RESULTS: 155 patients in one tertiary hospital and 152 patients in second care hospitals were recruited into study. Total lost follow-up ratio is significantly higher in second care hospitals than that in the tertiary hospital (8.6% vs 3.2%, x2= 3.928, P= 0.047). The therapeutic success rate for cough was 38.7% (controlled in 31 and improved in 29 patients) at first-step therapy, 32.3% (controlled in 26 and improved in 24 patients) at second-step therapy and 20.0% (controlled in 8 and improved in 23 patients) at third-step therapy in the tertiary hospital. It was comparable to corresponding 50.0% (controlled in 30 and improved in 46) at first-step therapy, 30.9% (controlled in 21 and improved in 26) at second-step therapy and 3.9% (controlled in 2 and improved in 9 patients) at third-step therapy in second care hospitals respectively. Furthermore, the overall cough resolution rate was not significantly different (91.0% vs 84.9%, x2= 1.052, P= 0.305). However, the efficacy of the third-step therapy is higher (20.0% vs 3.9%, x2= 18.657, P= 0.000) in the tertiary hospital than in the second care hospitals.