Fatigue and daytime sleepiness in patients with severe mental illness (SMI) should prompt screening for obstructive sleep apnea (OSA), a very common condition in this patient population.

Results of a large systematic review and meta-analysis showed that OSA affects approximately 40% of patients with SMI and more than one third of those with schizophrenia, major depressive disorder (MDD), or bipolar disorder.

“These findings underscore the importance of improved recognition and systematic assessment of OSA in people with severe mental illness,” the investigators, led by Constanza Sommerhoff, with the Bipolar and Depressive Disorders Unit at Hospital Clínic of Barcelona in Barcelona, Spain, wrote.

The study was published online in the November issue of Acta Psychiatrica Scandinavica.

Measuring Prevalence and Risk

OSA symptoms — including fatigue, daytime sleepiness, cognitive impairment, and poor sleep — may be mistaken for psychiatric symptoms or medication effects, delaying diagnosis and treatment.

Previous research suggests that OSA is common in those with SMI, but prevalence estimates varied widely across settings and diagnostic methods. Earlier meta-analyses also lacked pooled comparisons with healthy control participants, leaving the extent of excess risk unclear.

For the study, the researchers combed PubMed, Web of Science, Scopus, and PsycINFO for studies reporting either diagnosed OSA or high risk for OSA in patients with schizophrenia, bipolar disorder, or MDD.

A total of 55 studies were included in the systematic review and 47 in the meta-analysis. Together, the studies included 10,494 patients with SMI — 5016 with schizophrenia, 3968 with MDD, 905 with bipolar disorder, and 605 with mixed SMI diagnoses.

Across 26 studies involving 3858 patients, the pooled prevalence of objectively measured OSA was 41.2%. Prevalence was higher in sleep-referral settings and lower in clinical settings. OSA prevalence was similar across psychiatric diagnoses: 38.3% in schizophrenia, 38.2% in MDD, and 35.6% in bipolar disorder.

Overall, 46.7% of patients with SMI were classified as being at high risk for OSA based on screening questionnaires, including 51.8% of those with schizophrenia, 42.9% with MDD, and 34.2% with bipolar disorder.

Excessive daytime sleepiness was present in 39.2% of patients with SMI and 44.8% of those with MDD. Patients with bipolar disorder had significantly greater daytime sleepiness than healthy control participants, an association that remained significant after adjustment for multiple comparisons.

In the two studies that directly compared patients with SMI with healthy control participants, SMI was associated with almost seven times the odds of OSA (odds ratio, 6.96; adjusted P = .003). The authors cautioned, however, that the comparison was based on limited data.

Older age and recruitment from a sleep-referral setting were associated with higher OSA prevalence. In contrast, BMI, sleep-study method, and use of at-home testing were not significant moderators.

Potential Risk Factors

The investigators noted that several factors may contribute to the association between OSA and psychiatric illness, including sedentary behavior, smoking and alcohol use, unhealthy diet, and psychotropic medications that can cause sedation, respiratory depression, weight gain, and metabolic abnormalities.

At the same time, recognizing OSA can be challenging, they acknowledged, because fatigue, poor sleep, irritability, mood changes, and cognitive difficulties overlap with symptoms of psychiatric disease.

Daytime sleepiness may be attributed to psychotropic medications, while snoring and nighttime apneas may go unnoticed in patients without a bed partner.

The researchers encouraged clinicians not to automatically attribute daytime somnolence, fatigue, or disrupted sleep to psychiatric illness or medications without considering OSA.

Untreated OSA carries increased risk for hypertension, cardiovascular and cerebrovascular disease, diabetes, cognitive decline, and accidents, while existing evidence suggests that OSA treatment can improve sleep quality, daytime sleepiness, cognition, and quality of life.

The authors noted substantial heterogeneity across studies, including differences in diagnostic methods and OSA thresholds. Sleep-referral samples may also have inflated prevalence estimates.

Medication effects could not be adequately separated from those of comorbidities. About one third of studies were rated as poor quality, most were cross-sectional, and several comparative analyses relied on small samples.

Looking Beyond Insomnia

In an accompanying editorial, Seiji Nishino, MD, PhD, of Stanford University School of Medicine in Stanford, California, emphasized that the findings do not establish a causal relationship between OSA and psychiatric illness.

Much of the evidence is cross-sectional, and the association is likely influenced by shared factors including obesity, age, smoking, alcohol use, cardiometabolic disease, lifestyle, and psychotropic medications. Still, “uncertainty regarding causality does not diminish the importance of detecting and treating OSA,” he wrote.

A major challenge is symptom overlap, Nishino noted. Fatigue, low energy, poor concentration, irritability, reduced motivation, memory difficulties, and sleep disturbance may readily be attributed to psychiatric illness or medication effects. But fragmented or nonrestorative sleep and daytime sleepiness “should not automatically be regarded as psychiatric symptoms alone. They may also signal the presence of OSA,” he wrote.

He cautioned that although obesity is an important risk factor, OSA can also affect younger adults and patients without marked obesity. Clinicians should therefore not rule it out based on BMI alone.

Polysomnography for every patient is neither feasible nor necessary Nishino noted. Rather, he recommended identifying those whose symptoms and risk factors warrant further evaluation.

Screening tools such as STOP-Bang and the Epworth Sleepiness Scale can help identify patients who need further evaluation, but neither replaces diagnostic testing. A lack of reported sleepiness should not rule out OSA, Nishino noted.

Clinicians should not assume that patients with SMI cannot successfully use continuous positive airway pressure, he added. Education, careful mask fitting, and behavioral support can help patients with treatment initiation and adherence.

Ultimately, Nishino noted, this review should encourage psychiatrists to look beyond insomnia and sleep duration and ask a more basic question: “Is this patient breathing normally during sleep?”

The study had no commercial funding. Sommerhoff and Nishino declared having no conflicts of interest.