BACKGROUND: New York Heart Association (NYHA) classification correlates with quality of life and is useful in tracking changes in status in clinical trials. We explored methods to determine NYHA class in multicenter trials where double-blind conditions could not be maintained. METHODS AND RESULTS: A questionnaire was developed containing 7 major questions based on the standard definitions of NYHA classes. The questionnaire was administered to 116 patients with varying degrees of heart failure. When comparing NYHA determined by physicians at the site to NYHA assigned by 3 independent physician graders, there was an approximate 60% concordance. Concordance between independent reviewers was approximately 75%. Results of repeat grading of 30 randomly selected questionnaires indicated that graders provided the same score 90% of the time. Thus, although there were some differences from the site determination of NYHA class, the questionnaire had good inter- and intragrader reproducibility. In a second group of 103 patients enrolled in an ongoing device-intervention trial, we demonstrated that it is feasible to employ the questionnaire in a multicenter trial. Finally, NYHA class was correlated with quality of life and peak exercise oxygen consumption. CONCLUSIONS: A standardized questionnaire provided an approximate 60% concordance in assigning NYHA classification compared to the site assessment with approximately 90% reproducibility. This approach may be useful to determine NYHA classification within the context of clinical trials where blinded conditions are not possible.