Only Severely Limited, Premorbid Functional Status Is Associated With Short- and Long-term Mortality in Patients With Pneumonia Who Are Critically III A Prospective Observational Study

Background: Severe pneumonia requiring ICU admission has not been well characterized with respect to long-term outcomes or predictors thereof. We examined the association between premorbid functional status and mortality in patients with severe pneumonia. Methods: From 2000 to 2002, a population-based cohort of adults with pneumonia who were critically ill was enrolled and prospectively followed. Short-term (30-day) and long-term (1-year) mortality were examined using multivariable Cox regression models. Results: The final cohort included 271 patients, mean age 61 years, 59% men, and 16% from nursing homes. The mean Pneumonia Severity Index was 113 (71% class IV or V), and the mean Acute Physiology and Chronic Health Evaluation II score was 17. Overall, 121 (45%) patients were functionally independent, 115 (42%) had limited mobility, and 35 (13%) were completely dependent. Mortality was 11% at 30 days and 27% at 1 year; by functional status mortality was 6% at 30 days and 17% at 1 year for patients who were independent, 10% and 31% for patients with limited mobility, and 39% and 48% for patients who were dependent. Mortality was greater for patients who were completely dependent when compared with patients who were independent (adjusted hazard ratio [aHR], 5.3; 95% CI, 2.0-14.1; P<.001 at 30 days; and aHR, 3.0; 95% CI, 1.5-6.1; P=.002 at 1 year) or with patients who had limited mobility (aHR, 4.8; 95% CI, 2.0-11.2, P<.001 at 30 days; and aHR, 2.3; 95% CI, 1.3-4,4, P=.007 at 1 year). There were no mortality differences between patients with limited mobility and patients who were independent. Conclusions: One-quarter of patients with pneumonia who are critically ill are dead within 1 year. Severely limited premorbid functional status was associated with mortality; this should be considered at presentation for prognosis and at discharge for targeted follow-up. CHEST 2011;139(0:88-94