ERJ Open Research (May 2023)

Prevalence, treatment and determinants of obstructive sleep apnoea and its symptoms in a population-based French cohort

  • Pauline Balagny,
  • Emmanuelle Vidal-Petiot,
  • Adeline Renuy,
  • Joane Matta,
  • Justine Frija-Masson,
  • Philippe Gabriel Steg,
  • Marcel Goldberg,
  • Marie Zins,
  • Marie-Pia d'Ortho,
  • Emmanuel Wiernik

DOI
https://doi.org/10.1183/23120541.00053-2023
Journal volume & issue
Vol. 9, no. 3

Abstract

Read online

Background Obstructive sleep apnoea (OSA) is associated with increased morbidity and mortality. Although the disorder has been well studied in selected high-risk populations, few data exist on its prevalence in the general population. We aimed to assess the prevalence and determinants of OSA in France. Methods Data from participants of the French population-based CONSTANCES cohort aged 18–69 years at inclusion and being treated for sleep apnoea or screened for OSA in 2017 using the Berlin Questionnaire were analysed. Weighted analyses were performed to provide recent and representative results in the general population. Results Among 20 151 participants, the prevalence of treated sleep apnoea was 3.5% (95% CI 3.0–3.9%). The prevalence of untreated subjects with a positive Berlin Questionnaire was 18.1% (95% CI 17.3–19.2%) for a total weighted prevalence of treated sleep apnoea or high risk of OSA of 20.9% (95% CI 20.0–21.9%). Regarding prevalence of OSA symptoms, it was 37.2% (95% CI 36.1–38.3%) for severe snoring and 14.6% (95% CI 13.8–15.5%) for hypersomnolence. In multivariable logistic regression analysis, male sex, age, previous cardiovascular events, smoking, low educational level, low physical activity and depressive symptoms were associated with having either treated sleep apnoea or a positive Berlin Questionnaire. Conclusion In this large French population-based cohort, one in five participants had a high likelihood of OSA, whereas only 3.5% were treated for the disorder, suggesting major underdiagnosis in the general population. OSA diagnosis should be considered more often in people with risk factors such as depressive symptoms as well as unhealthy behaviours and socioeconomic conditions.