Prevalence of Sleep Disturbance and correlates in Chinese Older Adults with type 2 diabetes

Objective To examine the prevalence of sleep disturbances, the socio-demographic and clinical correlates in Chinese older adults with type 2 diabetes (T2DM). Methods A cross-sectional survey of 195 patients with T2DM was conducted using standardized instruments. The prevalence of any type of sleep disturbances was 38.5%, the incidence of DIS, DMS, and EMA were 23.6%, 25.6% and 26.2%, respectively. No statistically signicant difference was observed in socio-demographic data between participants with and without sleep disturbances except age. Sleep disturbances were associated with more severe depressive symptoms and lower quality of life (QOL). Only 32% of patients suffering sleep disturbance received treatment. Sleep disturbance is common in Chinese older patients with T2DM, but the rate of treatment appears low. More aggressive measures should be implemented to improve the assessment and treatment of sleep disturbances in patients with T2DM.


Results
There are 195 older adults participated in the study with a response rate of 100%. The frequency of at least one type of sleep disturbance, DIS, DMS, and EMA was 38.5%, 23.6%, 25.6% and 26.2%, respectively. The ratio of one, two, and three types of sleep disturbances were 15.9%, 8.2%, and 14.4%, respectively. Only 32% of subjects suffering sleep disturbance received treatment.
Demographic and Clinical Characteristics of the sample were displayed in Table 1. No statistically signi cant difference was observed in socio-demographic data between participants with and without sleep disturbances except age. Further test indicated that percentage of 70s age group was different signi cantly from the others at the 0.05 level.
The interrater reliability of the judgement of sleep disturbances between the interviewers obtained in examining 20 subjects was satisfactory (kappa values > 0.9).
All subjects were recruited if they were aged 50 years or above, diagnosed as diabetes by endocrinologists, and were able to communicate adequately and comprehend the purpose of the study. In this research, subjects aged 50 years or above were classi ed as "older adults." The same cutoff point was set in other research too [18].

Data Collection
Sleep disturbance during the past month was examined targeting three basic forms: "Do you have di culties in falling sleep?" for di culty initiating sleep (DIS); "Do you have di culties in maintaining sleep and wake up often?" for di culty maintaining sleep (DMS); and "Do you wake up in the midnight or early morning and have di culties in falling sleep again?" for early morning awakening (EMA). At least one "often" answer to the three questions quali ed the participant as "having sleep disturbance." Treatment for sleep disturbances during the past month was also recorded. This assessment of sleep disturbances has also been used in other studies [19,20].
Depressive symptom was measured using Montgomery-Asberg Depression Rating Scale (MADRS) [21]. MADRS is an interviewer-rated instrument, comprising 10 items, each measured on a 6-point scale (scores range from 0-60 with higher scores depicting greater symptom severity).
Quality of Life(QOL)was measured with the Chinese version of the World Health Organization Quality of Life-BREF (WHOQOL-BREF) [22]. The WHOQOL-BREF includes 26 items covering four domains: physical health, psychological health, social relationships, and environmental factors. A higher score re ects better QOL.

Statistical analysis
Database was setup via the software of EPIData, and the operation about data input was carried by two persons. Then the database was transferred to IBM® SPSS® Statistics Version 24 (IBM SPSS Statistics for Windows, IBM Corp, Armonk, NY). The signi cance level was set at 0.05, two-tailed Student's t tests were employed to compare continuous data between interviewees with and without different type of sleeping disturbance, and Chi square tests were implemented in the nominal and ordinal factors to verify the differences of subsets grouped from the original sample. Con dential intervals of 95% were calculated in the procedure of binomial tests.
All categorical factors were converted to dummy variables, indicators were speci ed as following: male, 50s-aged, living with marital spouse, having a religious faith, earning monthly less than or equal to 3000 RMB, illiterate or primary-school level, smoking by now, owning a health insurance. The coe cients of indicators in the model were supposed as zero by default. The pseudo determination coe cients of logistics regressions, Nagelkerke R2, were presented at last.
Quantitative variables and scale scorings were checked by the Shapiro-Wilk test to verify the normality. Chi square tests were performed between subsets with and without any sleep disturbance. Descriptive data are presented as mean values ± SD. The analyses of Binary Multifactor Logistic Regressions were performed to estimate the relative risk power in form of odd ratios (OR) which equalled natural constant to the power of B (eB).  Table 4 showed the independent relationships between each type of sleep disturbance and the socio-demographic and clinical characteristics. Advanced age, more severe depressive symptoms and female independently contributed to sleep disturbance.

Discussion
Within the realm of sleep research, the term "sleep disorder" is more common than the term "sleep disturbance". These two terms are different and should not be used interchangeably. Sleep disorder is diagnosed, using strict criteria, and may be classi ed into distinct categories, such as insomnia and sleep disorder. The term "sleep disturbance" in the context of diabetes was rst mentioned in 1992 [23], mainly used to represent insomnia symptoms, might be universally experienced by anyone at some point in their lives. Sleep disturbance was most commonly characterized by di culty initiating sleep (DIS), di culty maintaining sleep (DMS) [24][25][26][27], and early morning awakening (EMA).
According to an internet survey of over 7 thousand people in the United States with T2D, three-quarters of participants suffered from sleep symptoms, and one-quarter of them were actually diagnosed with a sleep disorder [28]. A study in Ethiopia showed that about 55.6% of diabetic patients were affected by sleep disturbance [29]. Other research revealed that the prevalence of sleep disturbance in people with T2DM ranged from 55-71% [30,31]. In this study, the prevalence of DIS, DMS, EMA and any type sleep disturbance was 23.6%, 25.6%, 26.2% and 38.5%, respectively. The nding of this study was lower than studies above maybe due to the different social-cultural backgrounds. In Chinese society, having mental disorders is a shame [32]. Mental diseases, including sleep disturbance, is considered a stigma for many Chinese, which leads to low medical service utilization and treatment [33]. An epidemiological study of 4596 individuals aged 18 years or above living in Beijing, China found that people believed that most members of their society have negative attitudes towards people with mental disorders [34]. Stigma of mental disorders is a barrier and may prevent people seeking [35,36]. And our study also found that only 32% of patients suffering sleep disturbance received treatment. On the other hand, due to the worry concerns related to drug dependence and side effects, patients often choose non-pharmacological treatments [37]. Acupuncture, as an important component of traditional Chinese medicine (TCM), has been widely applied to treat sleep disturbances [38]. In this study, 15.8% of people in the non-sleep disturbance group used a sleep aid occasionally, including Chinese herbal medicine and acupuncture, to improve their sleep. Maybe that's why some people did not report their sleep disturbances.
Research revealed that sleep disturbances have a negative impact on glucose metabolism [39] and there's a U-shape association between short and long sleep duration in individuals identi ed as risk factors for developing T2DM [8, 40,41]. Sleep duration of less than 6 h or > 9 h is associated with increased cardiometabolic risk in people with T2DM [42]. Previous research also found that people with T2DM with approximately 39% sleeping less than 6.5 h per night [43]. In this study, the average sleep time was 5.7 hours in sleep disturbance group and 6.86 hours in non-sleep disturbance group, suggested that shorter sleep time was more closely related to diabetes. In this trial, we also found that incidence of sleep disturbances increases with age, which is consistent with previous studies [44][45][46].
Depressive symptoms in people with diabetes can have a detrimental impact on engagement with diabetes management, and on glycemic control [47], as well as on health-related outcomes [48]. Adults with diabetes are more likely to report signi cant depressive symptoms than those without diabetes [49]. The estimated incidence of clinically signi cant depressive symptoms among adults with diabetes was 10.6 to 50.6% [49,50]. And the association between depression and sleep disturbance is close [51,52]. As such, the high prevalence of this comorbidity is accompanied by high rates of morbidity and mortality worldwide [53]. In this study, depressive symptom in the sleep disturbance group was more obvious than that of those without sleep disturbance, and the more symptoms of sleep disturbance, the more severe depressive symptoms.
Sleep disturbance is one of the factors impairing QOL in people with T2D [54][55][56][57][58], and posing an increased nancial burden to both individuals and society [59,60]. Diabetes and depression can also seriously affect an individual's quality of life [61]. Reports show that the QOL of people with diabetes is lower than that of people without diabetes [28,[62][63][64][65][66]. In this study, we also found that QOL in the sleep disturbance group was worse than that of those without sleep disturbance, and the more symptoms of sleep disturbance, the worse QOL. Our study also found the aging population will have a negative impact on QOL, which is consistent with previous study [67].

Conclusion
Sleep disturbance is common in Chinese older patients with T2DM, but the rate of treatment appears low. Sleep disturbance is related to depression and decreased quality of life in patients with T2DM. More aggressive measures should be implemented to improve the assessment and treatment of sleep disturbances in patients with T2DM.

Limitations and Future Lines of Research
The present study has several limitations. Firstly, only subjective sleep measure was used in this study, recall bias could affect research results. Objective sleep assessment should be added to future research. Secondly, the small sample size and convenience sampling limited the generalizability of the study ndings. Thirdly, more information related to sleep disturbances, such as blood-glucose levels, complications, should have been collected.

Declarations
Ethics approval and consent were approved by the Research Ethics Committee of the Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences (Reference number: GDREC2018543H (R1)). (Reference number: GDREC2018543H (R1)). Written informed consents were obtained from all participants before the data collection. The participants were assured of the con dentiality of their data.
Con rmation: We con rmed that the study protocol and all methods have been performed in accordance with the relevant guidelines and regulations in the Declaration of Helsinki.
Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no con ict of interest in regard to this work.
Funding: Natural Science Foundation of Guangdong Province (No.: 2018A03013816) supported this study in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Authors' contributions: Hou CL and Gong YX analyzed and interpreted the patient data regarding sleep disturbance and type 2 diabetes. Gong YX was a major contributor in writing the manuscript. FU-Jun Jia designed this study. Huang ZH, Wang F, and Yang JJ were responsible for sample collection and psychological evaluation. Wang SB participated in statistical analysis of data and prepared the tables. All authors read and approved the nal manuscript.
Acknowledgements: Thank the Endocrinology Department of Foshan rst people's Hospital for its strong support for this study.