Cough variant asthma (CVA) has been recognized as a precursor of asthma or a pre-asthmatic state because of the mildly heightened bronchial responsiveness and efficacy of bronchodilator therapy. Nevertheless, the accumulating evidence indicates that the pathophysiology is different between CVA and bronchial asthma. The most fundamental physiologic feature is a heightened cough response to methacholine-induced bronchoconstriction in CVA, while this response is rather reduced in bronchial asthma. The sensitivity of cough receptors located in the superficial layer of the airway wall is normal in CVA as well as bronchial asthma, but heightened in atopic cough. The pathologic feature of CVA is eosinophilic inflammation of the central to peripheral airway, reflected by eosinophilia in induced sputum, biopsied bronchial mucosa, and bronchoalveolar lavage fluid. The diagnosis of CVA has been commonly made based on therapeutic diagnostic procedures, while pathophysiologic diagnosis is ideal. The reason is that measurements of the sensitivity of cough receptors to inhaled capsaicin and cough response to induced bronchoconstriction are not possible at most chest clinics in the world. The efficacy of a beta2-agonist for a patient's coughing is evaluated to make a diagnosis of CVA. When the bronchodilator therapy is judged as efficacious, a tentative diagnosis of CVA is made. Then, induction therapy is initiated for resolution of the cough. The induction therapy consists of beta2-agonists, leukotriene receptor antagonists, and inhaled corticosteroids. In some patients whose cough does not subside with the therapy, short-burst oral corticosteroids (1 to 3 weeks) may be added. If the cough still does not subside with the therapy, the patient should be referred to cough specialists. When the cough subsides with the induction therapy, long-term management is recommended using inhaled corticosteroids, because 30% of patients develop typical bronchial asthma within several years. Problems with the therapeutic diagnosis are as follows: spontaneous relief of cough leading to a false positive result, and resistance to the therapy, leading to a false-negative result. Thus, a pathophysiologic diagnostic procedure should be established in the future.