We evaluated the costs and clinical outcomes of episodes of suspected sepsis in hematological patients. A propensity score-matched study was planned, comparing a retrospective cohort managed with standard assays and a prospective cohort managed with the addition of a molecular assay. Diagnostic procedures and therapy were considered as costs variables. The primary clinical endpoint was sepsis-related mortality, whereas the length of each suspected sepsis episode was investigated as a secondary endpoint. A total of 137 and 138 episodes in the prospective and the retrospective cohorts were studied, respectively; 101 pairs of highly matched episodes were analyzed, evidencing a trend of higher mortality in the retrospective cohort. No difference in length of suspected sepsis episode was observed. Significant savings were observed in the prospective cohort, especially due to reduced costs in antifungal therapy. The apparently more expensive molecular assay favored a more rational use of economic resources without influencing, and probably improving, the clinical outcome.
Keywords: cost-effectiveness study; health technology assessment; molecular diagnosis; real-time PCR; sepsis.