To test the relative merits of administering questionnaires with previous responses available (the informed condition) or unavailable (the blind condition), we administered blind and informed versions of a quality of life questionnaire to two groups of patients. One, a group of 43 stable subjects with chronic cardiorespiratory disease, were seen three times at fortnightly intervals; a separate series of 13 patients with chronic lung disease were evaluated before and after optimization of therapy. In the stable patients the informed strategy resulted in substantial decrease in the variance in the measurement of dyspnea, fatigue, and of emotional function. Large improvements in dyspnea, fatigue, and emotional function seen in patients undergoing treatment optimization were comparable using blind and informed methods. These results suggest that by letting patients see their previous responses we can decrease the sample size needed to detect changes in quality of life in clinical trials.