Effect of age and residential status on the predictive performance of CURB-65 score.

Background: The CURB-65 score was designed to predict 30-day mortality and decide the site of care of pneumonia. It is uncertain how age and residential status affect the accuracy of CURB-65 score in older patients. Aim: This study aimed to evaluate the effect of age and residential status on the performance of CURB-65 score in predicting 30-day all-cause mortality of pneumonia in older patients. Methods: We used combined data from two prospective observational cohorts of adult patients with pneumonia. Older patients aged ≥ 65 years were analysed. Results: Of 1644 patients, 1200 were living in the community and 464 residing in nursing homes. The 30-day mortality rates of community-acquired pneumonia (CAP) and nursing home-acquired pneumonia (NHAP) were 10.7% and 14.2% respectively. In each age group of CAP and NHAP, mortality increased with pneumonia severity. Mortality decreased with advancing age in NHAP, and the reverse was true for CAP. In NHAP but not CAP, the proportion of highly severe pneumonia (CURB-65 score ≥ 3) was lower in older patients. The overall predictive performance of CURB-65 score was comparable in CAP and NHAP. The accuracy deteriorated with advancing age in NHAP but not CAP. Using the cut-off score of ≥3 (highly severe pneumonia), the CURB-65 score achieved high negative predictive values (>90%) in CAP and NHAP. Conclusions: Advancing age adversely affected the predictive performance of CURB-65 score in NHAP but not CAP. The score was not helpful in guiding the site of care at admission in older patients with pneumonia.