Back

Impact of Sepsis Identification and Treatment Time on In-Hospital Mortality, Length of Stay, and Hours in the Intensive Care Unit

Sudat, S. E.

2021-05-30 intensive care and critical care medicine
10.1101/2021.05.26.21257894 medRxiv
Show abstract

ObjectiveThe objective of this study was to examine the impact of timely treatment and identification of sepsis on patient outcomes at Sutter Health, a mixed-payer healthcare system in northern California, US. MethodsThis observational, retrospective analysis considered electronic health record (EHR) data for individuals who presented with sepsis during 2016-17 at any of Sutter Healths 22 emergency departments (ED). Impacts were assessed for the timing of broad-spectrum antibiotic and intravenous (IV) fluid initiation, first vital signs, sepsis screening, and lactate results. Outcomes were in-hospital mortality, hospital length of stay (LOS) and intensive care unit (ICU) hours for patients discharged alive. ResultsThe final sample size was 35,847 (N=9,638 severe sepsis, N=5,309 septic shock). Early fluid initiation had the largest estimated impacts: a mortality reduction of 2.85%[2.03%,3.68%] overall and 2.94%[1.44%,4.48%] for severe sepsis (within 1 hour of sepsis presentation), and 14.66%[9.23%,20.07%] for septic shock (within 3 hours); reduced LOS (days) 1.39[1.08,1.71] overall, 2.30[1.31,3.21] severe sepsis, 3.07[1.21,4.94] septic shock; and fewer ICU hours 25.93[16.95,34.66] overall, 35.06[14.7,56.99] severe sepsis, 41.99[15.70,70.68] septic shock (within 3 hours). Sepsis screening within 30 minutes was also associated with mortality reductions (3.88%[2.96%,4.90%] overall, 1.74%[0.08%,3.50%] severe sepsis, 6.78%[3.12%,10.33%] septic shock). The greatest improvement opportunity was estimated for joint initiation of antibiotics and IV fluids, with a modest additional mortality reduction of 0.80%[0.47%,1.17%] overall, 0.77%[0.34%,1.19%] severe sepsis, 2.94%[1.83%,3.97%] septic shock; LOS reduction of 0.37[0.28,0.46] overall, 0.29[0.17,0.43] severe sepsis, 0.25[0.01,0.51] septic shock (within 1 hour); ICU hours reduction of 4.85[3.26,6.57] overall, 5.07[2.55,7.67] severe sepsis, 3.85[1.69,6.24] septic shock (within 3 hours).

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.