Walking the Tightrope with a Safety Net Blood Transfusion Process FMEA AnMed Health AnMed Health, located in Anderson, South Carolina, is one of the largest and most technologically advanced health systems in the state. A not-for-profit private corporation, AnMed Health - with,00 employees - is the county s largest employer. A regional healthcare provider, AnMed Health delivers care at approximately 0 locations, providing a wide range of diagnostic, treatment and surgical services to residents in northwest South Carolina and northeast Georgia. In addition to four hospitals - the -bed AnMed Health Medical Center, -bed AnMed Health Women s and Children s Hospital, 0-bed AnMed Health Rehabilitation Hospital and a -bed AnMed Health Wellspring chemical dependency hospital and the system includes a comprehensive cancer center, a cardiac and orthopedic center, a network of nine medical clinics and comprehensive home care division. Outpatient diagnostic, treatment, surgical and rehabilitation services are performed at both AnMed Health Medical Center and the Oglesby Center, a state-of-the-art ambulatory facility located on the AnMed Health Campus. Emergency medical care is available anytime day or night at the medical center s Emergency Department and Level II Trauma Center. History The Failure Mode and Effects Analysis (FMEA) tool was first developed and utilized by the United States Military in the late 190 s. It original use was in the evaluation of system and equipment failures. FMEA has also been used quite successfully in many industries. The aerospace industry began utilizing FMEA s in the mid-190 s. The FMEA process became more readily used within the healthcare industry during the late 1990 s. Healthcare organizations have recognized the benefits of FMEA to help mitigate potential risks and develop control strategies in high risk processes. RCA vs. FMEA Root cause analysis (RCA) and FMEA are different quality improvement tools that are both beneficial and needed to improve quality within a healthcare organization. The root cause analysis process is a reactive approach to improving quality due to the fact that RCA is performed after some type of event has occurred. The FMEA is a proactive tool that should be used before any type of event has occurred. Another difference between the two systems is that RCA focuses on an individual event where as the FMEA generally focuses on a complete process. Due to the fact that FMEA is applied prior to an event, there is more unbiased and open input about how processes may fail. Failure Mode and Effects Analysis To begin the FMEA, a high-level process flowchart should be developed. Once the flowchart has been developed, appropriate knowledge resource experts should be assembled into a FMEA project team. A FMEA knowledge expert should be utilized to train team members on the process. Once training has been completed, the team should start to build the Failure Mode and Effects Analysis for the selected process. From the high-level flowchart, the team should identify the process functions and determine the scope of the project.
Once the process functions have been identified the team should start to brainstorm potential failure modes for that process step. The brainstorming should be centered on how the process can fail or has failed. The team should list all of the potential failure modes for that process step. It is important the team not exclude any of the potential failure modes. After the list of potential failure modes has been identified, the team should then determine the potential effects for each of the potential failure modes. It is important for the team members to understand that the key component in this step involves assessing the wide range of potential effects. Having identified the potential effects of each failure the team should determine the seriousness for each failure. This should be accomplished by ranking the severity for each identified effect. The ranking is based on a 1 to 10 scale (see attachment A). The team should agree in advance on an evaluation criteria and ranking system. This evaluation criteria and ranking system may need to be adjusted to the type of process being examined. After ranking the seriousness of each failure, the team should brainstorm all of the reasons why the failure may occur. The team should list all conceivable potential causes for the failure. Once again, the emphasis should be placed on the potential causes. The second ranking that the team should place on each failure is based upon the frequency related to that specific failure. This ranking is again based on a 1 to 10 scale (see attachment B). The team should agree in advance on the ranking scale for the likelihood of occurrence. Process controls need to be identified as the activities that help prevent, stop or mitigate each potential failure. Process controls can be identified as human activities; such as visual inspections, double-checks and adherence to policy and procedures. Technology is being utilized more and more in regards to process controls in the healthcare industry. Following the identification of process controls, the team should determine, through ranking, how good their process controls are in detecting potential failures. Once again, this ranking is based on a 1 to 10 scale (see attachment C). The team should agree in advance on the ranking scale that will be utilized for this specific process. Now that rankings have been established for severity, occurrence and detection, the team should now calculate the risk priority number (RPN). This value is determined by the following formula: RPN= Severity x Occurrence x Detection. The range of the RPN values can be from 1 to 1000. Now with each RPN assigned to each potential failure mode, the team should develop priorities based upon the highest RPN s. Regardless of the RPN value, the team should investigate any high severity rankings. Corrective actions should be developed from the prioritized list of potential failures. The aim of the corrective actions should be to reduce the severity, occurrence and/or detection rankings. The FMEA should be reviewed on a regular basis to ensure the effectiveness of the corrective actions, or any event that cause a change in one of the previous rankings. Rationale for selecting Blood Transfusion process AnMed Health selected the Blood Transfusion process due to many factors. One of major factors was that it was a high risk, high volume process within our organization. Another reason was that the Joint Commission on Accreditation of Healthcare Organizations had issued a sentinel event alert for this process. A sentinel event is
an unexpected event that may involve patient death or serious physical or psychological injury. Other reasons were based upon staff interest and engagement in wanting to make a safe process, safer. Blood Transfusion FMEA Summary After much time and effort by the multidisciplinary team the FMEA rankings were completed. Overall, there were a total of 0 potential failure modes identified with specific causes. The average Risk Priority Number (RPN) was, with three () failure modes having a RPN greater than 00. The team identified a threshold RPN value of 00 that would assist in establishing priorities. Priorities and Corrective Actions After the project team had completed identifying and ranking each of the potential failures the team used the RPN s to establish priorities. There were three areas of focus that the team selected to address: 1. Pre-transfusion specimen drawn on the wrong patient. Wrong patient label placed on specimen. Transfusion started with a patient receiving dextrose. For each of the areas of focus, the team examined the current process and developed corrective action plans to reduce the risk priority number of the failure. The corrective action plans developed included: 1. Organizational wide patient identification and specimen labeling policy. Process for validation-typing for patients who had no prior blood typing history records in our organization.. Scripts developed for staff that may need to collect the additional validation-typing specimen.. Deployment of hand-held scanning and mobile printing technology for specimen procurement staff to allow facilitate patient identification and specimen labeling at the point of collection. (to be implemented in spring 00). Development and utilization of an alert sticker Infuse with normal saline only that is placed upon each unit of blood dispensed from Blood Bank. Training for all clinical staff regarding the new patient identification and specimen labeling policy.. Communication to organizational medical staff committees. Of all of the corrective actions taken by the team, the two having the biggest impacts for reducing risk were the development of the patient identification and specimen labeling policy and conducting validation-typing on patients who have no prior testing history. The recommended actions taken have resulted in a reduction in the risk priority number (RPN) by over %. AnMed Health has seen as much as 0 % of its patients having no prior testing history. However, only 8% of these patients have needed to have an additional specimen obtained due to the fact that a prior independent blood specimen had been collected and retained for testing. Factors to success This tool has proven to be an effective means of reducing risk and driving continual improvement efforts within AnMed Health. The FMEA tool has also been utilized for other high risk, high volume processes inside the organization with much success. Some of the processes that FMEA has been applied to include: surgical specimen, PCA morphine pump, home infusion and insulin processes. The factors for success deal with having a knowledgeable multidiscipline team who are committed to continuous improvement. The team must be open to discussing potential
failures related to their own processes. They must also be willing to develop corrective actions that they embrace and implement. Lastly, there would have been no success for this continuous improvement initiative if not for management support and involvement.
Attachment A SEVERITY DESCRIPITON RANKING Hazardous Very High High Moderate Low Category I: An error occurred that may contribute to or resulted in the patient s death Category H: An error occurred that required intervention necessary to sustain life Category G: An error occurred that may have contributed to or resulted in permanent patient harm Category F: An error occurred that may have contributed to or resulted in temporary harm to the patient and required initial or prolonged hospitalization Category E: An error occurred that may have contributed to or resulted in temporary harm to the patient and required intervention Category D: An error occurred that reached the patient and required monitoring to confirm that it resulted in no harm to the patient and/or required intervention to preclude harm Category C: An error occurred that reached the patient but did not cause patient harm Category B: An error occurred but the error did not reach the patient (An error of omission does reach the patient) Category A: Circumstances or events that have the capacity to cause error 10 9 8 1 National Coordinating Council for Medication Error Reporting and Prevention
Attachment B Likelihood of Occurrence More than once per day Ranking 10 Once every - days 9 Once per week 8 Once per month Once every months Once every months Once per year Once every 1 years Once every years Once every + years 1
Attachment C Probability Detection Ranking Detection not possible at any point in system Remote 0 of 10 10 9 Low Low likelihood that error will be detected before error reaches patient 1 of 10 of 10 8 Moderate Moderate likelihood of detection before error reaches patient of 10 of 10 High Error likely to be detected before error reaches patient Very High System will always detect error of 10 9 of 10 1