Background The current diagnostic algorithms for chronic cough require the establishment of the primary presumptive causes followed by the confirmation of diagnosis with the specific therapies.The aim of the study was to investigate the discrepancy between presumptive and definite causes and its clinical implication.Methods A total of 109 patients with chronic cough underwent laboratory investigations to identify the cause of cough; including sinus computerized tomography (if needed),histamine bronchial provocation,induced sputum cytology and 24-hour esophageal pH or multi-channel intraluminal impedance combined with pH monitoring.The presumptive causes were confirmed by treating them sequentially.The difference between presumptive and definite causes of chronic cough was compared.Results Single cause was more frequent in the definite diagnosis than in the presumptive diagnosis (78.9％ vs.54.1％,x2=15.01,P=0.0001).In contrast,multiple causes were significantly fewer in definite diagnosis than in the presumptive diagnosis (15.6％ vs.37.6％,x2=13.53,P=0.0002).There was a discrepancy between definite and presumptive causes in 30 patients (27.5％).Compared with the presumptive causes,definite upper airway cough syndrome (24.8％ vs.11.9％,x2=6.0,P=0.01) and gastroesophageal reflux disease (6.4％ vs.0,x2=7.23,P=0.007) was more frequent as a single cause of chronic cough while cough variant asthma plus gastroesophageal reflux disease (3.7％ vs.11.9％,x2=5.17,P=0.02) and upper airway cough syndrome plus nonasthmatic eosinophilic bronchitis (0 vs.9.2％,x2=10.48,P=0.001) were fewer as multiple causes of chronic cough.Conclusions A discrepancy was common between presumptive and definite causes of chronic cough.To treat presumptive causes sequentially may be a suitable solution for avoidance of erroneous multiple causes and possible over-treatment.