Objective To analyze diagnostic approach to severe acute respiratory syndrome (SARS) according to the diagnostic criteria issued by the Ministry of Health of China (MHC). Methods The clinical data and the diagnostic results of 108 cases of SARS were retrospectively reviewed according to the MHC criteria. Results There were 55 men and 53 women,with a median age of 34.5 years (range,12-78 years). The interval between their first visit and clinical diagnosis was 3 days (range,0-14 days). The diagnosis was made at the first visit in 7 (6.5%,7/108) cases with a history of exposure to SARS patients and infiltrates on chest radiograph. Eighty-nine (82.4%) and 12 (11.1%) patients were categorized as probable cases and suspected cases respectively at their first visit and a clinical diagnosis of SARS was made subsequently. The interval between first visit and reaching the final diagnosis was 1-3 days in 72 (66.7%) cases and 4 days in 29 (26.9%) cases. The final diagnosis was made in 0-14 days (median,2 days) for those (n=59,54.6%) with a history of close contact with SARS patients and 2-8 days (median,3 days) for those (n=49,45.4%) living in Beijing but without such a history ( P =0.03). The chest radiograph was interpreted as unremarkable in 26 (24.1%) cases at their first visit,and the diagnosis was made in 4 days (range 2-8 days),which was significantly longer compared with other cases ( P <0.001). In patients without a history of close contact with SARS patients,all the five criteria were met after combination antibiotic therapy had failed. Conclusions A chest radiograph without infiltrates at the early stage of SARS is an important factor responsible for delayed diagnosis. In patients without a history of close contact with SARS cases,antibiotic effect was a major factor influencing doctors’ diagnosis. Objective To analyze diagnostic approach to severe acute respiratory syndrome (SARS) according to the diagnostic criteria issued by the Ministry of Health of China (MHC). Methods The clinical data and the diagnostic results of 108 cases of SARS were retrospectively reviewed according to the MHC There are 55 men and 53 women, with a median age of 34.5 years (range, 12-78 years). The interval between their first visit and clinical diagnosis was 3 days (range, 0-14 days). The diagnosis was made at the first visit in 7 (6.5%, 7/108) cases with a history of exposure to SARS patients and infiltrates on chest radiograph. Eighty-nine (82.4%) and 12 (11.1%) patients were categorized as probable cases and suspected cases were respectively at their first visit and a clinical diagnosis of SARS was made subsequently. The interval between first visit and reaching the final diagnosis was 1-3 days in 72 (66.7%) cases and 4 days in 29 (26.9%) cases . The final diagnosis was made in 0-14 days ( median, 2 days) for those (n = 59,54.6%) with a history of close contact with SARS patients and 2-8 days (median, 3 days) for those (n = 49,45.4%) living in Beijing but without Such a history (P = 0.03). The chest radiograph was interpreted as unremarkable in 26 (24.1%) cases at their first visit, and the diagnosis was made in 4 days (range 2-8 days), which was significantly longer than with other cases (P <0.001). In patients without a history of close contact with SARS patients, all the five criteria were met after combination antibiotic therapy had failed. Conclusions A chest radiograph without infiltrates at the early stage of SARS is an important factor responsible for delayed diagnosis. In patients without a history of close contact with SARS cases, antibiotic effect was a major factor influencing doctors’ diagnosis.