Abstract
Sepsis is a common, life-threatening inflammatory condition that
can occur as a consequence of an active infection. In patients with
sepsis, early identification and treatment are key components of
reducing morbidity and mortality. Unfortunately, there is no
standardized way to identify patients with sepsis in the Emergency
Medical Services (EMS) setting, potentially delaying identification
and live-saving treatment. The goal of this project was to derive
and validate a predictive model and clinical risk prediction score
for EMS identification of severe sepsis. We performed a
retrospective cohort study of sequential, adult, at-risk patients
transported by a city-wide Emergency Medical Services (EMS) system
to a 900-bed, urban, public hospital between 2011 and 2012. At-risk
patients were defined as having all 3 of the following criteria
present in the EMS setting: heart rate >=90 beats per minute, 2)
respiratory rate >=20 breaths per minute, and 3) systolic blood
pressure >=110 mmHg. Among 66,439 EMS encounters, 555 patients
were included for analysis, of which 14% (n=75) had severe sepsis.
Severe sepsis (including septic shock) was defined by review of
clinical documentation. The cohort was randomly divided into
derivation (80%) and validation (20%) subgroups, and logistic
regression was performed to determine which EMS characteristics
were associated with a diagnosis of severe sepsis. The following
six risk factors were found to be EMS predictors of severe sepsis:
older age, EMS transport from a nursing home, Emergency Medical
Dispatch (EMD) 9-1-1 chief complaint category of "Sick Person", hot
tactile temperature, low systolic blood pressure, and low oxygen
saturation. The final predictive model showed good discrimination
in both derivation and validation subgroups (AUC 0.832 and 0.803,
respectively). Sensitivity of the final model was 91% in the
derivation subgroup and 78% in the validation subgroup; specificity
was 34% and 26%, respectively. Finally, the final predictive model
was converted into a prehospital severe sepsis (PreSS) risk
prediction score. A PreSS score of >=2 points performed with a
sensitivity of 86% and a specificity of 47%. Further validation of
the PreSS score is needed before determining the potential benefit
of its use.
Table of Contents
INTRODUCTION..............1 BACKGROUND................3
METHODS...................6 RESULTS..................11
DISCUSSION...............14 FIGURES AND TABLES.......18 Table
1..........18 Figure 1.........19 Table 2..........20 Table
3..........21 Table 4..........22 Table 5..........23 Table
6..........24 Table 7..........26 Table 8..........28 Figure
2.........29 Table 9..........30 Figure 3.........31 Table
10.........32 REFERENCES...............33
About this Master's Thesis
Rights statement
- Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
| School |
|
|---|
| Department |
|
|---|
| Degree |
|
|---|
| Submission |
|
|---|
| Language |
|
|---|
| Research Field |
|
|---|
| Parola chiave |
|
|---|
| Committee Chair / Thesis Advisor |
|
|---|
| Committee Members |
|
|---|