Collecting Data for a Registry

The first step is to begin methodically collecting a minimum set of SCA data in a registry to measure survival rates and system characteristics related to SCA response. Tracking the performance of the EMS system and the victim’s outcome requires the EMS agency to collaborate with receiving hospitals to obtain information about survival to hospital discharge for each case. When the information is used for quality improvement measures, hospitals should enthusiastically provide outcome information to the EMS system on an ongoing basis. Receiving hospitals may require a simple data sharing agreement with the EMS agency.

The first step is to begin methodically collecting a minimum set of SCA data in a registry to measure survival rates and system characteristics related to SCA response. Tracking the performance of the EMS system and the victim’s outcome requires the EMS agency to collaborate with receiving hospitals to obtain information about survival to hospital discharge for each case. When the information is used for quality improvement measures, hospitals should enthusiastically provide outcome information to the EMS system on an ongoing basis. Receiving hospitals may require a simple data sharing agreement with the EMS agency.

Sources of Data for the Registry

EMS systems often collect a multitude of patient-level information. Determining which pieces of information to use may appear daunting at first. Early adopters of cardiac arrest registries developed a list of essential and optional data elements called the Utstein style, named for the historic abbey in Norway where these pioneers first met.14Data for an SCA registry can be obtained from:

  1. EMS patient care records
    1. Patient fields: demographic data, location of arrest, witnessed status, bystander CPR
    2. Treatment fields: defibrillation, CPR, medications, IVs, ventilation methods, therapeutic hypothermia
    3. Time fields: restoration of spontaneous circulation, arrival at hospital
  2. 911 dispatch information
    1. Treatment fields: dispatcher assisted CPR
    2. Time fields: time of initial 911 call, time EMS/first responders dispatched, time EMS/first responders arrived at location
  3. EMS equipment (heart monitor/defibrillators, AEDs, end tidal CO2 monitors)
    1. Patient fields: complete ECG recording including initial rhythm
    2. Treatment fields: time of first defibrillation, CPR quality metrics, ventilation rates, end-tidal CO2 values
  4. Hospital medical records
    1. Patient fields: therapeutic hypothermia, emergent diagnostic heart catheterization, emergent percutaneous coronary intervention, emergent coronary artery bypass grafting, implantable cardiovertor defibrillator placement
    2. Patient outcome: admit to hospital, discharge alive, cerebral performance category (CPC) score
    3. Discharge disposition: died, discharge alive to home, discharge alive to rehabilitation, discharge alive to extended care facility (ECF)

Several of the variables noted above may be recorded in multiple locations. When this is so, the most accurate source of data should be used consistently to populate the registry. A complete list of suggested data fields is available online at HeartRecue Project and in the Appendix section of this guide.

Case Definition for a Prehospital SCA Registry

Cases meeting these criteria should be entered into the registry:

  • Out-of-hospital non-traumatic cardiac arrest
  • Victim assessed by organized EMS personnel
  • Victim received either

– External defibrillation by lay responders or emergency personnel, or
– Chest compressions by EMS personnel

Maintain a Consistent Case Definition

The precise definition of a “case” included in the registry is very important. Included cases should follow a standard cardiac arrest case definition to allow benchmarking across systems. Selection criteria for cases that are included in the registry must remain the same over time to allow improvements in survival rates to be tracked accurately.

Historically, cases were excluded from the registry if thorough review of the available medical information revealed a primary cause of the event in a body system outside of the heart.14 For example, cardiac arrest may be caused by drowning. Cases remaining were most likely due to a primary heart problem; they were of presumed cardiac etiology.

There is considerable variation in classification of the cause of OHCA depending upon the resources available to review hospital records and upon the training of the abstractors of the information. For example, with the need to elicit voluntary cooperation from receiving hospital employees, EMS systems may not be able to request thorough hospital chart review to determine the most likely cause of the event. It may be simpler and more consistent to include all cases except those obviously caused by external injury.