0
Original Investigation |

Role of Sleep Duration and Quality in the Risk and Severity of Type 2 Diabetes Mellitus FREE

Kristen L. Knutson, PhD; Armand M. Ryden, MD; Bryce A. Mander, BA; Eve Van Cauter, PhD
[+] Author Affiliations

Author Affiliations: Department of Medicine, University of Chicago (Drs Knutson, Ryden, and Van Cauter, and Mr Mander); and Institute for Neuroscience, Northwestern University (Mr Mander); Chicago, Ill.


Arch Intern Med. 2006;166(16):1768-1774. doi:10.1001/archinte.166.16.1768.
Text Size: A A A
Published online

Background  Evidence from laboratory and epidemiologic studies suggests that decreased sleep duration or quality may increase diabetes risk. We examined whether short or poor sleep is associated with glycemic control in African Americans with type 2 diabetes mellitus.

Methods  We conducted a cross-sectional study of volunteers with type 2 diabetes interviewed at the University of Chicago Hospitals, Chicago, Ill. The final analysis included 161 participants. Glycemic control was assessed by hemoglobin A1c (HbA1c) level obtained from medical charts. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI). Perceived sleep debt was calculated as the difference between preferred and actual weekday sleep duration.

Results  The mean ± SD sleep duration was 6.0 ± 1.6 hours, and 71% of the participants were classified as having poor quality sleep (PSQI score >5). We excluded patients with sleep frequently disrupted by pain (n = 39). In patients without diabetic complications, glycemic control was associated with perceived sleep debt but not PSQI score. The predicted increase in HbA1c level for a perceived sleep debt of 3 hours per night was 1.1% above the median. In patients with at least 1 complication, HbA1c level was associated with PSQI score but not perceived sleep debt. The predicted increase in HbA1c level for a 5-point increase in PSQI was 1.9% above the median.

Conclusions  In our sample, sleep duration and quality were significant predictors of HbA1c, a key marker of glycemic control. Combined with existing evidence linking sleep loss to increased diabetes risk, these data suggest that optimizing sleep duration and quality should be tested as an intervention to improve glucose control in patients with type 2 diabetes.

Figures in this Article

Chronic partial sleep loss due to bedtime restriction and sleep complaints are increasingly prevalent in modern society. During the past few years, evidence from laboratory and epidemiologic studies has accumulated, suggesting that decreased sleep duration and/or quality may adversely effect glucose regulation and increase the risk of type 2 diabetes mellitus.

Two published laboratory studies have reported alterations in glucose regulation during partial sleep restriction. In the first study, exposure to 5 days of 4-hour sleep durations was associated with a 40% reduction in glucose tolerance to intravenous glucose and a 30% reduction in the acute insulin response to glucose.1 These findings were confirmed in another study that used a randomized cross-over design with 2 nights of sleep restriction or extension (4-hour vs 10-hour bedtimes).2 After the second night of each condition, caloric intake was replaced by constant intravenous glucose infusion, and blood samples were collected every 20 minutes. After sleep restriction, morning glucose levels were higher and insulin levels were lower than after sleep extension.2 Preliminary data from an ongoing study revealed a marked reduction in glucose tolerance and insulin sensitivity after 8 nights of 5-hour bedtimes compared with 8-hour bedtimes.3 The consistency of these findings, despite differences in experimental design, suggests that sleep restriction has adverse effects on glucose metabolism.

A much larger number of epidemiologic studies, summarized in Table 1, have explored the relationship between sleep duration and/or quality and diabetes. The prospective studies, which involved different geographical locations, were remarkably consistent, indicating that short or poor sleep may increase the risk of developing type 2 diabetes. Evidence from cross-sectional studies suggests that a diabetic condition may involve a reduction in sleep duration or an impairment of sleep quality. Neuropathic pain and nocturia have been suggested as 2 possible causes of decreased sleep quality.14

Table Graphic Jump LocationTable 1. Summary of Epidemiologic Studies That Examined the Association Between Sleep and Diabetes

While there is extensive laboratory and epidemiologic evidence for an adverse effect of sleep-related breathing disorder (SRBD) on insulin sensitivity and risk for the metabolic syndrome,1521 there is less published data on the incidence of SRBD in patients with type 2 diabetes. One study from Sweden had indicated that the prevalence of severe obstructive sleep apnea in men with type 2 diabetes is 36%, more than double that found in normoglycemic subjects.22

The rapidly accumulating evidence for a relationship between impaired sleep and diabetes risk raises the possibility that an association between reduced sleep duration or quality and the severity of an existing diabetic condition may exist. We therefore present the results of a study that examined self-reported sleep duration and quality and glycemic control as assessed by hemoglobin A1c (HbA1c) levels in African Americans with type 2 diabetes.

The study involved a 30- to 45-minute interview and access to medical charts. The protocol was approved by the University of Chicago institutional review board, and all participants gave written informed consent. The participants received $20 for participating. Patients with type 2 diabetes were recruited at the University of Chicago Hospital, Chicago, Ill. A total of 298 patients (216 African Americans, 54 whites, 3 Hispanics, 1 Asian, and 24 of other or undetermined ethnicity) completed the study. The sample composition was consistent with the ethnic distribution of the local patient population.

Our study planned to exclude patients from the analysis who were diagnosed as having type 2 diabetes within the past year because it is unlikely that they had achieved stable glycemic control (n = 24). Patients who provided incomplete responses to the sleep questionnaire (n = 7), patients without a glycated hemoglobin (GHb) or HbA1c measurement within 3 months of the interview (n = 38), patients claiming to prefer 12 or more hours of sleep per night (n = 7), and patients of unknown ethnicity (n = 18) were also excluded. The final sample included 204 individuals (161 African Americans, 38 whites, and 5 of other ethnicities). There are well-documented ethnic differences in both diabetes risk and sleep.23,24 In an exploratory analysis, the interaction between our measure of sleep quality and race was a highly significant predictor of glycemic control (P<.001), indicating that the analysis needed to be stratified by race. Ethnic groups other than African Americans are, however, underrepresented in our sample. The present analysis therefore focuses only on African American patients.

During the interview, waist-hip ratio was measured, and the patients reported height and weight from which we calculated body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared). The questions on diabetes and its complications were part of the University of Chicago Diabetes/Quality of Life survey study,25 and thus this information was all self-reported. Frequency of checking glucose level, diabetes medications, and frequency of exercise were assessed. Subjects were categorized as either taking insulin (alone or in combination with oral antidiabetic agents) or not taking insulin. The presence of major complications of diabetes (neuropathy, retinopathy, nephropathy, coronary artery disease, and peripheral vascular disease) was also assessed. This variable was dichotomized into absence of complications and presence of 1 or more complications.

The interview included the Pittsburgh Sleep Quality Index (PSQI),26 a validated 19-item questionnaire that produces a global sleep quality score that ranges from 0 to 21, derived from 7 component scores.27,28 A global score greater than 5 distinguishes poor sleepers from good sleepers.26 The question “how often have you had trouble sleeping because you had pain?” identified patients with sleep disturbed by pain as those who responded “3 or more times per week.” We excluded these individuals in our analyses of the association between sleep and glycemic control because chronic pain is a likely confounder. The PSQI also included the actual number of hours of sleep obtained on weekdays and on weekends. A weekly average was calculated using the following formula:

(5/7 × Weekday Sleep Duration) + (2/7 × Weekend Sleep Duration).

We also asked preferred sleep duration and calculated perceived sleep debt as the difference between weekday sleep duration and preferred sleep duration. We created a modified PSQI score by removing the sleep duration component to assess sleep quality independently from sleep quantity.

The PSQI does not assess the presence of SRBD, which is frequent in type 2 diabetes.22 However, the PSQI includes questions about breathing and snoring, which we used to estimate risk of SRBD. Subjects were classified as at high risk of SRBD if they indicated on the PSQI that their sleep was disturbed 3 or more times per week because of difficulty breathing or coughing or snoring. Subjects who responded that their bed partners had noticed loud snoring or pauses in breathing during sleep 1 or more times per week were also classified as at high risk. Finally, this high-risk group included 12 patients who indicated during the interview that they had SRBD. All other subjects were classified as at low risk of SRBD.

The interview included the 20-item Center for Epidemiologic Studies Depression Scale (CES-D).29 Scores range from 0 to 60, where a higher score indicates greater depressive symptoms.

We obtained total GHb or HbA1c values from the medical charts. Midway through the study, the University of Chicago Laboratories switched from measuring GHb (by Bio-Rad Variant Classic boronate affinity-automated high-performance liquid chromatography [HPLC]) to HbA1c (by Bio-Rad Variant II ion exchange automated HPLC) (Bio-Rad Laboratories, Hercules, Calif). The intra-assay coefficient of variation was 0.5% to 1.0%, and the interassay coefficient of variation was 2.2% to 2.4% for both methods. We converted GHb values to HbA1c using the following equation:

HbA1c = 0.72 × GHb + 1.35.30

All statistical analyses were computed using SPSS 13.0 (SPSS Inc, Chicago, Ill). Hemoglobin A1c had a right-skewed distribution. Therefore, we used the natural log of HbA1c (lnHbA1c) in correlation and general linear model (GLM) analysis. Perceived sleep debt was used as the primary variable to quantify the impact of sleep duration because it incorporates individual differences in sleep need. We also repeated the analyses using weekly sleep duration rather than perceived sleep debt. We used the modified PSQI score as a marker of sleep quality. Finally, we calculated GLM for the dependent variable, lnHbA1c, including both categorical and continuous variables. The first model included 2 sleep variables as well as covariates and all first-degree interactions. Interactions with a P level of >.05 were removed one at a time by deleting the interaction with the largest P value. Because a significant interaction term that includes a sleep variable and a categorical variable indicates that the association between sleep and glycemic control is different for the different categories, we planned to stratify our regression analyses according to any significant interactions that included a categorical variable and a sleep variable.

The sample included 42 men and 119 women. Table 2 provides descriptive statistics. The mean HbA1c level was 8.3%, and 26% of the sample had an HbA1c level below 7%, which is the recommended optimal upper level of HbA1c.31 Only 6% of patients reported obtaining at least 8 hours of sleep on weeknights and only 22% obtained at least 7 hours. Approximately 71% of the patients exceeded the cutoff for poor sleep (ie, global PSQI score >5). We separated the 39 patients who reported sleep frequently disrupted by pain from the remainder of the sample. Sleep quantity was lower in the group with pain, indicated by shorter sleep durations and greater perceived sleep debt (Table 2). Modified PSQI scores were also higher in the group with frequent pain, even after excluding the pain question from the score.

Table Graphic Jump LocationTable 2. Characteristics of Patients*

The proportion of patients with poor sleep quality (global PSQI score >5) remained high (67%) after subjects with pain-disturbed sleep were excluded. The remainder of our analysis was restricted to patients without pain-disturbed sleep (35 men and 87 women). The only significant sex difference in sleep was for weekend sleep duration, which was significantly longer in women (6.6 ± 1.8 hours vs 5.9 ± 1.3 hours; P = .03).

In bivariate analyses, higher lnHbA1c levels were associated with higher scores on the modified PSQI (r = 0.28; P = .002), ie, lower sleep quality, a greater perceived sleep debt (r = 0.30; P = .001), and shorter weekly sleep durations (r = −0.17; P = .03). Higher depression scores were also associated with higher lnHbA1c levels (r = 0.27; P = .003). As expected, the depression score and the modified PSQI score were significantly correlated (r = 0.63; P<.001). Body mass index, waist-hip ratio, frequency of exercise, frequency of checking glucose, and duration of diabetes were not significantly related to lnHbA1c levels. Finally, younger participants had poorer glycemic control (r = −0.27; P = .002).

We performed a GLM analysis of variance, with lnHbA1c level as the dependent variable and perceived sleep debt, modified PSQI score, age, sex, BMI, insulin use, presence of diabetic complications, and 6 first-degree interactions as predictors. After controlling for age, sex, BMI, diabetic complications, and insulin use, the sleep variables were significant (perceived sleep debt, P = .01; modified PSQI score, P = .04) and had significant interactions with diabetic complications (diabetic complications × modified PSQI score, P = .001) and insulin use (insulin use × perceived sleep debt, P = .03; insulin use × modified PSQI score, P = .02). Adding the CES-D depression score as a predictor only marginally increased the variance accounted for by the model (from unadjusted r2 = 0.40 to r2 = 0.41). The depression score did not emerge as a significant predictor or have significant interactions with other predictors. Replacing perceived sleep debt with weekly sleep duration revealed a significant main effect of sleep duration (P = .02) and a trend for modified PSQI score (P = .06), and the interactions between the sleep variables and the presence of complications or the use of insulin remained significant. Because of the significant interactions between the sleep variables and diabetic complications or insulin use, regression analyses were stratified by diabetic complications and by insulin use (Table 3).

Table Graphic Jump LocationTable 3. Predictors of lnHbA1c From GLM Regression Analysis Stratified by Diabetic Complications and Insulin Use*

In patients without complications, perceived sleep debt but not subjective sleep quality was associated with lnHbA1c levels. In contrast, in patients with at least 1 complication, modified PSQI score, but not perceived sleep debt, was a significant predictor after controlling for covariates. Figure 1 presents bivariate associations between HbA1c level and the 2 sleep variables in patients without and with diabetic complications.

Place holder to copy figure label and caption
Figure 1.

Scatterplots between hemoglobin A1c (HbA1c) and modified Pittsburgh Sleep Quality Index (PSQI) score and perceived sleep debt stratified by diabetic complications (A, no diabetic complications; B, ≥1 diabetic complications). Note the interindividual variability of HbA1c levels at any given level of modified PSQI score or perceived sleep debt value.

Graphic Jump Location

Since we used lnHbA1c in regression analyses, the β coefficients of the GLM represent the proportional change in HbA1c level for an absolute change in main effect.32Figure 2 presents the predicted increase in HbA1c level for hypothetical patients with low (ie, corresponding to the 25th percentile of HbA1c values), average (ie, corresponding to the 50th percentile of HbA1c values), and high (ie, corresponding to the 75th percentile of HbA1c values) HbA1c levels. For example, in patients without complications, a perceived sleep debt increase of 3 hours per night for an individual with an HbA1c level of 7.5% predicts an HbA1c level of 8.6% (ie, an increase by 1.1%). In patients with at least 1 complication, a 5-point increase in PSQI for an individual with an HbA1c of 8.7% predicts an HbA1c level of 10.6% (ie, an increase by 1.9%). Because the β coefficient is proportional, changes in sleep variables have a larger effect at larger values of HbA1c.

Place holder to copy figure label and caption
Figure 2.

Predicted increase in hemoglobin A1c (HbA1c) resulting from an increase in perceived sleep debt (A) or a decrease in sleep quality (ie, an increase in Pittsburgh Sleep Quality Index [PSQI] score) (B) in a hypothetical subject who would have a baseline HbA1c in the low (25th percentile), median (50th percentile), or high (75th percentile) for the subgroup under consideration. The predicted change is based on an increase in perceived sleep debt of 3 hours for the group with no complications and the group not taking insulin. The predicted change is based on an increase in PSQI score by 5 points for the group with complications and the group taking insulin.

Graphic Jump Location

Table 3 also presents results for patients stratified by insulin use. For patients not taking insulin, perceived sleep debt, but not PSQI, was a significant predictor. For patients taking insulin, PSQI score, but not perceived debt, was significantly associated with lnHbA1c level.

Of the 122 patients, 23 (19%) were classified as at high risk of SRBD. This subgroup had a higher mean HbA1c level compared with those at low risk of SRBD (9.7 vs 7.9%; P<.01) but were not more likely to have complications or to be treated with insulin. The bivariate correlations between HbA1c level and sleep variables were essentially unchanged after excluding subjects at high risk of SRBD (lnHbA1c level and perceived sleep debt, r = 0.30 [P = .003]; lnHbA1c level and modified PSQI, r = 0.23 [P<.05]).

Recent laboratory and epidemiologic studies have indicated that insufficient sleep may result in decreased glucose tolerance and increased diabetes risk.1,4,6,9,33 The findings also raise the possibility that reduced sleep duration or quality could adversely affect glucose control in an existing diabetic condition. To our knowledge, the present study is the first to address this hypothesis by examining self-reported sleep duration and quality and HbA1c levels in patients with type 2 diabetes. Our analyses reveal that a higher perceived sleep debt or lower sleep quality are associated with poorer glucose control, after controlling for age, sex, BMI, insulin use, and the presence of major complications.

The direction of causality cannot be inferred from our analyses. Poor glycemic control in patients with diabetes could impair subjective sleep quality even in the absence of pain. For example, nocturia could play a role in the observed relationship between sleep and glycemic control.14 One of the limitations of our study is that the questionnaires did not include assessment of the severity of nocturia. Perceived sleep debt could partly reflect the inability to achieve sufficient sleep rather than a voluntary reduction of bedtime. The fact that most patients (83%) without pain-disrupted sleep did not report longer sleep during the weekend relative to weekdays is consistent with this notion. Thus, poor diabetes control could contribute both to a higher perceived sleep debt and lower sleep quality. On the other hand, evidence from previous laboratory and epidemiologic studies supports the opposite direction of causality, that is, that short or poor sleep impairs glucose control. The prospective studies showing a relationship between sleep duration and the development of symptomatic diabetes are particularly relevant to the present findings. Intervention studies are needed to test the hypothesis that paying a sleep debt and/or improving sleep quality may improve glucose control in patients with type 2 diabetes. The strength of the associations in our sample suggests that such studies would be warranted from a clinical standpoint. Our analyses indicated that in patients without complications, a perceived sleep debt of 3 hours per day was associated with an increase in HbA1c level by 1.1% above the median, while in patients with at least 1 complication, a 5-point increase in PSQI score (decreased sleep quality) was associated with an increase in HbA1c level by 1.9% above the median. The magnitude of these effects is comparable to those of widely used oral antidiabetic agents.3436 The fact that perceived sleep debt is significant for those without complications or those not taking insulin, while sleep quality, as measured by the modified PSQI, is significant for those with complications or those taking insulin, may be due to differences in severity of disease.

The present study introduces the construct of “perceived sleep debt” that captures habitual sleep time and subjective sleep need. In exploratory bivariate analyses, we noted that perceived sleep debt was more strongly correlated with levels of HbA1c compared with sleep duration (r = +0.30 vs r = −0.17), which suggested that perceived sleep debt may be a better predictor of glycemic control because it represents sleep loss in relation to subjective sleep need, which likely varies between individuals. We verified that the results of the regression model would have been qualitatively similar had we used sleep duration rather than perceived sleep debt but noted that perceived sleep debt accounted for a larger proportion of the variance compared with sleep duration.

Approximately 20% of our participants (17% of men and 27% of women) were identified as at high risk for SRBD. Others have found a 36% prevalence of SRBD in men with type 2 diabetes,22 while the prevalence in the general population has been estimated to be 24% in men and 9% in women.37 Our method of identifying SRBD using questions from the PSQI has not been validated, and it is possible that we have underestimated the prevalence of SRBD in this sample. When patients classified as being at high risk for SRBD were excluded from our analysis, the associations between HbA1c levels and sleep variables persisted, indicating that SRBD risk as estimated in our study is not the primary mediator of the relationship between glucose control and sleep.

The present study identifies sleep as a potential factor influencing glucose control in a specific population of patients with type 2 diabetes. African Americans represent approximately 18% of the 19 million Americans with type 2 diabetes.38 Similar associations between sleep and glucose control are likely to exist in other ethnic groups. A cohort study that included mostly whites found a relationship between sleep duration and the development of symptomatic diabetes.4 Additional research is needed to determine whether optimizing sleep duration and quality may improve glucose control in patients with type 2 diabetes. Sleep curtailment has become increasingly prevalent in modern society, and it cannot be excluded that this behavior has contributed to the current epidemic of type 2 diabetes.

Correspondence: Eve Van Cauter, PhD, Department of Medicine, University of Chicago, 5841 S Maryland Ave, MC1027 AMB M239, Chicago, IL 60637 (evcauter@medicine.bsd.uchicago.edu).

Accepted for Publication: June 23, 2006.

Author Contributions: The senior author had full access to the study data and takes responsibility for its integrity and the accuracy of the analysis.

Financial Disclosure: None reported.

Funding/Support: This work was supported by grants from the MacArthur Foundation, Chicago, Ill; the American Diabetes Association, Alexandria, Va; and the National Institutes of Health (RO1 DK-41814 and PO1 AG-11412), Bethesda, Md.

Role of the Sponsors: The sole role of the sponsors was financial support.

Spiegel  KLeproult  RVan Cauter  E Impact of sleep debt on metabolic and endocrine function. Lancet 1999;3541435- 1439
PubMed
Spiegel  KKnutson  KLeproult  RTasali  ECauter  EV Sleep loss: a novel risk factor for insulin resistance and type 2 diabetes. J Appl Physiol 2005;992008- 2019
PubMed
Leproult  RHolmback  UVan Cauter  E Marked decrease in insulin sensitivity following one week of partial sleep deprivation with or without circadian misalignment. Diabetes 2006;55 ((suppl 1)) A323- A324
Ayas  NTWhite  DPAl-Delaimy  WK  et al.  A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care 2003;26380- 384
PubMed
Kawakami  NTakatsuka  NShimizu  H Sleep disturbance and onset of type 2 diabetes. Diabetes Care 2004;27282- 283
PubMed
Nilsson  PMRoost  MEngstrom  GHedblad  BBerglunc  G Incidence of diabetes in middle-aged men is related to sleep disturbances. Diabetes Care 2004;272464- 2469
PubMed
Mallon  LBroman  JEHetta  J High incidence of diabetes in men with sleep complaints or short sleep duration: a 12-year follow-up study of a middle-aged population. Diabetes Care 2005;282762- 2767
PubMed
Bjorkelund  CBondyr-Carlsson  DLapidus  L  et al.  Sleep disturbances in midlife unrelated to 32-year diabetes incidence: the prospective population study of women in Gothenburg. Diabetes Care 2005;282739- 2744
PubMed
Meisinger  CHeier  MLoewel  H Sleep disturbance as a predictor of type 2 diabetes mellitus in men and women from the general population. Diabetologia 2005;48235- 241
PubMed
Yaggi  HKAraujo  ABMcKinlay  JB Sleep duration as a risk factor for the development of type 2 diabetes. Diabetes Care 2006;29657- 661
PubMed
Gislason  TAlmqvist  M Somatic diseases and sleep complaints: an epidemiological study of 3,201 Swedish men. Acta Med Scand 1987;221475- 481
PubMed
Hyyppa  MTKronholm  E Quality of sleep and chronic illnesses. J Clin Epidemiol 1989;42633- 638
PubMed
Sridhar  GRMadhu  K Prevalence of sleep disturbances in diabetes mellitus. Diabetes Res Clin Pract 1994;23183- 186
PubMed
Lamond  NTiggemann  MDawson  D Factors predicting sleep disruption in type II diabetes. Sleep 2000;23415- 416
PubMed
Ip  MSLam  BNg  MMLam  WKTsang  KWLam  KS Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med 2002;165670- 676
PubMed
Punjabi  NMShahar  ERedline  SGottlieb  DJGivelber  RResnick  HE Sleep-disordered breathing, glucose intolerance, and insulin resistance: the Sleep Heart Health Study. Am J Epidemiol 2004;160521- 530
PubMed
Punjabi  NMSorkin  JDKatzel  LIGoldberg  APSchwartz  ARSmith  PL Sleep-disordered breathing and insulin resistance in middle-aged and overweight men. Am J Respir Crit Care Med 2002;165677- 682
PubMed
Vgontzas  ANPapanicolaou  DABixler  EO  et al.  Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. J Clin Endocrinol Metab 2000;851151- 1158
PubMed
Coughlin  SRMawdsley  LMugarza  JACalverley  PMWilding  JP Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome. Eur Heart J 2004;25735- 741
PubMed
Svatikova  AWolk  RGami  ASPohanka  MSomers  VK Interactions between obstructive sleep apnea and the metabolic syndrome. Curr Diab Rep 2005;553- 58
PubMed
Tassone  FLanfranco  FGianotti  L  et al.  Obstructive sleep apnoea syndrome impairs insulin sensitivity independently of anthropometric variables. Clin Endocrinol (Oxf) 2003;59374- 379
PubMed
Elmasry  ALindberg  EBerne  C  et al.  Sleep-disordered breathing and glucose metabolism in hypertensive men: a population-based study. J Intern Med 2001;249153- 161
PubMed
Disdier-Flores  OMRodriguez-Lugo  LAPerez-Perdomo  RPerez-Cardona  CM The public health burden of diabetes: a comprehensive review. P R Health Sci J 2001;20123- 130
PubMed
Durrence  HHLichstein  KL The sleep of African Americans: a comparative review. Behav Sleep Med 2006;429- 44
PubMed
Meltzer  DEgleston  B How patients with diabetes perceive their risk for major complications. Eff Clin Pract 2000;37- 15
PubMed
Buysse  DJReynolds  CF  IIIMonk  THBerman  SBKupfer  DJ The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989;28193- 213
PubMed
Buysse  DJMonk  THReynolds  CFMesiano  DHouck  PRKupfer  DJ Patterns of sleep episodes in young and elderly adults during a 36-hour constant routine. Sleep 1993;16632- 637
PubMed
Carpenter  JSAndrykowski  MA Psychometric evaluation of the Pittsburgh Sleep Quality Index. J Psychosom Res 1998;455- 13
PubMed
Radloff  L The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1385- 401
Toney  LJMcLawhon  RWZibrat  SJ Evaluation and comparison of hemoglobin HbA1c on the Bio-Rad Variant II to total glycated hemoglobin on the Bio-Rad Variant Hemoglobin Testing Systems [abstract]. Clin Chem 2001;47A4
American Diabetes Association, Standards of medical care in diabetes—2006. Diabetes Care 2006;29 ((suppl 1)) S4- 42
PubMed
Gujarati  D Basic Econometrics. 3rd ed. New York, NY McGraw-Hill, Inc1995;
Gottlieb  DJPunjabi  NMNewman  AB  et al.  Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med 2005;165863- 868
PubMed
Duckworth  WMarcelli  MPadden  M  et al.  Improvements in glycemic control in type 2 diabetes patients switched from sulfonylurea coadministered with metformin to glyburide-metformin tablets. J Manag Care Pharm 2003;9256- 262
PubMed
Garber  AJDonovan  DS  JrDandona  PBruce  SPark  JS Efficacy of glyburide/metformin tablets compared with initial monotherapy in type 2 diabetes. J Clin Endocrinol Metab 2003;883598- 3604
PubMed
Moses  R Repaglinide in combination therapy. Diabetes Nutr Metab 2002;1533- 38
PubMed
Young  TPalta  MDempsey  JSkatrud  JWeber  SBadr  S The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;3281230- 1235
PubMed
Mokdad  AHFord  ESBowman  BA  et al.  Diabetes trends in the US: 1990-1998. Diabetes Care 2000;231278- 1283
PubMed

Figures

Place holder to copy figure label and caption
Figure 1.

Scatterplots between hemoglobin A1c (HbA1c) and modified Pittsburgh Sleep Quality Index (PSQI) score and perceived sleep debt stratified by diabetic complications (A, no diabetic complications; B, ≥1 diabetic complications). Note the interindividual variability of HbA1c levels at any given level of modified PSQI score or perceived sleep debt value.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.

Predicted increase in hemoglobin A1c (HbA1c) resulting from an increase in perceived sleep debt (A) or a decrease in sleep quality (ie, an increase in Pittsburgh Sleep Quality Index [PSQI] score) (B) in a hypothetical subject who would have a baseline HbA1c in the low (25th percentile), median (50th percentile), or high (75th percentile) for the subgroup under consideration. The predicted change is based on an increase in perceived sleep debt of 3 hours for the group with no complications and the group not taking insulin. The predicted change is based on an increase in PSQI score by 5 points for the group with complications and the group taking insulin.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable 1. Summary of Epidemiologic Studies That Examined the Association Between Sleep and Diabetes
Table Graphic Jump LocationTable 2. Characteristics of Patients*
Table Graphic Jump LocationTable 3. Predictors of lnHbA1c From GLM Regression Analysis Stratified by Diabetic Complications and Insulin Use*

References

Spiegel  KLeproult  RVan Cauter  E Impact of sleep debt on metabolic and endocrine function. Lancet 1999;3541435- 1439
PubMed
Spiegel  KKnutson  KLeproult  RTasali  ECauter  EV Sleep loss: a novel risk factor for insulin resistance and type 2 diabetes. J Appl Physiol 2005;992008- 2019
PubMed
Leproult  RHolmback  UVan Cauter  E Marked decrease in insulin sensitivity following one week of partial sleep deprivation with or without circadian misalignment. Diabetes 2006;55 ((suppl 1)) A323- A324
Ayas  NTWhite  DPAl-Delaimy  WK  et al.  A prospective study of self-reported sleep duration and incident diabetes in women. Diabetes Care 2003;26380- 384
PubMed
Kawakami  NTakatsuka  NShimizu  H Sleep disturbance and onset of type 2 diabetes. Diabetes Care 2004;27282- 283
PubMed
Nilsson  PMRoost  MEngstrom  GHedblad  BBerglunc  G Incidence of diabetes in middle-aged men is related to sleep disturbances. Diabetes Care 2004;272464- 2469
PubMed
Mallon  LBroman  JEHetta  J High incidence of diabetes in men with sleep complaints or short sleep duration: a 12-year follow-up study of a middle-aged population. Diabetes Care 2005;282762- 2767
PubMed
Bjorkelund  CBondyr-Carlsson  DLapidus  L  et al.  Sleep disturbances in midlife unrelated to 32-year diabetes incidence: the prospective population study of women in Gothenburg. Diabetes Care 2005;282739- 2744
PubMed
Meisinger  CHeier  MLoewel  H Sleep disturbance as a predictor of type 2 diabetes mellitus in men and women from the general population. Diabetologia 2005;48235- 241
PubMed
Yaggi  HKAraujo  ABMcKinlay  JB Sleep duration as a risk factor for the development of type 2 diabetes. Diabetes Care 2006;29657- 661
PubMed
Gislason  TAlmqvist  M Somatic diseases and sleep complaints: an epidemiological study of 3,201 Swedish men. Acta Med Scand 1987;221475- 481
PubMed
Hyyppa  MTKronholm  E Quality of sleep and chronic illnesses. J Clin Epidemiol 1989;42633- 638
PubMed
Sridhar  GRMadhu  K Prevalence of sleep disturbances in diabetes mellitus. Diabetes Res Clin Pract 1994;23183- 186
PubMed
Lamond  NTiggemann  MDawson  D Factors predicting sleep disruption in type II diabetes. Sleep 2000;23415- 416
PubMed
Ip  MSLam  BNg  MMLam  WKTsang  KWLam  KS Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med 2002;165670- 676
PubMed
Punjabi  NMShahar  ERedline  SGottlieb  DJGivelber  RResnick  HE Sleep-disordered breathing, glucose intolerance, and insulin resistance: the Sleep Heart Health Study. Am J Epidemiol 2004;160521- 530
PubMed
Punjabi  NMSorkin  JDKatzel  LIGoldberg  APSchwartz  ARSmith  PL Sleep-disordered breathing and insulin resistance in middle-aged and overweight men. Am J Respir Crit Care Med 2002;165677- 682
PubMed
Vgontzas  ANPapanicolaou  DABixler  EO  et al.  Sleep apnea and daytime sleepiness and fatigue: relation to visceral obesity, insulin resistance, and hypercytokinemia. J Clin Endocrinol Metab 2000;851151- 1158
PubMed
Coughlin  SRMawdsley  LMugarza  JACalverley  PMWilding  JP Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome. Eur Heart J 2004;25735- 741
PubMed
Svatikova  AWolk  RGami  ASPohanka  MSomers  VK Interactions between obstructive sleep apnea and the metabolic syndrome. Curr Diab Rep 2005;553- 58
PubMed
Tassone  FLanfranco  FGianotti  L  et al.  Obstructive sleep apnoea syndrome impairs insulin sensitivity independently of anthropometric variables. Clin Endocrinol (Oxf) 2003;59374- 379
PubMed
Elmasry  ALindberg  EBerne  C  et al.  Sleep-disordered breathing and glucose metabolism in hypertensive men: a population-based study. J Intern Med 2001;249153- 161
PubMed
Disdier-Flores  OMRodriguez-Lugo  LAPerez-Perdomo  RPerez-Cardona  CM The public health burden of diabetes: a comprehensive review. P R Health Sci J 2001;20123- 130
PubMed
Durrence  HHLichstein  KL The sleep of African Americans: a comparative review. Behav Sleep Med 2006;429- 44
PubMed
Meltzer  DEgleston  B How patients with diabetes perceive their risk for major complications. Eff Clin Pract 2000;37- 15
PubMed
Buysse  DJReynolds  CF  IIIMonk  THBerman  SBKupfer  DJ The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989;28193- 213
PubMed
Buysse  DJMonk  THReynolds  CFMesiano  DHouck  PRKupfer  DJ Patterns of sleep episodes in young and elderly adults during a 36-hour constant routine. Sleep 1993;16632- 637
PubMed
Carpenter  JSAndrykowski  MA Psychometric evaluation of the Pittsburgh Sleep Quality Index. J Psychosom Res 1998;455- 13
PubMed
Radloff  L The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1385- 401
Toney  LJMcLawhon  RWZibrat  SJ Evaluation and comparison of hemoglobin HbA1c on the Bio-Rad Variant II to total glycated hemoglobin on the Bio-Rad Variant Hemoglobin Testing Systems [abstract]. Clin Chem 2001;47A4
American Diabetes Association, Standards of medical care in diabetes—2006. Diabetes Care 2006;29 ((suppl 1)) S4- 42
PubMed
Gujarati  D Basic Econometrics. 3rd ed. New York, NY McGraw-Hill, Inc1995;
Gottlieb  DJPunjabi  NMNewman  AB  et al.  Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med 2005;165863- 868
PubMed
Duckworth  WMarcelli  MPadden  M  et al.  Improvements in glycemic control in type 2 diabetes patients switched from sulfonylurea coadministered with metformin to glyburide-metformin tablets. J Manag Care Pharm 2003;9256- 262
PubMed
Garber  AJDonovan  DS  JrDandona  PBruce  SPark  JS Efficacy of glyburide/metformin tablets compared with initial monotherapy in type 2 diabetes. J Clin Endocrinol Metab 2003;883598- 3604
PubMed
Moses  R Repaglinide in combination therapy. Diabetes Nutr Metab 2002;1533- 38
PubMed
Young  TPalta  MDempsey  JSkatrud  JWeber  SBadr  S The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;3281230- 1235
PubMed
Mokdad  AHFord  ESBowman  BA  et al.  Diabetes trends in the US: 1990-1998. Diabetes Care 2000;231278- 1283
PubMed

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 133

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Topics
PubMed Articles
Continuous glucose monitors: current status and future developments. Curr Opin Endocrinol Diabetes Obes 2013;20(2):106-11.
JAMAevidence.com

Users' Guides to the Medical Literature
Clinical Resolution

Users' Guides to the Medical Literature
Clinical Scenario