Forty percent of hospitalizations after severe sepsis are potentially preventable

Abstract

Rationale: Patients are frequently rehospitalized in the 90 days after severe sepsis. The rate of readmission exceeds patients' baseline rate of hospitalization, and also exceeds the rate after matched non-sepsis hospitalizations. We sought to determine the most common readmission diagnoses after severe sepsis, the extent to which readmissions may be preventable, and if the pattern of readmission diagnoses differs from that of non-sepsis hospitalizations. Methods: We studied participants in the U.S. Health and Retirement Study with linked Medicare claims (1998-2010). Using validated methods, we identified severe sepsis and non-sepsis hospitalizations, then measured 90-day readmissions in each cohort. Using Healthcare Cost & Utilization Project's Clinical Classification Software, we determined the ten most common readmission diagnoses after severe sepsis. We measured rates of "potentially preventable" readmissions using published definitions. We compared rates of all-cause, potentially preventable, and cause-specific hospitalizations between survivors of severe sepsis and non-sepsis hospitalizations using chi-squared tests. Results: We identified 3,494 severe sepsis and 41,524 non-sepsis hospitalizations, of which 2,843 (81.4%) and 40,252 (96.9%) survived to discharge, respectively. In the next 90 days, 43.6% of severe sepsis survivors were re-hospitalized, compared to 35.0% of non-sepsis survivors, p<0.001. The top readmission diagnoses following severe sepsis (Table 1) included several recognized potentially preventable diagnoses: heart failure, pneumonia, exacerbation of chronic obstructive pulmonary disease (COPD), and urinary infection. Also common were readmissions for sepsis, acute renal failure, and aspiration pneumonitis, diagnoses that could plausibly be prevented or treated early to prevent hospitalization. Patterns of readmission differed in severe sepsis survivors; rates of readmission for sepsis, renal failure, respiratory failure, and aspiration pneumonitis were higher and accounted for a greater proportion of the total readmissions. Potentially preventable hospitalizations-infection (sepsis, pneumonia, urinary tract, and skin or soft tissue), heart failure, COPD exacerbation, acute renal failure, and aspiration pneumonitis-accounted for 40.7% (95% CI: 38.4%-43.1%) of all readmissions after severe sepsis (compared to 26.1% (95% CI: 25.4%-26.8%) following non-sepsis admission, p<0.001), and 18.1% (95% CI: 16.7%-19.5%) of severe sepsis survivors experienced a readmission for one of these diagnoses (compared to 10.2% (95% CI: 9.9%-10.5%) following a non-sepsis admission, p<0.001). Conclusions: Forty percent hospitalizations after severe sepsis occur for diagnoses that may be preventable. A few disease categories account for a relatively large proportion of the hospitalizations after severe sepsis, suggesting the feasibility of tailoring post-discharge interventions to patient's personalized risk for these common post-sepsis diagnoses. (Table Presented).