. Rumble R, Morgan K. Hypnotics, sleep, and mortality in elderly people. J Am Geriatr Soc 1992; 40: 787–91.[Medline]
  33. Bliwise DL. Sleep in normal aging and dementia. Sleep 1993; 16: 40–81.[Medline]
  34. Baker FC, Maloney S, Driver HS. A comparison of subjective estimates of sleep with objective polysomnographic data in healthy men and women. J Psychosom Res 1999; 47: 335–41.[CrossRef][Medline]
  35. Buysse DJ, Reynolds CF III, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989; 28: 193–213.[CrossRef][Medline]
  36. Dew MA, Reynolds CF III, Monk TH, Buysse, Hoch CC, Jennings R, Kupfer DJ. Psychosocial correlates and sequelae of electroencephalographic sleep in healthy elders. J Gerontol 1994; 49: P8–18.[Abstract/Free Full Text]
  37. Dew MA, Reynolds CF III, Buysse DJ, Houck PR, Hoch CC, Monk TH, Kupfer DJ. Electroencephalographic sleep profiles during depression: effects of episode duration and other clinical and psychosocial factors in older adults. Arch Gen Psychiatry 1996; 53: 148–56.[Abstract/Free Full Text]
  38. Gierz M, Campbell SS, Gillin JC. Sleep disturbance in various nonaffective psychiatric disorders. Psychiatr Clin North Am 1987; 10: 565–81.[Medline]
  39. Giles DE, Rush AJ, Roffwarg HP. Sleep parameters in bipolar I, bipolar II, and unipolar depressions. Biol Psychiatry 1986; 21: 1340–3.[CrossRef][Medline]
  40. Hoch CC, Reynolds CF III, Houck PR, Hall F, Berman SR, Buysse DJ, Dahl E, Kupfer DJ. Predicting mortality in mixed depression and dementia using EEG sleep variables. J Neuropsychiatry Clin Neurosci 1989; 1: 366–71.[Abstract/Free Full Text]
  41. Rosansky SJ, Johnson KL, Hutchinson C, Erdel S. Blood pressure changes during daytime sleep and comparison of daytime and nighttime sleep-related blood pressure changes in patients with chronic renal failure. J Am Soc Nephrol 1993; 4: 1172–7.[Abstract]
  42. Tandon R, Shipley JE, Taylor S, Greden JF, Eiser A, DeQuardo J, Goodsen J. Electroencephalographic sleep abnormalities in schizophrenia: relationship to positive/negative symptoms and prior neuroleptic treatment. Arch Gen Psychiatry 1992; 49: 185–94.[Abstract/Free Full Text]
  43. Vanable PA, Aikens JE, Tadimeti L, Caruana-Montaldo B, Mendelson WB. Sleep latency and duration estimates among sleep disorder patients: variability as a function of sleep disorder diagnosis, sleep history, and psychological characteristics. Sleep 2000; 23: 71–9.[Medline]
  44. Ancoli-Israel S. Insomnia in the elderly: a review for the primary care practitioner. Sleep 2000; 23 (Suppl 1): S23–30.
  45. Lustberg L, Reynolds CF III. Depression and insomnia: questions of cause and effect. Sleep Med Rev 2000; 4: 253–62.[CrossRef][Medline]
  46. Hoch CC, Dew MA, Reynolds CF III, Monk TH, Buysse DJ, Houck PR, Machen MA, Kupfer DJ. A longitudinal study of laboratory- and diary-based sleep measures in healthy "old old" and "young old" volunteers. Sleep 1994; 17: 489–96.[Medline]
  47. Hoch CC, Dew MA, Reynolds CF III, Buysse DJ, Nowell PD, Monk TH, Mazumdar S, Borland MD, Miewald J, Kupfer DJ. Longitudinal changes in diary- and laboratory-based sleep measures in healthy "old old" and "young old" subjects: a three-year follow-up. Sleep 1997; 20: 192–202.[Medline]
  48. Monk TH, Buysse DJ, Reynolds CF III, Kupfer DJ. Inducing jet lag in older people: adjusting to a 6-hour phase advance in routine. Exp Gerontol 1993; 28: 119–33.[CrossRef][Medline]
  49. Monk TH, Buysse DJ, Reynolds CF III, Kupfer DJ, Houck PR. Circadian temperature rhythms of older people. Exp Gerontol 1995; 30: 455–74.[CrossRef][Medline]
  50. Reynolds CF III, Kupfer DJ, Taska LS, Hoch CC, Spiker DG, Sewitch DE, Zimmer B, Marin RS, Nelson JP, Martin D. EEG sleep in elderly depressed, demented, and healthy subjects. Biol Psychiatry 1985; 1985: 20: 431–42.[CrossRef][Medline]
  51. Reynolds CF III, Kupfer DJ, Hoch CC, Stack JA, Houck PR, Berman SR. Sleep deprivation in healthy elderly men and women: effects on mood and on sleep during recovery. Sleep 1986; 9: 492–501.[Medline]
  52. Reynolds CF III, Buysse DJ, Kupfer DJ, Hoch CC, Houck PR, Matzzie J, George CJ. Rapid eye movement sleep deprivation as a probe in elderly subjects. Arch Gen Psychiatry 1990; 47: 1128–36.[Abstract/Free Full Text]
  53. Reynolds CF III, Monk TH, Hoch CC, Jennings JR, Buysse DJ, Houck PR, Jarrett DB, Kupfer DJ. Electroencephalographic sleep in the healthy "old old": a comparison with the "young old" in visually scored and automated measures. J Gerontol 1991; 46: M39–46.[Abstract/Free Full Text]
  54. Endicott J, Spitzer RL. A diagnostic interview: the Schedule for Affective Disorders and Schizophrenia. Arch Gen Psychiatry 1978; 35: 837–44.[Abstract/Free Full Text]
  55. Folstein MF, Folstein SE, McHugh PR. "Mini-Mental State": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–98.[CrossRef][Medline]
  56. Hamilton M. Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol 1967; 6: 278–96.[Medline]
  57. National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. Bethesda (MD): National Heart, Lung, and Blood Institute; 1998.
  58. World Health Organization. Manual of the international statistical classification of diseases, injuries, and causes of death. 9th ed. Geneva: World Health Organization; 1977.
  59. Miller MD, Paradis CF, Houck PR, Mazumdar S, Stack JA, Rifai AH, Mulsant B, Reynolds CF III. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res 1992; 41: 237–48.[CrossRef][Medline]
  60. Schlernitzauer M, Bierhals AJ, Geary MD, Prigerson HG, Stack JA, Miller MD, Pasternak RE, Reynolds CF III. Recruitment methods for intervention research in bereavement-related depression: five years’ experience. Am J Geriatr Psychiatry 1998; 6: 67–74.[Medline]
  61. Monk TH, Reynolds CF III, Kupfer DJ, Buysse DJ, Coble PA, Hayes AJ, Machen MA, Petrie SR, Ritenour AM. The Pittsburgh Sleep Diary (PsgSD). J Sleep Res 1994; 3: 111–20.
  62. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques, and scoring system for sleep stages of human subjects. Washington DC: Department of Health, Education, and Welfare; 1965. National Institutes of Health Publication No. 204.
  63. Campbell PI, Reynolds CF III, Berman SR, Hoch CC, Stack JA, Houck PR. Assessment of sleep fragmentation in the elderly: a reliability study. Sleep Res 1987; 16: 547.
  64. Hall M, Baum A, Buysse DJ, Prigerson HG, Kupfer DJ, Reynolds CF III. Sleep as a mediator of the stress-immune relationship. Psychosom Med 1998; 60: 48–51.[Abstract/Free Full Text]
  65. Irwin M, Smith TL, Gillin JC. Electroencephalographic sleep and natural killer activity in depressed patients and control subjects. Psychosom Med 1992; 54: 10–21.[Abstract/Free Full Text]
  66. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale (NJ): Erlbaum; 1988.
  67. Kripke DF, Ancoli-Israel S, Fell RL, Mason WJ, Klauber MR, Kaplan O. Health risk of insomnia. In: Peter JH, Penzel T, Podszus T, Von Wichert P, editors. Sleep and health risk. Berlin: Springer-Verlag; 1991. p. 547–54.
  68. Cox DR, Oakes D. Analysis of survival data. 2nd ed. London: Chapman & Hall; 1984.
  69. Tabachnick BG, Fidell LS. Using multivariate statistics. 4th ed. New York: HarperCollins; 2001.
  70. Kraemer HC, Kazdin AE, Offord DR, Kessler RC, Jensen PS, Kupfer DJ. Coming to terms with the terms of risk. Arch Gen Psychiatry 1997; 54: 337–43.[Abstract/Free Full Text]
  71. Prinz PN, Vitiello MV, Raskind MA, Thorpy MJ. Geriatrics: sleep disorders and aging. N Engl J Med 1990; 323: 520–6.[Medline]
  72. Reynolds CF III, Hoch CC, Buysse DJ, Monk TH, Houck PR, Kupfer DJ. REM sleep in successful, usual, and pathological aging: the Pittsburgh experience 1980–1993. J Sleep Res 1993; 2: 203–10.[Medline]
  73. Hoch CC, Reynolds CF III, Kupfer DJ, Houck PR, Berman SR, Stack JA. The superior sleep of healthy elderly nuns. Int J Aging Hum Dev 1987; 25: 1–9.[Medline]
  74. Krueger JM, Toth LA, Floyd R, Fang J, Kapos L, Bredow S, Obal F Jr. Sleep, microbes and cytokines. Neuroimmunomodulation 1994; 1: 100–9.[CrossRef][Medline]
  75. Sherin JE, Shiromani PJ, McCarley RW, Saper CB. Activation of ventrolateral preoptic neurons during sleep. Science 1996; 271: 216–9.[Abstract]
  76. Sherin JE, Elmquist JK, Torrealba F, Saper CB. Innervation of histaminergic tuberomammillary neurons by GABAergic and galaninergic neurons in the ventrolateral preoptic nucleus of the rat. J Neurosci 1998; 18: 4705–21.[Abstract/Free Full Text]
  77. Steriade M, Contreras D, Curro-Dossi R, Nunez A. The slow (<1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. J Neurosci 1993; 13: 3284–99.[Abstract]
  78. Nofzinger EA, Mintun MA, Wiseman M, Kupfer DJ, Moore RY. Forebrain activation in REM sleep: an FDG PET study. Brain Res 1997; 770: 192–201.[CrossRef][Medline]
  79. Nofzinger EA, Price JC, Meltzer CC, Buysse DJ, Villemagne VL, Miewald JM, Sembrat RC, Steppe DA, Kupfer DJ. Towards a neurobiology of dysfunctional arousal in depression: the relationship between beta EEG power and regional cerebral glucose metabolism during NREM sleep. Psychiatry Res 2000; 98: 71–91.[Medline]
  80. Edinger JD, Fins AI, Sullivan RJ, Marsh GR, Dailey DS, Hope TV, Young M, Shaw E, Carlson D, Vasilas D. Sleep in the laboratory and sleep at home: comparisons of older insomniacs and normal sleepers. Sleep 1997; 20: 1119–26.[Medline]
  81. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O’Connor GT, Rapoport DM, Redline S, Robbins J, Samet JM, Wahl PW. The Sleep Heart Health Study: design, rationale, and methods. Sleep 1997; 20: 1077–85.[Medline]
  82. Bliwise DL, Bliwise NG, Partinen M, Fell R, Stepnowsky C, Estline E, Khazini N, Chinn A. Sleep apnea and mortality in an aged cohort. Am J Public Health 1988; 78: 544–7.[Abstract/Free Full Text]
  83. Ancoli-Israel S, Kripke DF, Klauber MR, Fell R, Stepnowsky C, Estline E, Khazeni N, Chinn A. Morbidity, mortality and sleep-disordered breathing in community dwelling elderly. Sleep 1996; 19: 277–82.[Medline]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
C. F. Reynolds III, L. Serody, M. L. Okun, M. Hall, P. R. Houck, S. Patrick, J. Maurer, S. Bensasi, S. Mazumdar, B. Bell, et al.
Protecting Sleep, Promoting Health in Later Life: A Randomized Clinical Trial
Psychosom Med, February 1, 2010; 72(2): 178 - 186.
[Abstract] [Full Text] [PDF]

Home page
 AJPHHome page
O. M. Buxton, L. M. Quintiliani, M. H. Yang, C. B. Ebbeling, A. M. Stoddard, L. K. Pereira, and G. Sorensen
Association of Sleep Adequacy With More Healthful Food Choices and Positive Workplace Experiences Among Motor Freight Workers
Am J Public Health, November 1, 2009; 99(S3): S636 - S643.
[Abstract] [Full Text] [PDF]

Home page
 Eur Heart JHome page
A. Steptoe
Night shift work and the cardiovascular health of medical staff
Eur. Heart J., November 1, 2009; 30(21): 2560 - 2561.
[Full Text] [PDF]

Home page
 FocusHome page
K. Doghramji, R. Grewal, and D. Markov
Evaluation and Management of Insomnia in the Psychiatric Setting
Focus, October 1, 2009; 7(4): 441 - 454.
[Abstract] [Full Text] [PDF]

Home page
 FocusHome page
S. Ancoli-Israel and L. Ayalon
Diagnosis and Treatment of Sleep Disorders in Older Adults
Focus, January 1, 2009; 7(1): 98 - 105.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
W. M. Troxel, J. M. Cyranowski, M. Hall, E. Frank, and D. J. Buysse
Attachment Anxiety, Relationship Context, and Sleep in Women With Recurrent Major Depression
Psychosom Med, September 1, 2007; 69(7): 692 - 699.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
E. M. Friedman, G. D. Love, M. A. Rosenkranz, H. L. Urry, R. J. Davidson, B. H. Singer, and C. D. Ryff
Socioeconomic Status Predicts Objective and Subjective Sleep Quality in Aging Women
Psychosom Med, September 1, 2007; 69(7): 682 - 691.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
C. Cirelli
Sleep disruption, oxidative stress, and aging: New insights from fruit flies
PNAS, September 19, 2006; 103(38): 13901 - 13902.
[Full Text] [PDF]

Home page
 Arch Intern MedHome page
M. R. Irwin, M. Wang, C. O. Campomayor, A. Collado-Hidalgo, and S. Cole
Sleep deprivation and activation of morning levels of cellular and genomic markers of inflammation.
Arch Intern Med, September 18, 2006; 166(16): 1756 - 1762.
[Abstract] [Full Text] [PDF]

Home page
 CJASNHome page
M. L. Unruh, D. J. Buysse, M. A. Dew, I. V. Evans, A. W. Wu, N. E. Fink, N. R. Powe, K. B. Meyer, and for the Choices for Healthy Outcomes in Caring for
Sleep Quality and Its Correlates in the First Year of Dialysis
Clin. J. Am. Soc. Nephrol., July 1, 2006; 1(4): 802 - 810.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
E. M. Friedman, M. S. Hayney, G. D. Love, H. L. Urry, M. A. Rosenkranz, R. J. Davidson, B. H. Singer, and C. D. Ryff
Social relationships, sleep quality, and interleukin-6 in aging women
PNAS, December 20, 2005; 102(51): 18757 - 18762.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
S. J. Motivala, A. Sarfatti, L. Olmos, and M. R. Irwin
Inflammatory Markers and Sleep Disturbance in Major Depression
Psychosom Med, March 1, 2005; 67(2): 187 - 194.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
M. Ekstedt, T. Akerstedt, and M. Soderstrom
Microarousals During Sleep Are Associated With Increased Levels of Lipids, Cortisol, and Blood Pressure
Psychosom Med, November 1, 2004; 66(6): 925 - 931.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
M. Hall, R. Vasko, D. Buysse, H. Ombao, Q. Chen, J. D. Cashmere, D. Kupfer, and J. F. Thayer
Acute Stress Affects Heart Rate Variability During Sleep
Psychosom Med, January 1, 2004; 66(1): 56 - 62.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
D. F. Kripke
Sleep and Mortality
Psychosom Med, January 1, 2003; 65(1): 74 - 74.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow An erratum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dew, M. A.
Right arrow Articles by Reynolds, C. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dew, M. A.
Right arrow Articles by Reynolds, C. F., III
Related Collections
Right arrow Other Epidemiology
Right arrow Sleep and Biological Rhythms

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS