Objective This study analyzed the medical expenditure and its influential factors, and compared the clinical effectiveness and medical expenditure of three major drugs. Methods We designed the cohort study to compare the difference of medical and pharmaceutical expenditures between patients with and without underlying diseases. Multi-linear regression was applied to analyze the influential factors. Incremental expenditure-effectiveness ratio was applied to study three clinically important drugs. Results The curing rate of non-critical patients was statistically significant than critical patients (73.68%, 99.38%, P=0.000) .The curing rate of non-critical patients without underlying diseases was statistically significant than those with underlying diseases in the cohort (96%, 99.66%, P=0.001 6). No significance was identified in the critical patients cohort. The medical expenditure of non-critical patients with and without underlying diseases were 7 879.22 and 7 172.23 RMB per capita, respectively. Accordingly, the medical expenditure in critical patients was 24 912.89 and 26 433.53 RMB per capita. No significance was identified in the two cohorts. Medical expenditure was positively correlated with age and disease severity, with its equation y=4585.71+79.04X1+17188.87X2 (X1: age, X2: disease severity). Regarding the clinical effectiveness and medical expenditure, no significance was identified in critical patients who administered small and medium dose of Methylprednisolone. The expenditure-effectiveness ratios of Ribavirin that was administered by non-critical patients without underlying dissuades were 6 107 and 4 225 RMB, respectively. Accordingly, the expenditure-effectiveness ratios of Thymosin were 11 651 and 6 107 RMB. Conclusions The curing rate of non-critical patients without underlying diseases was higher than the counterpart in the cohort. No influence of underlying diseases was found in the critical patient cohort. Medical expenditure was positively correlated with age and disease severity. Small-and-medium dose of Methylprednisolone might not influence the curing rate and medical expenditure in critical patients. The effectiveness of Thymosin for non-critical patients with and without underlying diseases was not significantly different. However, additional 5 877 RMB occurred if Thymosin was administrated. Likewise, the effectiveness of Ribavirin for non-critical patients remains the same. However, additional 1 082 RMB was consumed in Ribavirin-administrated patient.