The measurement of health-related quality of life (HRQOL) plays a very important role in the study of health outcomes, particularly in effectiveness studies. Nevertheless, the nature of these studies implies a series of challenges in the use of HRQOL instruments, due to the type of comparisons made, the heterogeneity of subjects included in terms of both the presence of comorbid conditions and the ability to understand and complete questionnaires, and because there is a need to collect the data in conditions of usual clinical practice. The aim of this paper is to discuss some of these issues, as well as other topics of current interest in the field, such as the use of individualized measures, applications of Item Response Theory, and response shift phenomenon research. With regard to the use of HRQOL measures in effectiveness studies, the main conclusions are that: a) instrument selection probably needs to be even more careful than in controlled clinical trials, and needs to take into account time pressures and variability in patients' ability to respond; b) that instrument selection should centre on the use of disease-specific measures; c) that generic measures may play an important role in capturing the broader effects of co-morbidity, and may be useful in controlling certain types of bias, and d) that more comparisons are required between disease-specific instruments for the same condition, particularly with respect to their discriminative power and their sensitivity to change.