The CURB-65 score is widely implemented as a prediction tool for identifying patients with community-acquired pneumonia (cap) at increased risk of 30-day mortality. However, since most ingredients of CURB-65 are used as general prediction tools, it is likely that other prediction tools, e.g. the British National Early Warning Score (NEWS), could be as good as CURB-65 at predicting the fate of CAP patients.
To determine whether NEWS is better than CURB-65 at predicting 30-day mortality of CAP patients.
This was a single-centre, 6-month observational study using patients' vital signs and demographic information registered upon admission, survival status extracted from the Danish Civil Registration System after discharge and blood test results extracted from a local database.
The study was conducted in the medical admission unit (MAU) at the Hospital of South West Jutland, a regional teaching hospital in Denmark.
The participants consisted of 570 CAP patients, 291 female and 279 male, median age 74 (20-102) years.
The CURB-65 score had a discriminatory power of 0.728 (0.667-0.789) and NEWS 0.710 (0.645-0.775), both with good calibration and no statistical significant difference.
CURB-65 was not demonstrated to be significantly statistically better than NEWS at identifying CAP patients at risk of 30-day mortality.