Abstract
The primary driver of health care costs for patients with community-acquired pneumonia (CAP) is the hospital length of stay (LOS). Unfortunately, hospital LOS comparisons are difficult to make from phase III CAP trials because of their structured designs and prespecified treatment durations. However, an opportunity still exists to draw inferences about potential LOS differences between treatments through the use of surrogates for hospital discharge. The intent of this study was to quantify the time to a clinical response, a proxy for the time to discharge readiness, among hospitalized CAP patients who received either ceftaroline or ceftriaxone in two phase III CAP FOCUS clinical trials. On the basis of the Infectious Diseases Society of America and American Thoracic Society CAP management guidelines and recent FDA guidance documents for community-acquired bacterial pneumonia, a post hoc adjudication algorithm was constructed a priori to compare the time to a clinical response, a proxy for the time to discharge readiness, between patients who received ceftaroline or ceftriaxone. Overall, 1,116 patients (ceftaroline, n=562; ceftriaxone, n=554) from the pooled FOCUS trials met the selection criteria for this analysis. Kaplan-Meier analyses showed that ceftaroline was associated with a shorter time, measured in days, to meeting the clinical response criteria (P=0.03). Of the patients on ceftaroline, 61.0, 76.1, and 83.6% achieved a clinical response by days 3, 4, and 5, compared to 54.3, 69.8, and 79.3% of the ceftriaxone-treated patients. In the Cox regression, ceftaroline was associated with a shorter time to a clinical response (HR, 1.16, P=0.02). The methodology employed here provides a framework to draw comparative effectiveness inferences from phase III CAP efficacy trials. (The FOCUS trials whose data were analyzed in this study have been registered at ClinicalTrials.gov under registration no. NCT00621504 and NCT00509106.).