Sepsis, Patient Deterioration
Conditions
Keywords
Sepsis, Patient Deterioration, AI, Artificial Intelligence, Machine learning, Decision support, Clinical decision support, Early warning systems
Brief summary
This study is designed to test two new risk scores - one designed to predict a patient's four-hour risk of developing sepsis and one designed to predict a patient's four-hour risk of deterioration (cardiac arrest, death, unplanned ICU transfer, or rapid response team call). The goal of this study is to improve provider awareness of a patient's risk of these two negative outcomes by providing them with new risk scores. The primary outcome will be the time from when the risk score becomes elevated to when vital signs such as heart rate or blood pressure are measured, suggesting an increased awareness.
Detailed description
The University of Maryland Medical System is conducting a study designed to test two new risk scores - one designed to predict a patient's four-hour risk of developing sepsis and one designed to predict a patient's four-hour risk of deterioration (death, cardiac arrest, unplanned ICU transfer, or rapid response call). The goal of this study is to improve provider awareness of a patient's risk of these two negative outcomes, sepsis and deterioration, by providing them with new risk scores. To assess provider awareness of risk, this study will measure time from elevated risk score to measurement of vital signs, such as heart rate or blood pressure with the hope that increased awareness of risk will decrease time to vital sign measurement, indicating closer monitoring. The first risk score, the RESCUE Score, produces a highly accurate 1-4 color-coded risk score where a 1 indicates that a patient has a low risk of having a deterioration event in the next four hours and a 4 indicates that the patient has a high risk of having a deterioration event in the next four hours. The second risk score, the SEPSys Score, produces a highly accurate 1-4 color coded risk score where 1 indicates that a patient has a low risk of developing sepsis in the next 4 hours and a 4 indicates that the patient has a high risk of developing sepsis in the next 4 hours. Both of these scores are predictive of events that may happen in the future but are not detective or diagnostic of events that are actively occurring. In this study, hospitals will be randomized for the order in which they cross-over from the control arm (no risk score) to either having the SEPSys or RESCUE Score visible for their inpatients. After a few months with an individual risk score, the second risk score will be added. Thus, at the beginning of the study, no hospitals with have the new risk scores available, for a period in the middle of the study each hospital will have one risk score available, and by the end of the study all hospitals will have both risk scores available. Regardless of which risk scores are available, the clinicians will be free to recommend treatments and implement care plans using their own judgement. No procedures impacting patient care will be specified for this study.
Interventions
DEVICESEPSys
The SEPSys Score is a clinical decision support tool for predicting the risk of developing sepsis in the next 4 hours. The SEPSys Score produces a 1-4, color-coded risk, where 4 (red) is the highest risk and 1 (green) is the lowest risk.
DEVICERESCUE
The RESCUE Score is a clinical decision support tool for predicting the risk of patient experiencing clinical deterioration (cardiac arrest, rapid response team call, death, or unplanned increase in care) in the next 4 hours. The RESCUE Score produces a 1-4, color-coded risk, where 4 (red) is the highest risk and 1 (green) is the lowest risk.
Sponsors
University of Maryland, Baltimore
Study design
Allocation
RANDOMIZED
Intervention model
FACTORIAL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Subject)
Intervention model description
This is a 2-intervention + control cluster-randomized factorial stepped wedge design clinical trial
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Hospital Unit Inclusion Criteria: * Units are part of a member hospital of the University of Maryland Medical System (UMMS) * Units treat inpatient patients * Units agree to participate in the study General Encounter Inclusion Criteria: * Encounter is on a participating inpatient unit * Encounter is associated with a bedded patient in the hospital * Encounter is associated with an adult (at least 18 years of age at time of admission) Additional SEPSys Encounter Inclusion Criteria: • Encounter has a SEPSys Score Additional RESCUE Encounter Inclusion Criteria: • Encounter has a RESCUE Score
Exclusion criteria
Hospital Unit
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Time To Vital Sign Measurement | From time of elevation of either the Sepsis Early Prediction System (SEPSys) or RESCUE Score to measurement of vital signs, assessed continually during the hospitalization (up to 24 hours after the elevation)). | Time to vital sign measurement after elevation of either the Sepsis Early Prediction System (SEPSys) or RESCUE Score. An elevated SEPSys or RESCUE Score is defined as a current SEPSys or RESCUE Score of 3 or 4 (highest risk) when the previous Score was a 1 (lowest risk) or 2. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Sepsis diagnosis | Beginning at time of hospital admission and assessed continually until hospital discharge (average of 5 days). | The patient develops sepsis during the hospitalization following elevation of the SEPSys score. An elevated SEPSys or RESCUE Score is defined as a current SEPSys or RESCUE Score of 3 or 4 (highest risk) when the previous Score was a 1 (lowest risk) or 2. |
| Time to sepsis-specific treatment | From time of elevation of the SEPSys Score to initiation of sepsis-specific antibiotics or fluid bolus, assessed continually during the hospitalization (up to 96 hours). | Time from first elevated SEPSys Score to order/administration of sepsis indicated antibiotics or fluid bolus. An elevated SEPSys or RESCUE Score is defined as a current SEPSys or RESCUE Score of 3 or 4 (highest risk) when the previous Score was a 1 (lowest risk) or 2. |
| Deterioration Events | Beginning at time of hospital admission and assessed continually until hospital discharge (average of 5 days). | Binary outcome of deterioration event (cardiac arrest, death, rapid response team call, unplanned increase in level of care). An elevated SEPSys or RESCUE Score is defined as a current SEPSys or RESCUE Score of 3 or 4 (highest risk) when the previous Score was a 1 (lowest risk) or 2. |
| Frequency of laboratory orders | From time of elevation of either the Sepsis Early Prediction System (SEPSys) or RESCUE Score to measurement of laboratory orders, assessed continually during the hospitalization (up to 24 hours). | Frequency of laboratory orders or measurements after an elevated SEPSys or RESCUE Score. An elevated SEPSys or RESCUE Score is defined as a current SEPSys or RESCUE Score of 3 or 4 (highest risk) when the previous Score was a 1 (lowest risk) or 2. |
Countries
United States
Contacts
Primary ContactSamuel A Tisherman, MD
Backup ContactJason Heavner, MD
Outcome results
None listed