Facing the challenge: Decreasing case fatality rates in severe sepsis despite increasing hospitalizations* Viktor Y. Dombrovskiy, MD, PhD, MPH; Andrew A. Martin, MD; Jagadeeshan Sunderram, MD; Harold L. Paz, MD S epsis remains a life-threaten- ing disease associated with a high mortality rate and is in- creasing in incidence in the United States (1, 2). Severe sepsis—that is, sepsis associated with organ dysfunc- tion—is of particular concern. More than 750,000 cases of severe sepsis were re- ported in the United States in 1995, and this number was expected to increase at a rate of 1.5% per yr (3). Patients with severe sepsis represent 6% to 15% of all patients in intensive care units (ICUs) and consume 40% of ICU resources (3– 5). Their mortality is 1.5–2.5 times greater than overall mortality in the ICU and has not changed essentially for the past 25 yrs (6). Comprehensive epidemiologic analy- sis is important to determine the extent of severe sepsis found in the community, to identify risk factors of disease, to de- velop rational treatment for hospitalized patients, and to allocate resources through efficient healthcare policy. An- gus and colleagues (3), using the admin- istrative data from seven states, including New Jersey, estimated the incidence, mortality, and hospital resource use for severe sepsis in the United States in 1995. Researchers in the United States and other countries determined the incidence and mortality for severe sepsis within sin- gle centers at various time periods (4, 7–14). Martin et al. (2) determined the incidence and hospital mortality of sepsis in general over a large time period. Until now, there have been no studies of the incidence and mortality of severe sepsis in a large geographic area over time. During the period under study (1995– 2002), the perspective on management of sepsis and its complications changed sig- nificantly. First, maximizing cardiac out- put was shown to have no survival benefit in two studies (15, 16). Second, the ap- proach to mechanical ventilation for acute respiratory distress syndrome be- gan to change to emphasize lower tidal volumes and a tolerance of high PCO 2 levels to avoid barotrauma and ventilator- induced lung injury (17, 18). Subse- quently, these changes in ventilator strat- *See also p. 2700. From the Division of Pulmonary and Critical Care Medicine, Department of Medicine, UMDNJ–Robert Wood Johnson Medical School, New Brunswick, NJ Dr. Dombrovskiy, Dr. Martin, and Dr. Sunderram have disclosed that they have no financial relationships or interests in any commercial companies. Dr. Paz has disclosed that he is a consultant/advisor for Johnson & Johnson. Address requests for reprints to: Viktor Dom- brovskiy, MD, PhD, MPH, UMDNJ-Robert Wood John- son Medical School, 125 Paterson Street, Suite 1400, New Brunswick, NJ 08903– 0019. E-mail: dombrovy@umdnj.edu Copyright © 2005 by Copyright © 2005 by Lip- pincott Williams & Wilkins DOI: 10.1097/01.CCM.0000186748.64438.7B Objective: To determine recent trends in severe sepsis–related rates of hospitalization, mortality, and hospital case fatality in a large geographic area and to determine the impact of age, race, and gender on these outcomes. Design: Trend analysis for the period of 1995 to 2002. Setting: Acute care hospitals in New Jersey. Patients: Subjects >18 yrs of age with severe sepsis who were hospitalized in New Jersey during the period of 1995 to 2002. Interventions: None. Measurements and Main Results: We analyzed data from the 1995–2002 New Jersey State Inpatient Databases (SID) developed as part of the Healthcare Cost and Utilization Project (HCUP), covering all acute care hospitals in the state. On the basis of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes for septicemia and organ dysfunc- tion, we identified 87,675 patients with severe sepsis. The per- centage of patients with severe sepsis among all hospitalized patients with sepsis grew steadily, from 32.7% to 44.7% (p < .0001), during these years. The crude rate of hospitalization with severe sepsis increased 54.2%, from 135.0/100,000 population in 1995 to 208.2/100,000 population in 2002 (p < .0001). Over time, the crude mortality rate rose by 35.8% (p < .0001), whereas the crude case fatality rate (number of deaths/number of cases) fell from 51.0% to 45.0% (p < .0001). For any given year, the rates of hospitalization and mortality were greater among older patients. After adjustment by age, the rates among blacks were greater than among whites, and they were greater among males than females. At the same time, there was no significant difference in the age-adjusted hospital case fatality rates with regard to gender and race. There was a significant increase in age-adjusted gen- der- and race-specific rates for hospitalization and mortality from 1995 to 2002. Blacks were more likely than whites to be admitted to the intensive care unit: for males, odds ratio  1.19 (95% confidence interval, 1.13–1.26), and for females, odds ratio  1.35 (95% confidence interval, 1.29 –1.42). However, although case fatality rate was increased among patients admitted to the intensive care unit, this was not reflected in an increased case fatality among blacks. In addition, age-adjusted gender- and race-specific case fatality rates declined during 1995–2002. Conclusions: In spite of increasing rates of hospitalization and mortality, there is a decreasing case fatality rate for severe sepsis. These data suggest that advances in critical care practice before and during the study period have resulted in improved outcomes for this population. (Crit Care Med 2005; 33:2555–2562) KEY WORDS: sepsis; severe sepsis; hospitalization; mortality; case fatality; trend 2555 Crit Care Med 2005 Vol. 33, No. 11