Association between nighttime-daytime sleep patterns and chronic diseases in Chinese elderly population: a community-based cross-sectional study

Chronic diseases constitute a major challenge to global public health with respect to quality of life and longevity [1]. In China, more than 80% of deaths and 70% of disability-adjusted life-years lost were reported due to chronic diseases in 2008 [2, 3]. Chinese populations are rapidly aging in recent decades, which intensifies the impact of chronic diseases [4, 5]. The number of people diagnosed with chronic diseases increases with age and over the years. It was reported that the proportion of those seniors aged 65 years or older suffering from one or more chronic diseases rose from 86.9 to 92.2% in the US from 1998 to 2009 [6]. In addition, several chronic diseases often occur concurrently in elderly people, which makes the management of these disease more challenging [7]. A large population-based study conducted in China showed significant correlations between chronic diseases, such as hypertension and dyslipidemia, dyslipidemia and diabetes, diabetes and arthritis, which makes the intervention and health promotion more difficult to carry out among the elderly populations [8].

Sleep physiology undergoes significant changes across the lifespan, and the distributions of sleep duration vary with age [9, 10]. Numerous health problems prevalent in the elderly may also influence sleep duration [9]. Thus, the relationship between sleep duration and chronic diseases differs for the elderly and middle-aged adults. Sleep habits may be an important indicator of chronic diseases including diabetes, hypertension, dyslipidemia, coronary heart diseases and cancer, though research findings have been mixed [11‐14]. Previous studies often emphasized on the all-day-long sleep duration with consideration of the distribution of nighttime and daytime [15]. Daytime napping is regarded as one component of a healthy lifestyle, especially in Latin America and Mediterranean countries [16, 17]. In China, daytime napping is a popular sleep behavior. Fang et al. reported that the prevalence of daytime napping was about 68.6% in a middle-aged Chinese population [16]. Further, significant relationship was found between nighttime sleep and daytime napping [18]. Therefore, more research is needed to understand what patterns of nighttime and daytime sleep are healthy, and for whom.

Research design is very critical in exploring the relationship between sleep patterns and chronic diseases [19‐21]. Altered sleep may affect or result from a chronic disease. Elucidating the association between specific sleep patterns, and different chronic diseases, has significant implication in public health and management of chronic diseases. Here, we reported a large community-based cross-sectional study which investigated the relationship between nighttime and daytime sleep patterns and chronic diseases in a Chinese elderly population.

The present cross-sectional study analyzed the prevalence of different chronic diseases by sleep patterns and found some associations between nighttime-daytime sleep patterns and chronic diseases. The prevalence of any of chronic diseases in our study population was 83.8%. CVD had the highest prevalence (64.3%) followed by dyslipidemia (31.3%), diabetes (20.0%), arthritis (14.6%) and cancer (4.3%). We found that compared to those with a sleep pattern of “short nighttime sleep with daytime napping”, those with “long nighttime sleep without daytime napping” had higher prevalence of diabetes and lower prevalence of arthritis. Those with “long nighttime sleep with daytime napping” had higher prevalence of diabetes and lower prevalence of cancer. Then in the habitual nappers, participants were grouped by nighttime-daytime duration patterns, people with “long nighttime sleep with short daytime napping” still had higher prevalence of diabetes, while lower prevalence of any of the chronic disease, cancer and arthritis, and when compared to those with “short nighttime sleep with long daytime napping”.

Inconsistent associations were found between sleep duration and prevalence of diabetes in elderly Chinese. Several cross-sectional and prospective studies suggest a U-shaped association between sleep duration and diabetes [25‐28], but one found no association [11]. The relationship between sleep patterns and cancer was also mixed. A meta-analysis of 13 cohorts suggests a positive association between long sleep duration and colorectal cancer, but an inverse association was seen with hormone related cancers, like in the breast [29]. Other studies have shown that short sleep duration, defined as sleeping ≤6 h per night, may be associated with cancers [30‐32]. For daytime napping, results were also mixed. A meta-analysis of several prospective studies on the relationship between daytime napping and risk of type 2 diabetes (T2D) found a 17% increased risk of T2D when comparing habitual nappers with non-nappers (RR = 1.17, 95% CI 1.08–1.27) [33]. For excess daytime napping, the meta-analysis evaluated a dose-response relationship, and the results suggested an 11% (95% CI 7–16%) increase in T2D risk for each increment in daytime napping of 30 min/day [33].

With the inconsistent relationship between sleep patterns and chronic diseases, it is important to evaluate the association in a more specific way by considering the inter-relationship between nighttime sleep and daytime napping. However, in our study, there was no significant association between total sleep duration and any of chronic diseases or individual chronic diseases. Significant associations were found between different nighttime-daytime sleep patterns and nighttime-daytime sleep duration patterns. Moreover, we found inverse associations between nighttime-daytime sleep duration patterns and diabetes prevalence as well as cancer prevalence. There was limited research exploring the different effect of sleep patterns on the prevalence of individual chronic diseases. A meta-analysis of 12 studies, involving 130,068 subjects, 49,791 nappers, and 19,059 deaths showed that daytime napping was associated with an increased risk of death from all causes, but no significant association was discovered between daytime napping and cancer mortality [34], indicating some differences in these conditions. Different association of nighttime-daytime sleep patterns with diabetes and cancer can be explained by other factors, such as diet. Diabetes is more susceptible than cancer to the effect of eating habits and nutrient intake [35]. Studies have reported that short sleep duration (often ≤6 h/night) is associated with obesity [36] and diabetes [26]. Insufficient sleep and metabolic disruption may affect a person’s diet and metabolism. Experimental studies and clinical trials showed that sleep restriction had unfavorable impacts on the appetite-related hormones which could further influence energy intake and weight, indicating that chronic metabolic abnormality’s link to chronic disease may be explained by specific sleep patterns [37‐40]. For some types of cancer, especially hormone-related cancer, a protective effect of longer nighttime sleep duration was observed. Melatonin had anti-proliferation and anti-angiogenesis effect [41]. Melatonin has been described to be involved in inhibitory influences on sex hormone levels [42], which have been reported to be associated with cancers of the breast, endometrium, ovary, prostate, and thyroid [43]. In addition, melatonin has shown a dose-dependent anti-oxidative effect, acting as a free radical scavenger to provide protection against damage from carcinogenic substance [44]. Melatonin level is positively related to sleep duration [45]. Therefore, longer sleepers may have higher melatonin, which subsequently reduce the risk of certain cancers.

What’s more, we also have some findings on the association between sleep patterns and arthritis. Compare to those with short nighttime sleep and daytime napping/long daytime napping, the pattern “long nighttime sleep without daytime napping/short daytime napping” got both lower proportion of the participants. Results in our study may be consistent with previous researches about the association between arthritis and nighttime sleep duration, which, in details, arthritis was associated with less nighttime sleep [46, 47]. However, little research has focused on the association between arthritis and daytime napping or the sleep pattern combined by nighttime and daytime sleep.

One of the limitations of the present study is that sleep duration was self-reported. Self-reported sleep patterns were susceptible to reporting bias, and objective measures of sleep duration are the best approach. Studying the relationship between subjective and objective measures of sleep, Lauderdale et al. found a moderate correlation between the two measures (r = 0.45) and suggested that new questions be added to the commonly used questions [48]. Another limitation of this study is the lack of a temporal relationship. This is an inherent limitation to cross-sectional studies, which only detect associations between variables studied and offer no information on whether the health condition affects sleep or whether sleep is part of the process causing the condition. Besides, few studies have focused on relationship between sleep patterns and pain. However, many studies have found that poor sleep quality was associated with pain, which may be caused by musculoskeletal pain, arthritis, or cancer treatment [49‐52]. In our study, it’s really a limitation that we have failed to include questions about detailed pain relating questions. We only had questions about arthritis in our questionnaire and we conceived that we should take consideration of the pain-related problems in our future research. Despite the limitations, our study has several strengths. Firstly, this is a large community-based study, which has limited selection bias. Secondly, our study focused on an elderly population in China. Life expectancy in Shanghai resident was higher than the average level of Chinese people, and in 2015, and that already reached 83.2 [53]. Few studies have focused on a population of with an average age of 74 years old as ours, which includes the older population as well as the older old population. Daytime napping is a common habit among many Chinese in China, and is traditionally viewed as a healthy lifestyle [18]. This practice is different from daytime napping in western countries, where napping is less common and is often unplanned and prompted by sleepiness likely caused by aging, deteriorating health status, or nighttime insomnia [54, 55]. Thirdly, few studies have focused on sleep patterns [56], mostly only on sleep duration or sleep quality. Except the distribution of different sleep patterns, we even found distinct associations between sleep patterns and different chronic diseases, which may be an indication for precision management of chronic diseases. Finally, previous investigations often focused on nighttime sleep duration, which is a commonly surveyed sleep variable in large observational studies. However, we evaluated sleep in four aspects including nighttime sleep, daytime napping, total sleep and the nighttime-daytime sleep pattern.

We found that nighttime-daytime sleep patterns in our study had distinct relationship with chronic disease, especially for diabetes, cancer and arthritis. Considering that daily sleep habits have important implication in health, further research is needed to improve our understanding on this subject, especially nighttime-daytime sleep patterns.