Health Care Utilization in the Aftermath of Severe Sepsis

By Betty T. Tran, MD, MSc

Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago

Dr. Tran reports no financial relationships relevant to this field of study.

This article originally appeared in the October 2014 issue of Critical Care Alert. It was edited by David J. Pierson, MD, and peer reviewed by William Thompson, MD. Dr. Pierson is Professor Emeritus, Pulmonary and Critical Care Medicine, University of Washington, Seattle, and Dr. Thompson is Associate Professor of Medicine, University of Washington, Seattle. Drs. Pierson and Thompson report no financial relationships relevant to this field of study.

SYNOPSIS: This observational cohort study of survivors of severe sepsis found that the post-discharge needs of this population are substantial. Severe sepsis survivors spent more days admitted to facilities after their acute hospitalization than prior and had greater mortality, a steeper decline in days at home, and a greater increase in proportion of days alive in a facility compared to survivors of non-sepsis hospitalizations.

SOURCE: Prescott HC, et al. Increased 1-year healthcare use in survivors of severe sepsis. Am J Respir Crit Care Med 2014;190:62-69.

Short-term mortality rates for severe sepsis have declined over the last 20 years, which translates to more patients surviving to hospital discharge, but little is known about their clinical course afterwards.¹ Using Medicare-linked data from participants in the Health and Retirement Study (1998-2005), Prescott and colleagues measured health care use among survivors of severe sepsis in two ways: 1) comparing their post-discharge inpatient health care use to their pre-sepsis use, and 2) comparing their post-discharge inpatient health care use to those of matched survivors of non-sepsis hospitalizations. Matching was done on age, sex, Charlson Comorbidity Index score, baseline disability as measured by deficiencies of activities of daily living, hospitalization length, and intensive care use. Severe sepsis was defined based on claims documenting infection and acute organ dysfunction. Using Medicare claims, the investigators were also able to determine the daily inpatient location of each patient (hospital, long-term acute care, or skilled nursing facility) in the 2 years surrounding their acute hospitalization. Patients who were known to be alive and not admitted to an inpatient facility were assumed to be at home, with the caveat that patients residing in nursing homes could not be identified on a daily basis.

After matching sepsis to non-sepsis hospitalizations, 1083 severe sepsis hospitalizations (93.2% of all severe sepsis hospitalizations without inpatient mortality) and 1083 non-sepsis hospitalizations were included in the analysis. The unmatched severe sepsis patients (n = 77) tended to be younger, had longer hospitalizations, were more likely to use intensive care services, and had greater post-sepsis resource use and mortality.

Severe sepsis survivors spent more days (median, 16 vs. 7 days, P < 0.001) and had a higher proportion of days alive (median, 9.6% vs. 1.9%, P < 0.001) admitted to an inpatient facility in the year after compared to the year prior to their severe sepsis hospitalization. Most of this utilization occurred in the first 90 days after discharge. This increase in facility days was also observed in the non-sepsis cohort. However, because the severe sepsis survivors had higher 90-day (27.5% vs. 15.5%, P < 0.01) and 1-year mortality (44.2% vs. 31.4%, P < 0.01), they had a steeper decline in the number of days spent at home (difference-in-differences, -38.6 days, P < 0.001) and a steeper rise in proportion of days spent in an inpatient facility (difference-in-differences, 5.4%, P < 0.001).

COMMENTARY

This study is unique in presenting the degree of resource consumption among severe sepsis survivors in a straightforward, but relevant manner. Post-discharge inpatient health care utilization can be accurately captured, and is clinically significant not only to third-party payers, but also to treating physicians, patients, and their families. In addition, the degree of resource consumption presented here is likely an underestimate, given the inability to measure non-inpatient services, such as nursing home usage and home care needs, and due to the exclusion of the unmatched severe sepsis patients who were sicker overall.

Although we tend to regard severe sepsis as a highly acute process, especially in the critical care setting, findings from this study suggest it behaves like any other chronic disease in terms of health care utilization if patients survive to hospital discharge. The within-person (pre- and post-sepsis) results highlight that a single episode of severe sepsis is often a pivotal event in terms of the patient’s health trajectory and, subsequently, health care consumption. Furthermore, although inpatient health care utilization rises similarly after non-sepsis hospitalizations, because severe sepsis survivors have markedly higher mortality, they have a dramatic increase in the proportion of days alive spent in an inpatient facility when compared to their non-sepsis peers.

The question remains as to what accounts for this excess post-discharge morbidity and mortality experienced by survivors of severe sepsis. This knowledge could direct targeted interventions and more focused attention during the critical 90-day transition period. However, it would also be worthwhile to ask whether we can prevent or effectively alter the course of many of these events. In this study, the severe sepsis population tended to be older, have a moderate comorbidity burden, and have at least mild-to-moderate functional limitations already at baseline. Therefore, a targeted, post-discharge intervention may also need to include palliative care consultation if we are truly dedicated to high quality, post-hospital care for survivors of severe sepsis.

REFERENCE

1. Stevenson EK, et al. Two decades of mortality trends among patients with severe sepsis. Crit Care Med 2014;42:625-631.