This study utilized a cross-sectional design with an internetbased survey aimed at individuals clinically diagnosed with narcolepsy. A total of 235 participants were recruited from St. Vincent’s Hospital in Suwon, South Korea. All participants had previously received a narcolepsy diagnosis and were undergoing treatment at the time of the survey. The study protocol received approval from the Institutional Review Board (IRB) of St. Vincent’s Hospital (No.VC22QISE0225), and written informed consent was obtained from each participant.

Diagnostic latency was defined as the interval (in years) between the age at symptom onset and the age at formal diagnosis, calculated by subtracting the age at symptom onset from the age at diagnosis. Diagnostic delay was operationally defined as a time interval exceeding 12 months between symptom onset and formal diagnosis. This threshold was based on prior epidemiological studies indicating that most narcolepsy patients are diagnosed more than 1 year after their symptoms begin. It represents a critical period for timely clinical recognition and has been established as a standard for differentiating prompt intervention from the extended latency typically seen in narcolepsy [19]. Specifically, previous study found that over 80% of patients experienced delays in diagnosis, with pediatric onset being a significant predictor of these delays. By adopting a cutoff of more than 1 year, our classification aligns with established literature and allows for comparison across studies [19]. Family history was defined as a record of major sleep abnormalities, such as narcolepsy, REM sleep-related disorders, hypersomnia, or insomnia, among first-degree relatives. Participants were classified post hoc into two groups based on diagnostic latency: a non-delay group (latency ≤1 year) and a delay group (latency >1 year). Data collection was conducted from September to December 2024.

Statistical analyses were conducted using SPSS software (version 27.0; IBM Corp.). Continuous variables are reported as means±standard deviations, and categorical variables are shown as frequencies and percentages. The primary objective was to compare clinical and demographic characteristics between two groups based on diagnostic latency: the non-delay group (diagnostic latency of ≤1 year) and the delay group (diagnostic latency of >1 year). Multivariate logistic regression analysis was employed to identify independent predictors of diagnostic delay, while controlling for potential confounders.

For univariate comparisons, independent t-tests were used for continuous variables, including onset age, diagnosis age, and clinical scores (e.g., Adult ADHD Self-Report Scale). Chisquare tests, or Fisher’s exact tests when appropriate, were employed to compare categorical variables such as sex distribution, narcolepsy type, family history of sleep disorders, and the prevalence of clinical symptoms (e.g., EDS, cataplexy, hallucinations).

To identify independent predictors of diagnostic delay and calculate adjusted effect sizes, a multivariate logistic regression analysis was conducted. To identify independent predictors of diagnostic delay, variables demonstrating a trend toward significance (p<0.10) in the univariate analyses (onset age and EDS) were included in the model. Additionally, sex and family history were included a priori as covariates, regardless of their univariate significance. This decision was based on their established clinical relevance in prior literature regarding sleep disorder presentation, family history, and healthcare-seeking behaviors [5,8,13]. The model was constructed using the enter method. The results are presented as odds ratios (ORs) along with their corresponding 95% confidence intervals (CIs).

Initial univariate comparisons revealed clear clinical differences between the two groups. The age at symptom onset was significantly lower in the delay group (15.13±4.88 years) compared to the non-delay group (17.89±7.91 years; p=0.002), indicating that younger patients are at a substantially higher risk for prolonged diagnostic latency (Table 2).

In terms of clinical symptoms, the prevalence of EDS was significantly higher in the delay group (98.9%) than in the nondelay group (90.6%; p=0.007). While EDS is a common symptom of narcolepsy, patients in the delayed diagnosis group more frequently reported extreme daytime sleepiness at symptom onset, suggesting that diagnosis often occurred only after the symptom burden became significant. Notably, there was no significant difference in family history of sleep disorders (p=0.241), implying that having an affected relative does not necessarily expedite the diagnostic process. Clinical trends approaching significance were observed in cataplexy prevalence (p=0.084), indicating a need for further multivariate investigation (Table 2).

Although the association between cataplexy and diagnostic delay did not reach statistical significance, a notable trend was observed (p=0.084) (Table 2). The prevalence of cataplexy was higher in the non-delay group (54.7%) than in the delay group (40.1%). This trend aligns with the distribution of narcolepsy subtypes shown in Table 1, where the proportion of patients with NT1—characterized by the presence of cataplexy—was higher in the non-delay group (71.7%) compared to the delay group (59.9%). These findings suggest that overt clinical markers like cataplexy may facilitate earlier diagnosis. Other REM-related symptoms, such as sleep paralysis and hypnagogic hallucinations, did not demonstrate an association with diagnostic latency (Table 2).

To identify independent predictors of diagnostic delay while controlling for confounding variables, we conducted a multivariate logistic regression analysis. The model included onset age, EDS, family history, and sex as independent variables (Table 3).

The analysis revealed that onset age is a significant independent predictor of diagnostic delay (OR=0.91, 95% CI [0.85- 0.97], p=0.003). Specifically, patients who developed symptoms at a younger age were increasingly likely to experience diagnostic delays; for each year earlier the symptoms began, the likelihood of a delayed diagnosis increased by nearly 10%. Additionally, EDS remained a significant predictor (OR=9.21, 95% CI [1.68-50.45], p=0.011). This finding indicates that diagnoses often occur only after excessive sleepiness becomes severe enough to disrupt daily functioning, suggesting that milder or less obvious symptoms may not prompt timely clinical evaluation.

In a subgroup analysis of 31 patients, we compared objective MSLT findings between the non-delay (n=11) and delay (n=20) groups. Although the subgroup size was insufficient to detect statistically significant differences, this analysis aimed to objectively assess whether diagnostic delay was associated with variations in disease severity. The non-delay group had a higher proportion of NT1 (72.7%) compared to the delay group (45.0%), which supports the notion that overt symptoms like cataplexy may facilitate earlier diagnosis. Regarding objective sleep parameters, the non-delay group exhibited a numerically shorter mean sleep latency (3.01±2.61 minutes) and a slightly higher mean SOREM count (3.09±1.30) compared to the delay group (3.82±3.26 minutes and 2.85±1.57 minutes, respectively). Although these differences were not statistically significant, the results indicate that patients in the delay group demonstrated objective sleep pathology of severity comparable to that of patients diagnosed promptly (Table 4).

An earlier age at symptom onset was the strongest predictor of delayed diagnosis in this cohort. Patients who developed narcolepsy symptoms at a younger age were significantly more likely to experience prolonged diagnostic latency, with the risk of delay increasing progressively as the age of onset decreased. This finding aligns with previous reports indicating that narcolepsy starting in childhood or adolescence is especially prone to delayed recognition [19]. Delayed recognition may occur because the normal sleep phase delay in adolescents, combined with early school schedules, often leads to chronic sleep deprivation. This can result in pathological sleepiness being misattributed to insufficient sleep [23].

From a psychiatric perspective, this vulnerability may reflect the nonspecific and heterogeneous clinical presentation of narcolepsy during developmental periods. In children and adolescents, EDS may manifest as irritability, emotional lability, inattention, behavioral issues, or academic decline, rather than as overt sleep attacks [24].

Moreover, symptoms that emerge early in life can become integrated into an individual’s perceived baseline functioning. When sleepiness and fatigue gradually develop during childhood, they are often interpreted as personality traits or temperament rather than as signs of an underlying disorder [25]. This tendency is reinforced when similar patterns are observed among family members, contributing to the normalization of symptoms and deferral of clinical attention until functional impairment becomes pronounced in adolescence or adulthood. This contributes to a prolonged diagnostic delay [26].

Cataplexy is a key characteristic of narcolepsy, yet this study found no significant difference between the delayed and nondelayed diagnosis groups regarding its presence. However, a descriptive trend emerged, indicating a higher proportion of NT1 in the non-delay group. This suggests that cataplexy might lower the threshold for clinical evaluation in some patients.

The lack of statistical significance implies that cataplexy alone cannot guarantee a timely diagnosis. One reason for this could be the limited awareness of cataplexy among patients, caregivers, and non-specialist clinicians. Episodes of sudden muscle weakness triggered by emotions may be misinterpreted as anxiety-related reactions, behavioral issues, or psychological phenomena, especially in pediatric and adolescent populations. Consistent with this observation, the delayed diagnosis group exhibited a higher proportion of NT2, while NT1 was more common in the non-delay group. This further suggests that the absence of overt cataplexy may contribute to delayed recognition. As a result, even the characteristic features of narcolepsy may fail to prompt appropriate diagnostic consideration, diminishing the potential diagnostic advantage of cataplexy.

Contrary to expectations, a family history of sleep disorders did not serve as a protective factor against diagnostic delays. This indicates that familial aggregation does not automatically lead to greater symptom awareness or prompt help-seeking behavior.

One possible explanation is that symptoms may be normalized within families. When excessive sleepiness or similar complaints are common among relatives, they may be viewed as familial traits instead of signs of a medical issue [27,28]. Such normalization may reduce the perceived severity of illness and delay psychiatric or medical consultation. In this context, a family history may paradoxically obscure the recognition of symptoms rather than facilitate diagnosis [29].

This finding is clinically important for clinics, as family history is often viewed as an indicator for increased vigilance. These findings suggest that relying solely on family-reported symptom history is insufficient; clinicians should systematically assess functional impairment and symptom burden independent of familial context [30]. Educational initiatives should be redesigned to specifically address families with a history of sleep issues, helping them differentiate between “familial traits” and clinical pathology. Ultimately, the early application of objective diagnostic tools, such as polysomnography and the Multiple Sleep Latency Test, in suspected juvenile cases remains the most reliable means of overcoming symptom normalization and ensuring timely intervention [31,32].

The burden of narcolepsy is twofold: it includes direct clinical disabilities, such as cataplexy-related injuries and severe daytime sleepiness, as well as long-term socioeconomic liabilities, which encompass impaired education, reduced employment opportunities, and decreased work productivity [33]. When diagnosis is delayed by over a decade, as observed in our juvenile-onset group, patients miss critical developmental windows. Early recognition is not just a clinical goal but a socioeconomic necessity; effective treatment during adolescence can preserve a patient’s academic trajectory and social integration, fundamentally improving their long-term quality of life [27,34].

This study examines adolescent-onset narcolepsy, a clinically vulnerable and often overlooked population that experiences significant diagnostic delays. By analyzing symptom profiles and family history within a real-world clinical cohort, the research offers valuable insights into the factors that facilitate timely recognition of the condition. The use of a stratified design and a relatively large sample size enhances the clinical relevance of the findings. Additionally, subgroup analyses that incorporate objective sleep study data provide biological validation for the clinical observations. Notably, this research is the first in South Korea to investigate whether family history and clinical features can serve as predictive factors for delayed versus timely diagnosis of narcolepsy.

Several limitations must be acknowledged. The timing of diagnosis and symptom onset relied on self-reported data, which may introduce recall bias; specifically, patients may systematically underestimate or overestimate the interval between symptom onset and formal diagnosis, potentially attenuating or inflating the magnitude of observed group differences. The crosssectional design limits causal inferences, and family history was assessed through binary self-report rather than through objective or genetic measures.

In conclusion, earlier age at symptom onset was the strongest independent predictor of diagnostic delay in narcolepsy within this cohort. The presence of cataplexy or a family history of sleep disorders did not consistently result in an earlier diagnosis. This indicates that even well-known symptoms and family background often do not prompt timely clinical recognition. These findings underscore the need for proactive consideration of narcolepsy in adolescents who present with EDS in psychiatric settings, as early diagnosis is crucial for minimizing delays and reducing long-term functional and psychosocial challenges.