A comparative study of community-acquired pneumonia patients admitted to the ward and the ICU.

BACKGROUND Limited information is available on the health-care utilization of hospitalized patients with community-acquired pneumonia (CAP) depending on the location of care. Our aim was to compare the clinical characteristics, etiologies, and outcomes of patients with CAP who were admitted to the ICU with those admitted who were to the ward service. METHODS A retrospective cohort study, at two tertiary teaching hospitals, one of which was a Veterans Affairs hospital, and the other a county hospital. Eligible subjects had been admitted to the hospital with a diagnosis of CAP between January 1, 1999, and December 31, 2001, had a confirmatory chest radiograph, and a hospital discharge International Classification of Diseases, ninth revision, diagnosis of pneumonia. Subjects were excluded from the study if they had designated "comfort measures only" or had been transferred from another acute care hospital or were nursing home patients. Bivariate and multivariable analysis evaluated 30-day and 90-day mortality as the dependent measures. RESULTS Data were abstracted on 730 patients (ICU, 145 patients; wards, 585 patients). Compared to ward patients, ICU patients were more likely to be male (p = 0.001), and to have congestive heart failure (p = 0.01) and COPD (p = 0.01). ICU patients also had higher mean pneumonia severity index scores (112 [SD, 35] vs 83 [SD, 30], respectively; p = 0.02). Patients admitted to the ICU had a longer mean length of hospital stay (12 days [SD, 10 days] vs 7 days [SD, 17 days], respectively; p = 0.07), and a higher 30-day mortality rate (23% vs 4%, respectively; p < 0.001) and 90-day mortality rate (28% vs 8%, respectively; p < 0.001) compared to ward patients. CONCLUSIONS ICU patients present with more severe disease and more comorbidities. ICU patients stay longer in the hospital and have a much higher mortality rate when compared to ward patients. Management strategies should be designed to improve clinical outcomes in ICU patients.