1.1 Historical notes

Over the past few decades, the field of diagnosing and treating sleep disorders has evolved into a distinct and autonomous discipline within healthcare. Sleep medicine has developed along a trajectory similarly to other major medical specialties, progressing through stages of scientific discovery and clinical refinement. The formal recognition of any medical specialty typically follows the emergence of relevant new concepts in medical science, and sleep medicine is no exception. The milestones of fundamental research in sleep neurobiology and molecular mechanisms of circadian rhythmicity have been extensively described in other papers (Huang 2018; Schulz 2022), and are beyond the scope of this review.

Sustained basic and clinical research in sleep science has generated a wealth of knowledge, firmly establishing sleep medicine as a medical discipline in its own right (Shepard et al. 2005). Initially, sleep disorders were poorly understood and addressed with generalised treatments and broad recommendations. Today, specific diagnostic and therapeutic options are available for clearly defined conditions. Sleep medicine now boasts its own taxonomy (AASM 2023), accompanied by standardised diagnostic criteria and targeted treatment protocols, consolidating its place in modern healthcare. Moreover, the name ‘sleep medicine’ has increasingly been mentioned in the title of medical journals since its first appearance in 1988 (Lavie 1988) (Figure 1). The term ‘paediatric sleep medicine’ first appeared in a paper title at the turn of the century (Owens 2001), and has since been published under this designation at a rate of 1–4 items per year.

Groundbreaking biomedical discoveries across various disciplines have contributed to our understanding of sleep mechanisms and related pathologies. Table 1 presents a selected anthology of these accomplishments. In this paper, we aim to provide a concise overview of significant advances in insomnia, neurology, psychiatry, respiratory medicine, and paediatric sleep medicine below.

1.2 Insomnia Disorder

Acute, transient insomnia affects most people at least once in their lifetime, while chronic insomnia impacts up to 10% of the general population. Given its high prevalence, it is arguably the most common sleep disorder. Chronic insomnia is coupled to a high disease burden and a distinctly reduced quality of life (Morin et al. 2015).

Historical accounts of insomnia date back to ancient times, with various causes and treatments described in both the medical and lay literature. In pre-modern times, a sinful life was often considered to be a major cause of sleeplessness. Remedies ranged from substances like opiates and alcohol to lifestyle modifications—early forms of what we now recognise as sleep hygiene (Ahlheim 2018).

In modern medical classification, insomnia is defined by subjective symptoms, including at least one sleep complaint—such as difficulty falling asleep, maintaining sleep, or waking too early—accompanied by significant daytime impairment (e.g., fatigue, poor concentration, mood disturbances, and heightened anxiety) (Riemann et al. 2023).

Until recently, insomnia was primarily regarded as a symptom secondary to other medical or psychiatric conditions, with treatment focused on addressing the potential underlying disorder. However, with the publication of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM) (APA 2013) and the third edition of the International Classification of Sleep Disorders (ICSD) published by the American Academy of Sleep Medicine (AASM) (AASM 2014), the term ‘insomnia disorder’ emerged, acknowledging growing evidence that insomnia is an independent condition warranting its own diagnostic and therapeutic approach.

Research has since confirmed that insomnia is a significant, independent risk factor for cardiometabolic diseases and psychiatric disorders. Studies have demonstrated that effective insomnia treatment is associated with a reduced incidence of depression and relapse prevention (Furukawa et al. 2024; Hertenstein et al. 2019).

Psychological models, such as the 3P model (predisposing, precipitating, and perpetuating factors) and cognitive models, have long formed the foundation of insomnia research (Spielman et al. 1987). More recently, the concept of 24-h hyperarousal—a state of mental overactivity—has been implicated in its pathophysiology (Dressle and Riemann 2023). While traditionally identified through psychological assessments, emerging evidence suggests underlying neurobiological mechanisms (Palagini et al. 2023; Riemann et al. 2025).

Insomnia treatment encompasses both pharmacological and non-pharmacological strategies (Riemann et al. 2023). Pharmacological treatments have evolved from classical benzodiazepines to benzodiazepine receptor agonists, melatonin, sedating antidepressants, and newer orexin receptor antagonists. Non-pharmacological treatments, particularly cognitive behavioural therapy for insomnia (CBT-I), focus on techniques such as stimulus control, sleep restriction, and cognitive restructuring. Sleep hygiene practices, including reducing caffeine, nicotine, and alcohol intake, also play a role. Current insomnia treatment guidelines strongly advocate CBT-I as the first-line therapy due to its efficacy and the potential risks associated with medication, such as dependence. However, due to a shortage of trained CBT-I therapists in many regions, digital CBT-I (dCBT-I) has emerged as a promising alternative, expanding access to evidence-based insomnia care.

1.3 Neurological Sleep Disorders

Neurological sleep disorders encompass a wide range of conditions associated with diseases of the central nervous system. Among these, narcolepsy, restless legs syndrome (RLS), and rapid eye movement (REM) sleep behaviour disorder (RBD) stand out as prominent and well-characterised syndromes. Each of these is briefly described below.

1.4 Sleep and Psychiatric Disorders

Following the discovery of REM sleep in 1953, a number of pioneering researchers began exploring sleep patterns in individuals with psychiatric disorders—most notably affective disorders and schizophrenia. Although early studies hoped to uncover specific sleep abnormalities tied to particular psychiatric conditions, these expectations were largely unmet. Nevertheless, the research was innovative in demonstrating that psychiatric phenomena could be linked to electrophysiological markers of brain activity during sleep, offering critical evidence for the biological basis of mental illness.

Interestingly, the first polygraphic sleep study in patients with affective disorders predated the discovery of REM sleep. Conducted in 1946 by Diaz-Guerrero et al., the study examined bipolar depression and noted ‘a greater proportion of sleep which is light and more frequent oscillations from one level of sleep to another than normally occurs’ (Diaz-Guerrero et al. 1946). However, subsequent studies revealed considerable variability in sleep features among patients. For instance, some individuals displayed signs of hypersomnia rather than insomnia (Kupfer et al. 1972). Other findings once thought specific to depression—such as shortened REM latency (Kupfer and Foster 1972)—were also found to lack diagnostic specificity and were not consistently present across all patients. These varying results helped shape the current view of psychiatric conditions as overlapping, transdiagnostic syndromes involving emotional and cognitive dysregulation. Genetic studies have since reinforced these observations, showing shared genetic influences among psychiatric diseases, for example, schizophrenia and bipolar disorder, as well as between sleep traits and psychiatric conditions such as insomnia and depression (Lane et al. 2017).

Several papers have acclaimed the German psychiatrist Johann Christian August Heinroth (1773–1843)—the first university professor of psychiatry—as a pioneer of sleep deprivation in the treatment of ‘melancholia’ (Steinberg and Hegerl 2014). The antidepressant effects of total sleep deprivation were later confirmed (Pflug and Tolle 1971), and laid the foundation for specific treatment protocols—some of which are still in use today, even though their exact mechanisms remain under investigation. Advances in circadian rhythm research furthered the therapeutic possibilities. Rosenthal et al. observed that reduced daylight exposure during winter contributed to depressive symptoms in some individuals, which could be alleviated with exposure to bright artificial light (Rosenthal et al. 1984). This finding marked the formal introduction of light therapy as a treatment for seasonal affective disorder, and more recently, also for non-seasonal depression (Tao et al. 2020).

Research into schizophrenia was initially motivated by the resemblance between daytime hallucinations in psychosis and hallucinatory dream experiences. However, the anticipated abnormalities in REM sleep among patients with schizophrenia were not consistently observed. In fact, Dement and others found in early polygraphic studies that REM periods were comparable between patients and healthy controls (Dement 1955). A reduction in slow-wave sleep in schizophrenia has been reported, but this finding lacks specificity (Igert and Lairy 1962). Using high-density EEG montages, researchers at the University of Wisconsin demonstrated a reduction in centroparietal EEG power in specific frequency bands, along with decreased sleep spindle activity in schizophrenia (Ferrarelli et al. 2007). These neural changes are associated with cognitive impairments and are considered endotypic features of the disorder.

Although sleep disturbances have been described in psychiatric illness for centuries, and were already mentioned in ancient accounts of melancholia, they only emerged into formal diagnostic criteria for psychiatric research in the 1970s (Feighner et al. 1972). The DSM-III formally included insomnia or hypersomnia as diagnostic criteria for mood disorders and recognised reduced need for sleep as a core symptom of mania in bipolar disorder (APA 1980). The critical role of sleep in mental health became even more evident through late 20th-century epidemiological studies, which showed that insomnia often precedes depression (Baglioni et al. 2011; Ford and Kamerow 1989). These findings spurred interest in treating insomnia as a preventative strategy. Contemporary research has since demonstrated the efficacy of CBT-I in patients with co-occurring mental disorders, particularly depression, posttraumatic stress disorder, and alcohol dependence (Hertenstein et al. 2022).

Today, sleep disturbances and psychiatric disorders are widely recognised as bidirectionally related (Baglioni et al. 2016; Riemann et al. 2020). Sleep problems can contribute to the onset, persistence, and recurrence of psychiatric disorders, while psychiatric conditions can likewise cause sleep disturbances. In some cases, the two may also occur independently. By eliminating the distinction between primary and secondary sleep disorders, the DSM-5 acknowledges sleep disorders as independent conditions that warrant separate diagnosis and treatment, even when psychiatric comorbidities are present.

1.5 Sleep-Disordered Breathing

Although previous cases had been published—some noting breathing pauses during sleep—a case report by Burwell et al. was key in establishing the link between morbid obesity, hypersomnolence, and sleep-disordered breathing (SDB) (Burwell et al. 1956). This report provided a detailed clinical account and introduced the term Pickwickian Syndrome, drawing a parallel between the patient's characteristics and Joe the Fat Boy, a character from Charles Dickens' novel The Pickwick Papers. While Burwell et al. observed snoring, they did not report episodes of breathing cessation during sleep. Instead, they attributed the hypersomnolence to alveolar hypoventilation with hypercapnia—all of which resolved with weight loss.

Subsequent studies, primarily by European researchers, included both daytime and nocturnal sleep analyses of Pickwickian patients, and revealed periodic respiration with frequent apnoeic episodes (Lavie 2008). Initially, hypersomnolence was thought to result from ‘carbon dioxide poisoning’, but researchers soon recognised that apnoea was the primary cause. This was confirmed when tracheostomy led to the immediate normalisation of hypersomnolence and respiratory blood gases, demonstrating that sleep-related upper airway obstruction was responsible for both disrupted breathing and excessive daytime sleepiness.

At Stanford University, Christian Guilleminault and colleagues expanded this understanding by demonstrating that sleep apnoea, even in the absence of obesity, was linked to a broad range of clinical symptoms. In their seminal publication ‘The Sleep Apnea Syndromes’, they made a comprehensive description of signs, symptoms, and laboratory findings in obstructive sleep apnoea (OSA) (Guilleminault, Tilkian, et al. 1976). They recommended that the diagnosis should be considered when at least 30 apnoeic episodes could be observed during 7 h of sleep. This work was pivotal as it provided evidence that previously unexplained medical conditions stemmed from OSA, which from then on was established as a distinct medical disorder.

While OSA can lead to severe, potentially life-threatening complications, treatment options in the 1960s and 1970s were limited to weight loss and tracheostomy. A breakthrough came in 1981 when Sullivan et al. introduced continuous positive airway pressure (CPAP) as a treatment for OSA (Sullivan et al. 1981). This revolutionary therapy induced complete resolution of respiratory disturbances and immediate symptom relief, even in the most severe cases. It quickly became the gold standard for OSA management. The success of CPAP not only prompted the mass production of CPAP devices but also accelerated the use of polysomnography for diagnostic purposes of OSA and other common sleep disorders. Given the high prevalence of OSA, sleep laboratories and multidisciplinary sleep centres mushroomed worldwide.

However, as CPAP therapy can be cumbersome and poorly tolerated by some patients, alternative treatments had to be developed. Pharyngoplasty, initially introduced to treat snoring (Fujita et al. 1981), often proved unreliable in OSA management (Epstein et al. 2009). Mandibular advancement devices (MRAs) have since become the most commonly used CPAP alternative (Schmidt-Nowara et al. 1995). While MRAs are less efficacious than CPAP in reducing the apnoea–hypopnoea index (AHI), they appear equally effective in improving symptom scores (Ramar et al. 2015).

Over the past decade, research has highlighted the complex pathophysiology of OSA, revealing that multiple factors contribute to its development (Eckert et al. 2013). This growing understanding has paved the way for personalised treatment strategies that attempt to target specific endotypic traits in individual patients (Edwards et al. 2019).

1.6 Paediatric Sleep Disorders

Sleep disorders in children remained largely overlooked in medical literature until the 1970s. By the 1980s, paediatric textbooks began addressing sleep disturbances as a distinct area of interest within paediatrics. Since then, the field has gained increasing recognition, leading to the publication of numerous medical and public books. However, it was not until 1995 that the first comprehensive textbook, Principles and Practice of Paediatric Sleep Medicine was published (Ferber and Kryger 1995). This milestone helped establish paediatric sleep medicine as an independent subdiscipline, often developing under the broader framework of adult sleep medicine. As interest grew among both general and specialised paediatricians, dedicated paediatric sleep centres began to emerge.

At the turn of the 21st century, sleep research expanded across multiple medical and social health fields, revealing distinct differences between paediatric and adult sleep disorders. Population studies highlighted the health risks associated with declining sleep duration in modern societies, particularly among children. Observational research in paediatric sleep physiology and pathophysiology led to key discoveries. For example, the identification of rapid eye movements in infants during paradoxical sleep—characterised by increased motility, irregular breathing, and elevated heart rate—contributed to the understanding of REM sleep, later confirmed in adults (Aserinsky and Kleitman 1953). Meanwhile, studies on NREM sleep demonstrated its role in typical paediatric parasomnias, which originate from slow-wave sleep rather than REM sleep as previously surmised (Broughton 1968).

Yoss and Daly provided the first systematic description of narcolepsy in children aged 3 to 14 years and emphasised the frequent misdiagnosis of the disorder (Yoss and Daly 1957).

Sleep was also linked to mortality risks such as sudden infant death syndrome (SIDS), which accounts for nearly one-third of deaths in infants aged 1–6 months. The research of André Kahn on respiratory events during sleep in infants was fundamental. He was among the first authors to observe apnoeas during sleep in future SIDS victims (Kahn 1979). He pioneered the development of ambulatory monitoring systems to detect respiratory spells in infants (Kahn and Blum 1982). Yet, despite extensive research, the precise causes of SIDS are not fully elucidated (AAP 1992).

Paediatric OSA was initially described by Guilleminault, Eldridge, et al. (1976). In a subsequent report of 50 patients, it was shown that the clinical phenotype of OSA in children is different from that in adults (Guilleminault et al. 1981). While excessive daytime sleepiness is a key symptom in adults, children with OSA or narcolepsy often present with hyperactivity or attention deficits instead (Chervin et al. 2002). Research on paediatric OSA has evolved over time, initially focusing on its impact on growth and development in the 1980s, shifting towards cognitive and behavioural concerns in the 1990s, and more recently exploring its associations with obesity, inflammation, and biomarkers (Bruni and Ferri 2015). Similarly, paediatric RLS, first described in 1994 (Walters et al. 1994), differs from the adult form. In children, RLS is often linked to growing pains and attention-deficit hyperactivity disorder (ADHD), necessitating the development of specialised diagnostic criteria (Picchietti et al. 2013). These distinctions underscore the unique characteristics of paediatric sleep disorders despite their similarities to adult conditions. In recognition of these differences, dedicated paediatric sections have been incorporated into the ICSD (AASM 2023).

Paediatric sleep medicine continues to thrive, driven by the unique nature of sleep throughout human development. Scientific advances in the field have been propelled by numerous researchers over the past decades (Bruni and Ferri 2015). Currently, PubMed records between 3000 and 4000 annual publications on sleep in children aged 1–18 years. The field's growth is also reflected in the establishment of specialised scientific societies. The European Paediatric Sleep Club (EPSC) was founded in 1991 under the umbrella of the European Sleep Research Society (ESRS). Thanks to the efforts of André Kahn and the pivotal contributions of Christian Guilleminault, the EPSC later evolved into the International Paediatric Sleep Association (IPSA), which was officially established during the World Association of Sleep Medicine meeting in Berlin on October 13, 2005 (Bruni and Ferri 2015).

Through ongoing research and collaboration, paediatric sleep medicine continues to expand, offering deeper insights into childhood sleep disorders and their long-term health implications.

4.1 Generic Issues in Sleep Medicine

Despite the universal relevance of sleep and the huge prevalence of sleep disorders in the population, sleep medicine all too often faces marginalisation and poor access to resources, particularly in terms of access to and funding of healthcare as well as clinical research.

A major cultural hurdle within the field is the misconception that, because everyone sleeps, everyone understands sleep. This oversimplification undermines the expertise required to manage sleep disorders and perpetuates misinformation. The rise of self-proclaimed sleep consultants and social media influencers further complicates the landscape. These unregulated sources often promote oversimplified or misleading advice, creating confusion among patients and undermining evidence-based care.

Recent years have also seen growing reliance on consumer-grade and portable diagnostic tools. While these technologies can increase access to the measurement of sleep variables, they are limited by a lack of critical features such as neurophysiologic signal recording needed for accurately measuring sleep stages and detecting subtle but clinically significant disturbances. Conversely, indications of disturbed nocturnal sleep may cause unnecessary health concerns in people without sleep complaints (Baron et al. 2017). Direct-to-consumer apps and over-the-counter devices bypass clinical interpretation, reducing sleep medicine to a consumer service model. This trend not only diminishes the role of qualified specialists but also weakens the multidisciplinary approach necessary for accurate diagnosis and effective treatment.

Several areas within sleep medicine remain marked by uncertainty and conceptual blind spots. While several sleep disorders retain diagnostic validity when viewed through robust theoretical frameworks, such as Koch's postulates (Hucklenbroich 2014; Spitzer et al. 1978) and the Bradford-Hill criteria (Hill 1965), others rest on less secure foundations. Some paradigms endure unchallenged for extended periods, potentially impeding progress. As such, emerging models that question traditional assumptions should be welcomed. The field still lacks tried-and-tested biomarkers to support many presumed diagnoses, and their discovery remains a critical goal. This ongoing search for stronger, evidence-based disease models is reflected in the subsequent updates of the ICSD described above.

Pharmacological therapies for sleep disorders have largely emerged from empirical observations rather than rigorous trials. For many conditions, such as RBD and other parasomnias, randomised controlled trials (RCTs) are lacking, often due to funding challenges. Even when RCTs exist, they rarely exceed 1 year in duration, and data on long-term side effects remain limited. Treatment guidelines often rely on placebo-controlled studies rather than comparisons with existing, cost-effective alternatives, particularly for newer agents targeting residual sleepiness in OSA or central hypersomnolence disorders (Dauvilliers et al. 2013). These gaps hinder the development of a robust, evidence-based pharmacological framework adaptable to diverse healthcare systems (Bassetti et al. 2021). An additional, underexplored concern is bioequivalence after drug patent expiration. Anecdotal reports suggest that some generic formulations of drugs, such as modafinil or sodium oxybate, may be less effective or safe than brand-name versions. Similar issues have been better documented in other fields like cardiology, epilepsy, and transplantation medicine (Dauvilliers et al. 2022; Lechat 2022).

4.2 Insomnia Disorder

A significant barrier in the management of chronic insomnia is limited engagement with professional healthcare services. Epidemiological data suggest that fewer than 20% of individuals suffering from chronic insomnia actively seek medical attention (Chalet et al. 2024). The majority remain undiagnosed and untreated or instead rely on non-clinical interventions such as internet-based resources, self-help literature, or over-the-counter medications obtained through pharmacies or alternative sources.

This presents a clinical concern, as proper evaluation through differential diagnosis is essential in cases of chronic insomnia. Evidence-based treatment options—both pharmacological and psychotherapeutic—are well established. However, when these are bypassed, patients are at increased risk for developing comorbid psychiatric conditions, particularly anxiety and depressive disorders (Baglioni et al. 2011; Hertenstein et al. 2019).

Given the established association between chronic insomnia and poor mental health outcomes, it is reasonable to assume that early and effective treatment may mitigate these risks (Boland et al. 2023). Thus, raising awareness about insomnia as a serious clinical condition is critical—not only among the general public but also within the healthcare system itself.

To address the problems of diagnosis and access to appropriate treatment, recent work has proposed a stepped-care model for the diagnosis and management of insomnia, offering a scalable framework that aligns clinical resources with patient needs (Baglioni et al. 2023). This approach emphasises initial low-intensity interventions (e.g., dCBT-I) followed by more specialised care for non-responders, thereby improving access while preserving clinical efficacy.

For clinicians, adopting structured, evidence-based care pathways and promoting early recognition of insomnia as a treatable disorder can significantly improve outcomes and reduce the long-term burden of comorbid mental health conditions.

Currently, no medications have been approved by the U.S. Food and Drug Administration (FDA) or the European Medicines Agency (EMA) for the long-term treatment of chronic insomnia disorder. While most chronic medical conditions have access to ‘on-label’ pharmacological therapies, the lack of approved long-term treatments for insomnia represents a significant gap in the therapeutic landscape. Due to the risks of tolerance and adverse effects, benzodiazepines and benzodiazepine receptor agonists are not recommended for use beyond 4 weeks (Riemann et al. 2023). As a result, long-term pharmacological management often relies on ‘off-label’ use of medications such as antidepressants or antiepileptics (Dujardin et al. 2022).

The introduction of dual orexin receptor antagonists (DORAs) marks a notable recent advancement in insomnia pharmacotherapy (Mignot et al. 2022; Riemann et al. 2023). Among these, daridorexant is the only compound thus far approved by the EMA. In a clinical trial, daridorexant at a 50 mg dose was shown to be safe and well tolerated, with sustained improvements in both sleep parameters and daytime functioning after 12 months of continuous use (Kunz et al. 2023). Nonetheless, further RCTs will be needed before DORAs can be formally approved for long-term treatment of chronic insomnia disorder.

4.3 Neurological Sleep Disorders

4.4 Sleep and Psychiatric Disorders

Psychological and psychiatric factors may play a contributing role in clinical manifestations of sleep disorders in general, in addition to the overlay of medication side-effects, pain, other comorbid medical conditions, and poor sleep hygiene. It is key for the sleep specialist to be able to assess the importance of all these factors when teasing out treatment strategies to manage disturbed sleep.

Sleep disturbances are both a cause and consequence of psychiatric disorders (Freeman et al. 2020); however, no unified theoretical framework currently exists to explain the bidirectional and disorder-specific interactions between sleep and psychopathology. It also remains unclear which individuals with sleep disturbances are at increased risk of developing psychiatric conditions, and conversely, which psychiatric presentations are most likely to precipitate chronic sleep dysfunction.

Identifying biomarkers linked to specific psychiatric disorders or underlying neuropsychological traits could enable both early intervention and the development of personalised treatment strategies. Moreover, tailoring psychiatric pharmacotherapy based on individual sleep–wake profiles may enhance treatment efficacy and reduce adverse effects (Paunio 2021).

Despite growing evidence of the clinical relevance of sleep in psychiatric populations, a major limitation lies in the disconnect between research and clinical practice. In most psychiatric care settings, sleep and circadian rhythm disturbances are not systematically assessed, monitored, or integrated into treatment planning. As a result, the full therapeutic potential of evidence-based sleep interventions—such as CBT-I or chronotherapeutic approaches—remains underutilised in routine care.

Closing this gap will require greater integration of sleep assessment into psychiatric diagnostics, education of mental health professionals on the relevance of sleep in psychopathology, and broader implementation of targeted, sleep-focused interventions (Martin et al. 2024).

4.5 Sleep-Disordered Breathing

The prevailing paradigm in OSA is that recurrent apnoeas and hypopnoeas lead to systemic disease manifestations in a dose-dependent manner (AASM 2023). Conventionally, the AHI has been used to categorise OSA disease severity as mild, moderate, or severe (AASM 1999). However, recent evidence has highlighted significant limitations in the AHI's metric properties. It is a poor predictor of key clinical outcomes such as EDS, hypertension, health-related quality of life, and treatment response. Consequently, the assumed dose–response relationship between the frequency of respiratory events and clinical impact appears to be flawed. Reliance on the AHI leads to an overestimation of OSA's clinical relevance (Pevernagie et al. 2020).

Scientific interest has meanwhile shifted from merely counting respiratory events to understanding the pathophysiologic exposures they induce, particularly their systemic effects. It is now presumed that biomarkers of exposure may offer improved prediction of OSA severity. Growing evidence highlights the central role of intermittent hypoxia in driving cardiovascular comorbidities. Emerging biomarkers, such as hypoxic burden, may provide a more accurate assessment of the clinical impact of OSA (Stanczyk et al. 2025). However, clinical observations at the individual patient level reveal considerable variability and unpredictability also in the relationship between the degree of hypoxic exposure and symptom burden. Although markers of hypoxia may outperform the AHI in some respects, their overall predictive power is likely to remain limited.

The frequently observed mismatch between OSA severity, as assessed by pathophysiological markers, and a patient's symptom profile suggests the involvement of additional, less apparent factors. These factors may elude detection through conventional physiological recordings, as they could stem from an individual's genomic constitution or from mechanisms that are not captured in routine clinical assessments of sleep. Recent evidence suggests that certain vulnerability factors modulate the association between performance on cognitive testing and the severity of OSA (Marchi et al. 2024). Such underlying influences may be relevant as, ultimately, they could determine a person's susceptibility vs. resilience to the harmful systemic effects of OSA. Nonetheless, research in this area remains in its early stages.

In light of these insights, it may be time to rethink the disease model of OSA. At its core, OSA is driven by repetitive physiological strain caused by respiratory events and their adverse systemic effects. This chronic, night-by-night strain gradually inflicts damage on the body. However, the extent and progression of end-organ injury are shaped not only by the magnitude of the strain but also by the intrinsic resilience of the affected organs. It is this dynamic interplay between chronic stress and individual resilience that ultimately determines the clinical manifestations of OSA in the course of time.

A repetitive strain injury (RSI) model of OSA, illustrated in Figure 2, conceptualises the disease mechanism as a cascade. The process begins with the burden of repetitive respiratory events—a starting point that, on its own, has proven to be a poor predictor of clinical outcomes. The next stage in the cascade is the systemic strain, representing the cumulative exposure to the physiological consequences of these events. The final stage is injury, reflected in the clinical signs and symptoms associated with OSA.

When applying this model to individual patients, two key uncertainties arise. First, in the ‘downward’ direction: how robust and consistent is the pathogenic cascade from respiratory events over pathophysiological exposure to systemic injury? What is the validity of this RSI model? Second, in the ‘upward’ direction: to what extent can observed symptoms and clinical signs be confidently attributed to OSA, given their often non-specific nature and potential overlap with other conditions?

Currently, these uncertainties pose significant challenges in clinical practice. While the ‘repetitive strain’ component can be measured with reasonable accuracy using existing technologies, the prediction of downstream ‘injury’ remains imprecise. To enhance clinical decision-making, additional biomarkers with validated prognostic value will be required. Similarly, a satisfying therapeutic response to CPAP or alternative treatments cannot be assumed unless it is clear that the symptoms stem from OSA. If the manifestations are unrelated, treatment may yield disappointing results. Therefore, there is also a need for the development and validation of markers that add specificity to the causal link between OSA and its clinical manifestations. Such markers would significantly improve the ability to predict treatment outcomes and personalise care.

4.6 Paediatric Sleep Disorders

Paediatric sleep medicine faces several challenges. Sleep disorders in children, including insomnia, parasomnias, and sleep-disordered breathing, are frequently underdiagnosed or misattributed to behavioural issues. This is partly due to limited paediatric sleep training among primary care providers, resulting in missed opportunities for early intervention. Adding to the problem is the scarcity of specialised paediatric sleep centres, and, in existing centres, the limited access to diagnostic and therapeutic services. PSG and CBT-I are not widely available, with long waiting times and few public health options. In addition, multidisciplinary care models remain rare, leading to fragmented and inefficient service delivery (Reynolds et al. 2023).

These challenges collectively delay accurate diagnoses, allow sleep problems to persist or worsen, and contribute to comorbid conditions. They also increase the burden on healthcare providers, who must correct misinformation and rebuild trust with families.

The lack of trained professionals further impairs the implementation of evidence-based treatments, making interventions less effective. Left untreated, paediatric sleep disorders can lead to increased healthcare utilisation, hinder cognitive and emotional development, impair academic performance, and contribute to long-term societal costs.

Furthermore, pharmacotherapy in paediatric sleep disorders has often followed extrapolation from adult data or anecdotal practice. Many existing studies are short in duration, underpowered, and focused on symptom relief rather than developmental outcomes.

Paediatric sleep medicine requires prioritisation at the public health level through improved medical training, equitable access to resources, and integrated care models to address systemic challenges efficiently. Finally, emerging technologies like AI hold promise for improving diagnostic accuracy, but their application in paediatric sleep medicine must be carefully validated to ensure safety and effectiveness (Schlarb et al. 2025).

5.1 Resolution of Areas of Uncertainty

Some sleep disorders, for example, parasomnias, remain poorly understood due to the challenges in diagnosing them—often only by exclusion—and the inconsistent clinical experiences reported globally (Hrozanova et al. 2019). Advancing our understanding in this area will require increased research funding, enhanced cross-disciplinary and interinstitutional collaboration, as well as a stronger emphasis from the medical community on recognising sleep as a fundamental component of overall health.

In OSA and other domains of sleep medicine, the main focus of investigation has been on male patients. This male-predominant bias has led to a stark underrepresentation of females and children in epidemiologic studies and clinical sleep medicine reports (Hasuneh et al. 2024; Magnusdottir and Hill 2024; Moscucci et al. 2025). It has become evident that these populations often present with distinct phenotypic features compared to adult males. For instance, while men with OSA tend to be sleepy and typically exhibit loud snoring and witnessed apneas, women are more likely to report symptoms such as insomnia, fatigue, morning headaches, and mood disturbances, which may lead to misdiagnosis or underdiagnosis (Jordan et al. 2014; Pavlova and Sheikh 2011). Similarly, children with OSA may exhibit behavioural problems, hyperactivity, or learning difficulties rather than classic respiratory symptoms (Marcus et al. 2012). Obviously, these knowledge gaps must be addressed to improve diagnostic accuracy and to develop personal therapeutic approaches for women and children.

Last but not least, there are gaps in knowledge regarding the diagnosis, management and clinical significance of sleep-disordered breathing in subjects with intellectual disability (Riha et al. 2024) and during different physiological states, for example, ageing and extreme old age as well as menopause and pregnancy (Johns et al. 2022). Work in these areas is sparse, and a more consolidated effort would be required to manage clinical implications and to resolve many questions that remain unanswered to date.

5.2 Technological Innovation

As in other domains of healthcare, sleep medicine is situated in a metamorphic environment, driven by rapid advancements in technology that are reshaping how disorders are being diagnosed, monitored, and treated. Several areas of interest are under development and tend to converge, that is, wearable technology, application of AI, remote monitoring, digital therapeutics, and advanced diagnostics. This evolution holds promise for enhancing accessibility, personalisation, and efficiency across the field. While several branches of academia and industry are involved worldwide, the Sleep Revolution project is worth mentioning in this context.

To address prevailing challenges in OSA, the Sleep Revolution project was funded in 2021 by the EU's Horizon 2020 research and innovation programme (Arnardottir et al. 2022; Sleep_Revolution 2021; Sleep_Revolution_EU 2021). The consortium consists of 39 partners from Europe and Australia, and combines multidisciplinary competencies and resources from academia, healthcare, and industry. The project aims to fundamentally change the clinical management of SDB by introducing a new diagnostic and digital management paradigm. It offers a comprehensive scope of diagnostic investigations, including a mobile application with a sleep diary and objective cognitive tests, a novel sleep questionnaire, a smart watch, and PSG recording in the home setting. A digital platform functions as a bridge between researchers, patients, and clinicians. Moreover, advanced telemedicine technology, novel machine learning algorithms for diagnosis, and a high degree of participatory patient involvement are combined in several clinical studies. The overall goal is to enable personalised treatment, enhance patient engagement, reduce diagnostic costs, and improve access to care. The initiative aims to establish new standardised guidelines in sleep medicine.

5.3 Organisation of Sleep Medicine as a Discipline

As described above, sleep medicine is founded on multidisciplinary interaction. Yet, interdisciplinary sleep healthcare remains inconsistent and fragmented. Many sleep practitioners focus routinely on a single aspect of sleep medicine, for example, CPAP treatment of OSA or CBT-I (Edinger et al. 2021). The establishment of a unifying training pathway, as discussed above, for all practitioners working in the field would contribute substantially to interdisciplinary work, better understanding of the pathophysiology of most relevant sleep disorders, and more holistic care of individual patients (De Backer et al. 2011; Pevernagie et al. 2009; Strohl 2011). Formal education in sleep medicine should already start in undergraduate and continue in postgraduate healthcare training.

In settings where multidisciplinary sleep medicine is practised, integrated healthcare can be assured by implementing applicable standards. Sleep centre accreditation ensures the quality of diagnosis and treatment in sleep medicine. Originally introduced in Europe in 2006 for centres performing inpatient PSG and vigilance tests (Pevernagie and Steering Committee of European Sleep Research 2006), the standards have since been updated to reflect changes in clinical practice, including the widespread use of ambulatory diagnostics. In the recently revised European accreditation guideline, a tiered system was introduced (Hartley et al. 2024). Levels 1 and 2 sleep centres provide full laboratory-based diagnostics, with level 1 typically being university-affiliated and involved in teaching and research. Level 3 and 4 centres offer a mix of inpatient and ambulatory testing, with level 3 performing PSG and level 4 focusing on polygraphy, often for detecting OSA. Accreditation emphasises the importance of trained medical teams, proper equipment, patient care pathways, and adherence to national and European guidelines.

Finally, advocacy for sleep medicine should be extended to regulatory administrations at a national level. Healthcare authorities should be briefed about the adverse clinical and socio-economic outcomes of untreated sleep disorders. Besides enhancing sleep medicine education among healthcare professionals across various medical disciplines, the establishment of sleep medicine as a discipline in its own right is the ultimate goal. Sleep medicine has become an authorised medical subspecialty in France and Germany. In many other European countries, sleep medicine is still off the grid in the national healthcare system. The ESRS fosters the accreditation of sleep medicine at a European level. An ad hoc taskforce has been appointed to approach the European Union of Medical Specialists (UEMS) and to accredit sleep medicine as a multidisciplinary joint committee (UEMS 2025). Being accredited by the UEMS will offer opportunities for official recognition of sleep medicine in other European countries and for international harmonisation of educational programmes.

5.4 Transition Towards Precision Medicine

The sleep medicine field is intrinsically complex. Diagnostic procedures are intensive and possibly already beyond their expiration date. The adoption of newer, more precise approaches to enhance both accuracy and accessibility is required. The demand for innovation necessitates not only technical development but also identification of novel markers that enhance diagnostic specificity and prognostic relevance.

The quest for suitable biomarkers is paramount in contemporary sleep medicine. A biomarker is defined as ‘a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacological responses to a therapeutic intervention’ (Biomarkers Definitions Working 2001; Califf 2018; Strimbu and Tavel 2010).

Hypocretin-1 can be measured in the cerebrospinal fluid. It is a sensitive and specific indicator, as a level below 110 pg/mL has high accuracy for the diagnosis of narcolepsy type 1 in adults. By contrast, the AHI fails as a biomarker in corroborating causality of clinical manifestations and severity of SDB. However, the AHI is a sensitive indicator of therapeutic adequacy, as it will usually drop to normal values if treatment with CPAP or MRA is efficacious. Therefore, the AHI should not be discarded from the diagnostic framework but may be repurposed as an outcome variable. This approach may also refine the concepts of therapeutic outcomes.

Using pathophysiologic indices such as the AHI for the assessment of therapeutic effects enables the discrimination of efficacy versus effectiveness of treatment. The terms efficiency, efficacy, and effectiveness are often used interchangeably, but they have distinct meanings, especially in the context of research, healthcare, business, and policy analysis (Figure 3) (Patel 2021). In the clinical setting, ‘efficacy’ means that the treatment is successful at controlling the pathogenetic mechanism (e.g., normalising the AHI in OSA). ‘Effectiveness’ means that the treatment also results in adequate improvement of symptoms and signs (e.g., reducing sleepiness and high blood pressure in OSA).

Distinguishing between these aspects is crucial, as treatments may differ in their physiological impact while yielding similar clinical outcomes. For example, treatment with MRA often provides comparable symptomatic relief to CPAP, despite achieving a smaller reduction in the AHI (Ramar et al. 2015). Confusion arises when efficacy—defined as the degree of pathophysiological control—is inconsiderately equated with effectiveness, which pertains to clinical benefit experienced by the patient. In OSA, the concept of ‘mean disease alleviation’ implies that a partial decrease of the AHI under treatment can signify an overall reduction in disease burden (Budhiraja et al. 2025). However, this assumption holds only if clinical symptoms improve in parallel. If symptoms such as EDS persist despite adequate AHI control, the concept becomes ambiguous. Indeed, it is well established that OSA treatment may be efficacious in physiological terms but ineffective in alleviating symptoms of sleepiness (Bonsignore et al. 2021). This potential disconnect between efficacy and effectiveness is particularly relevant for non-CPAP therapies aimed at specific endotypic traits in OSA. Whether such treatments, while being efficacious, also produce meaningful symptomatic relief remains subject to further investigation (Stanczyk et al. 2025).

Specificity enhancing biomarkers will have to be found in many areas of sleep medicine, for example, sleep deprivation, sleep disturbances in psychiatric disorders, SDB and chronic insomnia. However, relevant biomarkers for sleep disorders are hard to find (Borker and Strohl 2025). Taking OSA as an example, systematic reviews studying several candidate protein and genomic markers have found inconsistent findings and have not resulted in either sensitive or specific actionable markers to predict or intercede on comorbidity development.

In the future, precision medicine may be envisaged as a new paradigm to overcome the limitations of the current diagnostic workup. Precision medicine is defined as ‘treatments targeted to the needs of individual patients on the basis of genetic, biomarker, phenotypic or psychosocial characteristics that distinguish a given patient from other patients with similar clinical presentations’, the main objective being to ‘improve clinical outcomes for individual patients while minimizing unnecessary side effects for those less likely to respond to a given treatment’ (Jameson and Longo 2015). The rationale for embracing precision medicine is the observation that chronic diseases are ‘complex’ and ‘heterogeneous’. In this setting, ‘complex’ means that they have several components with nonlinear dynamic interactions, whereas ‘heterogeneous’ indicates that not all of these components are present in all patients or, in a given patient, at all timepoints. Obviously, many sleep disorders meet this definition of chronicity. While precision medicine is emerging on the horizon, the way to this destination will be long and uncertain.

In summary, sleep medicine stands at a crossroads. Its future will depend on redesigning the mission and vision of the field, remedying prevailing misconceptions, confronting structural challenges, embracing innovation, and validating its crucial role in overall health.