BACKGROUND: The CURB-65 (confusion, urea >7 mmol/l, respiratory rate ≥30 breaths/min, low blood pressure and age ≥65 years) score is a simple, wellvalidated tool for the assessment of severity in communityacquired pneumonia (CAP). It is unknown whether it is used routinely in China.

OBJECTIVE: To determine the frequency of use of the CURB-65 score in routine hospital practice and the consequences of non-implementation.

METHODS: A retrospective analysis of medical records from 1230 in-patients with CAP in a Chinese medical college-affiliated hospital.

RESULTS: No CAP patient underwent the CURB-65 test at admission. Based on the British Thoracic Society guidelines, the 716 (58.2%) in-patients with a CURB65 score of 0 and the 402 (32.7%) in-patients with CURB-65 score of 1 should have received ambulatory treatment, whereas the 14 (1.2%) patients with CURB65 scores of ≥3 should have been admitted to the critical care unit. The maximum excess total annual costs for managing CAP patients with CURB-65 scores of 0 and 1 were estimated at respectively US$94 383.12 and US$66 313.92 in the hospital.

CONCLUSIONS: The CURB-65 scoring tool in patients with CAP was not applied in routine hospital practice, resulting in inappropriate hospitalisation and excess costs.