Self-reported Sleep Quality is Associated With the ... - Semantic Scholar

SLEEP QUALITY AND METABOLIC SYNDROME

Self-reported Sleep Quality is Associated With the Metabolic Syndrome J. Richard Jennings,PhD1,2; Matthew F. Muldoon, MD3; Martica Hall,PhD1,2; Daniel J. Buysse, MD1; Stephen B. Manuck,PhD2 Department of Psychiatry, 2Department of Psychology, 3Department of Medicine, University of Pittsburgh, Pittsburgh, PA

1

Pittsburgh Sleep Quality Index were related significantly to the presence of the metabolic syndrome—an increase of the global sleep score of 2.6 points (approximately 1 SD) was associated with an odds of having the metabolic syndrome of 1.44 (p = .04, confidence interval = 1.01-2.06). Linear-regression results showed that the Pittsburgh Sleep Quality Index global sleep-quality score was related significantly to waist circumference, body mass index, percentage of body fat, serum levels of insulin and glucose, and estimated insulin resistance. Conclusions: Self-reported global sleep quality is significantly related to the metabolic syndrome and several of its core components. Keywords: Metabolic syndrome, obesity, sleep quality, insulin resistance, Pittsburgh sleep quality index Citation: Jennings JR; Muldoon MF; Hall M et al. Self-reported sleep quality is associated with the metabolic syndrome. SLEEP 2007;30(2):219223.

Study Objectives: To determine whether a simple, structured self-report of overall sleep quality is associated with the presence of the metabolic syndrome and its component risk factors. Design: An observational, cross-sectional study comparing global scores on the Pittsburgh Sleep Quality Index with concurrently collected measures of the components of the metabolic syndrome and presence or absence of the syndrome. The metabolic syndrome criterion of the American Heart Association/National Heart, Blood, and Lung Institute was adopted. Setting: University laboratory. Patients/Participants: Two hundred ten volunteers with a mean age of 46 years (57% men) screened for the presence of serious illness and related medications. Interventions: N/A. Measurements and Results: All analyses were adjusted for sex and age. Logistic regression showed that poor global sleep-quality scores on the

heterogeneous sample of clinic patients, Vorona and colleagues20 found an inverse relationship between body mass index and reported hours of sleep. Experimental evidence suggests that short sleep duration plays a causal role in obesity. In Spiegel and Van Cauter’s pioneering studies,16,21-23 sleep restriction to 4 hours per night for only a single week in young, normal-weight men both increased body weight and induced endocrine and metabolic changes consistent with presence of the metabolic syndrome. Relatedly, patients with obstructive sleep apnea very often have the metabolic syndrome, an observation that may or may not be mediated by obesity or reduced sleep quality.24,25 Despite these associations, not all studies have found sleep disruption and metabolic disease to be related. For example, one recent report failed to find a relationship between sleep disturbance in midlife in women and subsequent incidence of diabetes.26 Given the public-health significance of the metabolic syndrome and the evidence of its relationship with sleep, it is important to know whether normative variation in reported sleep quality is associated with the metabolic syndrome. The Pittsburgh Sleep Quality Index (PSQI) is a widely used measure of sleep quality that is well validated, reliable, and readily completed by most individuals.27,28 The PSQI provides an overall index of sleep quality, includes indexes for a number of components of sleep quality, and is suitable for large-scale epidemiologic investigations. Here we report available data on 210 individuals from a sample of community volunteers who completed the PSQI as part of an investigation of cardiovascular risk12 and its sources.

INTRODUCTION OBESITY IS A GROWING EPIDEMIC IN MUCH OF THE WORLD.1,2 AS RECOGNIZED BY REAVEN AND COLLEAGUES, 3 OBESITY TENDS TO CLUSTER WITH OTHER factors to form the “metabolic syndrome” that has been found to predict Type 2 diabetes mellitus, and relate as well to morbidity and mortality from cardiovascular disease, cancer, and arthritis.2,46 The metabolic syndrome is defined by a coexistence of several diagnostic factors.7-9 These factors include obesity (particularly abdominal adipose tissue), blood pressure elevation, high fasting serum concentrations of triglycerides and glucose, and low serum high-density lipoprotein cholesterol. Increased insulin levels and insulin resistance are defined by some as part of the metabolic syndrome but are considered by others as an underlying cause for the syndrome.1,3,10,11 When initially identified, the clustering of these risk factors was thought to be induced by dysregulation of insulin.3 However, the pathophysiologic basis of the metabolic syndrome remains in doubt;12,13 indeed, some have questioned the clinical value of identifying the syndrome despite the associations between the syndrome and disease incidence.11 Short sleep duration is associated with, and may contribute to, obesity and the metabolic syndrome.14-19 For example, in a large Disclosure Statement This was not an industry supported study. Dr. Buysse is a consultant for Actelion, Cephalon, Eli Lilly, GlaxoSmithKline, Merck, Neurocrine, Neurogen, Pfizer, Respironics, Sanofi-Aventis, Servier, Sepracor, and Takeda. Drs. Jennings, Muldoon, Hall, and Manuck have indicated no financial conflicts of interest.

METHODS Subjects Subjects were participants in the University of Pittsburgh’s Adult and Human Behavior Project and were recruited from Allegheny County, Pennsylvania, via mailed brochures. All were community volunteers, 30 to 54 years of age. Exclusion criteria included clinical history of atherosclerotic disease, cancer diag-

Submitted for publication June 1, 2006 Accepted for publication October 17, 2006 Address correspondence to: Dr. J. R. Jennings, E1329 WPIC, 3811 O’Hara St., Pittsburgh, PA 15213; Tel: (412) 246-6220; Fax: (412) 246-6210; E-mail [email protected] SLEEP, Vol. 30, No. 2, 2007

219

Sleep and Metabolic Syndrome—Jennings et al

nosis or treatment within the past year, chronic liver or kidney disease, as well as use of insulin, weight-loss, or psychotropic medications. The University of Pittsburgh Institutional Review Board approved the protocol, and subjects gave informed consent. A total of 231 participants completed the PSQI. Fifteen were excluded because of current use of statin medications (which precluded determination of metabolic syndrome criterion). An additional 6 individuals were excluded because of missing fasting blood-sample results. This left data from 210 participants for analysis.

Table 1— Description of the 210 Subjects Comprising the Sample Variable Age, y Men Caucasian PSQI, global score Blood Pressure, mm Hg Systolic Diastolic Glucose, mg/dL Triglycerides, mg/dL High density lipoprotein, mg/dL Body mass index, kg/m2 Fat, % of total body weight Waist circumference, cm Insulin, μIU/mL HOMA Beck Depression Inventory, score Presence of metabolic syndrome, no. Taking medications, no. Antihypertensive Dyslipidemic Oral hypoglycemic

Risk-Factor Assessments Subjects arrived at the project office between 7:30 am and 10:30 am after a 12-hour overnight fast. After subjects rested in the seated position for at least 10 minutes, a trained staff member obtained 2 blood pressure measurements from the right arm using a mercury sphygmomanometer and a regular, large, or extra large adult cuff, according to the subject’s arm circumference. The 2 readings were averaged. Then, phlebotomy was performed along with measurement of height, weight, waist circumference at the umbilicus, and lean body mass (bioelectrical impedance body composition analyzer, Tanita Corp. of America, Inc., Arlington Heights, IL). Determinations of standard serum lipids, glucose, and insulin were performed by the Heinz Nutrition Laboratory, University of Pittsburgh Graduate School of Public Health, as previously described.29 Insulin resistance was estimated from the homeostasis model assessment (HOMA-IR) = serum insulin (µIU/mL) × fasting blood glucose (mmol/L)/22.5.30 The metabolic syndrome was defined by the criterionof the American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement.7,9 The metabolic syndrome is considered to be present in any individual meeting at least 3 or more of the following 5 criteria: blood pressure ≥ 130 mm Hg systolic/85 mm Hg diastolic or taking antihypertensive medication, fasting glucose ≥ 100 mg/dL or taking diabetic medication, waist circumference > 102 cm for men or 88 cm for women, triglycerides ≥ 150 mg/dL or use of dyslipidemic medication, or high-density lipoprotein cholesterol < 40 mg/dL in men and 50 mg/dL in women or use of dyslipidemic medication.

114.3 ± 12.4 78.7 ± 9.4 97.9 ± 22.8 119.8 ± 86.0 56.1 ± 16.1 26.6 ± 5.2 27.6 ± 8.3 90.6 ± 16.0 13.2 ± 6.8 3.3 ± 2.2 3.1 ± 3.5 41 (20) 13 (6) 2 (1) 4 (2)

Data are presented as mean ± SD or number (%). HOMA refers to homeostasis model assessment, a measure of insulin resistance.

ues = .37-.73, p < .001). For this reason and to protect against experiment-wise error rate inflation, we focused solely on the global sleep-quality score. A logistic-regression analysis was performed using Statistica (6.1, StatSoft, Tulsa, OK), with presence or absence of the metabolic syndrome as the dependent variable and the PSQI global sleep-quality score as the predictor variable, controlling for sex and age. The same predictors were then used in a multiple-regression analysis relating global sleep quality to each of the metabolic-syndrome component risk factors, insulin levels, the HOMA index of insulin resistance,30 and fat as a percentage of body weight. Follow-up analyses assessed the robustness of the association of the PSQI global score and the metabolic syndrome when controlling for depression, smoking history, alcohol use, years of education, and use of hypertensive medication. RESULTS

Sleep-Quality Assessment

Table 1 describes the characteristics of the sample. The average global sleep-quality score indicated slightly poorer sleep quality than the average of the healthy controls used in developing the PSQI,1 and 30% of the current sample scored in the poor-sleeper range (global sleep-quality index > 5). Blood pressure and serum indexes were reasonably typical for healthy individuals in this age range. Forty-one (20%) of the participants met the criterion for the metabolic syndrome. The PSQI global sleep-quality score was associated significantly with the presence of the metabolic syndrome (Table 2). After adjusting for the independent effects of sex and age on the metabolic syndrome, for every increase of 2.6 points on the PSQI, an individual was 1.44 times more likely to meet the criterion for the metabolic syndrome. The strength of association between the PSQI and the metabolic syndrome (p = .02) was essentially the same in a more complex model that added smoking history, alcohol use, and years of education into the logistic model. Due

After consuming a snack, subjects completed several psychometric instruments that included the PSQI. The PSQI has been widely used and has good psychometric properties. The instrument, its reliability, and its validity are presented in Buysse et al.27,28 Nineteen items generate a global sleep-quality score, as well as scores on 7 components of sleep quality: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. A global sleep-quality score of greater than 5 is recommended as defining “poor sleepers.” Analysis Because glucose, triglycerides, insulin, and HOMA data were positively skewed, logarithmic transformations were conducted. Preliminary analyses revealed that the PSQI subscales all showed a substantial correlation with the global sleep-quality score (r valSLEEP, Vol. 30, No. 2, 2007

45.8 ± 6.0 119 (57) 200 (95) 4.6 ± 2.6

220


Table 2—Results of Logistic Regression Relating the Metabolic Syndrome to the PSQI Global Sleep Quality Index and Control Variables

Table 3—Multiple Regression Resultsa Relating Aspects of the Metabolic Syndrome to the Global Pittsburgh Sleep-Quality Index in 210 Subjects

Variable Odds Ratioa 95% CI Significanceb Age, y 1.13 1.0 to 1.22 .002 Sex 2.19 1.48 to 3.36 < .001 PSQI global sleep quality 1.44 1.01 to 2.06 .04

Variable Blood Pressure Systolic Diastolic Glucose Triglyceride HDL BMI Fat percentage Waist Circumference Insulin Insulin resistanceb

CI refers to confidence interval. a Odds ratio is the difference in odds of meeting the criterion for metabolic syndrome as associated with a variation of 1 SD in the Pittsburgh Sleep Quality Index (PSQI) (SD = 2.6) or with 1 year of age and male sex more likely to exhibit the metabolic syndrome. b Wald statistic.

to the possibility that depressed affect impacted reports of sleep quality, a separate model was run that included scores on the Beck Depression Inventory,31 as well as age and sex as covariates. In this analysis, depression did not contribute significantly to odds for meeting the criterion for the metabolic syndrome, but the statistical strength of the association between the PSQI global sleepquality score and the metabolic syndrome was decreased to p = .12. The relationship between the PSQI global score and individual features of the metabolic syndrome was examined with multiple regressions relating each factor to the PSQI global sleep-quality score, controlling for age and sex. As shown in Table 3, global sleep quality was related significantly to the 3 measures of obesity (body mass index, waist circumference, and fat percentage), and the 3 measures of insulin resistance (fasting insulin concentration, fasting glucose concentration, and estimated insulin resistance). Global sleep quality was unrelated to blood pressure or lipid measures.

PSQI Global Sleep Quality Index SE t p value

.04 .07 .14 .04 -.09 .19 .19 .12 .14 .17

.06 .06 .07 .07 .06 .07 .06 .06 .07 .07

.67 1.08 2.13 .66 -1.48 2.88 3.24 2.11 2.07 2.54

NS NS .03 NS NS .004 .001 .04 .04 .01

PSQI refers to Pittsburgh Sleep Quality Index; HDL, high-density lipoprotein; BMI, body mass index. See Table 1 for measurement units. a Controlled for sex and age. b Estimated from the homeostasis model of assessment, a measure of insulin resistance.

has also been related to the metabolic syndrome.25 In the current report, we examined the relationship of sleep quality with the risk factors comprising the metabolic syndrome and also insulin, estimated insulin resistance, body mass index, and percentage of body fat. The PSQI global sleep-quality score correlated positively with those metabolic syndrome components related to adiposity—waist circumference, body mass index, and percentage of body fat. This suggests that sleep quality may influence the metabolic syndrome through its relationship with degree of obesity. Significant relationships between sleep quality and glucose, fasting insulin levels, and insulin resistance were also identified. The initial conceptualization of the metabolic syndrome3 posited insulin regulation as a key underlying factor. Although it is now known that not all individuals with the metabolic syndrome are also insulin resistant,11 our results relate interestingly to a hypothesis relating sleep restriction to insulin resistance. The human experimental work on sleep restriction and metabolic factors11,16,21-23 has been interpreted to suggest that inadequate sleep time relates to increased appetite and subsequently to insulin resistance. These experiments were in normal-weight volunteers and, therefore, suggest that sleep time may be an initiating factor influencing body weight and the metabolic syndrome. If our current nonclinical sample suffers from mild disruptions of sleep similar to those induced in the experimental volunteers, then the current results may be consistent with this hypothetical mechanism. Note, however, that the PSQI is not directed at the subjective effects of experimental sleep restriction and that our participants were not all normal weight but, instead, show a substantial range of body mass indexes. Our results may be readily interpreted as suggesting that increased body mass alters both sleep quality and metabolic function. Our results are limited by a several factors. As noted above, our sample was largely Caucasian and extension to racially diverse groups would be required to establish generalizability to the overall population. The sample is cross-sectional, so we cannot judge the direction of influence between our variables, eg, sleep quality and the metabolic syndrome or body mass and sleep qual-

DISCUSSION In our analyses of data from middle-aged, generally healthy, Caucasian adults, poor global sleep quality, as measured by the PSQI, was related significantly to the metabolic syndrome. In this sample, the prevalence of the metabolic syndrome was 20%, whereas Ford1 found a prevalence of 34.5% for the metabolic syndrome among the 3501 men and women older than 20 years of age in the National Health and Nutrition Examination Survey (19992002). Therefore, the sample of community volunteers in the current study had somewhat more-favorable average values than national estimates of adult body composition and metabolic indexes. PSQI scores were well within a standard deviation of the control sample used to develop the scale27 but with a slightly greater mean score for global sleep quality. Thus, the descriptive results suggest that our sample is reasonably typical and that the association between PSQI reports and the metabolic syndrome may generalize to middle-aged, reasonably healthy, Caucasian individuals. Our sample of African Americans was too small to justify generalization to this group and too small to separately analyze. Notably, virtually all reported results were marginally stronger when these participants were excluded from the analyses. The current report complements other reports on sleep and facets of the metabolic syndrome. Prior epidemiologic reports have related amount of sleep time to obesity,14-17,19,20 hypertension,32 and to incident diabetes (in women).18 Obstructive sleep apnea SLEEP, Vol. 30, No. 2, 2007

β

221


the metabolic syndrome and each to self-reports of sleep quality.

ity. The reported strength of the relationship between global sleep quality and the metabolic syndrome may be conservative, since individuals with clinical atherosclerotic disease and those taking insulin or statins were excluded.31 Self-reports of sleep are valuable but do not provide the direct measures of sleep that electroencephalographic studies provide. Self-reports have value as a measure of sleep in usual circumstances and as an overall index rather than an index solely for particular nights using polysomnography. Self-reports are, however, subject to recall biases and do not provide measures of sleep architecture. We cannot determine the precise basis for the reports of poor sleep quality without physiologic studies. For example, obstructive sleep apnea is associated with the metabolic syndrome.25 An important issue is whether the current result predominantly reflects a relationship between sleep apnea and the metabolic syndrome. Three observations question this interpretation. First, based on available statistics from the general population in the appropriate age range, 6% of the participants could be expected to be diagnosed with obstructive sleep apnea.33 This prevalence compares to a prevalence of 20% for the metabolic syndrome in our sample. Thus, in the unlikely case that all participants with sleep apnea exhibited the metabolic syndrome,25 sleep apnea would not predict all cases of the metabolic syndrome. Second, Coughlin et al25 found, in regression analyses, that obstructive sleep apnea is significantly or marginally related to the following components of the metabolic syndrome: systolic and diastolic blood pressure, insulin level, HOMA, and triglyceride and high-density lipoprotein cholesterol levels. The PSQI results shared with sleep apnea only an association with the insulin and HOMA values. In contrast to the relationship between sleep apnea and components of the metabolic syndrome, PSQI global sleep quality also showed significant relationships with body mass index, waist circumference, fat percentage, and glucose. Finally, the PSQI global sleep-quality index is not closely associated with obstructive sleep apnea. In a polygraphic study of 435 patients, 58% of whom had sleep apnea, the PSQI global sleep index failed to significantly distinguish between those with and without apnea.33 In a smaller sample of survivors of sexual assault (n = 151) clinically diagnosed sleep-disordered breathing was correlated .24 with PSQI global sleep quality.34 In short, the relative prevalence of obstructive sleep apnea, the pattern of its relationship to the metabolic syndrome, and the weak relationships between sleep apnea and the PSQI in previous reports suggest that obstructive sleep apnea is only likely to explain a small portion of the relationship that we have observed between global sleep quality and the metabolic syndrome. Nonetheless, polysomnography studies that identify sleep apnea and relate it to both self-reported sleep and the metabolic syndrome are necessary to evaluate our inferences from the literature and gain further insight into the relationship of physiologically assessed sleep and the metabolic syndrome. In summary, the current investigation found that poor overall sleep quality is associated with increasing prevalence of the metabolic syndrome and several of its component risk factors. The practical significance of our finding is that a simple self-report instrument on sleep quality can alert individuals and their physicians to additional possible risks for the metabolic syndrome or, more generally, a set of risk factors for cardiovascular disease and diabetes. Future studies should characterize general-population samples with respect to specific types of sleep disturbances (particularly, obstructive sleep apnea) and the relationships of each to SLEEP, Vol. 30, No. 2, 2007

ACKNOWLEDGMENTS We gratefully acknowledge the support of National Institute of Health grants P01-HL-40962 and M01-RR-000056. REFERENCES 1. 2. 3. 4. 5. 6. 7.

8.

9.

10. 11.

12.

13.

14. 15. 16. 17. 18.

222

Ford ES. Prevalence of the metabolic syndrome defined by the International Diabetes Federation among adults in the U.S. Diabetes Care 2005;28:2745-9. Laaksonen DE, Niskanen L, Lakka HM, Lakka TA, Uusitupa M. Epidemiology and treatment of the metabolic syndrome. Ann Med 2004;36:332-46. Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes 1988;37:1595-607. Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardiovascular disease mortality in middleaged men. JAMA 2002;288:2709-16. Ford ES. Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence. Diabetes Care 2005;28:1769-78. Aguilar-Salinas CA, Rojas R, Gomez-Perez FJ, et al. High prevalence of metabolic syndrome in Mexico. Arch Med Res 2004;35:7681. Executive summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285:2486-97. Federation ID. IDF world-wide definition of the metabolic syndrome. Available at: www.idf.org/home/index.cfm?unode=112007EAACE-41D2-9FA0-BAB6E25BA072. Accessed on: October 1,2006 Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005;112:2735-52. Alberti KG, Zimmet P, Shaw J, Group IDFETFC. The metabolic syndrome—a new worldwide definition. Lancet 2005;366:105962. Kahn R, Buse J, Ferrannini E, Stern M. The metabolic syndrome: time for a critical appraisal. Joint statement from the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia 2005;48:1684-99. Muldoon MF, Mackey RH, Korytkowski MT, Flory JD, Pollock BG, Manuck SB. The metabolic syndrome is associated with reduced central serotonergic responsivity in healthy community volunteers. J Clin Endocrinol Metabol 2006;91:718-21. Muldoon MF, Mackey RH, Williams KV, Korytkowski MT, Flory JD, Manuck SB. Low central nervous system serotonergic responsivity is associated with the metabolic syndrome and physical inactivity. J Clin Endocrinol Metabol 2004;89:266-71. Wolk R, Gami AS, Garcia-Touchard A, Somers VK. Sleep and cardiovascular disease. Curr Probl Cardiol 2005;30:625-62. Bass J. Turek, F.W. Sleepless in America: a pathway to obesity and the metabolic syndrome? Archives of Internal Medicine 2005;165:15-6. Spiegel K, Knutson K, Leproult R, Tasali E, Van Cauter E. Sleep loss: a novel risk factor for insulin resistance and Type 2 diabetes. J Appl Physiol 2005;99:2008-19. Gangwisch JE, Malaspina D, Boden-Albala B, Heymsfield SB. Inadequate sleep as a risk factor for obesity: analyses of the NHANES I. Sleep 2005;28:1289-96. Ayas NT, White DP, Al-Delaimy WK, et al. A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care 2003;26:380-4. Sleep and Metabolic Syndrome—Jennings et al

19. Hasler G, Buysse DJ, Klaghofer R, et al. The association between short sleep duration and obesity in young adults: a 13-year prospective study.[see comment]. Sleep. 2004;27:661-6. 20. Vorona RD, Winn MP, Babineau TW, Eng BP, Feldman HR, Ware JC. Overweight and obese patients in a primary care population report less sleep than patients with a normal body mass index. Arch Intern Med 2005;165:25-30. 21. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet 1999;354:1435-9. 22. Spiegel K, Tasali E, Penev P, Van Cauter E. Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med 2004;141:846-50. 23. Spiegel K, Leproult R, L’Hermite-Baleriaux M, Copinschi G, Penev PD, Van Cauter E. Leptin levels are dependent on sleep duration: relationships with sympathovagal balance, carbohydrate regulation, cortisol, and thyrotropin. J Clin Endocrinol Metabol 2004;89:576271. 24. Poulain M, Doucet M, Major GC, et al. The effect of obesity on chronic respiratory diseases: pathophysiology and therapeutic strategies. CMAJ 2006;174:1293-9. 25. Coughlin SR, Mawdsley L, Mugarza JA, Calverley PMA, Wilding JPH. Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome. Eur Heart J 2004;25:735-41. 26. Bjorkelund C, Bondyr-Carlsson D, Lapidus L, et al. Sleep disturbances in midlife unrelated to 32-year diabetes incidence: the prospective population study of women in Gothenburg. Diabetes Care. 2005;28:2739-44. 27. Buysse DJ, Reynolds CF, 3rd, 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. 28. Carpenter JS, Andrykowski MA. Psychometric evaluation of the Pittsburgh Sleep Quality Index. J Psychosom Res 1998;45:5-13. 29. Muldoon MF, Nazzaro P, Sutton-Tyrrell K, Manuck SB. White-coat hypertension and carotid artery atherosclerosis: a matching study. Arch Intern Med 2000;160:1507-12. 30. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412-9. 31. Beck AT, Ward, CH. Mendelson, M., Mock, JE. Erbaugh, JK. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:56171. 32. Gangwisch JE, Heymsfield SB, Boden-Albala B, et al. Short sleep duration as a risk factor for hypertension: analyses of the first National Health and Nutrition Examination Survey. Hypertension 2006;47:833-9. 33. Gliklich RE, Taghizadeh F, Winkelman JW, Gliklich RE, Taghizadeh F, Winkelman JW. Health status in patients with disturbed sleep and obstructive sleep apnea. Otolaryngol Head Neck Surg 2000;122:542-6. 34. Krakow B, Melendrez D, Johnston L, et al. Sleep-disordered breathing, psychiatric distress, and quality of life impairment in sexual assault survivors. J Nerv Mental Dis. 2002;190:442-52.

SLEEP, Vol. 30, No. 2, 2007

223
