Wu · Sleep medicine 2024 · Cross-sectional study · n=2176

Association between age and comorbid insomnia and sleep apnea.

Cited 15 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional observational study

PubMed 39531785 · doi:10.1016/j.sleep.2024.11.011 · record verified 2026-08-29

What was done

A single-night polysomnography with video monitoring was conducted in 2,176 Chinese adults (ages 18–102) undergoing screening for obstructive sleep apnea (OSA). Insomnia was defined as difficulty initiating or maintaining sleep, or early morning awakening at least three times per week for more than three months with daytime symptoms. Comorbid insomnia and sleep apnea (COMISA) was defined as an apnea-hypopnea index (AHI) ≥5 with insomnia. Characteristics were compared between COMISA and OSA-only groups, and sex-adjusted logistic regression evaluated the association between age and COMISA.

What was found

Of the participants, 981 had COMISA and 433 had OSA only. Compared to OSA-only patients, COMISA patients were more often female (48% vs. 24%), had longer sleep onset latency (+5.8 min), shorter total sleep time (-14.0 min), lower AHI (-14.4 points), and higher mean peripheral oxygen saturation (+0.6%; all p < 0.05). The COMISA vs. OSA-only proportion was 51.9% vs. 48.1% in participants under 50 years, compared to 78.1% vs. 21.9% in those aged 70 or older. Logistic regression adjusted for sex showed a 3% increase in the odds of COMISA per additional year of age (OR 1.03, 95% CI: 1.02–1.04).

Why it matters

The study shows that insomnia frequently co-occurs with OSA, particularly in older adults where it comprises the vast majority of sleep apnea presentations, highlighting the need for dual screening and tailored management.

Limits

The study is cross-sectional, precluding causal or temporal inferences. Data relied on a single-night polysomnography subject to first-night effect. Statistical adjustment was limited to sex, leaving potential confounding by body mass index, medical and psychiatric comorbidities, and medication use unmeasured.

Cited by