Sepsis
Conditions
Keywords
Sepsis, Clinical Protocols, Emergency Medicine
Brief summary
Sepsis is a life-threatening complication of infection that can be difficult to recognize and treat promptly. Timely administration of antibiotics for emergency department (ED) patients with sepsis is challenging. The goal of this study is to determine the potential effectiveness and unintended consequences of reorganizing ED care for patients with suspected sepsis.
Interventions
OTHERCode Sepsis protocol (full implementation)
Implementation of a coordinated, structured, multidisciplinary team-based protocol for initial evaluation and treatment of ED patients with suspected sepsis.
Sponsors
National Institute of General Medical Sciences (NIGMS)
Intermountain Health Care, Inc.
Study design
Allocation
NON_RANDOMIZED
Intervention model
SINGLE_GROUP
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Outcomes Assessor)
Intervention model description
Patients presenting to the intervention emergency department after implementation of the care reorganization intervention - Code Sepsis protocol - will be compared to patients presenting prior to the intervention. Patients presenting to control emergency departments will receive usual care both before and after Code Sepsis implementation and are included in this study to control for changes in intervention-site outcomes over time unrelated to the tested intervention.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
(overall): 1. Adult (age ≥18 years) 2. ED patient at Intermountain Medical Center, Utah Valley Hospital, Dixie Regional Medical Center 3. Arrival to study emergency department (ED) during study period of 11/13/2018 to 2/12/2021.
Exclusion criteria
(overall): (1) Trauma patient
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Door-to-antibiotic Time | Up to 24 hours from ED arrival (an average of 3 hours) | Time from sepsis patients' emergency department arrival to intravenous (or equivalent) antibiotic initiation |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| All-cause 1-year Mortality | 1 year after ED arrival | Sepsis patient death on or before the 365th day after ED arrival |
| All-cause In-hospital Mortality | From the time of admission to hospital discharge (up to 1 year, average 14 days) | Sepsis patient death prior to hospital discharge |
| Hospital Charges | From the time of admission to hospital discharge (up to 1 year, average 14 days) | Amount charged to sepsis patient for their medical care during index ED visit and associated hospitalization |
| All-cause 30-day Mortality | 30 days after ED arrival | Sepsis patient death on or before the 30th day after ED arrival |
| Hospital Length of Stay | From the time of admission to hospital discharge (up to 1 year, average 14 days) | Length of time from sepsis patients' ED arrival until hospital discharge |
Other
| Measure | Time frame | Description |
|---|---|---|
| Antibiotic Spectrum (Sepsis Patients) | Up to 24 hours from ED arrival (an average of 6 hours) | Total antibiotic spectrum score for all unique antibiotics administered in first 24 hours using antibiotic spectrum scoring system developed by Stenehjem et al. (Clin Infect Dis 2016;63:1273-1280). Individual antibiotics have spectrum scores of 1 to 5, with the total score resulting from summation of the spectrum scores for each unique antibiotic administered during the first 24 hours. Minimum total 24-hour score is therefore 1 (administration of a single, minimum-spectrum antibiotic), with no maximum score. Higher values represent treatment with antibiotics with a broader spectrum. |
| Antibiotic Utilization' | Up to 24 hours from ED arrival (an average of 6 hours) | Fraction of all ED patients receiving antibiotics within 24 hours of ED arrival |
| Antibiotic Spectrum (All ED Patients) | Up to 24 hours from ED arrival (an average of 6 hours) | Total antibiotic spectrum score for all unique antibiotics administered in first 24 hours using antibiotic spectrum scoring system developed by Stenehjem et al. (Clin Infect Dis 2016;63:1273-1280). Individual antibiotics have spectrum scores of 1 to 5, with the total score resulting from summation of the spectrum scores for each unique antibiotic administered during the first 24 hours. Minimum total 24-hour score is therefore 1 (administration of a single, minimum-spectrum antibiotic), with no maximum score. Higher values represent treatment with antibiotics with a broader spectrum. |
| Adverse Effects of Antibiotics (Sepsis Patients) | From the time of ED arrival to hospital discharge (up to 1 year, average 14 days) | Measured as the fraction of sepsis patients with a discharge diagnosis code consistent with anaphylaxis or with an adverse reaction to antibiotics |
| Adverse Effect of Antibiotics (All ED Patients) | From the time of ED arrival to hospital discharge (up to 1 year, average 14 days) | Measured as the fraction of all ED patients with a discharge diagnosis code consistent with anaphylaxis or with an adverse reaction to antibiotics |
| New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients) | Beginning 72 hours after ED arrival to 90 days after ED arrival | Incidence of positive stool test for Clostridium difficile colitis between 72 hours and 90 days after ED arrival among ED sepsis patients and (2) all ED patients. |
| New Onset Clostridium Difficile Colitis Incidence (All ED Patients) | Beginning 72 hours after ED arrival to 90 days after ED arrival. | Incidence of positive stool test for Clostridium difficile colitis between 72 hours and 90 days after ED arrival among all ED patients. |
| Antibiotic Overtreatment Rate | Up to 24 hours from ED arrival (an average of 6 hours) | Fraction of ED patients with a primary hospital discharge diagnosis of congestive heart failure or venous thromboembolism and no discharge diagnosis consistent with infection who received antibiotics in the ED |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Pre-implementation Usual Care (Intervention Site) Adult patients age ≥18 years who received usual care after presenting with sepsis to the intervention hospital emergency department before implementation of sepsis care reorganization (Code Sepsis implementation). | 50,437 |
| Code Sepsis Post-implementation (Intervention Site) Adult patients age ≥18 years presenting to the intervention hospital emergency department after implementation of sepsis care reorganization (Code Sepsis implementation).
Code Sepsis protocol (full implementation): Implementation of a coordinated, structured, multidisciplinary team-based protocol for initial evaluation and treatment of ED patients with suspected sepsis. | 33,415 |
| Pre-implementation Usual Care (Control Sites) Adult patients age ≥18 years who received usual care after presenting to the control hospital emergency department with sepsis before implementation of sepsis care reorganization (Code Sepsis implementation) at the intervention hospital. | 56,192 |
| Post-implementation Usual Care (Control Sites) Adult patients age ≥18 years who receive usual care after presenting to the control hospital emergency department after implementation of sepsis care reorganization (Code Sepsis implementation) at the intervention hospital. | 40,358 |
| Total | 180,402 |
Baseline characteristics
| Characteristic | Pre-implementation Usual Care (Intervention Site) | Total | Post-implementation Usual Care (Control Sites) | Pre-implementation Usual Care (Control Sites) | Code Sepsis Post-implementation (Intervention Site) |
|---|---|---|---|---|---|
| Age, Continuous | 43 years | 47 years | 47 years | 49 years | 45 years |
| Ethnicity (NIH/OMB) Hispanic or Latino | 8710 Participants | 26429 Participants | 5176 Participants | 6189 Participants | 6354 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 40120 Participants | 147872 Participants | 33541 Participants | 48753 Participants | 25458 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 1607 Participants | 6101 Participants | 1641 Participants | 1250 Participants | 1603 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 660 Participants | 2416 Participants | 567 Participants | 771 Participants | 418 Participants |
| Race (NIH/OMB) Asian | 1012 Participants | 2651 Participants | 362 Participants | 475 Participants | 802 Participants |
| Race (NIH/OMB) Black or African American | 1456 Participants | 3582 Participants | 471 Participants | 704 Participants | 951 Participants |
| Race (NIH/OMB) More than one race | 110 Participants | 448 Participants | 111 Participants | 115 Participants | 112 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 1802 Participants | 4499 Participants | 703 Participants | 895 Participants | 1099 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 3713 Participants | 10711 Participants | 2141 Participants | 1788 Participants | 3069 Participants |
| Race (NIH/OMB) White | 41684 Participants | 156095 Participants | 36003 Participants | 51444 Participants | 26964 Participants |
| Region of Enrollment United States | 50437 Participants | 180402 Participants | 40358 Participants | 56192 Participants | 33415 Participants |
| Sex: Female, Male Female | 27762 Participants | 96021 Participants | 20624 Participants | 30458 Participants | 17177 Participants |
| Sex: Female, Male Male | 22675 Participants | 84381 Participants | 19734 Participants | 25734 Participants | 16238 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk | EG003 affected / at risk |
|---|---|---|---|---|
| deaths Total, all-cause mortality | 2,085 / 50,437 | 1,426 / 33,415 | 2,520 / 56,192 | 1,760 / 40,358 |
| other Total, other adverse events | 2 / 50,437 | 3 / 33,415 | 1 / 56,192 | 2 / 40,358 |
| serious Total, serious adverse events | 0 / 50,437 | 0 / 33,415 | 0 / 56,192 | 0 / 40,358 |
Outcome results
Primary
Door-to-antibiotic Time
Time from sepsis patients' emergency department arrival to intravenous (or equivalent) antibiotic initiation
Time frame: Up to 24 hours from ED arrival (an average of 3 hours)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Door-to-antibiotic Time | 160 minutes |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Door-to-antibiotic Time | 146 minutes |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Door-to-antibiotic Time | 139 minutes |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Door-to-antibiotic Time | 140 minutes |
Comparison: Differences-in-differences analysis using multivariable gamma regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: <0.00195% CI: [-18.6, -7]Regression, gamma
Secondary
All-cause 1-year Mortality
Sepsis patient death on or before the 365th day after ED arrival
Time frame: 1 year after ED arrival
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | All-cause 1-year Mortality | 467 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | All-cause 1-year Mortality | 483 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause 1-year Mortality | 603 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause 1-year Mortality | 628 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.195% CI: [0.67, 1.04]Regression, Logistic
Secondary
All-cause 30-day Mortality
Sepsis patient death on or before the 30th day after ED arrival
Time frame: 30 days after ED arrival
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | All-cause 30-day Mortality | 201 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | All-cause 30-day Mortality | 262 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause 30-day Mortality | 280 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause 30-day Mortality | 332 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.4595% CI: [0.68, 1.19]Regression, Logistic
Secondary
All-cause In-hospital Mortality
Sepsis patient death prior to hospital discharge
Time frame: From the time of admission to hospital discharge (up to 1 year, average 14 days)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | All-cause In-hospital Mortality | 113 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | All-cause In-hospital Mortality | 148 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause In-hospital Mortality | 149 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | All-cause In-hospital Mortality | 190 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.1595% CI: [0.53, 1.1]Regression, Logistic
Secondary
Hospital Charges
Amount charged to sepsis patient for their medical care during index ED visit and associated hospitalization
Time frame: From the time of admission to hospital discharge (up to 1 year, average 14 days)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure. Subjects who met sepsis criteria but were missing outcome data (N=80 overall) were also excluded from this analysis.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Hospital Charges | 7.3 US dollars * 1000 |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Hospital Charges | 15.0 US dollars * 1000 |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Hospital Charges | 7.6 US dollars * 1000 |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Hospital Charges | 15.2 US dollars * 1000 |
Comparison: Differences-in-differences analysis using multivariable gamma regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.7195% CI: [-1.4, 2.1]Regression, gamma
Secondary
Hospital Length of Stay
Length of time from sepsis patients' ED arrival until hospital discharge
Time frame: From the time of admission to hospital discharge (up to 1 year, average 14 days)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Hospital Length of Stay | 2.7 days |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Hospital Length of Stay | 3.0 days |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Hospital Length of Stay | 2.7 days |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Hospital Length of Stay | 2.9 days |
Comparison: Differences-in-differences analysis using multivariable gamma regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.1995% CI: [-0.4, 0.1]Regression, gamma
Other Pre-specified
Adverse Effect of Antibiotics (All ED Patients)
Measured as the fraction of all ED patients with a discharge diagnosis code consistent with anaphylaxis or with an adverse reaction to antibiotics
Time frame: From the time of ED arrival to hospital discharge (up to 1 year, average 14 days)
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Adverse Effect of Antibiotics (All ED Patients) | 259 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Adverse Effect of Antibiotics (All ED Patients) | 172 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Adverse Effect of Antibiotics (All ED Patients) | 287 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Adverse Effect of Antibiotics (All ED Patients) | 202 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for prespecified covariates.p-value: 0.995% CI: [0.78, 1.33]Regression, Logistic
Other Pre-specified
Adverse Effects of Antibiotics (Sepsis Patients)
Measured as the fraction of sepsis patients with a discharge diagnosis code consistent with anaphylaxis or with an adverse reaction to antibiotics
Time frame: From the time of ED arrival to hospital discharge (up to 1 year, average 14 days)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Adverse Effects of Antibiotics (Sepsis Patients) | 79 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Adverse Effects of Antibiotics (Sepsis Patients) | 66 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Adverse Effects of Antibiotics (Sepsis Patients) | 79 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Adverse Effects of Antibiotics (Sepsis Patients) | 63 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.7395% CI: [0.68, 1.75]Regression, Logistic
Other Pre-specified
Antibiotic Overtreatment Rate
Fraction of ED patients with a primary hospital discharge diagnosis of congestive heart failure or venous thromboembolism and no discharge diagnosis consistent with infection who received antibiotics in the ED
Time frame: Up to 24 hours from ED arrival (an average of 6 hours)
Population: Restricted to patients who had a primary hospital discharge diagnosis of congestive heart failure or venous thromboembolism and no discharge diagnosis consistent with infection
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Antibiotic Overtreatment Rate | 48 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Antibiotic Overtreatment Rate | 42 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Overtreatment Rate | 45 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Overtreatment Rate | 58 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for prespecified covariates.p-value: 0.5995% CI: [0.69, 1.93]Regression, Logistic
Other Pre-specified
Antibiotic Spectrum (All ED Patients)
Total antibiotic spectrum score for all unique antibiotics administered in first 24 hours using antibiotic spectrum scoring system developed by Stenehjem et al. (Clin Infect Dis 2016;63:1273-1280). Individual antibiotics have spectrum scores of 1 to 5, with the total score resulting from summation of the spectrum scores for each unique antibiotic administered during the first 24 hours. Minimum total 24-hour score is therefore 1 (administration of a single, minimum-spectrum antibiotic), with no maximum score. Higher values represent treatment with antibiotics with a broader spectrum.
Time frame: Up to 24 hours from ED arrival (an average of 6 hours)
Population: Patients who did not receive antibiotics (i.e., received no antimicrobials or received only antivirals or other non-antibiotic classes of antimicrobials) were excluded from this analysis.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Antibiotic Spectrum (All ED Patients) | 3 score on a scale |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Antibiotic Spectrum (All ED Patients) | 3 score on a scale |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Spectrum (All ED Patients) | 3 score on a scale |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Spectrum (All ED Patients) | 3 score on a scale |
Comparison: Differences-in-differences analysis using multivariable linear regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for prespecified covariates.p-value: 0.0295% CI: [0.03, 0.31]Regression, Linear
Other Pre-specified
Antibiotic Spectrum (Sepsis Patients)
Total antibiotic spectrum score for all unique antibiotics administered in first 24 hours using antibiotic spectrum scoring system developed by Stenehjem et al. (Clin Infect Dis 2016;63:1273-1280). Individual antibiotics have spectrum scores of 1 to 5, with the total score resulting from summation of the spectrum scores for each unique antibiotic administered during the first 24 hours. Minimum total 24-hour score is therefore 1 (administration of a single, minimum-spectrum antibiotic), with no maximum score. Higher values represent treatment with antibiotics with a broader spectrum.
Time frame: Up to 24 hours from ED arrival (an average of 6 hours)
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure. Patients who did not receive antibiotics (i.e., received no antimicrobials or received only antivirals or other non-antibiotic classes of antimicrobials) were excluded from this analysis.
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Antibiotic Spectrum (Sepsis Patients) | 5 score on a scale |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Antibiotic Spectrum (Sepsis Patients) | 5 score on a scale |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Spectrum (Sepsis Patients) | 5 score on a scale |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Spectrum (Sepsis Patients) | 4 score on a scale |
Comparison: Differences-in-differences analysis using multivariable ??? regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.0295% CI: [0.06, 0.58]Regression, Linear
Other Pre-specified
Antibiotic Utilization'
Fraction of all ED patients receiving antibiotics within 24 hours of ED arrival
Time frame: Up to 24 hours from ED arrival (an average of 6 hours)
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | Antibiotic Utilization' | 5404 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | Antibiotic Utilization' | 3834 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Utilization' | 6039 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | Antibiotic Utilization' | 4458 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for prespecified covariates.p-value: 0.00595% CI: [1.03, 1.17]Regression, Logistic
Other Pre-specified
New Onset Clostridium Difficile Colitis Incidence (All ED Patients)
Incidence of positive stool test for Clostridium difficile colitis between 72 hours and 90 days after ED arrival among all ED patients.
Time frame: Beginning 72 hours after ED arrival to 90 days after ED arrival.
Population: Patients with a positive stool test for Clostridium difficile colitis at baseline (from 14 days prior to ED arrival to 72 hours after ED arrival) were excluded from analysis.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (All ED Patients) | 104 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (All ED Patients) | 61 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (All ED Patients) | 172 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (All ED Patients) | 79 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for prespecified covariates.p-value: 0.195% CI: [0.93, 2.14]Regression, Logistic
Other Pre-specified
New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients)
Incidence of positive stool test for Clostridium difficile colitis between 72 hours and 90 days after ED arrival among ED sepsis patients and (2) all ED patients.
Time frame: Beginning 72 hours after ED arrival to 90 days after ED arrival
Population: Restricted to participants who met sepsis criteria (acute organ failure plus suspected or confirmed infection) before ED departure. Patients with a positive stool test for Clostridium difficile colitis at baseline (from 14 days prior to ED arrival to 72 hours after ED arrival) were excluded from analysis.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Pre-implementation Usual Care (Intervention Site) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients) | 28 Participants |
| Code Sepsis Post-implementation (Intervention Site) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients) | 20 Participants |
| Pre-implementation Usual Care (Control Sites) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients) | 44 Participants |
| Post-implementation Usual Care (Control Sites) - Sepsis Patients | New Onset Clostridium Difficile Colitis Incidence (Sepsis Patients) | 30 Participants |
Comparison: Differences-in-differences analysis using multivariable logistic regression to estimate the change in outcome after versus before intervention implementation at the intervention site adjusted for the change observed at control sites and for subjects' age, sex, Elixhauser Comorbidity Score, ED arrival via ambulance, English language preference, temperature, systolic blood pressure, Acute Physiology Score, Sequential Organ Failure Assessment score, and ED-diagnosed source of infection.p-value: 0.9495% CI: [0.49, 2.17]Regression, Logistic