Objectives: (1) To compare the continuous positive airway pressure (CPAP) requirement at the time of diagnosis (T-0), after 2 weeks (T-2), and after 4 weeks (T-4) of CPAP treatment, in patients with severe obstructive sleep apnea (OSA); and (2) to assess whether any alteration in CPAP requirement over the first 4 weeks of CPAP treatment would influence daytime alertness, subjective sleepiness, or mood. Design: A prospective, controlled, single-blind crossover study. Setting: University teaching hospital. Patients: Ten patients with newly diagnosed and previously untreated severe OSA (aged 52+/-9 years, apnea hypopnea index [AHI] of 99+/-31) and subsequently 10 control patients (aged 52+/-11 years, AHI 85+/-17). Measurements: Overnight polysomnography with CPAP titration to determine the CPAP requirement, which was standardized for body position and sleep stage, on all three occasions (T-0, T-2, T-4). Objective sleep quality, daytime alertness, subjective sleepiness (Epworth Sleepiness Scale), and mood (Hospital Anxiety and Depression Scale). Results: CPAP requirement decreased from T-0 to T-2 (median difference, 1.5 cm H2O, 95% confidence interval [CI], 1.1 to 2.7 cm H2O, p=0.0004) and did not differ between T-2 and T-4. Use of the lower CPAP pressure during T-0 to T-2 was associated with a decrease in Epworth scale (mean difference, 2.6, 95% CI, 1.2 to 4; p=0.01) and anxiety (median change, 2; 95% Ct, 0.5 to 2.9, p=0.03) scores, as compared with the first 2 weeks. Daytime alertness did not differ between T-0 to T-2 and T-2 to T-4. Conclusion: CPAP requirement falls within 2 weeks of starting CPAP treatment. A change to the lower required CPAP was not associated with any deterioration in daytime alertness but was associated with small subjective improvements in sleepiness and mood.