Sleep features and the risk of type 2 diabetes mellitus: a systematic review and meta-analysis.

Introduction

According to the International Diabetes Federation (IDF), approximately 537 million adults worldwide were affected by diabetes in 2021, with type 2 diabetes mellitus (T2DM) accounting for around 90% of these cases. This figure is projected to increase in the coming decades [1]. Prior studies have highlighted the significance of altering behavioral patterns to regulate blood metabolism [2,3]. Sleep, a fundamental biological behavior, is governed by the interplay of circadian rhythms, homeostatic mechanisms, and neurohormonal processes [4]. The duration and quality of sleep are closely linked to an individual’s metabolic health [5,6]. Existing studies indicate that individuals with diabetes are more likely to experience both insufficient and excessive sleep durations compared to their non-diabetic counterparts, which, in turn, are associated with poorer metabolic control [5,7]. Many studies have described a U-shaped relationship between sleep duration and the risk of developing T2DM, suggesting that both abbreviated and extended sleep are linked to a higher likelihood of T2DM [8]. However, other investigations have failed to yield consistent results regarding the correlation between sleep duration and T2DM [6,9,10].

Moreover, the majority of research has focused primarily on sleep duration and T2DM, with a notable lack of studies evaluating the relationship between com­prehensive sleep features and T2DM. Sleep quality, insomnia, daytime napping, and sleep chronotype all significantly impact glucose metabolism. Evidence indicated that difficulties initiating or maintaining sleep are associated with chronic inflammation and metabolic disorders, but the precise correlation between sleep quality and T2DM remains inconclusive [11,12].

The impact of daytime napping on diabetes is also a topic of controversy. Some studies have indicated that excessive daytime napping (over 30 min) may increase the risk of developing T2DM and related complications [13,14], while others have failed to confirm this correlation [15–18]. Additionally, several studies have indicated that individuals with an evening chronotype face a heightened risk of developing T2DM, likely due to a misalignment between their biological clocks and daily routines [19–21], which suggested that sleep chronotype was a critical sleep feature that associated with T2DM.

Numerous studies, including cross-sectional studies, longitudinal studies, clinical trials, and epidemiological surveys, have explored the association between various sleep features and T2DM. The cumulative evidence underscores the urgent need for a comprehensive meta-analysis to synthesize the existing data. Such a systematic review and meta-analysis would provide more definitive insights into the relationship between sleep patterns and the risk of T2DM.

Research design and methods

A systematic review and meta-analysis were conducted to investigate the correlation between sleep features and the risk of T2DM, with registration in the inter­national prospective register of systematic reviews (Registration# CRD42023457631).

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for conducting this systematic review and meta-analysis [22]. We conducted a comprehensive literature search across the PubMed, Embase, Web of Science, and Scopus databases up until October 1, 2024, to identify studies examining the relationship between sleep and Type 2 Diabetes Mellitus (T2DM). The search was limited to articles published in English. The search terms used are provided in Table S1, and the search was limited to English language publications. In addition, we manually searched references and relevant reviews to identify any additional eligible studies. Articles were initially screened based on their titles and abstracts, and those that met the inclusion criteria were retrieved for full-text review. Figure 1 illustrates the results of our literature search and study selection process.

Flow diagram of literature search and study selection.

T2DM: type 2 diabetes mellitus; IGT: impaired glucose tolerance.

Studies were included in this meta-analysis if they satisfied the following criteria: (1) the study design was observational; (2) the exposure of interest was sleep features, including nighttime sleep duration, nighttime sleep quality, sleep chronotype and daytime napping; (3) the outcome was T2DM; (4) the investigators reported the adjusted odds ratio (OR), related risk (RR), or hazard ratio (HR) with the 95% confidence interval (CI); and (5) the full text was available. We excluded (1) animal studies, clinical trials, reviews, commentaries, letters, meta-analyses, and studies that examined other associations; (2) duplicate reports; and (3) studies in which the reported data were incomplete or unavailable. Two authors independently determined the eligibility of the studies. Any differences in opinion about eligibility were resolved through consensus by a third author.

The duration and quality of nightly sleep were assessed using the Pittsburgh Sleep Quality Index (PSQI) or self-reported measures. Sleep quality was primarily classified as good, moderate, or poor. The sleep chronotype was determined using the Morningness-Eveningness Questionnaire (MEQ) or self-reports. The acquisition of daytime napping data was primarily based on self-reported data. The sources for obtaining a diagnosis of T2DM included the oral glucose tolerance test (OGTT) criteria, the American Diabetes Association (ADA) and World Health Organization (WHO) criteria, self-reports of physician diagnoses, hospital diagnoses, causes of death, recommendations from the Chinese Diabetes Society or glycosylated hemoglobin, type A1C (HbA1c) levels.

The data were extracted independently by H-YL and WY, and differences were resolved by discussion and consensus with either TH or HZ. We extracted the following information from each study: authors, year of publication, study name, study type, study location, sample size (total and numbers of males and females), participants’ age, T2DM assessment, sleep measure assessment, population source, categories of sleep duration, sleep quality, chronotype, daytime napping, covariates adjusted in the multivariable analysis, and the ORs/RRs/HRs (95% CIs) for all the sleep features categories.

Cross-sectional studies were evaluated using the Agency for Healthcare Research and Quality (AHRQ) [23], which comprises 11 components with a cumulative score of 11. The scores were categorized as low (1-3), medium (4-7), or high (8-11). The Newcastle–Ottawa Scale (NOS) [24] was used to evaluate cohort and case–control studies. The scale comprises a total score of 9, consisting of 3 points for outcome measures, 2 points for between-group comparability, and 4 points for choosing the research population. The scores for low, medium, and excellent quality were 0-3, 4-6, and 7-9, respectively.

All the statistical analyses were conducted using R 4.2.2 (meta package). This meta-analysis aimed to quantitatively assess the correlation between sleep features and T2DM. Since the incidence of T2DM is sufficiently low, the ORs were equivalent to the RRs or HRs in this meta-analysis. The extracted ORs (95% CIs) were pooled using the inverse variance method. Forest plots were used to display the individual and aggregate effect sizes. Heterogeneity was assessed using the Q statistic and quantified by I2 and P values. A fixed-effects model was chosen if I2 ≤ 50% and p ≥ 0.01, while a random-effects model was selected if I2 > 50% and p < 0.01. Subgroup analyses were conducted to explore potential latent population heterogeneity. For the association between nighttime sleep duration and T2DM, stratified analyses were performed based on gender, age, and ethnicity. When examining the relationship between nighttime sleep quality and T2DM, stratifications were made by age and BMI. Additionally, the effect of napping on T2DM was assessed with stratified analyses by age, gender, menopausal status, and BMI. Lastly, the association between chronotype and T2DM was analyzed with stratifications by age and BMI. Funnel plots and Egger’s regression tests were employed to identify the absence of significant bias. The absence of publication bias can be inferred if the value of P is greater than 0.05. The trim and fill method were employed to address potential publication bias, and the robustness of the pooled results was assessed through sensitivity analyses. The detailed R analysis code is provided in the Supplementary materials.

Results

The specific process of inclusion and exclusion of this study is depicted in Figure 1. As of October 1, 2024, a comprehensive search identified 9,930 studies from PubMed, 7,199 from Embase, 15,049 from Web of Science, and 14,639 from Scopus. In addition, six studies were included through manual literature tracing. Subsequently, a meticulous evaluation of the eligibility criteria led to the exclusion of 20,767 noncompliant studies. After that, a close reading of 289 documents was performed, and ultimately, seventy-three studies were selected for in-depth analysis. The relationship between nighttime sleep duration and T2DM was explored in fifty-three studies (Table S2) [8–11,15,17,25–71], while nighttime sleep quality and T2DM were analyzed in seven studies (Table S3) [11,12,54,55,60,65,72]. Furthermore, the association between sleep chronotype and T2DM was examined in eight studies (Table S4) [21,39,73–78]. Daytime napping and T2DM were examined in twenty-one studies (Table S5) [15–18,26,28–33,79–88]. Kappa statistics were calculated to assess the consistency of data screening and selection across studies (Tables S6, S7). The PRISMA assessment of the included literature was listed in Table S8. Followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for conducting this systematic review and meta-analysis

We conducted a comprehensive analysis of fifty-three studies involving a total of 1,478,297 participants. The prevalence of T2DM varied significantly, ranging from 1.84% to 27.33% in these studies. The subjects were categorized into long sleep (≥8 h) and short sleep (≤7 h) duration groups according to their original data [2]. The findings presented in Figure 2 demonstrate a significant elevation in the risk of T2DM for individuals in both the long sleep duration (OR = 1.13; 95% CI = 1.09, 1.18) and short sleep duration (OR = 1.18; 95% CI = 1.13, 1.23) groups, showing a U-shaped relationship (Figures S1, S2). However, the increased risk was slightly more pronounced among those with a short sleep duration. After stratification by age, the most significant association between either the extremely long sleep duration or short sleep duration groups and T2DM risk was observed among individuals aged 30-39 years. Notably, the influence of sleep duration on T2DM risk diminished with increasing age in individuals older than 30-39 years old (Figures S3–S4).

The association between sleep features and risk of T2DM.

Sleep features include sleep chronotype, daytime napping, nighttime sleep duration, and sleep quality.

The age subgroups were analyzed in accordance with the average age of the participants from the original study.

Subgroup analysis of the association between nighttime sleep and T2DM.

Nighttime sleep duration was stratified by gender, age, ethnicity and BMI.

BMI: body mass index. Gender subgroups were derived from studies incorporating gender stratification, and data were extracted and pooled after adjustment for covariates. Age and BMI subgroups were analyzed according to the average age and BMI of the participants in the original studies.

The association between sleep chronotype subgroups and risk of T2DM.

Sleep chronotype was stratified by age and BMI.

BMI: body mass index. The age and BMI subgroups were analyzed in accordance with the average age and BMI of the participants from the original study.

Subgroup analyses revealed sex differences in the effect of nighttime sleep duration on T2DM risk. The increased risk of T2DM associated with short sleep duration was consistent across sexes (men: OR = 1.13; 95% CI = 1.04, 1.24; women: OR = 1.10; 95% CI = 1.06, 1.14; Figures S5, S6). Conversely, long sleep duration in women was linked to a greater risk of T2DM (men: OR = 1.12; 95% CI = 1.07, 1.18; women: OR = 1.17; 95% CI = 1.07, 1.27; Figures S7,S8). More specifically, in men, the impact of a short sleep duration was marginally greater than that of a long sleep duration (OR = 1.13 vs. OR = 1.10), whereas for women, a long sleep duration had greater effects on the risk of T2DM (OR = 1.12 vs. OR = 1.17). Furthermore, the interaction between sex and age highlighted that extremely short sleep durations had a more pronounced effect on men aged 40-49 years (OR = 1.31; 95% CI = 1.17, 1.48). Similarly, the effect of a short sleep duration on the development of T2DM in men decreased with age thereafter. Minimal effects of a short sleep duration on men were observed in individuals over the age of 60 years, yet the influence of a longer sleep duration on this group remained high. Remarkably, both the long and short sleep duration groups only demonstrated an increased risk of T2DM in women aged 50 years and older, and this risk diminished with advancing age thereafter.

The association between daytime napping and risk of T2DM.

The daytime napping was stratified by gender, menopausal status, age and BMI.

BMI: body mass index. The age and BMI subgroups were analyzed in accordance with the average age and BMI of the participants from the original study.

The joint effect of nighttime sleep and daytime napping on the risk of T2DM.

Moreover, we conducted a race-stratified analysis and found that white individuals and Asian individuals had a significantly elevated risk of T2DM due to both long and short sleep durations, whereas no such effect was observed among black individuals in relation to either long or short sleep duration (Figure 3).

A total of seven studies including 57,757 people explicitly reported subjective sleep quality [11,54,55,60,65,72]. Figure 2 demonstrates a significant association between poor sleep quality and an increased risk of T2DM (OR = 1.50; 95% CI = 1.30, 1.72), while good sleep quality was associated with a lower risk of T2DM. In contrast, intermediate sleep quality did not appear to significantly impact the risk of T2DM (OR = 1.09; 95% CI = 0.94, 1.27). Precise stratified analyses of individuals with poor sleep quality revealed that the risk was greater for individuals aged above 50 years (OR = 2.39; 95% CI = 1.39, 4.14) or those with a BMI less than 25 (OR = 1.85; 95% CI = 1.39, 2.44; Figure 3 & Figure S9).

A total of eight studies focused on sleep chronotypes and their associations with T2DM [21,39,73–78]. Chronotypes can be categorized into three groups: morning type (M-type), intermediate type (I-type), and evening type (E-type) [89]. The risk of contracting T2DM was notably elevated in individuals characterized as I-type (OR = 1.31; 95% CI = 1.02, 1.68) or E-type (OR = 1.59; 95% CI = 1.18, 2.13). In contrast, the risk of developing T2DM was reduced by approximately 18% in individuals with the early M-type (OR = 0.82; 95% CI = 0.80, 0.85) (Figure 2 & Figure S10). Next, subgroup analysis revealed that the E-type posed a greater risk to individuals aged 55 years or younger, whereas the M-type offered greater protection for those aged >55 years. Furthermore, after stratification by BMI, individuals with the E-type and a BMI greater than 30 exhibited a significantly elevated risk of T2DM (OR = 8.40; 95% CI = 2.44, 28.89). However, the M-type demonstrated only protective benefits for individuals with a BMI less than 30, while no significant correlations were observed between this pattern and T2DM risk in individuals with a BMI greater than 30 (Figure 4).

A total of twenty-one studies involving 901,033 individuals examined various daytime napping durations and nap frequencies as well as the relationship between daytime napping and the risk of developing T2DM. Among these studies, three investigated the association between daytime napping as a dichotomous variable and T2DM risk [29,32,82]. Using fixed-effect models, the findings revealed that daytime napping was associated with an increased risk of T2DM compared to not daytime napping (OR = 1.24; 95% CI = 1.08, 1.44) (Figure 2 & Figure S11). Furthermore, seven studies classified daytime napping as either ≤1 or >1 h [15,16,26,31,33,79,86]. In comparison with no daytime napping, both daytime napping duration groups had an increased risk of T2DM, but the risk was even greater for daytime napping exceeding 1 h (OR = 1.18 vs. OR = 1.38) (Figure 2 & Figure S12). To further investigate the correlation between daytime napping duration and T2DM, eight studies categorized daytime napping into four additional groups: ≤30 min, 31-60 min, 61-90 min and ≥91 min [15,17,28,30,80,81,84,86]. Their findings revealed a significant correlation between a daytime napping duration exceeding 30 min and an increased risk of developing T2DM, with the risk increasing by 7-20% as the daytime napping duration increased (Figure 2 & Figure S13). Additionally, three studies reported a significant association between the frequency of daytime napping and the risk of developing T2DM [18,85,87]. Specifically, individuals who napped four or more times per week exhibited a significantly greater likelihood of developing T2DM than those who did not engage in daytime napping (OR = 1.39; 95% CI = 1.27, 1.53) (Figure 2 & Figure S14).

Subgroup analyses by sex revealed more details (Figure 5). Among men, only daytime napping for both ≤1 h and >1 h was significantly associated with increased risks of developing T2DM compared with no daytime napping. Other daytime napping durations were not significantly associated with T2DM in men. In contrast, the relationship between daytime napping and T2DM in women was consistent with that in the overall population. Specifically, the risk of T2DM increased with increasing daytime napping duration. Notably, women exhibited a greater risk of developing T2DM than men when engaging in daytime napping lasting longer than 1 h. Furthermore, it was observed that only daytime napping for ≥91 min was significantly associated with T2DM among women.

Subgroup analyses based on age also revealed that daytime napping was associated with an increased risk of developing T2DM in individuals well below the age of 65 years (OR = 1.27; 95% CI = 1.05, 1.54). More specifically, among participants aged ≥50 years, a more significant increase in this risk was observed for those who napped for more than one hour (OR = 1.45; 95% CI = 1.34, 1.57). The interaction between sex and age revealed a significant association between daytime napping for ≥61 min and the risk of developing T2DM in women under 50 years of age (OR = 1.69; 95% CI = 1.27, 2.25). However, for women, a more specific analysis revealed that only postmenopausal women who engaged in daytime napping for more than one hour were found to have a significant association with the risk of developing of T2DM.

The association between BMI and T2DM also exhibited a U-shaped correlation in the case of daytime napping lasting less than 1 h, indicating that individuals with a BMI less than 24 or greater than 28 had an increased risk of developing T2DM. In contrast, the relationship between daytime napping for more than 1 h and T2DM was less influenced by BMI status, with all BMI groups showing a significant association with an elevated risk and a peak risk for the 24-28 BMI group (Figure 4).

As mentioned above, extreme sleep durations (≤7 h or ≥8 h) significantly increased the likelihood of developing T2DM, while poor sleep quality exhibited an even stronger influence on T2DM risk. We also carried out a combined analysis of four studies to examine the integrative impact of nocturnal sleep duration and sleep quality on T2DM [54,55,60,72]. A significant correlation between extreme nighttime sleep duration accompanied by poor sleep quality and T2DM risk was observed compared to individuals who slept for 7-8 h with good sleep quality (poor quality and a sleep duration ≤7 h: OR = 1.69; 95% CI = 1.11, 2.58; poor quality and a sleep duration ≥8 h: OR = 2.15; 95% CI = 1.19, 3.91) (Figure 6). In addition, intermediate-quality sleep combined with less than 7 h of sleep also increased the risk of T2DM (OR = 1.32; 95% CI = 1.12, 1.56) (Figure 6).

As two important components of 24-hour total sleep, the interactions between daytime napping and nighttime sleep were mutual, suggesting that there might be a synergistic effect between the two on T2DM. Hence, a comprehensive analysis was conducted on the collective findings of four studies [28–31]. The results showed that, compared with individuals who slept for 7-8 h at night without taking a nap, individuals who slept for less than 7 h or more than 8 h at night and who took a nap for more than 1 h exhibited an increased risk of developing T2DM; the ORs were 1.71 (95% CI = 1.19, 2.45) and 1.63 (95% CI = 1.35, 1.97), respectively. Furthermore, T2DM risk was also significantly elevated in individuals who had a nighttime sleep duration of ≥10 h without engaging in any daytime napping (OR = 1.70; 95% CI = 1.18, 2.45; Figure 6).

The outcome of the sensitivity analyses demonstrated that the direction of the estimated parameters remained unchanged even after eliminating the studies, thereby corroborating the robustness of our findings (Figures S15–S26). All the studies were of moderate quality or higher, and detailed information can be found in Tables S6, S7. The funnel plot and Egger’s test revealed the presence of publication bias for studies investigating a sleep duration ≤7 h, a sleep duration ≥8 h, nap vs. non-nap, and E-type. Despite this, the trim and fill analysis revealed no missing studies (Figures S27, S38).

Discussion

To the best of our knowledge, this is the first comprehensive systematic review and meta-analysis to explore the existing body of evidence on the correlation between various aspects of sleep characteristics and the risk of T2DM.

This study revealed a U-shaped pattern for the relationship between nighttime sleep duration and T2DM risk, with extreme night sleepers (either ≥8 h or ≤7 h) having a significantly greater risk of T2DM. This result was in accordance with the acclaimed optimal 7-8 h as the lowest risk duration in the study by Shan et al. [2], and extreme sleep durations of ≤5-6 h and ≥ 8-9 h with increased risks of developing T2DM, as identified by Cappuccio et al. [90]. A meta-analysis of seven studies also revealed a U-shaped association between sleep duration and HbA1c levels [5]. A Japanese study of 4,402 individuals with T2DM found that both short and long sleep durations were associated with a higher likelihood of metabolic syndrome and insulin resistance [91]. Notably, the impact of extreme sleep durations on increasing T2DM risk was more pronounced in individuals aged 30 to 39 years, although this effect diminished with age. These findings suggest that the influence of sleep duration on blood glucose levels varies by age, aligning with previous research indicating differing sleep needs between older and younger adults [92]. Additionally, our study revealed that gender significantly modifies the relationship between extreme sleep durations and T2DM risk. Specifically, sleep deprivation had a stronger impact on T2DM risk in men, while the effect of extended sleep durations was more pronounced in women. These findings suggest that moderate sleep duration has an independent protective effect on blood glucose control and that the influence of sleep duration on T2DM risk should be assessed individually in different subgroups.

In addition to nighttime sleep duration, poor sleep quality was unambiguously identified as a risk factor for T2DM. Consistent with the findings of Anothaisintawee et al. and Lee et al. [5,93], our results demonstrated a significant association between poor sleep quality and an increased risk of T2DM. Similarly, research by Osonoi et al. reported a significant correlation between poor sleep quality and both HbA1c and fasting blood glucose levels, even after adjusting for age and gender [94]. Tang et al. further confirmed this relationship, showing a significant correlation after adjusting for age, gender, BMI, and disease history [95]. These studies highlight that perceived sleep insufficiency and poor sleep quality negatively impact blood glucose regulation.

The internal circadian rhythm synchronizes our biological sleep/wake cycle with the 24-hour day, aided by zeitgebers (‘time givers’) and neurohormonal pathways, including melatonin [96]. Sleep chronotype is a circadian characteristic that reflects an individual’s preference for morning (early to bed and early to rise), evening (late to bed and late to rise), or intermediate (neither) patterns [97]. Our results indicated that the risk of T2DM was significantly higher in individuals with evening chronotypes, while intermediate chronotypes appeared to have a more protective effect. Previous studies have consistently shown an association between chronotype preferences and T2DM risk [39,77,98]. For instance, individuals with an evening preference exhibited a 2.5-fold higher risk of T2DM compared to those who preferred morning chronotypes, regardless of their sleep duration or adequacy [39]. Research also suggests that disruptions in circadian rhythm may lead to impaired glucose tolerance, insulin resistance, and elevated levels of oxidative stress and inflammation, exacerbating metabolic disorders and increasing the risk of T2DM [73].

This study found that daytime napping for more than 30 min and napping more frequently than four times per week were associated with an elevated risk of T2DM. Jinjin Yuan et al. identified that long nap durations (≥60 min) were linked to poor blood glucose control [99]. A prospective study involving 435,342 non-diabetic participants from the UK Biobank, with a median follow-up of 9.2 years, also revealed that daytime napping is associated with an increased risk of T2DM [87]. Among 12,977 individuals with T2DM, those who frequently napped during the day were 1.3 times more likely to exhibit poor blood glucose control compared to those who did not have a napping habit [100]. However, the precise impact of daytime napping on glucose metabolism remains unclear. One potential mechanism could involve alterations in melatonin levels. During a 12-hour simulated night shift, a temporary increase in melatonin levels was observed following a 2-hour nap [101]. Melatonin is known to play a role in regulating insulin secretion and insulin resistance, which could, in turn, influence glucose control.

The effects of various sleep features on the risk of T2DM are not independent; rather, they interact with one another. Our research indicated that individuals experiencing both poor sleep quality and nighttime sleep durations exceeding 8 h had the highest prevalence of T2DM. Additionally, those with nighttime sleep durations of less than 7 h who also engaged in prolonged daytime napping (>1 h) demonstrated a significantly elevated risk of developing T2DM. While an adequate nighttime sleep duration is a prerequisite for optimal sleep quality, poor sleep quality can potentially result in insufficient rest and recovery for both the body and the brain, consequently affecting the duration of sleep [88]. A study including 109 patients with type 1 diabetes (mean duration of diabetes 17.7 years) revealed that poor sleep quality and short sleep duration were negatively correlated with the estimated glucose disposal rate and thus could be used as significant predictors of decreased insulin sensitivity [102]. Another study revealed robust associations between the quality and duration of sleep and HbA1c levels, indicating that both poor sleep quality and insufficient sleep duration work together to significantly elevate HbA1c levels [103]. Thus, it is not surprising that in our study, extreme nighttime sleep duration and poor sleep quality also exhibited considerable interactive increases in the risk of T2DM.

As the main component of 24-hour sleep, nighttime sleep duration is crucial to human health. However, the balance between nighttime sleep duration and daytime napping is also important. Excessive daytime napping may influence nighttime sleep, especially when the total sleep duration surpasses the regular range. Overly long daytime napping may disrupt the quality and duration of nighttime sleep, resulting in sleep cycle disruption and subsequent impacts on the body’s biological clock and metabolic processes [104]. A study involving self-reported sleep data from 398 participants revealed that individuals who slept less than 5 h and refrained from daytime napping exhibited a significantly greater risk of poor glycemic control than those who slept for 6-7 h during the night without taking daytime napping. Importantly, they also indicated that incorporating daytime napping into their routine could efficiently reduce the risk of poor glycemic control among individuals with short nighttime sleep durations [105]. Our findings confirmed that both excessive and insufficient nighttime sleep significantly affect the onset and progression of T2DM, and this risk could be exacerbated by the addition of protracted daytime napping duration. However, the risk of T2DM may be mitigated by achieving the recommended optimal total sleep duration through moderate catch-up sleep sessions. Nevertheless, how to supplement so-called sleep debt through the coordination of nighttime sleep and daytime napping needs further elucidation.

The impact of sleep duration and quality on insulin sensitivity in humans is both significant and direct [106,107]. Extreme sleep conditions, whether characterized by excess sleep or deficient sleep, can lead to a decrease in leptin levels and an increase in ghrelin levels [108]. This hormonal imbalance can trigger heightened appetite, raising the risk of overeating [109], which may subsequently increase BMI and fat accumulation, thereby exacerbating insulin resistance [110,111]. Poor sleep quality, characterized by frequent awakenings and interruptions, can compromise the body’s insulin responsiveness, enhancing insulin resistance [112] and worsening glycemic control [113]. The presence of poor sleep patterns, whether due to extreme sleep durations or compromised quality, has been scientifically linked to elevated levels of inflammation [114]. Evidence suggests that inadequate sleep duration is associated with increased levels of inflammatory markers, such as high-sensitivity C-reactive protein (CRP) and interleukin-6 (IL-6) [115,116]. Persistent inflammation can directly impair the functionality of pancreatic beta cells, leading to reduced insulin secretion and disrupting the regulation of blood glucose levels [117].

The risk of unhealthy eating patterns and weight gain, which exacerbate insulin resistance, is particularly pronounced among individuals with an evening (E-type) chronotype [118]. Due to their prolonged wakefulness at night, these individuals often experience poorer nighttime sleep quality and reduced sleep duration. This disruption can lead to hormonal imbalances that affect insulin production and efficacy [119]. Additionally, disrupted sleep cycles can result in elevated cortisol levels [120], which is another key factor contributing to insulin resistance and the development of T2DM [121]. Irregular sleep chronotypes, especially evening types, are also significantly associated with heightened inflammatory responses. Research has shown a positive correlation between the E-type chronotype and increased levels of interleukin-1β (IL-1β) and IL-6 in university students [122]. Lower scores on the Morningness-Eveningness Questionnaire (MEQ), which suggests an evening chronotype, are also positively correlated with elevated levels of C-reactive protein (CRP) [123]. Moreover, individuals with an E-type chronotype often struggle to maintain adequate levels of daytime physical activity [124]. This reduction in physical activity further limits energy expenditure, increasing the likelihood of weight gain [124]. This reduction in physical activity further limits energy expenditure, increasing the likelihood of weight gain [125]. These sleep patterns not only directly influence weight management but also indirectly increase the risk of developing T2DM [126].

Prolonged daytime napping can disrupt regular nocturnal sleep patterns, leading to reduced sleep quality and, consequently, an increase in insulin resistance [127]. Excessively long naps during the day may further disturb the body’s hormone secretion rhythms, including those of insulin and cortisol, both of which are essential for blood glucose regulation [128,129]. This disruption raises the risk of insulin resistance, a critical factor in the development of type 2 diabetes mellitus (T2DM). Moreover, daytime napping habits that shorten nighttime sleep duration or create irregular sleep patterns can exacerbate inflammation levels in the body [130]. Collectively, these findings suggest that disrupted sleep and inflammation contribute to the increased risk of T2DM and may also heighten the likelihood of diabetic complications [131]. The circadian rhythm plays a vital role in maintaining metabolic health by synchronizing various physiological processes, enabling the efficient metabolism of substances and energy at appropriate times [132]. Sleep pattern disruption interferes with the biological clock’s ability to regulate key metabolic functions, such as glucose and fat metabolism [133], leading to reduced efficiency and an elevated risk of metabolic disorders like T2DM [134]. Furthermore, disruptions to biorhythms, whether due to prolonged daytime naps or evening chronotypes (E-types), impair the body’s ability to make necessary physiological adjustments during periods of rest or activity [135–137], compromising overall health and increasing susceptibility to T2DM [19,20,138].

This study has several strengths. First, it conducted a comprehensive assessment of the relationship between four key sleep dimensions, including nighttime sleep duration, sleep quality, chronotypes, and daytime napping, and the risk of type 2 diabetes mellitus (T2DM). By examining not only the individual associations between these sleep characteristics and T2DM but also their interactive effects, the research provided substantial evidence-based support for understanding the role of sleep, a critical behavioral factor, in the onset of T2DM. Second, the methodology was rigorous, adhering to PRISMA guidelines for quality assessment and control of the included studies. Most of the studies analyzed were of high methodological quality, enhancing the reliability of the findings. Third, subgroup analyses based on population characteristics yielded valuable insights that could inform personalized, targeted screening and intervention strategies for high-risk groups in both clinical and public health settings.

However, this study also has some limitations. First, the literature search was restricted to published studies in English, which may have excluded relevant unpublished or non-English research. Second, sleep characteristics in all the included studies were based on self-reported data, highlighting the need for high-quality cohort studies utilizing objective measures of sleep, such as actigraphy or polysomnography. Third, as this meta-analysis is based on observational studies, directly controlling for residual or unmeasured confounding factors was impossible. However, we minimized the impact of confounders by using adjusted estimates from the multivariable models in each study and performing stratified analyses by age, gender, and BMI. Finally, while the limited data in certain subgroups resulted in fewer studies being included, the low heterogeneity suggests that the results remain fairly robust. Larger-scale studies are needed to further validate these findings.

Conclusion

This systematic review and meta-analysis identified a U-shaped relationship between nighttime sleep duration and the risk of type 2 diabetes mellitus (T2DM), with the lowest risk observed in individuals sleeping 7–8 h per night. Poor nighttime sleep quality, evening chronotypes, and daytime naps exceeding 30 min were also identified as potential risk factors for T2DM. Notably, the risk of T2DM was significantly higher in individuals who had poor sleep quality combined with more than 8 h of nighttime sleep, or in those who slept less than 7 h at night but engaged in over 1 h of daytime napping. These specific combinations of sleep characteristics worked synergistically, further increasing the risk of developing T2DM. This study provides strong public health evidence supporting targeted screening and interventions focused on sleep behaviors, particularly for high-risk groups prone to T2DM.