To test whether, in patients with chronic dyspnea, a diagnostic approach based on objective confirmation of suspected diagnoses would be superior to one based on clinical impression alone, we prospectively studied 85 patients with a primary complaint of dyspnea seen in a pulmonary subspecialty clinic. We achieved 100% success in determining the causes of dyspnea compared with only 66% accuracy based on clinical impression alone. Four groups of disorders, asthma, chronic obstructive pulmonary disease, interstitial lung diseases, and cardiomyopathy accounted for two thirds of the cases. Findings on the history and physical examination were too nonspecific to determine the specific diagnosis. Pulmonary function testing, including a methacholine bronchoprovocation challenge, were the most useful diagnostic tests, particularly for chronic obstructive pulmonary disease and asthma. Chest roentgenogram was most useful for interstitial lung disease, and comprehensive exercise testing for dyspnea due to psychogenic factors or deconditioning. Specific therapy was effective in reducing or eliminating dyspnea in the majority of cases. We conclude that a diagnostic approach to chronic dyspnea based on objective findings and verification, rather than clinical impression alone, will consistently lead to an accurate diagnosis and an improved therapeutic outcome.