The Utstein Cardiac Arrest Database

The Utstein Cardiac Arrest Database A custom-designed electronic database for real-time collection of information during CPR Medical Student, UCD & St. Vincent’s University Hospital Consultant in Emergency Medicine, St. Vincent’s University Hospital Department of Public Health Medicine and Epidemiology, University College Dublin Brian Hayes Mr. John Ryan Dr. Anthony Staines Topics for discussion Project background: Cardiac arrest  Resuscitation research and Utstein Style  Aims of the Utstein Cardiac Arrest Database  User interface of the Database  Other features of the Database  Conclusions  Sudden Cardiac Death  Common and often first manifestation of coronary heart disease  50% of all cardiovascular disease mortality in developed world  In Ireland, 11,000 deaths are caused by cardiovascular disease every year.  6,000 can be classified as “sudden” Survival after cardiac arrest  Usually defined as “survival until hospital discharge”  Varies greatly between different Emergency Medical Systems worldwide If the links in the chain come together rapidly and  Chances of survival efficiently, the probability of improved by well survival is increased. integrated “chain of survival” Classification of Cardiac Arrest Out-of-hospital vs In-hospital  Survival generally much worse in out-ofhospital arrests WHY? – Less likely to be witnessed by bystanders or HCPs – Delay in commencement of CPR especially if not witnessed – Delay in defibrillation – Delay in definitive care (airway mangement, ACLS drugs)  70% of cardiac arrests in Ireland occur outside the hospital setting  Overall survival rate to discharge for such patients is close to zero The Utstein Style  Outcomes after out-of-hospital cardiac arrest can be compared between EMS systems using the Utstein Style  Consensus Conference of international experts developed Utstein style in 1990 to standardise reporting of study results  Provides definitions for commonly used terms – E.g. “Response time”, “Return of Spontaneous Circulation”, “Survival”  Lists important variables that must be gathered for every patient – E.g. Arrest witnessed, Collapse-to-defibrillation interval, etc.  Recommends a template for reporting study results, in a flow-chart format The Utstein Results Template Break down arrests into groups and subgroups according to: •Aetiology •Witnessing •Initial rhythm •CPR •Discharge alive •Alive at one year •etc. Resuscitations attempted Non-cardiac aetiology Cardiac aetiology Arrest not witnessed Arrest witnessed (bystanders) Arrest witnessed (EMS personnel) Initial rhythm asystole Initial rhythm VF Initial rhythm VT Other initial rhythms Determine presence or absence of bystander CPR for each subset Never achived ROSC Any Return of Spontaneous Circulation Efforts ceased: a: Expired in the field b: Expired in ED Admitted to ICU/Ward Expired in Hospital a:total b:within 24 hours Discharged alive Seattle vs. New York Average Response time New York City Seattle 11.4 mins (median) EMTs: 2.0±1.7 min Paramedics: 6.0±2.9 min Overall survival to discharge 1.4% 28% Dublin ? ? Sources: Am J Cardiol 1990; 65:4B-12B, JAMA 1994; 271:678-683 Utstein Cardiac Arrest Database: Aims  To develop a user-friendly computerised database for the resuscitation room in SVUH ED which would: – Gather all pre-hospital data on each arrest according to the Utstein Criteria. – Gather all emergency department data in real-time, while resuscitation is in progress. – Gather outcome variable data days, weeks or months after the index arrest. Utstein Cardiac Arrest Database: Aims (continued) Requirements of the database: – Incorporate a straightforward and intuitive user interface – Make the process of data entry as easy as possible for the user – Follow the Utstein criteria  Improve the quality of data obtained for audit purposes Utstein Cardiac Arrest Database: Long-term aims  Enable rapid computer analysis of data  Allow comparison of SVUH ED with other hospitals and Emergency Medical Services (EMS) in Ireland and worldwide, through the Utstein template  Highlight deficiencies in pre-hospital and Emergency Department care and provide guidance for improved patient management and staff training at both levels Current auditing system in SVUH  Paper based and retrospective  Does not gather all of the recommended Utstein “core” data variables  E.g “Arrest witnessed by emergency personnel” “Time of arrest” entered in only 62% of cases “Initial AED rhythm” in only 48% of cases  Poorly completed    Why is this?    Retrospective data entry Poor layout of audit form Confusing use of language on audit form Utstein Cardiac Arrest Database      Programmed in Microsoft Access® 2000 and Microsoft Visual Basic® 6.0 Multitable relational database Gathers several hundred discreet data points Filters data as it is being entered to ensure its validity Locks data after it is entered to prevent it being altered without a password Design Process Iterative process  Close consultation with departmental staff  Presentation of progress to staff to obtain feedback and advice  Data Entry  Point-and-click – – Keyboard control also Barcode reader to be added Limit possible values and avoid typographic errors  Drop-down lists –  Free text entry  Greyed-out options  “Now” button – – – Time-sensitive process Hard to record accurately Synchronisation with atomic clock through internet ensuring accuracy and precision Data Entry   Option buttons Automatic calculation of values – Glasgow Coma Scale, Fluid balance pupils, drugs  Default values – Data Analysis Print-out Conclusions  Functioning database program created – Awaiting implementation in SVUH ED – Iterative developmental process to be continued through qualitative analysis over next few months  Observation of the program in action  Interviews with users to obtain feedback – Further improvement: integration of this database with the ED patient management system to share data Conclusions (continued) Electronic and paper-based systems to run in tandem until computer system firmly established  Database will grow over several years, data eventually to form the basis of an outcome study  Website to be established to distribute program as freeware  Questions and Discussion