Purpose: The objectives are to assess different approaches to measure the effect of nCPAP on quality of life and evaluate the responsiveness of generic and disease specific instruments in sleep apnea patients. Methods: We have used three quality of life questionnaires: SF-36, EuroQol 5D (EQ-5D) and the specific sleep diseases FOSQ. OSAS patients were interviewed at baseline prior to the start of the treatment and at three months follow-up. A first approach was based on distribution-based measures such as t-test comparisons, the effect size and the standardised response mean (SRM). Responsiveness was assessed using anchor-based measures such as the ROC curves and the minimally clinically important difference (MCID). Results: Complete data were available for 124 patients. Mean differences (95% parametric CI) at three months follow-up were 11.34 (8.51 to 14.16) for FOSQ 0.03 (0.02 to 0.04) for SF-36 and 0.04 (0.02 to 0.07) for EQ-5D. Effect size varies from small (0.21) with EuroQol 5D to moderate (0.51) with FOSQ. SF-36 obtained intermediate scores (0.35). The figures for SRM were respectively: EQ-5D 0.29, FOSQ 0.71 and SF-36 0.50. The calculation of the MCID produced different values: EQ-5D 0.09, FOSQ 13.64 and SF-36 0.05. One question from the SF-36 was used as a gold standard anchor variable. EQ-5D was the questionnaire that obtained the biggest area under the ROC curve followed by the SF-36. Nevertheless, the area size was small in all cases. Conclusions: nCPAP improves significantly the quality of life of OSAS patients. However, none of the instruments seemed to discriminate better than the others, therefore our assessment with the available instruments is only partially satisfactory. We were not able to find changes in score that correspond to moderate or large in terms of patient benefit. It is not currently clear the value thresholds to assess individual changes in order to maintain or stop the treatment. Current quality of life assessment should be enhanced with consensus methodology that may help to understand and interpret the results of future research.