Abstract
Objectives
To evaluate the association between sleep duration and the incidence of diabetes stratified by sleep-related factors among Chinese men.
Methods
This study included 34,825 men who provided information on sleep-related questions in the Shanghai Men’s Health Study, a population-based cohort study conducted in Shanghai, China from 2002 to 2011. Participants were excluded who had a history of diabetes or who were diagnosed with diabetes within 2 years of recruitment. Cox regression was employed to evaluate the influence of sleep duration and its interaction with sleep-related factors on diabetes risk.
Results
A total of 1521 incident cases were documented during a median of 5.6 follow-up years. Adjusted hazard ratios and 95% confidence intervals were 1.0 (0.9–1.1) and 1.2 (1.0–1.3) for men who slept <7 and ≥8 hours per day, respectively, compared with those who slept 7 hours per day (ptrend=0.01). Stratified analyses revealed that the association between sleep duration and risk of diabetes was only statistically significant among current smokers and regular drinkers, never tea drinkers, men with a high body mass index, hypertension or comorbidity, and men who did not work nightshift or who snored. A statistically significant interaction between tea drinking and sleep duration was observed (pinteraction=0.01). The above association patterns remained when daytime nappers were excluded from the analyses.
Conclusions
The data suggested that longer sleep duration, particularly among individuals already exhibiting factors linked to poor quality of sleep, was associated with diabetes. The association between sleep duration and diabetes may be modified by tea drinking, especially in older men or men with more sleep-related factors.
Keywords: Sleep duration and snoring, Diabetes incidence, Association, Lifestyle, Comorbidity
INTRODUCTION
Both sleeping quantity and quality play important roles in human health. The American Sleep Foundation has reported that a large majority (75%) of American adults complain of having had at least one symptom of a sleep problem [1] such as difficulty falling asleep, waking up too early and not being able to get back to sleep, or snoring. Both short and long sleep duration have been reported to be long-term risk factors for poor health, including obesity, diabetes, cardiovascular disease, and cancer [2–5].
However, data are inconsistent on the association between the duration and quality of sleep and the prevalence/incidence of diabetes. Results from cross-sectional studies and prospective studies have shown both short and long sleep duration, following a U-shaped pattern, to be associated with an increased risk of diabetes [6–8]. A Finish study reported this U-shaped association in women only [9]. Several other studies, however, found that either short sleep duration or long sleep duration, but not both, were associated with an incidence of diabetes [10–11]. Other studies have reported no association between sleep duration and diabetes [12–13].
Sleep duration may affect glucose metabolism by reducing glucose intolerance, insulin resistance, and acute insulin response to glucose [3]. The relationship between sleep duration and diabetes incidence may be influenced by many factors that influence the quality of sleep, such as body mass index (BMI), snoring and night shift work, and physical and economic conditions [2]. Furthermore, evidence suggests that the risk factors for diabetes may vary by ethnicity [14–15]. However, the vast majority of previous research studies on the association of sleep duration and diabetes risk have been conducted in developed countries, and few studies have focused on the effects of sleep-quality-related factors on this association.
Using data from the Shanghai Men’s Health Study, a population-based cohort study of middle-aged Chinese men, a systematic evaluation was conducted of the association between sleep duration and diabetes risk, and its modification by sleep-quality-related factors.
METHODS
Study population
The Shanghai Men’s Health Study (SMHS) is a population-based cohort study of 61,480 Chinese men aged 40–74 years and living in urban Shanghai, China. Participants were recruited to the study from April 2002 to June 2006. A total of 83,033 eligible men from eight communities in urban Shanghai were invited to participate by trained interviewers through in-person contact, and 61,480 men who had no prior history of cancer were ultimately enrolled in the study (response rate 74.0%). Reasons for non-participation were refusal (n=17,823, 21.5%), out of area during enrollment (n=2370, 2.9%), and other miscellaneous reasons including poor health or hearing problems (n=1360, 1.6%). Sleep-related questions were added to the study after initiation of study enrollment. Thus, sleep information was unavailable for 16,776 SMHS participants, who were excluded from the current study.
Approximately 91% of SMHS participants provided a spot urine sample for the study, on which a urine glucose test was conducted. Participants who had positive results to the urine glucose test or who did not complete a urine test (n=7367) were excluded from the current study. Additionally, the study excluded SMHS participants who reported a history of diabetes at the baseline survey or were possible diabetes cases (n=1881), who were diagnosed with diabetes within the first 2 years of study enrollment (n=256), or who had <2 years of follow-up (n=375) from the study. The final analytic data set comprised 34,825 men.
A structured questionnaire was used to collect information on demographic characteristics, disease history, and lifestyle factors, including physical activity level, smoking and drinking history, dietary intake, tea consumption and shiftwork. Anthropometric measurements (taken at the baseline survey by trained interviewers) included height, weight, and circumference of waist and hips. The BMI was calculated as weight (in kg) divided by the square of height (in m), and the waist hip ratio (WHR) was calculated as waist circumference divided by hip circumference.
Sleep-related information was collected by the following questions: “In the past year, how many hours on average did you sleep per day? (Including sleeping at night time and napping during the day time, but not including waking time during the sleep period)” and “In the past year, did you nap during the day at least once a week? (Not including time slept during the daytime if you worked nightshift).”
Ascertainment of type 2 diabetes
Information on type 2 diabetes (T2D) was self-reported and collected during the in-person follow-up surveys, which took place every 2–3 years after baseline enrollment. The current study considered individuals who reported having a diagnosis of T2D and who met at least one of the following criteria, recommended by the American Diabetes Association, as having the study outcome: 1) fasting glucose concentration ≥7 mmol/L on at least two separate occasions; 2) an oral-glucose-tolerance test (OGTT) performed at the doctor’s office with a value ≥11.1 mmol/L; and/or 3) use of hypoglycemic medication (ie, insulin or oral hypoglycemic drugs). Twelve self-reported diabetes cases that did not meet the above criteria, and 30 men who died of diabetes but had no information on diabetes diagnosis were excluded from the current study.
Statistical analysis
Sleep duration was categorized into three groups in the analysis: <7 hours, 7 hours, and ≥8 hours. ANOVA was used for comparing continuous variables and Chi-squared test was used for dichotomous variables between T2D cases and healthy men. Cox regression was applied to calculate hazard ratios (HRs) and 95% CIs for each sleep duration group, adjusted for potential confounders at the baseline survey. The study used 7 hours of sleep duration as reference in the analysis because this group has been previously suggested to be associated with the lowest incidence of diabetes [8]. Covariates, including age at interview (continuous), smoking (pack-years), alcohol consumption (drinks per day), current tea drinking (yes/no), energy intake (continuous), and physical activity (met-hours/year), snoring (yes/no), self-reported nightshift work history (yes/no), BMI (continuous), family history of diabetes (yes/no), hypertension (yes/no), and comorbidity (yes/no) were included in the COX model.
Tests for linear trend were performed by using an ordinal value for each sleep category (1 for <7 hours, 2 for 7 hours, and 3 for ≥8 hours) as the continuous variable in the model. Sleep-quality-related factors, such as physical inactivity, smoking, drinking, high BMI (≥25), comorbidity (≥1), hypertension (yes) and snoring (yes), were treated as the risk factors, and the total number of risk factors for each participant was tallied to calculate a ‘risk score’. The score was categorized to three categories: low risk (score=0–2), moderate risk (score=3–4) and high risk (score>4). All p-values were two-tailed and differences at p<0.05 were accepted as statistically significant. All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC).
RESULTS
The average age of cohort participants was 55.0 years (standard deviation (SD)=9.6 years). Average sleep duration was 7.1 hours (SD=1.3 hours). About 26.3% of participants in the study reported sleep duration of 7 hours per day, and 18,651 participants reported no daytime napping (nocturnal sleep only).
Table 1 shows selected demographic characteristics, lifestyle factors, and anthropometric measurements, as well as daytime napping and snoring by diabetes status. All data presented in Table 1 are age-adjusted. Over an average of 5.6 years of follow-up, a total of 1521 diabetes cases were documented from 2 years after study enrollment. Participants who developed diabetes were older, had higher BMI and WHR, and were more likely to have higher energy intake, higher prevalence of coronary heart disease and hypertension, and a family history of diabetes and snoring, compared with diabetes-free participants. There were no significant differences in education, income, or lifestyle factors between diabetes cases and healthy men.
Table 1.
Comparison of baseline characteristics of diabetes cases and diabetes-free men in the SMHS.*
| Characteristics | Diabetes (n=1521) | Healthy men (n=33,304) | p |
|---|---|---|---|
| Age (years, x̄ ± SE ) | 55.63 ± 0.25 | 54.93 ± 0.05 | <0.01 |
| BMI (kg/m2, x̄ ± SE ) | 25.76 ± 0.08 | 23.53 ± 0.02 | <0.01 |
| WHR (x̄ ± SE ) | 0.9294 ± 0.0014 | 0.8979 ± 0.0003 | <0.01 |
| Energy intake (kcal/day, x ± SE ) | 1964.06 ± 12.24 | 1915.40 ± 2.62 | <0.01 |
| Education (%) | |||
| Primary school or less | 6.38 | 6.33 | 0.18 |
| Middle school | 36.87 | 34.48 | |
| High school | 36.83 | 37.54 | |
| College or more | 19.91 | 21.65 | |
| Income (yuan/per capita/year, %) | |||
| <6000 | 14.07 | 12.92 | 0.29 |
| ≥6000 | 77.56 | 77.91 | |
| ≥10,000 | 8.37 | 9.17 | |
| Occupation (%) | |||
| Professional | 23.51 | 23.72 | 0.61 |
| Clerical | 22.50 | 21.33 | |
| Manual workers | 53.99 | 54.95 | |
| Disease history | |||
| Hypertension (%) | 43.44 | 27.00 | <0.01 |
| Coronary heart disease (%) | 6.40 | 4.17 | <0.01 |
| Myocardial infarction (%) | 1.02 | 0.69 | 0.16 |
| Comorbidity score ≥1 (%) | 28.64 | 26.88 | 0.13 |
| Family history of diabetes (%) | 28.13 | 16.10 | <0.01 |
| Current smoker (%) | 60.63 | 61.00 | 0.77 |
| Ever drank alcohol (%) | 33.19 | 34.51 | 0.29 |
| Current tea drinker (%) | 65.86 | 64.46 | 0.28 |
| Regular exercise participation (%) | 34.01 | 34.36 | 0.75 |
| Nightshift work (%) | 35.59 | 34.61 | 0.51 |
| Daytime napping (%) | 49.17 | 46.33 | 0.03 |
| Snoring (%) | 59.55 | 43.00 | <0.01 |
Means and proportions were adjusted for age at interview
Sleep duration was positively associated with diabetes risk, with HRs (95% CIs) of 1.00 (0.87–1.14) and 1.16 (1.02–1.32) for men who slept <7 and ≥8 hours per day, respectively, compared with those who slept 7 hours per day (pfor trend=0.01) (Table 2). Analyses stratified by selected factors that are known to be associated with sleeping duration or quality showed that long sleep duration and diabetes risk were more apparent in men who were smokers, drinkers, or never consumed tea, had a high BMI or comorbidity, who did not have nightshift work, or who snored. A significant interaction was observed between tea drinking and sleep duration. Specifically, among never tea drinkers, sleeping <7 hours per day was associated with a decreased risk of diabetes incidence [HR: 0.87 (95% CI: 0.69–1.10)], while sleeping ≥8 hours per day was associated with increased risk of diabetes incidence [HR: 1.23 (95% CI: 0.99–1.53)], compared with men who slept 7 hours per day (pfor trend<0.01). Sleep duration was not associated with diabetes risk among tea drinkers.
Table 2.
Associations between sleep duration and diabetes risk by selected lifestyles factors.*
| Sleep duration | <7 hours | 7–7.9 hours | ≥8 hours | plinear | pinteraction | |||
|---|---|---|---|---|---|---|---|---|
| n=11,608 | n=9149 | n=14,068 | ||||||
|
|
|
|
||||||
| cases | HR (95% CI) | cases | HR | cases | HR CI (95%) | |||
| Total | 506 | 1.00 (0.87, 1.14) | 382 | 1.00 | 633 | 1.16 (1.02, 1.32) | 0.01 | |
|
| ||||||||
| Age (years) | ||||||||
|
| ||||||||
| 40- | 175 | 0.93 (0.74, 1.16) | 141 | 1.00 | 225 | 1.18 (0.95, 1.46) | 0.02 | 0.54 |
|
| ||||||||
| 50- | 157 | 1.05 (0.82, 1.33) | 120 | 1.00 | 219 | 1.12 (0.89, 1.41) | 0.49 | |
|
| ||||||||
| 60- | 174 | 1.03 (0.82, 1.31) | 121 | 1.00 | 189 | 1.18 (0.93, 1.48) | 0.23 | |
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 339 | 1.07 (0.91, 1.27) | 250 | 1.00 | 406 | 1.12 (0.96, 1.31) | 0.54 | 0.01 |
|
| ||||||||
| No | 167 | 0.87 (0.69, 1.10) | 132 | 1.00 | 227 | 1.23 (0.99, 1.53) | <0.01 | |
|
| ||||||||
| Regular exercise participation | ||||||||
|
| ||||||||
| Yes | 179 | 1.04 (0.83, 1.30) | 134 | 1.00 | 222 | 1.23 (0.99, 1.53) | 0.09 | 0.70 |
|
| ||||||||
| No | 327 | 0.99 (0.84, 1.17) | 248 | 1.00 | 411 | 1.13 (0.96, 1.32) | 0.07 | |
|
| ||||||||
| Current smokers | ||||||||
|
| ||||||||
| Yes | 288 | 0.92 (0.77, 1.09) | 239 | 1.00 | 377 | 1.08 (0.92, 1.27) | 0.04 | 0.75 |
|
| ||||||||
| No | 218 | 1.11 (0.90, 1.38) | 143 | 1.00 | 256 | 1.27 (1.03, 1.57) | 0.15 | |
|
| ||||||||
| Ever alcohol drank | ||||||||
|
| ||||||||
| Yes | 173 | 0.80 (0.64, 1.02) | 122 | 1.00 | 207 | 1.00 (0.80, 1.26) | 0.04 | 0.65 |
|
| ||||||||
| No | 333 | 1.08 (0.91, 1.27) | 260 | 1.00 | 426 | 1.21 (1.04, 1.42) | 0.09 | |
|
| ||||||||
| BMI | ||||||||
|
| ||||||||
| <25 (kg/m2) | 220 | 1.10 (0.89, 1.36) | 145 | 1.00 | 264 | 1.24 (1.01, 1.52) | 0.19 | 0.52 |
|
| ||||||||
| ≥25 (kg/m2) | 286 | 0.94 (0.79, 1.12) | 237 | 1.00 | 369 | 1.13 (0.96, 1.34) | 0.02 | |
|
| ||||||||
| Comorbidity ≥1 | ||||||||
|
| ||||||||
| Yes | 153 | 1.13 (0.87, 1.46) | 98 | 1.00 | 190 | 1.43 (1.12, 1.83) | 0.03 | 0.45 |
|
| ||||||||
| No | 353 | 0.94 (0.80, 1.10) | 284 | 1.00 | 443 | 1.07 (0.92, 1.24) | 0.07 | |
|
| ||||||||
| Hypertension | ||||||||
|
| ||||||||
| Yes | 221 | 0.89 (0.73, 1.09) | 177 | 1.00 | 278 | 1.08 (0.89, 1.31) | 0.03 | 0.68 |
|
| ||||||||
| No | 285 | 1.09 (0.91, 1.31) | 205 | 1.00 | 355 | 1.25 (1.05,1.48) | 0.08 | |
|
| ||||||||
| Shift work | ||||||||
|
| ||||||||
| Yes | 187 | 0.95 (0.75, 1.19) | 128 | 1.00 | 222 | 1.13 (0.91, 1.41) | 0.08 | 0.98 |
|
| ||||||||
| No | 319 | 1.01 (0.85,1.19) | 254 | 1.00 | 411 | 1.17 (1.00, 1.37) | 0.04 | |
|
| ||||||||
| Snoring | ||||||||
|
| ||||||||
| Yes | 435 | 1.02 (0.88, 1.17) | 329 | 1.00 | 554 | 1.18 (1.02, 1.35) | 0.02 | 0.99 |
| No | 71 | 0.91 (0.62, 1.32) | 53 | 1.00 | 79 | 1.09 (0.76, 1.56) | 0.28 | |
Adjusted for age, BMI, energy intake, smoking (pack-years), drinking (drinks per day), current tea drinking, exercise (met-hours/year), hypertension, family history of diabetes, shiftwork, snoring, comorbidity, and napping (stratifying variable was not adjusted in the model). Excludes those cases that were diagnosed within 2 years of recruitment.
The current study further evaluated sleep duration and diabetes risk excluding individuals who took naps. As shown in Table 3, the risk pattern between sleep duration and diabetes risk remained, with pfor trend=0.04. Moreover, this association was also observed in men who did not exercise. However, the association was attenuated in men who had comorbidity, hypertension, who did not have nightshift work, or who snored when sleeping among those who had nocturnal sleep only. The interaction between tea drinking and sleeping duration remained, and the point estimate for both short and long sleeping duration changed: HRs were 0.89 (95% CI: 0.64–1.24) for sleep duration <7 hour per day and 1.32 (95% CI: 0.96–1.81) for sleep duration ≥8 hour per day (pfor trend=0.01). Associations between sleep duration and diabetes were also found among those people with unhealthy lifestyles, such as smoking and drinking (pfor trend<0.05). However, in the current study, HR was 1.13 (95% CI: 0.96–1.32) among non-exercisers with sleep duration ≥8 hours/day and this association remained among nocturnal sleepers [HR=1.28 (95% CI: 1.03–1.59), pfor trend=0.04]. An association between sleep duration and diabetes was found among men who snored (pfor trend=0.02) in the current study, but this association was attenuated among nocturnal sleepers only (pfor trend=0.07). No association between sleep duration and diabetes was found among men who did not snore. The analysis was also performed among participants who did not have shift work, and association patterns among them were similar with the data of nocturnal sleepers (data not shown).
Table 3.
Association of nocturnal sleep duration and diabetes incidence by selected lifestyles.*
| Nocturnal sleep duration | <7 hours | 7–7.9 hours | ≥8 hours | plinear | pinteraction | |||
|---|---|---|---|---|---|---|---|---|
| n=6553 | n=4940 | n=7158 | ||||||
|
|
|
|
||||||
| cases | HR (95% CI) | cases | HR | cases | HR (95% CI) | |||
| Total | 264 | 1.01 (0.84, 1.22) | 192 | 1.00 | 310 | 1.20 (1.00, 1.44) | 0.04 | |
|
| ||||||||
| Age (years) | ||||||||
|
| ||||||||
| 40- | 103 | 0.95 (0.70, 1.29) | 77 | 1.00 | 134 | 1.25 (0.93, 1.66) | 0.04 | 0.43 |
|
| ||||||||
| 50- | 84 | 0.96 (0.69, 1.34) | 64 | 1.00 | 103 | 1.19 (0.87, 1.64) | 0.15 | |
|
| ||||||||
| 60- | 77 | 1.08 (0.76, 1.56) | 51 | 1.00 | 73 | 1.25 (0.87, 1.80) | 0.43 | |
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 180 | 1.08 (0.86, 1.36) | 127 | 1.00 | 206 | 1.16 (0.92, 1.45) | 0.48 | 0.05 |
|
| ||||||||
| No | 84 | 0.89 (0.64, 1.24) | 65 | 1.00 | 104 | 1.32 (0.96, 1.81) | 0.01 | |
|
| ||||||||
| Regular exercise participation | ||||||||
|
| ||||||||
| Yes | 83 | 0.95 (0.67, 1.34) | 59 | 1.00 | 87 | 1.08 (0.77, 1.52) | 0.41 | 0.45 |
|
| ||||||||
| No | 181 | 1.05 (0.84, 1.32) | 133 | 1.00 | 223 | 1.28 (1.03, 1.59) | 0.04 | |
|
| ||||||||
| Current smokers | ||||||||
|
| ||||||||
| Yes | 152 | 0.93 (0.73, 1.18) | 125 | 1.00 | 196 | 1.18 (0.94, 1.49) | 0.02 | 0.27 |
|
| ||||||||
| No | 112 | 1.12 (0.82, 1.52) | 67 | 1.00 | 114 | 1.24 (0.91, 1.68) | 0.49 | |
|
| ||||||||
| Ever alcohol drank | ||||||||
|
| ||||||||
| Yes | 92 | 0.83 (0.59, 1.16) | 58 | 1.00 | 110 | 1.14 (0.82, 1.57) | 0.03 | 0.32 |
|
| ||||||||
| No | 172 | 1.08 (0.86, 1.36) | 134 | 1.00 | 200 | 1.21 (0.97, 1.50) | 0.29 | |
|
| ||||||||
| BMI | ||||||||
|
| ||||||||
| <25 (kg/m2) | 117 | 1.20 (0.90, 1.62) | 76 | 1.00 | 141 | 1.44 (1.08, 1.91) | 0.15 | 0.98 |
|
| ||||||||
| ≥25 (kg/m2) | 147 | 0.91 (0.71, 1.16) | 116 | 1.00 | 169 | 1.13 (0.89, 1.43) | 0.06 | |
|
| ||||||||
| Comorbidity ≥1 | ||||||||
|
| ||||||||
| Yes | 74 | 1.04 (0.71, 1.52) | 48 | 1.00 | 83 | 1.36 (0.95, 1.96) | 0.10 | 0.92 |
|
| ||||||||
| No | 190 | 0.98 (0.79, 1.22) | 144 | 1.00 | 227 | 1.17 (0.94, 1.44) | 0.08 | |
|
| ||||||||
| Hypertension | ||||||||
|
| ||||||||
| Yes | 106 | 1.00 (0.74, 1.35) | 80 | 1.00 | 115 | 1.16 (0.87, 1.55) | 0.29 | 0.61 |
|
| ||||||||
| No | 158 | 0.99 (0.77, 1.26) | 112 | 1.00 | 195 | 1.21 (0.96, 1.53) | 0.06 | |
|
| ||||||||
| Shift work | ||||||||
|
| ||||||||
| Yes | 109 | 1.00 (0.73, 1.38) | 63 | 1.00 | 115 | 1.26 (0.92, 1.74) | 0.09 | 0.83 |
|
| ||||||||
| No | 155 | 0.99 (0.78, 1.25) | 129 | 1.00 | 195 | 1.16 (0.93, 1.45) | 0.14 | |
|
| ||||||||
| Snoring | ||||||||
|
| ||||||||
| Yes | 224 | 1.00 (0.82, 1.22) | 172 | 1.00 | 269 | 1.18 (0.97, 1.43) | 0.07 | 0.92 |
|
| ||||||||
| No | 40 | 1.17 (0.65, 2.10) | 20 | 1.00 | 41 | 1.52 (0.87, 2.64) | 0.25 | |
Adjusted for age, BMI, energy intake, smoking (pack-years), drinking (drinks per day), current tea drinking, exercise (met-hours/year), hypertension, family history of diabetes, shiftwork, snoring, and comorbidity (stratifying variable was not adjusted in the model). Excludes those cases that were diagnosed within 2 years of recruitment.
Since age, lifestyle, health condition, BMI and snoring are sleep-related factors, and these factors may influence the modification effect of tea consumption, the current study calculated ‘risk score’ by sum of the total number of sleep-related factors for each participant, and evaluated the possible modifying effect of tea consumption stratified by age and ‘risk score’ (Table 4). No associations of sleep duration and diabetes were found among participants aged <60 years or with low/moderate risk scores. However, sleep duration was associated with diabetes among non-tea drinkers aged ≥ 60 years: HRs were 0.78 (95% CI: 0.55–1.10) for sleep duration <7 hours per day and 1.24 (95% CI: 0.88–1.74) for sleep duration ≥8 hours per day (pfor trend<0.01). This association was also found among non-tea drinkers with high ‘risk score’ (pfor trend<0.01). These associations were modified by tea consumption (pfor interaction<0.01 and 0.03, respectively).
Table 4.
Modification effect of tea intake on association of sleep duration and diabetes incidence by age and risk score.*
| Sleep duration | <7 hours | 7–7.9 hours | ≥8 hours | plinear | pinteraction | |||
|---|---|---|---|---|---|---|---|---|
| n=7351 | n=6213 | n=9185 | ||||||
|
|
|
|
||||||
| cases | HR CI (95%) | cases | HR | cases | HR CI (95%) | |||
| Age <60 (years) | ||||||||
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 241 | 0.97 (0.80,1.17) | 190 | 1.00 | 297 | 1.08 (0.89, 1.30) | 0.22 | 0.37 |
|
| ||||||||
| No | 91 | 1.01 (0.74,1.39) | 71 | 1.00 | 147 | 1.26 (0.94, 1.69) | 0.10 | |
|
| ||||||||
| Age ≥ 60 (years) | ||||||||
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 98 | 1.34 (0.97,1.85) | 60 | 1.00 | 109 | 1.18 (0.86, 1.63) | 0.41 | <0.01 |
|
| ||||||||
| No | 76 | 0.78 (0.55,1.10) | 61 | 1.00 | 80 | 1.24 (0.88, 1.74) | <0.01 | |
|
| ||||||||
| Low risk (score: 0–2) | ||||||||
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 82 | 1.08 (0.75, 1.56) | 50 | 1.00 | 78 | 1.20 (0.84, 1.73) | 0.54 | 0.43 |
|
| ||||||||
| No | 50 | 1.06 (0.65, 1.72) | 27 | 1.00 | 54 | 1.49 (0.92, 2.41) | 0.10 | |
|
| ||||||||
| Moderate risk (score: 3–4) | ||||||||
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 132 | 1.07 (0.82, 1.39) | 102 | 1.00 | 182 | 1.14 (0.89, 1.46) | 0.54 | 0.34 |
|
| ||||||||
| No | 67 | 0.95 (0.66, 1.37) | 62 | 1.00 | 95 | 1.22 (0.87, 1.72) | 0.12 | |
|
| ||||||||
| High (score: >4) risk | ||||||||
|
| ||||||||
| Currently drink tea | ||||||||
|
| ||||||||
| Yes | 125 | 1.00 (0.77, 1.31) | 98 | 1.00 | 146 | 1.08 (0.84,1.41) | 0.51 | 0.03 |
|
| ||||||||
| No | 50 | 0.68 (0.45, 1.04) | 43 | 1.00 | 78 | 1.17 (0.80, 1.72) | <0.01 | |
Adjusted for age, BMI, energy intake, smoking (pack-years), drinking (drinks per day), current tea drinking, exercise (met-hours/year), hypertension, family history of diabetes, shiftwork, snoring, and comorbidity (stratifying variable was not adjusted in the model). Excludes those cases that were diagnosed within 2 years of recruitment.
DISCUSSION
In this population-based prospective cohort study it was found that sleep duration was positively associated with diabetes risk, particularly among men who had ever drank alcohol or smoked regularly, who never drank tea, who had a high BMI, comorbidity, men who did not work nightshift or who snored. The association pattern remained and some point estimates increased when participants who napped during the day were excluded. A significant multiplicative interaction between tea consumption and sleep duration on diabetes risk was observed among participants aged ≥60 years or with a high-risk score.
The association between sleep duration and diabetes is inconsistent in the literature. In the HIPOP-OHP study, a prospective cohort study conducted in Japan with a median of 4.2 years follow-up time, a null association between risk of developing diabetes and sleep duration was found [13]. A report from 23,620 middle-aged participants of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study reported a similar result: sleep duration was not associated with the risk of diabetes [12]. However, a literature review that included 36 cohort studies (1,061,555 participants) reported pooled RRs for sleep ≤5 hours and ≥9 hours per day of 1.48 (95% CI: 1.25–1.76) and 1.36 (1.12–1.65), respectively, and concluded that the risk of developing diabetes associated with sleep duration is comparable to that of traditional risk factors [16]. A recent cross-sectional study of 18,121 Chinese participants aged ≥40 years found that only long sleep duration (>9 hours per day), not short duration, was associated with high HbA1c, impaired glucose regulation (IGR), and diabetes in China [17]. A cohort study conducted in China also reported similar findings [18]. The results from the current study are in line with those findings.
It is worth mentioning that the positive sleep and diabetes association persisted among those only sleeping at night, presumably for whom there was a better exposure assessment. In addition, to minimize the potential for reverse causation, diabetes cases that were diagnosed within 2 years of follow-up, along with all first 2-year observations were excluded. Further excluding the first 5 years of observation showed a similar risk pattern, although some of the statistics lost significance due to the reduced statistical power (data not shown).
The mechanisms underlying the association between sleep duration and diabetes risk is unclear. One possible explanation of the association is that participants with longer sleep duration were more likely to snore, compared with short sleepers. This sleep problem could have played a part in the association [10,19]. In the current study, proportions of snoring and daytime napping in long sleepers was 45.1% and 50.9%, respectively, and significantly higher than in short sleepers, which is consistent with the data in the Finish diabetes prevention study [20]. Thus, long sleep duration could also be reflective of sleep problems, such as sleep-disordered breathing, which were associated with insulin resistance and diabetes in the current study. Another possible explanation is that sleep duration is highly correlated with obesity, the latter of which has been found to have deleterious effects on both glucose homeostasis and ß-cell function, and with inducing insulin resistance [21]. Both BMI and WHR of long sleepers were significantly higher than in short sleepers in the current study, suggesting that obesity could be a mediation factor between sleep duration and diabetes risk.
Tea is the most widely consumed non-alcoholic beverage in the world, with especially high consumption in Asian countries [22]. Several studies have reported the beneficial effects of tea, not only on cardiovascular diseases but also on diabetes and other chronic diseases [23]. A retrospective study performed in Japan found a 33% risk reduction of developing diabetes among participants who consumed six or more cups of green tea daily, compared to those consuming less than one cup per week [24]. A big cohort study conducted in eight European countries reported that people who drink at least four cups of tea per day might have a 16% lower risk of developing diabetes than non-tea drinkers [25]. However, a limited number of clinical trials using tea and tea extracts, or its main ingredient catechin, have shown inconsistent results in controlling hyperglycemia in diabetes patients. MacKenzie et al. showed no significant difference in glucose control after 3 months of ingestion of decaffeinated green tea extracts in diabetes patients [26]. However a recent study concluded that chamomile tea has some beneficial effects on glycemic control and serum lipid profile in T2D patients [27].
The effect of tea on sleep duration and diabetes risk is complicated. On one hand, tea consumption, especially green tea, has been shown to be associated with a significantly reduced risk of diabetes [28], due to its high flavonoid content, which may act as an antioxidant and anti-inflammatory agent to reduce the oxidative stress and inflammation that is often associated with the progressive impairment of pancreatic b-cell function in the development of diabetes [29–30]. On the other hand, tea consumption is negatively associated with sleep duration, due to its caffeine content, which can increase sleep onset latency and decrease total sleep time [31]. These functions could attenuate the association between sleep duration and diabetes risk among tea drinkers. The current study found that sleep duration was only associated with risk of diabetes among men who were not tea drinkers. The interaction effect of tea consumption and sleep duration on diabetes risk was statistically significant.
In studies published since 1990, current smokers were found to have a 1.2–2.6 times higher risk of diabetes than nonsmokers [32–33]. Smoking not only impairs insulin action in both diabetes patients and healthy people, through enhancing ethanol-induced pancreatic injury, but also increases lipase activity, which is linked with increased insulin resistance [34–35]. Several epidemiological studies have reported obesity as a key player in diabetes development and an elevated FFA level as a risk marker for long-term development of glucose intolerance and progression to diabetes [36]. Drinking is also a complex factor in the association with diabetes: moderate alcohol consumption reduces the risk of diabetes by 20–40% [37], while high alcohol consumption is associated with higher diabetes risk [38]. Meanwhile, smoking, drinking and obesity are related to the quality of sleep [2,4,6], and thus could modify the association between sleep duration and diabetes risk. In the current study, however, no modification effect of these factors was found on the association.
Day napping is a more common lifestyle habit in China than in other countries [39]. The prevalence of habitual day napping has been reported to be 61.7% among Chinese men aged ≥60 years [40]. In the current study, the percentage of napping was 53.6%. Picarsic et al. observed no difference between nappers and non-nappers in terms of reported nighttime sleep duration, sleep-onset latency, or sleep efficiency [41]. However, the current study found that the proportion of nappers was higher among participants who slept ≥9 hours per day compared with those who slept <6 hours (56.1% vs 42.8%, p<0.001), and napping was associated with diabetes risk. The current data were in line with another prospective study in which it was reported that duration of day napping from 1996–1997 was associated with a higher risk of diabetes after the year 2000 in a dose-response manner [42]. And among participants who reported no napping, only short sleepers had a higher risk of diabetes, whereas among those napping >1 hour/day, both long and short sleeping was associated with higher risk of diabetes. This evidence suggests that napping could be an independent risk factor for diabetes, and napping habits may modify the association between nocturnal sleep duration and diabetes [42]. However, the current study also detected a relationship between night sleep duration and diabetes risk in the group of men who had only nighttime sleep. There was no difference in association of sleep duration with risk of diabetes between all participants and participants who slept only at night.
Snoring is a sleep problem, and a common and early symptom of obstructive sleep apnea [43]. Snoring affected 43.7% men in the current study, which is a similar proportion reported by Stradling [44]. The current study found an association between sleep duration and diabetes risk (OR=1.23. 95% CI: 1.01–1.49) among snorers with sleep duration ≥9 hours per day, but this association was not found among non-snorers. A recent study of 6522 participants aged ≥20 years from the National Health and Nutrition Examination Survey 2005–2008 reported that snoring was associated with diabetes risk with OR of 1.44 (95% CI: 1.16–1.79), an association that was also found in a meta-analysis [45–46]. The mechanisms underlying this association have not been elucidated completely. It could be that nocturnal intermittent hypoxia and hypercapnia contribute to increased sympathetic nervous activity and increased oxidative stress, which ultimately lead to insulin resistance, or relate to activation of pro-inflammatory cytokine production [47]. The current study found no reports on the relation of snoring and sleep duration to diabetes risk. However, snoring and longer sleep duration could have an additive effect on diabetes risk among the Chinese population. Further studies need to be conducted on this topic.
Strengths of the current study include its large sample size, and prospective and population-based design. In addition, a wide range of factors related to sleep duration and quality were incorporated into the study. The study also had several limitations. First, the information about sleep duration was self-reported, which may have lead to misclassification. However, a reasonable agreement between self-reported and actigraph-measured sleep duration has been previously reported [48]. Second, although snoring was included in the study, there were no other direct measurements of quality of sleep, such as obstructive sleep apnea or insomnia, which could have provided more information to help understand the association between sleep duration and risk of diabetes. Third, due to lack of information on length of afternoon napping time, the current study was unable to evaluate the relationship between napping duration and diabetes risk.
In conclusion, the current study found that sleeping duration was positively associated with diabetes risk, particularly among overweight individuals, but also those with comorbidity, hypertension, and snoring. The results suggest that targeting these higher-risk individuals for more rigorous diabetes screening and preventive interventions could provide a public health benefit.
Highlights.
The Shanghai Men’s Health Study, a large population-based cohort study, was used to assess association between sleep duration and risk of type 2 diabetes incidence. The study found a J-shaped association between sleep duration and diabetes risk (ie, higher risk of diabetes was associated with longer sleep duration). This association was also observed among people who already had factors linked to poor quality of sleep, such as smoking, drinking, obesity, hypertension and comorbidity. Tea drinking could modify this association. The results suggest that targeting these higher-risk individuals for more rigorous diabetes screening and prevention could provide public health benefits.
Acknowledgments
This work was supported by grants from the US National Institutes of Health and the Intramural Research Program of the National Institutes of Health, National Cancer Institute (R01 CA082729 and UM1 CA173640 to Xiao-Ou Shu) and partially by grands from the State Key Project Specialized for infectious Disease, China (2008ZX10002-015 and 2012ZX10002008-002 to Yong-Bing Xiang). The authors thank all participants and staff members of the Shanghai Men’s Health Study for their important contributions and Ms. Nan Kennedy for editing the manuscript.
Footnotes
The authors have no conflicts of interest to disclose.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.US-Sleep-Foundation. Sleep in America poll: Summary of findings. 2005 http://www.sleepfoundation.org/sites/default/files/2005_summary_of_findings.pdf.
- 2.Tu X, Cai H, Gao YT, et al. Sleep duration and its correlates in middle-aged and elderly Chinese women: the Shanghai Women's Health Study. Sleep Med. 2012;13:1138–45. doi: 10.1016/j.sleep.2012.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cappuccio FP, D’Elia L, Strazzullo P, Miller MA. Quantity and quality of sleep and incidence of type 2 diabetes: a systematic review and meta-analysis. Diabetes Care. 2010;33:414–20. doi: 10.2337/dc09-1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cappuccio FP, Cooper D, D’Elia L, Strazzullo P, Miller MA. Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. Eur Heart J. 2011;32:1484–92. doi: 10.1093/eurheartj/ehr007. [DOI] [PubMed] [Google Scholar]
- 5.Blask DE. Melatonin, sleep disturbance and cancer risk. Sleep Med Rev. 2009;13:257–64. doi: 10.1016/j.smrv.2008.07.007. [DOI] [PubMed] [Google Scholar]
- 6.Kachi Yuko, Ohwaki Kazuhiro, Yano Eiji. Association of sleep duration with untreated diabetes in Japanese men. Sleep Med. 2012;13:307–9. doi: 10.1016/j.sleep.2011.04.008. [DOI] [PubMed] [Google Scholar]
- 7.Gangwisch JE, Heymsfield SB, Boden-Albala B, et al. Sleep duration as a risk factor for diabetes incidence in a large U.S. sample. Sleep. 2007;30:1667–73. doi: 10.1093/sleep/30.12.1667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gottlieb DJ, Punjabi NM, Newman AB, et al. Association of sleep time with diabetes mellitus and impaired glucose tolerance. Arch Intern Med. 2005;165:863–7. doi: 10.1001/archinte.165.8.863. [DOI] [PubMed] [Google Scholar]
- 9.Tuomilehto H, Peltonen M, Partinen M, et al. Sleep duration is associated with an increased risk for the prevalence of type 2 diabetes in middle-aged women - The FIN-D2D survey. Sleep Med. 2008;9:221–7. doi: 10.1016/j.sleep.2007.04.015. [DOI] [PubMed] [Google Scholar]
- 10.Mallon L, Broman JE, Hetta 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;28:2762–7. doi: 10.2337/diacare.28.11.2762. [DOI] [PubMed] [Google Scholar]
- 11.Yaggi HK, Araujo AB, McKinlay JB. Sleep duration as a risk factor for the development of type 2 diabetes. Diabetes Care. 2006;29:657–61. doi: 10.2337/diacare.29.03.06.dc05-0879. [DOI] [PubMed] [Google Scholar]
- 12.Von Ruesten A, Weikert C, Fietze I, Boeing H. Association of Sleep Duration with Chronic Diseases in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study. PloS ONE. 2012;7:e30972. doi: 10.1371/journal.pone.0030972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hayashino Y, Fukuhara S, Suzukamo Y, et al. Relation between sleep quality and quantity, quality of life, and risk of developing diabetes in healthy workers in Japan: the high-risk and population strategy for occupational health promotion (HIPOP-OHP) study. BMC Public Health. 2007;7:129–37. doi: 10.1186/1471-2458-7-129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ferdinand ZZ, Abhishek P, Renee MB, et al. Race/Ethnicity, Sleep Duration, and Diabetes Mellitus: Analysis of the National Health Interview Survey (NHIS) Am J Med. 2012;125:162–7. doi: 10.1016/j.amjmed.2011.08.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shadyab AH, Kritz-Silverstein D, Laughlin GA, Wooten WJ, Barrett-Connor E, Araneta MR. Ethnic-specific associations of sleep duration and daytime napping with prevalent type 2 diabetes in postmenopausal women. Sleep Med. 2015;16:243–9. doi: 10.1016/j.sleep.2014.11.010. [DOI] [PubMed] [Google Scholar]
- 16.Anothaisintawee T, Reutrakul S, Van Cauter E, Thakkinstian A. Sleep disturbances compared to traditional risk factors for diabetes development: Systematic review and meta-analysis. Sleep Med Rev. 2015;30:11–24. doi: 10.1016/j.smrv.2015.10.002. [DOI] [PubMed] [Google Scholar]
- 17.Zheng Y, Wang A, Pan C, et al. Impact of night sleep duration on glycemic and triglyceride levels in Chinese with different glycemic status. J Diabetes. 2015;7:24–30. doi: 10.1111/1753-0407.12186. [DOI] [PubMed] [Google Scholar]
- 18.Han X, Liu B, Wang J, et al. Long sleep duration and afternoon napping are associated with higher risk of incident diabetes in middle-aged and older Chinese: the Dongfeng-Tongji cohort study. Ann Med. 2016;48:216–23. doi: 10.3109/07853890.2016.1155229. [DOI] [PubMed] [Google Scholar]
- 19.Resnick HE, Redline S, Shahar E, et al. Diabetes and sleep disturbances: findings from the Sleep Heart Health Study. Diabetes Care. 2003;26:702–9. doi: 10.2337/diacare.26.3.702. [DOI] [PubMed] [Google Scholar]
- 20.Tuomilehto H, Peltonen M, Partinen M, et al. Sleep duration, lifestyle intervention, and incidence of type 2 diabetes in impaired glucose tolerance: The Finnish Diabetes Prevention Study. Diabetes Care. 2009;32:1965–71. doi: 10.2337/dc08-1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Greenberg AS, McDaniel ML. Identifying the links between obesity, insulin resistance and beta-cell function: potential role of adipocyte-derived cytokines in the pathogenesis of type 2 diabetes. Eur J Clin Invest. 2002;32(Suppl 3):24–34. doi: 10.1046/j.1365-2362.32.s3.4.x. [DOI] [PubMed] [Google Scholar]
- 22.Graham HN. Green tea composition, consumption, and polyphenol chemistry. Prev Med. 1992;21:334–50. doi: 10.1016/0091-7435(92)90041-f. [DOI] [PubMed] [Google Scholar]
- 23.Wierzejska R. Tea and health--a review of the current state of knowledge. Przegl Epidemiol. 2014;68:501–6. [PubMed] [Google Scholar]
- 24.Iso H, Date C, Wakai K, Fukui M, Tamakoshi A JACC Study Group. The relationship between green tea and total caffeine intake and risk for self-reported type 2 diabetes among Japanese adults. Ann Intern Med. 2006;144:554–62. doi: 10.7326/0003-4819-144-8-200604180-00005. [DOI] [PubMed] [Google Scholar]
- 25.Rafraf M, Zemestani M, Asghari-Jafarabadi M. Effectiveness of chamomile tea on glycemic control and serum lipid profile in patients with type 2 diabetes. J Endocrinol Invest. 2015;38:163–70. doi: 10.1007/s40618-014-0170-x. [DOI] [PubMed] [Google Scholar]
- 26.Mackenzie T, Leary L, Brooks WB. The effect of an extract of green and black tea on glucose control in adults with type 2 diabetes mellitus: double–blind randomized study. Metabolism. 2007;56:1340–4. doi: 10.1016/j.metabol.2007.05.018. [DOI] [PubMed] [Google Scholar]
- 27.Zemestani M, Rafraf M, Asghari-Jafarabadi M. Chamomile tea improves glycemic indices and antioxidants status in patients with type 2 diabetes mellitus. Nutrition. 2016;32:66–72. doi: 10.1016/j.nut.2015.07.011. [DOI] [PubMed] [Google Scholar]
- 28.Bhatti SK, O'Keefe JH, Lavie CJ. Coffee and tea: perks for health and longevity? Curr Opin Clin Nutr Metab Care. 2013;16:688–97. doi: 10.1097/MCO.0b013e328365b9a0. [DOI] [PubMed] [Google Scholar]
- 29.Rietveld A, Wiseman S. Antioxidant effects of tea: evidence from human clinical trials. J Nutr. 2003;133:3285S–92S. doi: 10.1093/jn/133.10.3285S. [DOI] [PubMed] [Google Scholar]
- 30.Steptoe A, Gibson EL, Vuononvirta R, et al. The effects of chronic tea intake on platelet activation and inflammation: a double-blind placebo controlled trial. Atherosclerosis. 2007;193:277–82. doi: 10.1016/j.atherosclerosis.2006.08.054. [DOI] [PubMed] [Google Scholar]
- 31.Hindmarch I, Rigney U, Stanley N, Quinlan P, Rycroft J, Lane J. A naturalistic investigation of the effects of the day-long consumption of tea, coffee and water on alertness, sleep onset and sleep quality. Psychopharmacology. 2000;149:203–16. doi: 10.1007/s002130000383. [DOI] [PubMed] [Google Scholar]
- 32.Shi L, Shu XO, Li H, et al. Physical activity, smoking, and alcohol consumption in association with incidence of type 2 diabetes among middle-aged and elderly Chinese men. PLoS One. 2013;8:e77919. doi: 10.1371/journal.pone.0077919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Patja K, Jousilahti P, Hu G, Valle T, Qiao Q, Tuomilehto J. Effects of smoking, obesity and physical activity on the risk of type 2 diabetes in middle-aged Finnish men and women. J Intern Med. 2005;258:356–62. doi: 10.1111/j.1365-2796.2005.01545.x. [DOI] [PubMed] [Google Scholar]
- 34.Kong C, Nimmo L, Elatrozy T, et al. Smoking is associated with increased hepatic lipase activity, insulin resistance, dyslipidaemia and early atherosclerosis in type 2 diabetes. Atherosclerosis. 2001;156:373–8. doi: 10.1016/s0021-9150(00)00664-x. [DOI] [PubMed] [Google Scholar]
- 35.Hartwig W, Werner J, Ryschich E, et al. Cigarette smoke enhances ethanol-induced pancreatic injury. Pancreas. 2000;21:272–8. doi: 10.1097/00006676-200010000-00009. [DOI] [PubMed] [Google Scholar]
- 36.Jung CH, Kang YM, Jang JE, et al. Fatty liver index is a risk determinant of incident type 2 diabetes in a metabolically healthy population with obesity. Obesity. 2016;24:1373–9. doi: 10.1002/oby.21483. [DOI] [PubMed] [Google Scholar]
- 37.Koppes LL, Dekker JM, Hendriks HF, Bouter LM, Heine RJ. Moderate alcohol consumption lowers the risk of type 2 diabetes: a meta-analysis of prospective observational studies. Diabetes Care. 2005;28:719–25. doi: 10.2337/diacare.28.3.719. [DOI] [PubMed] [Google Scholar]
- 38.Cullmann M, Hilding A, Ostenson CG. Alcohol consumption and risk of pre-diabetes and type 2 diabetes development in a Swedish population. Diabet Med. 2012;29:441–52. doi: 10.1111/j.1464-5491.2011.03450.x. [DOI] [PubMed] [Google Scholar]
- 39.Soldatos CR, Allaert FA, Ohta T, Dikeos DG. How do individuals sleep around the world? Results from a single–day survey in ten countries. Sleep medicine. 2005;6:5–13. doi: 10.1016/j.sleep.2004.10.006. [DOI] [PubMed] [Google Scholar]
- 40.Liu L, Tang J, Liu X, Liu Z, Zhao G, Wang G. Sleep behavior of the urban elderly: A cross-sectional study. Chin J Behav Med Sci. 2002;11:310–13. [Google Scholar]
- 41.Picarsic JL, Glynn NW, Taylor CA, et al. Self-Reported Napping and Duration and Quality of Sleep in the Lifestyle Interventions and Independence for Elders Pilot Study. J Am Geriatr Soc. 2008;56:1674–80. doi: 10.1111/j.1532-5415.2008.01838.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Xu Q, Song Y, Hollenbeck A, Blair A, Schatzkin A, Chen H. Day napping and short night sleeping are associated with higher risk of diabetes in older adults. Diabetes Care. 2010;33:78–83. doi: 10.2337/dc09-1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea: a population health perspective. Am J Respir Crit Care Med. 2002;165:1217–39. doi: 10.1164/rccm.2109080. [DOI] [PubMed] [Google Scholar]
- 44.Stradling JR, Crosby JH. Predictors and prevalence of obstructive sleep apnoea and snoring in 1001 middle aged men. Thorax. 1991;46:85–90. doi: 10.1136/thx.46.2.85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Sabanayagam C, Teppala S, Shankar A. Markers of Sleep Disordered Breathing and Diabetes Mellitus in a Multiethnic Sample of US Adults: Results from the National Health and Nutrition Examination Survey (2005–2008) Int J Endocrinol. 2012:879134. doi: 10.1155/2012/879134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Xiong X, Zhong A, Xu H, Wang C. Association between Self-Reported Habitual Snoring and Diabetes Mellitus: A Systemic Review and Meta-Analysis. J Diabetes Res. 2016:1958981. doi: 10.1155/2016/1958981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Iiyori N, Alonso LC, Li J, et al. Intermittent hypoxia causes insulin resistance in lean mice independent of autonomic activity. Am J Respir Crit Care Med. 2007;175:851–7. doi: 10.1164/rccm.200610-1527OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lockley SW, Skene DJ, Arendt J. Comparison between subjective and actigraphic measurement of sleep and sleep rhythms. J Sleep Res. 1999;8:175–83. doi: 10.1046/j.1365-2869.1999.00155.x. [DOI] [PubMed] [Google Scholar]