SummaryTrauma staff at The Alfred Hospital use a computerized decision support system to guide the care of patients during the critical first 30 minutes of resuscitation. Known as the Trauma Reception and Resuscitation System, this program generates prompts based on more than 40 algorithms and real-time clinical data, including patient vital signs and information entered by a trauma nurse. Displayed on a large overhead monitor, these prompts are used by clinicians to direct the care of trauma patients and to facilitate documentation and communication. The program reduced overall medical errors, along with the incidence of several specific types of mistakes, including aspiration pneumonia (caused by entrance of foreign materials into the bronchial tree) and errors during management of shock.Strong: The evidence consists of a comparison of overall medical errors, patients receiving error-free treatment, incidence of aspiration pneumonia, and shock management errors in patients participating in the program to two similar control groups—one evaluated at baseline and a second randomized to receive usual care during a 2-year trial of the program.
Developing OrganizationsThe Alfred Hospital
Melbourne, Victoria, Australia
Date First Implemented2006
Problem AddressedSeverely injured patients face a high risk of morbidity and mortality due to medical errors by time-pressed, highly stressed trauma staff in the emergency department (ED), especially during the critical first 30 minutes of treatment. Other industries have reduced errors in high-stress environments through decision support systems that guide critical decisions, yet few hospitals or trauma centers use this approach.
- High risk of errors, preventable death: Multidisciplinary trauma teams face tremendous time pressures, often have to perform multiple tasks simultaneously based on memory, and frequently experience breakdowns in communication, especially during the initial minutes after patients arrive at the ED. Despite the existence of guidelines, protocols, and continuous performance improvement initiatives, medical errors still occur frequently in these situations (with these errors sometimes leading to death), even in established trauma centers with experienced trauma care professionals.1 For example, an expert panel found that a quarter of trauma deaths in Victoria, Australia could have been prevented if medical errors had not been made, with the ED phase of care being responsible for the greatest number of mistakes (an average of 7.5 per patient).2
- Unrealized potential of decision support: Systems that provide guidance in high-stress situations with a large number of variables have reduced human errors in other industries, including aviation and energy. For example, computerized prompts built into flight control systems provide immediate feedback that helps avoid errors. However, few hospitals or trauma centers use these types of decision support systems to guide trauma care.3
Description of the Innovative ActivityTrauma staff at The Alfred Hospital use a computer decision support system to guide the care of patients during the critical first 30 minutes of resuscitation. Known as the Trauma Reception and Resuscitation System, this program generates prompts based on more than 40 algorithms and real-time clinical data, including patient vital signs and information entered by a trauma nurse. Displayed on a large overhead monitor, these prompts are used by clinicians to direct the care of trauma patients and to facilitate documentation and communication. Key program elements include the following:
- Entry of clinical information into system: When emergency medical services personnel notify the trauma center of an incoming patient, the trauma nurse leader, known informally as the "nurse scribe," enters preliminary information on the patient's status into the system. Once the patient arrives, the system uses physiologic monitors to track clinical data such as heart rate, blood pressure, peripheral oxygen saturation, the amount of carbon dioxide exhaled by the patient (if an endotracheal tube is in place), and body temperature. The nurse manually enters additional information (e.g., blood glucose level, level of consciousness, neurological assessment) as it becomes available.
- Algorithm-based prompts for physicians: The system contains more than 40 algorithms covering five major subcategories of trauma resuscitation: airway management, ventilation and chest decompression, management of shock, generic trauma, and specialty situations (neurotrauma, burns, spinal cord injury, and orthopedic trauma). Since interactions during trauma care tend to be complex, dozens of algorithms typically run simultaneously, meaning that many prompts may appear at once. The prompts appear in order of urgency in a drop-down menu format on a large overhead monitor on the wall of the trauma bay; the monitor also displays cumulative physiologic, diagnostic, and treatment data. The medical trauma team leader responds to these computer-generated diagnostic and intervention prompts, and the nurse scribe enters these. If the team ignores a prompt, a menu appears with selections for why it has been ignored, thus ensuring staff know they have decided against that pathway. The prompts serve as guides and do not replace the medical staff's expertise. Examples include the following:
- Simple prompts: Prompts often remind trauma staff to perform a single task, such as to roll the patient to examine the back or measure the distance between the end of the endotracheal tube and gums.
- Complex prompts: Some prompts are more complex. For example, prompts might remind staff working with a patient with abnormally low blood pressure to control external hemorrhage, check for related tachycardia, use intravenous crystalloid to restore hemodynamics, and/or trigger activation of the air entry algorithm to ensure that tension pneumothorax, hemothorax, or pericardial tamponade are not contributing to low blood pressure.
- Linkages to facilitate communication, documentation: The system connects to the hospital's computer system, thus making it easier for trauma staff to communicate with each other and with staff elsewhere in the hospital, and to document what happens to the patient. The system has links to the following:
- Printers: The system connects to a network printer, enabling users to print interim and final summaries that contain all information entered during the resuscitation. This feature frees the trauma nurse leader from writing, thus allowing greater focus on operating the system and communicating with the rest of the trauma team. Printed summaries then travel with the patient through the hospital.
- Electronic patient file: All information entered into the system during a resuscitation becomes part of an electronic patient data file with a unique number, stored on a separate server.
- Compact disc (CD) burner: The system connects to a CD burner, thus allowing for storage of each patient's information on a CD, including the patient data file, a video recording of the session, and a screen capture video of the monitor.
Context of the InnovationThe Alfred Hospital is a tertiary academic medical center in Melbourne, Victoria, Australia with a Level I adult trauma center with four trauma bays. Teams consisting of six medical and nursing personnel care for each patient, with an emergency physician serving as team leader. The impetus for this program came from the aforementioned study finding that a quarter of trauma deaths in Victoria could have been prevented if medical errors had not occurred, with the ED phase of care being the most error prone. As a result, a multidisciplinary team began researching approaches for reducing medical errors during trauma treatment, which in turn led to development of the program.
ResultsThe program reduced overall medical errors, along with the incidence of specific problems such as aspiration pneumonia and errors in management of shock.
Strong: The evidence consists of a comparison of overall medical errors, patients receiving error-free treatment, incidence of aspiration pneumonia, and shock management errors in patients participating in the program to two similar control groups—one evaluated at baseline and a second randomized to receive usual care during a 2-year trial of the program.
- Fewer overall medical errors: The 435 patients whose treatment was guided by the system during a 2-year period had an average of 2.1 medical errors, below the 2.5 average in a "baseline" control group (i.e., a group of 300 similar patients evaluated at program implementation) and the 2.3 average in the randomized group of 436 similar patients receiving usual care during the 2-year trial. Overall, 22 percent of those who participated in the program received error-free care, well above the 16-percent error-free rate in the baseline control group.
- Fewer cases of aspiration pneumonia: Only 2.5 percent of those trauma patients served by the system during the 2-year trial experienced aspiration pneumonia, less than half the 5.3 percent rate experienced by those in the randomized control group over this time period.
- Fewer shock management errors: The average patient served by the system experienced 0.55 shock management errors during treatment, below the 0.58 average in the randomized control group during the trial and the 0.75 average in the baseline control group. The program had a particularly large impact on the use of pressure dressings to control hemorrhage, an intervention that was associated with a significant reduction in the amount of blood products transfused during the trial.
Planning and Development ProcessKey steps included the following:
- Team formation: In April 2004, the hospital formed the Trauma Reception and Resuscitation Team to research ways to reduce medical errors during trauma treatment. The team included 33 emergency, anesthesiology, surgical, and critical care medical and nursing staff, along with staff from the hospital's information technology (IT) department. The team decided to develop and implement a computer system that provided prompts based on trauma algorithms and real-time patient information.
- Research: The team spent 9 months analyzing current practice and the medical literature on trauma reception and resuscitation. The review encompassed emergency medical, radiological, anesthesiology, surgical, and nursing texts. Based on this research, the team identified key decision points in initial trauma resuscitation, and compiled a list of hundreds of published algorithms that guide many resuscitation-related tasks and decisions.
- Software development: In 2005, the hospital partnered with IT specialists from Swinburne University of Technology, who began developing the system's software. Working together, staff from the hospital and university adapted algorithms from the research to create more than 40 draft algorithms specifically for the computer system. The draft algorithms underwent several levels of testing that focused on interfaces, screen displays, and content. The team reached a consensus on the final algorithms written into the software.
- Staff training: In advance of implementation, the hospital conducted a series of training sessions for all staff who would use the system. Training included online education packages, simulated trauma cases, and real cases in which trainees were paired with staff who had completed the training.
- Preliminary, small-scale trial: From October through December 2005, the team tested the system in two of the trauma center's four bays. Based on this small-scale test, the team made minor improvements, after which they deemed the system ready for a larger-scale trial.
- Comprehensive trial: From January 2006 through February 2008, the hospital conducted the aforementioned randomized study comparing key outcomes in trauma treatment in patients receiving system-guided care (in two bays) with similar metrics in a group receiving usual care (in the other two bays) and a baseline group of patients.
- Full implementation: After seeing the positive results from the comprehensive trial, hospital leaders expanded the program to all four bays in November 2008.
- Ongoing improvement: Hospital staff regularly update the software based on newly published data and their own experiences in the trauma center. The algorithms receive regular scrutiny, discussion, and evaluation by trauma specialists.
Resources Used and Skills Needed
- Staffing: Four hospital employees worked full-time over a 20-month period to develop and test the system, and many nursing and medical staff contributed expertise while continuing with their normal job responsibilities. System implementation led to the hiring of about two critical care nursing project officers. Currently, approximately 50 physicians and 36 critical care nurses use the system as part of their regular job responsibilities.
- Costs: The cost of developing the system totaled roughly $1.7 million (U.S.), while annual maintenance averages about $50,000 (U.S.).
Funding SourcesVictorian Transport Accident Commission
The Victorian Transport Accident Commission provided a 5-year research grant of about $900,000, with the hospital covering the additional costs.
Tools and Other ResourcesScreen shots of several algorithm-based prompts and additional background are available on the Trauma Reception and Resuscitation Web site: http://www.trrproject.com/.
A video of a local news station's report on the computer system is available on YouTube at: http://www.youtube.com/watch?v=_PmDaDkS-7Q&feature=player_embedded#at=46.
Getting Started with This Innovation
- Rely on available information: Existing research, information systems, policies, processes and checklists can provide the basis for this type of system. Set aside sufficient time to systematically review and assimilate this information.
- Give clinicians flexibility: Staff will be more likely to embrace the system if its role is to guide—rather than dictate—treatment decisions.
- Use simulation before application: Before rolling out the system, make sure all staff practice using it through simulated scenarios that closely approximate the hectic atmosphere of the ED.
Sustaining This Innovation
- Regularly update algorithms and computer system: Newly available research may make existing algorithms and prompts outdated. Consequently, staff need to stay abreast of current research and update the system accordingly. Similarly, technological advances can help the system run more smoothly, so work with IT specialists to update relevant software and hardware as appropriate.
- Monitor impact to stimulate quality improvement: Staff should collect, critically analyze, and regularly discuss data on the program's impact on medical errors and outcomes, and then use this information to improve the program over time.
Contact the InnovatorMark Fitzgerald, MB, BS
Director, Trauma Service
The Alfred Hospital
55 Commercial Rd
Melbourne, Victoria 3004
Phone: (011) 613 9076 5325
Innovator DisclosuresDr. Fitzgerald reported having no financial interests or business/professional affiliations relevant to the work described in this profile.
References/Related ArticlesFitzgerald M, Cameron P, Mackenzie C, et al. Trauma resuscitation errors and computer-assisted decision support. Arch Surg. 2011 Feb;146(2):218-25. [PubMed]
Houshian S, Larsen MS, Holm C. Missed injuries in a level I trauma center. J Trauma. 2002 Apr;52(4):715-9. [PubMed]
2 McDermott FT, Cordner SM, Tremayne AB. A "before and after" assessment of the influence of the new Victorian trauma care system (1997-1998 vs 2001-2003) on the emergency and clinical management of road traffic fatalities in Victoria. Victoria, Australia: Report of the Consultative Committee on Road Traffic Fatalities. December 2003.
Burke CS, Salas E, Wilson-Donnelly K, et al. How to turn a team of experts into an expert medical team: guidance from the aviation and military communities. Qual Saf Health Care. 2004;13 Suppl 1:i96-i104. [PubMed]
|Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.|
Original publication: November 09, 2011.
Original publication indicates the date the profile was first posted to the Innovations Exchange.
Last updated: November 20, 2013.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.
Date verified by innovator: November 08, 2012.
Date verified by innovator indicates the most recent date the innovator provided feedback during the annual review process. The innovator is invited to review, update, and verify the profile annually.