Functional status determined by Barthel Index predicts community acquired pneumonia mortality in general population

Abstract

Barthel Index (BI) measures functional status. Our aim was to analyze if BI and other factors not included in Pneumonia Severity Index (PSI) predict mortality in general population with community acquired pneumonia (CAP). Prospective observational study including all patients with CAP diagnosed in 2006. Endpoint of study: 30-day mortality. Variables not included in PSI as BI were analyzed. Strength of association was determined by odds ratio (OR) with 95% confidence interval. 550 patients, mean age of 60.3 ± 20.8, were included. 32 were lost during follow-up and 518 patients were finally analyzed. 44 (8.5%) patients died in the first 30 days after CAP diagnosis. In bivariate analysis, mortality was significantly more frequent in patients with PSI ≥ IV (19.2% vs 1.9%), BI≤80 points (23.9% vs 2.9%), multilobar infiltrate (20% vs 6%), diabetes mellitus (14.9% vs 6.5%), influenza vaccination (11.9% vs 6.6%) and pneumococcal vaccination (16.7% vs 6%). In multivariate analysis, mortality independently associated factors were: BI ≤80, OR: 3.9(CI95% 1.4-10.5; p < 0.001); PSI ≥ IV OR: 3.9(1.2-12.7; p < 0.05); and multilobar infiltrate OR: 2.9(1.1-7.3; p = 0.05). A BI score ≤80 is associated with a higher mortality in patients with CAP independently of the PSI. BI can be a useful tool to predict CAP mortality in general population.