Objective: To elucidate the predictive role of age and other pre-treatment, putative confounding factors on compliance with nasal continuous positive airway pressure (nCPAP) therapy.Patients and methods: This study was designed as a prospective cohort study in the setting of a sleep laboratory in a teaching hospital at Saint Antoine, Paris. One hundred and sixty-three patients referred to the sleep laboratory with complaints of snoring and excessive daytime sleepiness for whom nCPAP had been prescribed for obstructive sleep apnea syndrome (OSAS; defined as an apnea-hypopnea index (AHI) of >15/h of sleep during a polysomnographic recording) were followed for a median period of 887 days. The main outcome measure was the risk ratio for elderly patients associated with nCPAP compliance.Results: Four patients, who remained under treatment, died before the end of the study, and 50 patients stopped their nCPAP therapy for reasons other than death (insomnia, equipment too noisy, etc.). When compliance curves were compared by univariate analysis (log-rank test), the oldest group (57/163 patients, >60 years old) was significantly less compliant with nCPAP than the youngest (P=0.01). However, in the Cox's proportional hazards model, age did not exert any independent effect on compliance with nCPAP after controlling for confounding factors (adjusted relative risk, 1.09, 0.5-2; P=0.70). On the other hand, female sex (adjusted relative risk, 2.8, 1.4-5.4; P=0.002), a body mass index (BMI) of </=30 kg/m(2) (adjusted relative risk, 2.2, 1.2-4; P=0.006), an Epworth sleepiness scale (ESS) score of </=15 (adjusted relative risk, 3.2, 1.1-8.9; P=0.025), an AHI of </=30/h (adjusted relative risk, 2.2, 1.2-4; P=0.01) and a nCPAP of >/=12 cmH(2)O (adjusted relative risk, 2.3, 1.2-4.4; P=0.011) were predictive factors for non-compliance.Conclusion: This study suggests that there is no independent effect of age on compliance with nCPAP therapy.