Benchmarking 101: Using Trauma Data to Improve Care



Benchmarking 101: Using Trauma Data to Improve Care
Benchmarking 101: Using Trauma Data to Improve Care Andrew Bernard, MD FACS Paul A. Kearney Endowed Chair in Trauma Surgery Trauma Medical Director UK Healthcare State Trauma Manager’s Council NaDonal AssociaDon of State EMS Officials’ Fall MeeDng Louisville, Kentucky October 12, 2015 ObjecDves 1.  Describe the purpose, design and use of registries 2.  List some sample system analyses with direct relevance to improving care 3.  Contrast opportuniDes with challenges in using state data registries Discharge Databases Trauma Registries Finance/Billing Data The Major Trauma Outcome Study • 
ACS COT 1982-­‐1987 139 North American hospitals 80,544 paDents MOI data Mortality 9.0% Survival probability norms StaDsDcal comparisons of actual and expected numbers of survivors made for each insDtuDon. Champion HR et al. The Major Trauma Outcome Study: establishing naDonal norms for trauma care. J Trauma. 1990; 30:1356–65. Est 1994 900 centers > 3 million entries NaDonal Trauma Registry of Canada Trauma Registry of the German Society for Trauma Surgery NaDonal Trauma Registry ConsorDum (Australia and New Zealand) Key Registry Elements to Consider 1. 
Collected variables Inclusion and exclusion criteria Registry sogware Registry staff and training Data cleaning strategies NaDonal Trauma Data Standard • 
Demographics MOI (external cause of injury codes-­‐‘e-­‐codes’) Procedures Clinical (ICD-­‐10) LOS DisposiDon Mortality AIS/ISS Charges Payers ComplicaDons Commonly missing: longer-­‐term mortality and funcDonal outcomes Sources of Data Variability • 
Prehospital/field or ED deaths PaDents who are discharged from the ED Isolated hip fractures Drownings? SuffocaDon? ElectrocuDon? Burns? Registrars •  2 staffing models: Code Agreement Among 3 Registrars –  Dedicated registrars –  Nurse coordinator/registrar combo •  Un-­‐sung heroes •  Much manual entry •  Quality can vary 29,000 cases 12 ICU’s Misset B, et al. Reliability of diagnosDc coding in intensive care paDents. Crit Care. 2008; 12:R95. Academic Emergency Medicine 2006; 13:314-­‐324. Errors skew study results Missing data are inevitable Mechanisms needed to idenDfy and fix these errors –imputaDon is an opDon Significant variability in data points remains. Example: statewide registries employ different strategies to esDmate missing Dme of injury. • 
OpDons? • 
Dispatch Dme for EMS 5, 10, 15, or 20 minutes before the EMS dispatch Dme EMS arrival to the ED as the indicator of the Dme of injury Cri,cal Limita,on: Data Linking ACADEMIC EMERGENCY MEDICINE 2006; 13:69–75 ….a valid method for matching ambulance records to a trauma registry without the use of pa,ent iden,fiers; however, the sensi,vity of iden,fying true matches is cri,cally dependent on the number and type of common variables included in the analysis. 1. 
Age (years) Gender Date Out-­‐of-­‐hospital Dmes Call Dme Hospital arrival Dme IniDal out-­‐of-­‐ hospital vital signs GCS 9.  SBP, HR, RR 10.  Out-­‐of-­‐hospital intubaDon 11.  MOI 12.  blunt vs penetraDng injury 13.  Race 14.  Alcohol in-­‐ toxicaDon 15.  County 16.  Hospital 17.  Rural versus non-­‐rural injury semng Injured adults and children who will be transferred should NOT undergo abdominal imaging.
Abdominal imaging should only be performed if it will change the course of treatment at a given
facility. When performed, the following guidelines should be considered:
1.  Imaging of the abdomen without the use of IV contrast has little diagnostic value in both adults
and children and should not be performed except in unusual circumstances.
2.  Oral contrast is not necessary in abdominal imaging for trauma in most cases.
3.  Abdominal CT scanning is generally NOT appropriate in trauma victims exhibiting signs of shock.
4.  Children are particularly vulnerable to the harmful effects of radiation.
5.  Contrary to what is sometimes believed, the evaluation of a child with possible abdominal trauma
CAN be performed clinically, without imaging, in most cases.
6.  If transferred, send all imaging electronically or on CD (DICOM format) with patient.
Abdominal Trauma Possible
Do other injuries exceed capabilities of local hospital?
Initiate Trauma Center transfer
process immediately.
Child (age < 16)
Age > 2
GCS > 8
Norm BP for Age
Norm Exam
AST < 200
HCT > 30
Norm CXR
Do Not Image Further
Unless required for
immediate treatment.
Transfer to
Trauma Center ASAP.
1.  Send images on CD.
2.  Document name and
contact # of
interpreting radiologist
Adult (age ≥ 16)
Concerning Hx or PE
Yes but if imaging
is neg, will admit.
No and if imaging is
neg, will admit.
CT Abdomen/Pelvis
with IV Contrast
*The exclusion of IV contrast in
adbominal imaging significantly reduces
diagnostic utility. Repeat imaging will be
necessary in most cases, increasing
total radiation exposure and cost.
quality improvement clinical research injury prevenDon policy development Kentucky’s 1st Level 4 Trauma Center September 13, 2010; Marcum and Wallace Memorial Hospital, Irvine VerificaDon Team Lisa Fryman, RN Dick Bartler, KHA J.D. Richardson Andrew Bernard September 13, 2010; Marcum and Wallace Memorial Hospital, Irvine ED Length of Stay
Marcum and Wallace Trauma Center
2008 Prior
Single-­‐center data 7.6% vs 9.5% 69 hospitals in 14 states Resources Hospital discharge records NaDonal Death Index NEJM, 2006; 354:366-­‐378 Trauma System Effects on Crash Mortality Crash Mortality Change % (also restraints, EtOH and speed limit) 8
Trauma Sys.
restraint laws
speed limit
*1993 Inventory of Trauma Systems *Fatality Analysis ReporDng System *NaDonal Safety Council *NaDonal Highway Traffic Safety AdministraDon Nathens JAMA 2000 Using Data in PrevenDon NaDon/World Region State County City Community CriDcisms •  Limited pre-­‐hospital death data •  Limited post-­‐discharge data •  ‘StaDc’-­‐epidemiologic analyses should be conDnuous Lynne More, PhD Hôpital de l'Enfant-­‐Jésus Québec American Journal of Public Health | August 2014, Vol 104, No. 8 The American Association for the Surgery of Trauma 2015 Annual Meeting
THE IMPACT OF SHORT PREHOSPITAL TIMES ON TRAUMA CENTER PERFORMANCE BENCHMARKING James P Byrne MD MSc (c) N Clay Mann PhD, Christopher J Hoeg MA, John P Hunt MD MPH, Avery B Nathens MD PhD Div. of General Surgery, Sunnybrook Health Sciences Center, University of Toronto, Canada Background
Trauma System Emergency Medical Services Injury Trauma Center Arrival in ED Prehospital ,me Outcome Processes of Care The Impact of Short Prehospital Times on Trauma Center Performance Benchmarking 27 Background
20M GSW L chest Massive Hemothorax SBP 70 mmHg •  Trauma centers with short prehospital Dmes: •  Early arrival in ED à Lifesaving intervenDon and survival •  More unsalvageable paDents arriving alive à Poor outcomes The Impact of Short Prehospital Times on Trauma Center Performance Benchmarking 28 Objective
To determine the impact of EMS prehospital Dme on the risk-­‐adjusted rate of ED death and overall hospital mortality at urban trauma centers across the United States The Impact of Short Prehospital Times on Trauma Center Performance Benchmarking 29 Methods
•  RetrospecDve, ecologic study design •  EMS data: NaDonal EMS InformaDon System (NEMSIS) •  Trauma centers: ACS Trauma Quality Improvement Program (TQIP) •  Linkage: DesDnaDon hospital zip code Prehospital ,mes NEMSIS TQIP EMS acDvaDons Trauma Center The Impact of Short Prehospital Times on Trauma Center Performance Benchmarking 30 Prehospital Time and Risk-adjusted ED Death
SHORT PHT The Impact of Short Prehospital Times on Trauma Center Performance Benchmarking 31 FIELD TRIAGE DECISION SCHEME:
Measure vital signs and level of consciousness
Glasgow Coma Scale
Systolic blood pressure
Respiratory rate
< 14 or
< 90 or
< 10 or > 29 (< 20 in infant < one year)
Take to a trauma center. Steps 1 and 2 attempt to identify
the most seriously injured patients. These patients should be
transported preferentially to the highest level of care within
the trauma system.
Assess anatomy of injury
• All penetrating injuries to head, neck, torso, and extremities
proximal to elbow and knee
• Flail chest
• Two or more proximal long-bone fractures
• Crushed, degloved, or mangled extremity
• Amputation proximal to wrist and ankle
• Pelvic fractures
• Open or depressed skull fracture
• Paralysis
Take to a trauma center. Steps 1 and 2 attempt to identify
the most seriously injured patients. These patients should be
transported preferentially to the highest level of care within
the trauma system.
Assess mechanism of injury and
evidence of high-energy impact
• Adults: > 20 ft. (one story is equal to 10 ft.)
• Children: > 10 ft. or 2-3 times the height of the child
High-Risk Auto Crash
• Intrusion: > 12 in. occupant site; > 18 in. any site
• Ejection (partial or complete) from automobile
• Death in same passenger compartment
• Vehicle telemetry data consistent with high risk of injury
Auto v. Pedestrian/Bicyclist Thrown, Run Over, or with
Significant (> 20 MPH) Impact
Motorcycle Crash > 20 MPH
Transport to closest appropriate trauma center, which
depending on the trauma system, need not be the highest
level trauma center.
Assess special patient or
system considerations
• Older Adults: Risk of injury death increases after age 55
• Children: Should be triaged preferentially to pediatric-capable
trauma centers
Anticoagulation and Bleeding Disorders
• Without other trauma mechanism: Triage to burn facility
• With trauma mechanism: Triage to trauma center
Time Sensitive Extremity Injury
End-Stage Renal Disease Requiring Dialysis
Pregnancy > 20 Weeks
EMS Provider Judgment
Contact medical control and consider transport to a trauma
center or a specific resource hospital.
Transport according to protocol
When in doubt, transport to a trauma center.
For more information, visit:
U.S. Department of Health and Human Services
Centers for Disease Control and Prevention
(317) 233-7315
May 30, 2012
INDIANAPOLIS—Hoosiers who suffer traumatic injuries are now more likely to be
treated at verified trauma centers immediately following an injury, thanks to the
adoption of the Triage and Transport Rule by the state’s Emergency Medical Services (EMS) Commission. The rule aims to better provide critical quality care
during the first 60 minutes following a traumatic injury, known as the “golden hour” of care in emergency medicine.
"In our efforts to create a statewide trauma system, this rule is foundational,” said State Health Commissioner Gregory Larkin, MD. “It should mean more severely injured patients will be taken directly to trauma centers, and it should also mean
more hospitals will become trauma centers."
The rule advises that once EMS personnel have assessed the condition of each
patient at the scene of an incident, they should then determine if the patient falls
into the most severe categories of trauma injury as defined by the Centers for
Disease Control and Prevention. If so, the patient must be taken to the nearest
trauma center. The exceptions may include if transporting the patient would take
more than 45 minutes, a patient’s right to decide or if the patient’s life will be endangered if care is delayed by going to the nearest trauma care center. In these
scenarios, the patient should be taken to the nearest appropriate hospital, as
determined by the EMS provider protocols.
Traumatic injury is the number one killer of Hoosiers under the age of 45.
“Our number one priority is ensuring patients with traumatic injuries receive the lifesaving care they need, when they need it,” said Arthur L. Logsdon, director, Trauma and Injury Prevention at the Indiana State Department of Health. “Trauma
centers utilize specific medical equipment and have skilled medical professionals
available to handle the most severe, blunt force and puncture wounds. Trauma
centers are proven to save the lives of those who have suffered the most
devastating injuries. I am pleased the EMS Commission has voted positively on this
In fact, nationally, there is a 25 to 30 percent greater chance of survival when
severely injured patients are taken to a trauma center versus an emergency room.
Trauma centers differ from hospital emergency rooms in some significant ways.
Emergency departments are not trauma centers, as the typical emergency
department treats broken legs, concussions, back sprains, lacerations, injuries
resulting from motor vehicle rear-end crashes and trips on the sidewalk. Trauma
center “typical patients” include those with multiple fractures, brain injuries, paralysis, punctured lungs, handgun and stab wounds, car rollovers and ejections
J Trauma Acute Care Surg 2015; 79 Sept :372-­‐7. Beware: Data Validity 1. 
Missing entries Mistakes in transcribing or coding Duplicate entries Redundant data unnecessarily introduced •  2003 NTDB dataset: 25% discarded •  Data validaDon protocols are essenDal Summary •  Resources are innumerable •  Ask relevant quesDons •  Find the right data source to answer the quesDon •  Enlist the right person/people to pull and analyze the data •  Always quesDon data validity J Trauma Acute Care Surg 2014;77:280-­‐5. 

Similar documents