Objective: To evaluate the impact of implementing an institutional policy for detection of severe sepsis and septic shock.Methods: Study before (stage I), after (stage II) with prospective data collection in a 195 bed public hospital.. Stage I: Patients with severe sepsis or septic shock were included consecutively over 15 months and treated according to the Surviving Sepsis Campaign guidelines. Stage II: In the 10 subsequent months, patients with severe sepsis or septic shock were enrolled based on an active search for signs suggesting infection (SSI) in hospitalized patients. The two stages were compared for demographic variables, time needed for recognition of at least two signs suggesting infection (SSIΔt), compliance to the bundles of 6 and 24 hours and mortality.Results: We identified 124 patients with severe sepsis or septic shock, 68 in stage I and 56 in stage II. The demographic variables were similar in both stages. The Δt-SSI was 34 ± 54 hours in stage I and 7 ± 8.4 hours in stage II (p <0.001). There was no difference in compliance to the bundles. In parallel there was significant reduction of mortality rates at 28 days (54.4% versus 30%, p <0.02) and hospital (67.6% versus 41%, p <0.003). Conclusion: The strategy used helped to identify early risk of sepsis and resulted in decreased mortality associated with severe sepsis and septic shock.