Waking Up to the Importance of Sleep in Type 2 Diabetes Management: A Narrative Review

Although lifestyle modifications are fundamental therapeutic components in the management of type 2 diabetes, national and international guidelines have predominantly focused on pharmaceutical therapies. The prominence and integration of lifestyle modifications into guidelines and resources for clinical decision-making have paled in comparison, epitomized by fewer citations and less defined clinical impact. Now, for the first time, the latest American Diabetes Association/European Association for the Study of Diabetes (ADA/EASD) consensus guidelines have broken with tradition by incorporating a growing body of evidence linking health outcomes associated with type 2 diabetes to the movement behavior composition over the whole 24-h day (1,2). In this context, a 24-h day comprises a sequence of movement behaviors distributed on a continuum ranging from limited/no movement to high-intensity activities. The five S’s (sleep, sitting, stepping, sweating, and strengthening) encapsulate these physical behaviors, and their inclusion represents an important milestone in bridging the gap between current knowledge around 24-h behaviors and clinical care. Of note, the importance of sleep as a key lifestyle component in the management of type 2 diabetes is promulgated, which complements recent reviews (3–5), and is placed, for the first time, on a level playing field with the importance of physical activity. This is important key because despite sleep occupying approximately one-third of the day for most people and modulating a variety of metabolic, endocrine, and cardiovascular processes, it only appears in ∼40% of clinical practice guidelines (6). Given the fundamental role of sleep in the health and well-being of those living with type 2 diabetes, there is a need to increase its exposure and applicability for health care professionals, policymakers, and individuals with lived experience. As such, we highlight some of the key evidence underpinning the importance of sleep in the management of type 2 diabetes, outline practical advice about initiating conversations in clinical care, and propose future directions for sleep research in the management of type 2 diabetes. As a frame of reference for the reader, the common terminology and definitions used throughout this article can be found in Table 1.

Sleep is a metabolically active and dynamic process that involves complex behavioral and physiological processes. Sleep has a typical underlying architecture underpinned by a rhythmic alternation between nonrapid eye movement (NREM) and rapid eye movement (REM) stages (7). In particular, the deeper stages of NREM sleep (also termed stage N3) are the most refreshing and restorative, allowing the body to repair cells, tissues, and muscles (8). Over the course of the night, total sleep is made up of several rounds of the sleep cycle, which is composed of four individual stages. A visual representation (hypnogram) and description of the various stages within a sleep cycle can be found in Fig. 1 and Table 2.

Despite healthy sleep consisting of adequate duration, quality, timing, and regularity, along with the absence of sleep disturbances or disorders, guidelines and recommendations have typically focused on duration. General recommendations suggest at least 7 h of sleep per night regularly for optimal health (9,10). Specific recommendations for those living with type 2 diabetes have not been developed and as such should defer to these global recommendations.

Historically, the research in type 2 diabetes has focused on sleep disorders and deficiencies, with the most prevalent reported in Table 1. Such disturbances can increase the risk of developing type 2 diabetes and its associated micro- and macrovascular complications (e.g., neuropathy, nephropathy, and retinopathy), alongside several health-related quality of life domains (4,11). However, a range of outcome measures (i.e., beyond sleep disorders) can be used to characterize “optimal sleep” (Table 1). Discussion of each characteristic is beyond the scope of this article; therefore, we primarily focus on the three overarching constructs outlined in the latest ADA/EASD consensus report—quantity, quality, and timing (i.e., chronotype) of sleep—as they represent important and underrecognized components of type 2 diabetes management (1,2). However, we also acknowledge that these sleep behaviors usually coexist and interact with each other in a compensatory manner. For ease of interpretation, we highlight each construct in turn, while outlining their role in the incidence of type 2 diabetes and effects on glycemic control, cardiovascular disease (CVD) risk, and mortality. These outcomes were chosen as they represent the most robust evidence to date (1,2). That said, we recognize that sleep also impacts on other markers of interest (e.g., depression), which is likely to become more prominent as the evidence base evolves.

An overall summary of the meta-analyses published within each construct can also be found in Table 3. However, due to the varying level of available evidence this narrative review presents data across the research hierarchy spectrum (i.e., from systematic reviews/meta-analysis to single, cohort studies).

There is now established evidence for a U-shaped association between sleep duration and type 2 diabetes incidence, with the nadir typically occurring at 7 h per day, with short (typically defined as <6 h) and long (typically defined as >9 h) sleep duration having up to a 50% increase in the risk of type 2 diabetes, including progression from prediabetes (12). Dose-response analysis has also demonstrated in comparison with 7 h/night, each hour decrease or increase in sleep is associated with a 9–14% increase in risk of type 2 diabetes (13–16)

Despite allowing significant advancement in our knowledge, studies to date have mostly used self-reported and single time point assessments of sleep. Use of objective measures of sleep duration or genetics-based analyses has continued to support an association for sleep as a risk factor for metabolic dysfunction but has produced equivocal evidence of an association for long sleep. In the UK Biobank cohort studies, for example, although use of objective measures of sleep duration demonstrated that sleeping >8 h/night is associated with increased CVD, cerebrovascular, and mood disorders, there was no evidence of a relationship with type 2 diabetes (17). A recent meta-analysis, alongside Mendelian randomization studies, also failed to support a higher risk of cardiometabolic dysfunction with longer sleep (12,18,19) with some evidence suggesting that longer sleep may even be protective (18). This suggests that deleterious associations with longer sleep may be explained by confounding or reverse causation, whereas associations with shorter sleep may be causal in nature and represent a target for intervention. Indeed, this is supported by emerging interventional research that has shown that sleep extension interventions in short sleepers result in improved insulin sensitivity and reduced daily energy intake (20,21).

Subjectively quantified sleep is associated (U-shaped) with HbA_1c and fasting plasma glucose in those with type 2 diabetes, with both long (>8 h/night) and short (<6 h/night) sleep durations adversely influencing glycemic control (16). Recent clinical evidence also extends beyond glycemic control in demonstrating that short sleep duration is also associated with CVD risk and mortality in those living with type 2 diabetes (20,21). For example, data from the UK Biobank cohort demonstrated that short (≤5 h/night) sleep duration was associated with a 42–70% higher risk of ischemic stroke and CVD mortality in comparisons with 7 h/night (22). These results mirror the J-shaped association previously observed between all-cause and CVD mortality and sleep duration, where ≤4 h sleep was associated with a 41% increased risk of all-cause mortality and 54% increased risk of CVD mortality vs. 7 h sleep (21).

Sleep quality has recently been defined as “an individual's self-satisfaction with all aspects of the sleep experience” (23). This is underpinned by four attributes: sleep efficiency, sleep latency, sleep duration, and wakefulness after sleep onset (WASO) (Table 1). Many factors (e.g., environmental, behavioral, psychological, and physiological) can contribute to poor or insufficient sleep quality that impact health outcomes, including those associated with type 2 diabetes.

A recent analysis using seven cycles of National Health and Nutrition Examination Survey (NHANES) data (n = 16,517) demonstrated that sleep quality has declined from 2005–2006 to 2017–2018 and that the highest prevalence of diabetes was consistently observed in the low sleep quality group (24). These findings corroborate previous meta-analysis, where poor sleep quality was associated with a 40–84% increased risk of developing type 2 diabetes (14,25).

Epidemiological studies have suggested that there are associations between sleep disturbances and glycemic control in those living with type 2 diabetes. For example, Knutson et al. (26) demonstrated that sleep quality was a significant predictor of HbA_1c in those individuals with at least one diabetes-related complication. Indeed, the predicted increase in HbA_1c level for a 5-point increase in Pittsburgh Sleep Quality Index (PSQI) score in those with complications and taking insulin was 1.9% (0.7 mmol/mol) (i.e., moving from 8.7 to 10.6% [71.6 to 92.4 mmol/mol]). As the change is proportional, the increase in PSQI score may have a greater effect at a higher HbA_1c. A 2017 meta-analysis that included a small number of studies (n = 5) also showed that in those with type 2 diabetes who had difficulty in initiating or maintaining sleep, poorer sleep quality was associated with higher HbA_1c levels (16).

There is limited evidence on the association between sleep disturbance and risk of incident CVD in those living with type 2 diabetes. However, data from a large, single cohort study (n = 36,058) demonstrated that disturbances in sleep were associated with increased risk for all CVD (hazard ratio [HR] 1.24 [95% CI 1.06–1.46]) and coronary heart disease (1.24 [1.00–1.53]) events in those living with newly diagnosed type 2 diabetes (<6 months) (27).

Studies examining sleep disturbances in relation to mortality have shown significantly increased risk (27,28). More specifically, in a recent study with use of prospective data from UK Biobank (n = 487,728) investigators examined associations of frequent sleep disturbances, diabetes, and risk of all-cause mortality (28). In examination of sleep disturbances (28% prevalence) and diabetes, the presence of both was associated with increased risk of all-cause mortality (HR 1.87 [95% CI 1.75–2.01]) in comparisons with subjects who had either or neither condition (28).

The subset of sleep management concerning an individual’s chronotype, or what most people understand as being an early bird or a night owl, can influence sleep timing and consistency. In humans, the circadian clock is divided into two distinct parts, the master clock in the suprachiasmatic nucleus of the hypothalamus and peripheral clocks, situated in the peripheral tissues (29). From a biological perspective, sleep timing depends on two processes: sleep debt (i.e., the difference between the amount of sleep we need and the amount we get) and an internal circadian clock that synchronizes biological sleep/wake rhythms to our 24-h day, aided by zeitgebers (“time givers”) and neurohormonal pathways (including melatonin). However, many facets of modern life, such as work schedules (i.e., night/rotating shifts), can lead to sleep/wake schedules that are misaligned relative to our internal biological clock. Our bodies appear to have evolved to cope with day-to-day variability in sleep timing within certain thresholds, beyond which unfavorable responses occur (Fig. 2). This misalignment (termed social jet lag) occurs when different endogenous circadian rhythms are not synchronized with one another and/or with external cues or social pressures, such as work pattern (30).

Chronotype preference has been linked with many chronic diseases, including type 2 diabetes (31–34). For example, for those with a preference for evenings (i.e., going to bed late and getting up late) there was a 2.5-fold higher odds ratio for type 2 diabetes as compared with morning types (i.e., going to bed early and getting up early), independent of sleep duration and sleep sufficiency (34). Moreover, investigators of a recent cohort analysis showed that after accounting for multiple lifestyle and sociodemographic factors, middle-aged nurses with an evening chronotype demonstrated a 19% increased diabetes risk compared with morning chronotypes (35).

There is also a growing body of evidence suggesting a chronotype-dependent association between work hours (i.e., shift work) and metabolic disease (36–38). For instance, findings of a meta-analysis of observational studies indicate that individuals exposed to shift work have up to a 10% increased risk of type 2 diabetes compared with those with no shift work experience (39,40). Similarly, in those with established type 2 diabetes who work night shifts, glycemic control is more likely to be impaired compared with people with type 2 diabetes performing day work (41). Findings of a 2015 article also showed that women who were late chronotypes without any history of rotating night shift work had a 1.5-fold increased risk of type 2 diabetes (odds ratio 1.51 [95% CI 1.13–2.02]) (42). Interestingly, the investigators observed an interaction between chronotype and shift work, where late chronotypes had a significant increase in type 2 diabetes risk only when their shift schedule did not involve night work. Conversely, although early chronotypes had a lower risk of type 2 diabetes, this increased with the length of rotating night shift work, possibly driven by more circadian misalignment during night shifts (42). Therefore, these results suggest that if work times interfere with sleep timing, shift and day workers may be at an increased risk for type 2 diabetes.

Although, limited by high heterogeneity between studies, a recent systematic review and meta-analysis demonstrated that social jet lag (vs. no social jet lag) is associated with higher HbA_1c levels in those living with type 2 diabetes (0.42% mean difference [95% CI 0.12–0.72]) (43). When chronotype, social jet lag, and glycemic control are considered together, there also appears to be a significant relationship between later chronotype and HbA_1c levels, but only for patients with >90 min of social jet lag (44). Other sleep disturbances (insufficient sleep, poor sleep quality, and sleep apnea) influencing glycemic control may also induce a degree of social jet lag. For instance, objectively measured variability in sleep duration, which may reflect partial sleep deprivation alternating with sleep compensation, was most strongly associated with HbA_1c in 172 individuals with type 2 diabetes (when compared with total sleep duration, subjective sleep quality, and sleep efficiency) (45). The difference in average HbA_1c between participants in the lowest and highest quartile of variability in sleep duration was 1.0% (45).

Findings of a prospective cohort analysis including those living people with type 2 diabetes (n = 3,147) demonstrated that disrupted circadian-activity rest rhythms (defined with use of accelerometer-derived average activity during wake and sleep) were associated with higher risks of CVD (HR 1.38 [95% CI 1.03–1.84]), ischemic heart disease (2.49 [1.71–3.64]), and CVD-related mortality (3.98 [1.76–9.00]). Similar associations were also observed for all-cause mortality (1.75 [1.14–2.71]) (46). Although not confined to those with type 2 diabetes, definite evening types have been shown to have a significantly increased risk of all-cause mortality (HR 1.10 [1.02–1.18]) compared with definite morning types (47). This analysis, conducted in 433,268 UK Biobank individuals over a 6.5-year period, also demonstrated that the effect size across different chronotypes was similar to the effect observed for BMI, renal, musculoskeletal, and gastrointestinal/abdominal disorders (47).

The bidirectional link between sleep complaints and type 2 diabetes likely occurs via multiple physiological and behavioral mechanisms (Fig. 2). Indeed, sleep restriction is associated with several hormonal changes that are known to impact insulin resistance and insulin secretion. For example, melatonin and cortisol, which are produced from the hypothalamic-pituitary-adrenal axis modulate the sleep-wake cycle and display an inverse relationship (48). However, these patterns change with short sleep duration (both habitual and experimental) as sleep restriction causes elevated markers of sympathetic activation and catecholamines (49). As a result, cortisol levels become higher in the evening (50) and display a lower rate of decline (51). Such changes can lead to a lower response of β-cells to glucose and reduce insulin sensitivity (driven by lower glucagon-like peptide 1 levels) (52). Short sleep duration and sleep deprivation are also associated with elevated levels of proinflammatory cytokines, changes in adipokines secreted from adipose tissue (53,54), enhanced lipolysis (55), and increased hunger and appetite, largely driven by changes in leptin (decrease) and ghrelin (increase) (56).

Circadian misalignment also plays an important role in the etiology of type 2 diabetes. The circadian rhythm is driven by circadian clock genes, controlling physiological and behavior processes over a 24-h period (57). Indeed, polymorphisms in many of the well-established core clock genes (e.g., CLOCK,BMAL1, CRYO) have been shown to increase the risk of type 2 diabetes (58), with gene-behavior interaction studies also demonstrating interactions between diet and clock gene mutations that affect fasting glucose (59), insulin resistance (60), and type 2 diabetes (61).

Endogenous rhythms are produced by the suprachiasmatic nucleus, alongside cells in peripheral organs (e.g., skeletal muscle) that also have an intrinsic circadian clock (29) (Fig. 2). These anticipatory rhythms are synchronized to light cycles, but as we stay awake for longer and are exposed to more artificial light, we subject our body to various behaviors that may exacerbate cardiometabolic abnormalities. For example, we recently showed that evening chronotypes (i.e., preferring to go to bed late and getting up late) engage in lower levels of moderate-to-vigorous physical activity levels (approximately −10 min/day, −56%) compared with morning chronotypes (i.e., preferring to go to bed early and getting up early) (62). Therefore, an advancement of the internal circadian rhythm through informed timing of physical activity and time-restricted eating may be a useful adjunct therapeutic strategy (alongside sleep interventions) to foster metabolic improvements (63,64) and chronobiological homeostasis and better align internal rhythms with the environment and standard social schedules.

Although the aforementioned mechanisms may increase the risk and impair the management of type 2 diabetes, associated comorbidities also demonstrate a bidirectional relationship with sleep. For instance, sleep complaints often present prior to the onset of a new or recurrent episode of depression (65), suggesting that sleep may be involved in its pathogenesis. Conversely, symptoms of depression have also been shown to be an important correlate of suboptimal sleep quality, with as many as 90% of those with depression also having issues with sleep quality (66), making differentiating cause-and-effect relationships problematic.

There is no single measure of sleep, as the construct spans multiple dimensions and levels of analysis. As a result, sleep can be quantified through a variety of subjective, objective/physiological, and behavioral observation methodologies. A visual summary is provided in Fig. 3. For example, questionnaires can capture chronotype, quantity, latency to sleep onset, duration, and level of daytime sleepiness. Examples include the Munich Chronotype Questionnaire, Morning-Eveningness Questionnaire, PSQI, and Epworth Sleepiness Scale (ESS) (67–70). Such subjective reports of sleep have informed most of the evidence base to date and are important in a clinical setting, as they can help determine whether further screening and/or treatment for a sleep complaint might be justified.

Polysomnography (PSG), which is the current gold standard for the objective measurement of sleep, provides insights into nocturnal physiology, including a recording of objective sleep architecture and measures of cardiopulmonary function (71). Actigraphy also allows physical behaviors (including sleep duration, sleep timing, and WASO) to be measured over a 24-h period (72). Indeed, the evolution of wearable technology provides individuals with an array of readily available self-assessed outcomes (e.g., duration, sleep schedules, sleep stages). A major strength is that they allow for recordings over a number of nights and in ecologic conditions. However, there are clinical implications, as although all wearables harness similar types of technology, variation in accuracy across devices and metrics exists. That said, if an individual is asleep, most wearables are 90–95% accurate for identifying this behavior (73,74). However, in determining sleep onset latency and WASO, the wearables only perform at a medium level of accuracy (∼60%), which diminishes further in trying to determine specific stages of sleep (∼50%) (74).

Other wearable and nearable examples include smart mattresses, pulse oximeters, and radar-based devices. Of particular note, electroencephalograms (EEG) are a widely used noninvasive method for monitoring neuronal action within the brain during sleep. Although usability and reliability issues with the use of EEG exist, recent advances in wearable devices mark an important step forward. Specific commercially available examples include the Dreem, Muse, and BrainBit headbands, alongside in-ear EEG devices (75). Although these devices are still to be validated at scale, they offer promising alternatives to PSG for long-term monitoring of sleep stages.

To reduce misclassification and capture sleep that is representative, individuals should be encouraged to obtain data over multiple nights (including both weekdays and weekends). Such data, although not a substitute for medical evaluations, may be used as an adjunct to an appropriate clinical evaluation or to facilitate behavior change.

Several behavioral and pharmacological sleep interventions exist; however, the majority of the evidence in type 2 diabetes focuses on the treatment of sleep disorders (in particular obstructive sleep apnea [OSA]) as opposed to diabetes-related outcomes (e.g., glycemia) per se. Discussion of them all in turn is beyond the scope of this review; therefore, we highlight some of the more prominent interventions that have been used to aid sleep management in those with type 2 diabetes. Definitions can also be found in Table 1.

Understandably, cognitive behavioral therapy (CBT) and cognitive behavioral therapy for insomnia (CBTi) are often confused. CBT is most commonly used to treat the symptoms of anxiety and depression, while CBTi is specifically designed for insomnia (76). As such, CBTi is recommended as first-line treatment for both short- and long-term insomnia (77) and includes elements of sleep hygiene, education, and stimulus control. Results of meta-analyses have shown CBTi to produce clinically meaningful improvements in sleep parameters, including sleep latency, sleep efficiency, total sleep time, WASO, and the number of awakenings (78,79). However, to date, in the majority of studies in those with type 2 diabetes investigators have used CBT, not CBTi, per se. Indeed, results of a 2021 randomized controlled trial, undertaken in 1,033 individuals living with type 2 diabetes, demonstrated improvements in glycemic control and sleep quality following lectures/discussions with trained general practitioners. Both HbA_1c (−0.17% at 6 months, −0.43% at 12 months) and sleep quality (−0.50 and −0.90 lower PSQI score at 6 and 12 months, respectively) were improved (80). This study, among others, was encapsulated in a 2022 meta-analysis (32 studies, n = 7,006), which included only one study specifically looking at CBTi. Nevertheless, the ability of CBT to reduce HbA_1c and fasting glucose was pooled at −0.14% and 0.9 mmol/L, respectively (81). However, the results of this meta-analysis and others should be interpreted with caution as individuals with type 1 diabetes and gestational diabetes mellitus were also included (82,83).

Sleep education can play a fundamental role in ensuring individuals understand the relationship between sleep and overall well-being. In particular, it may include information on sleep health, sleep cycles, or consequences of insufficient sleep or sleep hygiene tips and may be delivered using a variety of methods (e.g., group-based education, webinars, apps). However, there is a paucity of evidence, particularly relating to glycemic outcomes in those living with type 2 diabetes. In a randomized pilot study in 31 adults with type 2 diabetes who did not sleep before midnight, a combination of conventional diabetes education and sleep education reduced HbA_1c, fasting glucose, and insulin resistance more than diabetes education alone (84). The potential utility of sleep hygiene as an intervention to improve sleep quality and glycemic control in prediabetes and diabetes has also been demonstrated (85). The intervention, which included sleep tips outlined by the American Academy of Sleep Medicine, resulted in improvements in sleep quality, time, and efficiency (10). For instance, PSQI score was 3.6 points lower in comparison with the control group. Moreover, the intervention resulted in reductions in HbA_1c of 0.39% and 0.66% at 3 and 6 months, respectively (85).

Although continuous positive airway pressure (CPAP) is the current gold standard treatment for OSA (86), the impact on markers of glycemic control are negligible. For example, a systematic review and meta-analysis (6 studies, n = 581) showed no benefit for glycemic control (HbA_1c and fasting glucose) versus placebo (87). That said, CPAP should always be offered to those living with type 2 diabetes who present with OSA (regardless of the impact on diabetes-specific outcomes, given the symptomatic benefits, along with improvements in quality of life) (88).

A number of alternative therapies to CPAP also exist, including (but not limited to) lifestyle modification (e.g., weight loss), mandibular advancement devices, positional therapy, surgical procedures (e.g., upper airway), and hypoglossal nerve stimulation (see Table 1 for definitions). With such therapies improvements have been shown in apnea-hypopnea index, polysomnographic-based outcomes, daily function depressive symptoms, and quality of life, along with reduced arousals and rate of oxygen desaturation (89–92). Further research will determine whether these non-CPAP therapies are viable treatment alternatives for diabetes-specific outcomes.

The body naturally produces melatonin, but recently there has been interest in external sources (e.g., supplements) to aid sleep management. In those living with type 2 diabetes and insomnia, short-term use of prolonged-release melatonin has been shown to improve sleep maintenance and sleep quality (93). However, the direct impact of melatonin on glycemic parameters appears ambivalent, with a systematic review and meta-analysis showing potential benefits for fasting glucose and insulin sensitivity but not HbA_1c (94). That said, the potential of melatonin to influence proinflammatory pathways as well as oxidative stress state in those with diabetes (95) means that more clinical trials are needed that use combinations of melatonin with current therapeutic agents for the treatment of diabetes.

Despite health care professionals routinely asking about important indicators of health (e.g., weight, diet, and medication status), questions about sleep are often overlooked. Sleep should always be discussed as part of a holistic approach to lifestyle behavior in diabetes care. To aid discussions in clinical care, we provide a visual summary of the different methodologies, targets, and health impacts pertaining to the three sleep constructs: quantity, quality, and timing (i.e., chronotype) (Fig. 4).

Clinicians and health care professionals are also encouraged to follow the five S’s for the management of sleep in clinical practice (Survey, Support, Shared decision making, Solutions, Signpost [Fig. 5]). Information gathered during the survey stage should inform the decision-making approach. If there is an indication that there may be an underlying sleep disorder, the patient could be fast-tracked straight to the signpost stage. If the main contributing factors to inadequate sleep health appear to be related to behavioral or lifestyle choices, they could be addressed using the processes outlined in the support, shared decision-making, and solutions stages. This may involve targeted recommendations/interventions that should be tailored to what is available in the local setting. The consultations may also need to be extended to multiple disciplines and specialties (e.g., sleep specialists, psychologists, diabetes care and education specialists).

Specific sleep behavior goals should be agreed on between the person living with type 2 diabetes and the clinical care team; shared decision-making is fundamental. SMART (specific, measurable, achievable, realistic, time-bound) goals can be used to facilitate conversations and action plans, where even small changes in sleep behavior should be acknowledged (e.g., turning off the television 15 min earlier). Realistic expectations should also be guided by self-efficacy and illness perceptions. Considering aspects of sleep, well-being, and their association with diabetes (e.g., diabetes-related distress) could support health care professionals in formulating diabetes management plans that are better tailored to the needs of the individual.

Sleep hygiene improvements can often aid the quantity and quality of sleep. It is important to emphasize that these recommendations are fundamental for the maintenance of healthy sleep, not solely as a treatment for sleep complaints. To facilitate behavior change, it is imperative that the importance of self-monitoring sleep behavior is emphasized, whether this be through objective or subjective methodologies (specific examples are highlighted in Fig. 3). Regular reviews of the data (by both the clinician and the person living with diabetes) are also pivotal to reinforce sleep behavior goals.

As we shift toward device-based measures of physical behaviors, future research should continue to integrate and assess the impact of sleep on glycemic control and overall health in type 2 diabetes. Given that interventional work is still in its infancy, there remain many unanswered questions the answers to which have the potential to reiterate and elevate the importance of sleep in diabetes clinical care. For example, how do lifestyle (e.g., weight loss, exercise [including timing]), and behavioral interventions affect sleep outcomes, physiological functioning, and well-being? Although weight loss results in a significant and clinically relevant improvement in OSA in a dose-response manner (96), the impact of newer classes of diabetes drugs on sleep-associated outcomes (e.g., OSA) is unknown. This is an important area for future research, particularly given their influence on body weight (especially compared with lifestyle interventions alone).

Health care professional training is pivotal, supported by established referral pathways into sleep (and other movement-based) therapies. Such training should also include appropriate behavior-change techniques and strategies to initiate and maintain healthy sleep, alongside an active lifestyle. Data management systems, coupled with self-monitoring tools and mobile health apps, are also important to capture, inform, and tailor advice around sleep. Qualitative methodologies may also shed more light on barriers to and facilitators of obtaining sufficient sleep. Ultimately, shared decision-making should contribute to the patient-centered holistic approach to diabetes management, but the recognition of the importance of sleep presents multiple interventional opportunities to induce glycemic and overall health benefits.

In summary, the latest ADA/EASD consensus report highlights sleep as a central component in the management of type 2 diabetes, placing it on a par with more traditional lifestyle factors, such as diet and exercise. Indeed, there is a plethora of observational data showing the associations between sleep and health outcomes. In contrast, although improvement in sleep characteristics (e.g., sleep extension) appears achievable in those with sleep disturbances, the impact that this has on type 2 diabetes and its associated outcomes is largely unknown. Therefore, this review should act as a timely reminder to incorporate sleep into clinical consultations and ongoing diabetes education.

Acknowledgements. The authors thank Mike Bonar and his creative team (Charlie Franklin and Shehnaz Jamal), who assisted with the conception and execution of figures.

Funding. The research was supported by the NIHR Leicester Biomedical Research Centre.

The views expressed are those of the authors and not necessarily those of the National Health Service, National Institute for Health and Care Research, or the U.K. Department of Health and Social Care.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. M.J.D. conceptualized the study. J.H. and A.C. researched data. J.H. wrote the manuscript. All authors reviewed and edited the manuscript and approved the final version of this manuscript.

Prior Presentation. Parts of this study were presented in abstract form at the 58th Annual Meeting of the European Association for the Study of Diabetes, 19–23 September 2022, Stockholm, Sweden.