Need assistance?
By Sree Roy
Narcolepsy type 1 (NT1) is defined by clear biomarker criteria—loss of orexin-producing neurons and the hallmark symptom of cataplexy—while narcolepsy type 2 (NT2) and idiopathic hypersomnia (IH) have often been treated as blurrier hypersomnias, potentially evolving into different diagnoses with time.
Yet an investigation using Danish nationwide administrative registries linked to expert-verified clinical cohorts is upending clinical assumptions about which patients experience the most severe downstream consequences. Across comprehensive measures of healthcare costs, labor force dropouts, and psychiatric comorbidity, patients diagnosed with NT2 and IH experienced greater disruption to their socioeconomic lives than those with NT1—often starting years before an official diagnosis was rendered.
“That surprised me somewhat…maybe the underlying pathophysiology of narcolepsy type 2 is more complex than we deem it to be,” says study co-author Niels Christian Haubjerg Østerby, MD, a research associate and PhD student in the department of neurology at Bispebjerg Hospital and the Danish Center for Sleep Medicine at Rigshospitalet in Denmark. “When we started out with the study, I was quite confident that we would see a stronger signal in the type 1 group.”
The surprising findings, published in Sleep Health and building on earlier work by Østerby and co-investigators published in SLEEP Advances, underscore the necessity of recognizing the vulnerability of NT2 and IH patients and looking beyond basic sleep metrics when evaluating therapeutic success.
Dissecting the Subtype Differences
The study examined a well-characterized cohort of 433 patients diagnosed with central disorders of hypersomnolence at the Danish Center for Sleep Medicine between 2004 and 2022, matched in a 1-to-4 ratio with population controls from Danish national registries based on age, sex, marital status, and municipality. Unlike purely administrative claims analyses, every patient had undergone comprehensive in-person expert clinical assessment, overnight polysomnography (PSG), daytime Multiple Sleep Latency Tests (MSLT), and cerebrospinal fluid (CSF) orexin analysis.
The researchers analyzed outcomes across a five-year observation window spanning two years before to two years after formal diagnosis. Across nearly every parameter, the financial and social burden was heavily weighted toward NT2.
Total annual healthcare expenditures peaked in the year leading up to diagnosis across all three conditions, but the magnitude varied sharply:
- NT2: Healthcare costs reached €9,846 compared with €1,568 in matched controls.
- IH: Healthcare costs peaked at €8,713 compared with €2,357 in controls.
- NT1: Expenditures peaked lower, at €6,011 compared with €2,143 in controls.
Moreover, NT2 was the only cohort with significantly elevated healthcare costs two years prior to formal diagnosis (€4,647 vs. €1,198 for controls), demonstrating a protracted, resource-intensive trajectory toward an accurate evaluation.
Personal earned income demonstrated a similar gradient. Two years post-diagnosis, patients with NT2 earned an average annual income of €19,502 compared with €30,834 among controls. While patients with NT1 and IH also earned less than their respective control groups, the differences did not reach statistical significance. Public social safety-net transfers, including disability allowances and sick leave, told an identical story: NT2 transfers peaked two years after diagnosis at €12,463 versus €6,430 for controls.
Psychiatric Signal Before Diagnosis
One of the most notable findings in the registry linkage was the role of psychiatric utilization before a sleep diagnosis was ever established. Two years before diagnosis, NT2 patients generated an average of €1,224 in inpatient and outpatient psychiatric care costs, compared with just €61 among matched population controls. In NT1, psychiatric care costs never significantly differed from controls at any point in the five-year window.
These findings mirror the group’s 2024 analysis published in SLEEP Advances, which tracked psychiatric diagnoses in the 10 years preceding sleep diagnosis. In that study, NT2 and IH showed high odds of total psychiatric comorbidity relative to controls, whereas NT1 was not significantly associated with elevated total psychiatric comorbidity. Patients with NT2 displayed elevated associations across schizophrenia spectrum disorders, mood disorders, neurotic disorders, and personality disorders.
This raises a central clinical conundrum: Does this pattern indicate true psychiatric comorbidity, or are patients who have central hypersomnolence misdiagnosed by mental health professionals unaware of underlying sleep pathology?
“My personal opinion is that it’s both,” Østerby says.
Østerby emphasizes that this overlap demands clinical vigilance rather than mutual exclusivity. “It makes it so important that every time you have a psychiatric patient where the extent of the hypersomnolence is beyond explainable by the psychiatric disorder, or perhaps you have already treated it quite effectively and you are left with a large residual hypersomnolence, it’s important to suspect that it could be both at play at the same time,” he explains. “Of course, it’s always possible that it is mainly one or the other, or that perhaps some of the patients are misclassified or misdiagnosed with psychiatric disorders. But personally, I feel quite confident that they coexist.”
Persistent Unemployment Post-Diagnosis
Clinicians often anticipate that establishing a diagnosis and initiating wake-promoting pharmacotherapy will stabilize patients’ vocational trajectories. However, the Danish registry data reveal a sobering reality: receiving a diagnosis did not reverse employment loss.
Instead, labor market disengagement climbed across all three groups:
- NT2: Unemployment rose from 16.9% two years pre-diagnosis to 28.6% two years post-diagnosis, carrying an adjusted odds ratio of unemployment of 4.6.
- IH: Unemployment escalated from 9.6% to 25.5%, with an adjusted odds ratio of 2.2 at two years post-diagnosis.
- NT1: Unemployment rose from 9.5% to 12.4%, with an adjusted odds ratio of 2.8.
Why are current clinical interventions falling short of keeping patients in the workforce?
“The major player is probably insufficient treatment,” Østerby says. “It’s heterogenic. Some patients respond very well to some of the central stimulants, and some of the patients don’t. And I think a large part lack a good response to the medication.”
Pharmacological limitations are compounded by workplace expectations and cultural misconceptions about sleep disorders. “Besides just the medication, I think it can be very difficult for employers and workspaces to accept patients with these disorders because they look perhaps on the surface to be lazy; there’s a lot of social stigma around this disorder,” Østerby adds. “I feel that’s a part of it—that the patients have difficulty fitting into the workforce, even though they want to.”
However, Østerby points to upcoming pharmacological classes as a source of cautious optimism. “I’m very hopeful for the orexin agonists,” he says. “I think that curve will look very different in the future. I at least hope so.”
Navigating Diagnostic Ambiguity
In evaluating idiopathic hypersomnia, the researchers noted another anomaly: while inpatient and outpatient costs tended to subside after diagnosis in NT1 and NT2, somatic outpatient expenditures among IH patients remained persistently high and even increased slightly two years post-diagnosis (€1,967 versus €758 for controls).
While this pattern could point to refractory sleepiness, Østerby suggests broader somatic complications may be responsible. “In our comorbidity analysis, we showed that they have a very high odds of having abnormal clinical findings on routine laboratory testing,” he notes, citing the elevated odds ratio of 4.8 for unspecific laboratory and clinical findings. “Since they have this kind of unspecific finding…it could be that they were sent for more workup in different disciplines of medicine.”
This ongoing search highlights the longstanding diagnostic instability between NT2 and IH. Both lack definitive biological markers, and repeat testing on the MSLT frequently shows diagnostic drift. When asked whether the heavy burden observed in NT2 might partly stem from diagnostic classification instability rather than a distinct biological subtype, Østerby acknowledges the challenge while defending the clinical validity of the cohort.
“It’s very good to be critical of the classification because the classification is mainly clinical to some extent,” Østerby says. “When we get a patient at our center, usually what we would expect is that they have been through evaluations and optimizations in psychiatric care, other internal medicine ambulatories, or what have you, and no one has resolved the sleepiness.”
To assemble the study cohort, Østerby personally reviewed the electronic patient charts of roughly 1,100 evaluated individuals, all of whom underwent lumbar puncture, PSG, and MSLT. A substantial portion was eliminated from the final central disorders of hypersomnolence groups.
“There are not really any places to send them,” Østerby explains regarding patients who fail to meet strict criteria. “We have a diagnostic code called secondary hypersomnia…usually that’s because we say it is overwhelmingly certain that it’s a different cause…and without clear paraclinical signs that they have a true hypersomnia. A lot of patients will also present with a subjective hypersomnia, which we cannot objectify.”
Even amid the phenotypic ambiguity between NT2 and IH, Østerby argues that clinical diagnostic efforts remain essential: “You can be cynical and say ‘you might just be overdiagnosing and attributing diagnostics to unspecific clinical findings.‘ That’s a very true possibility. But I think that most of the patients will have an improvement to some extent, at least by getting expert help and by trying the medications. In my opinion, that justifies the uncertainty.”
Expanding Success Metrics Beyond Sleep Latency
In modern sleep clinics, therapeutic success is overwhelmingly gauged through standardized instruments like the Epworth Sleepiness Scale or physiological metrics from the MSLT and Maintenance of Wakefulness Test. Yet the Danish study’s findings suggest these tools may fail to capture the real-world operational impairments patients experience every day.
When asked whether vocational and healthcare utilization outcomes should be systematically integrated into clinical follow-up, Østerby drew parallels to other neurological subspecialties.
“Definitely,” he says. “We do this often in other disciplines of neurology—headache patients, migraine patients,” Østerby continues. “It’s such an important question to ask the patients: Does it affect your work ability? Does it affect your social life? And these are markers that should definitely be incorporated in the outpatient clinics and the follow-up for the patients.”
For team members—including sleep techs administering daytime testing and dental sleep medicine practitioners screening for excessive daytime sleepiness that persists despite oral appliance therapy—recognizing when a patient’s sleepiness is disproportionate to primary findings is vital.
NT2, Østerby says, “is very underestimated.” He adds, “To an extent, I feel sometimes that narcolepsy type 1 can be underestimated as well. It’s so rare that we tend to not really give it so much thought as a differential diagnosis. If you work in a busy clinic, you have to dig pretty far for many clinicians before you get to the possibility of the patient having a central hypersomnolence disorder.”
He advises clinicians against dismissing symptoms when routine explanations fall short: “The main thing in a clinical context would be to refer the patient for a sleep study if you think the proportion of complaint is out of proportion with what you suspect it to be. People always have a go-to explanation of, ‘Oh, it’s just your sleeping pattern,’ or perhaps poorly treated hypothyroidism...It would always be good to refer the patient for a sleep study.”
More in Understanding the Hypersomnia Burden:
ID 80513592© Paulus Rusyanto| Dreamstime.com
Advertisement
