Do things right the first time
Understand your high-risk processes
By Patrice Spath, ART
Brown-Spath and Associates
Forest Grove, OR
Many of the same root-cause analysis (RCA) techniques used to evaluate a sentinel event once it occurs can also be used to avoid potential future problems. The goal of health care organizations should be to do things right the first time. There must be activities to fix current problems; however, there must also be an emphasis placed upon what could be done differently in the future to improve the underlying processes that are creating the types of problems that occur most frequently.
An emphasis on "fire prevention" can lead to less "fire fighting." The secret to preventing errors is understanding your high-risk processes. Where are the problems likely to occur? How can the process be changed to reduce or eliminate the likelihood of an undesirable event?
Error prevention starts with effective ongoing monitoring of high-risk processes. A high-risk process is one that, if performed incorrectly or inadequately, could result in significant patient harm. Performance measures should provide information on how well the process is working and whether it is consistently followed. This type of data can lead to an understanding of the failure rate over time and the cause of these failures. Questions to ask of your high-risk processes include:
· Is the process consistent from day to day? Is it, for example, in a state of statistical control?
· If the process is in control, what is the average common cause rate of errors or process failures?
· What are the most frequent errors or failure types over the last several months?
· What process improvements should be made to reduce these errors/failures?
What health care processes are high risk? There is no easy answer to this question. Recent studies of undesirable occurrences in hospitals found that 10% to 20% of patients experience an adverse event.1 Drug-related errors are one of the common causes of adverse events, with the medication ordering and administration processes appearing to be at greatest risk for errors.2 At the other end of the incidence spectrum are anesthesia-related deaths, which in one study were found to be 8.6 per 10,000 anesthetics administered.3 While drug-related events are more common than anesthetic deaths, the processes involved may be equally risky.
Another source of information about common undesirable events in health care is the Joint Commission on Accreditation of Healthcare Organizations in Oakbrook Terrace, IL. From October 1996 through mid-September 1997, these were the most common events that caused the Joint Commission to conduct a sentinel event review at a health care facility:
· 24 events relating to medication errors;
· 18 potentially preventable suicides;
· 8 incidents involving misidentification of surgical site;
· 7 deaths related to delay in treatment;
· 5 deaths of patients in restraints;
· 3 transfusion-related deaths;
· 2 events involving oxygen distribution systems;
· 2 infant abductions.
In the absence of mandatory reporting requirements, it is impossible to know if these events are the most frequently occurring types of events or merely ones that received the most media attention. This same data reliability problem is found in malpractice claim databases. While malpractice claims data may appear to be a good source of information about high-risk processes and medical errors, research by the Harvard Medical Practice Study group has shown that negligent adverse events in hospitalized patients occur at a rate of 7.6 times that of actual claims.4 Claims databases may only be telling you about the tip of the medical error iceberg.
The delivery of health care services is a series of complex and interrelated processes. It is difficult to single out the most important processes that, if performed incorrectly, are likely to result in a sentinel event. Because drug-related adverse events are being studied widely, there are a number of sources of information on medication delivery processes that are most problem-prone.
One such source is United States Pharma co poe ia's (USP) Medication Errors Reporting Program (MERP). This is a voluntary reporting program in which health care professionals who encounter actual or potential medication errors can report their experiences to a centralized databank. Copies of the MERP reports are routinely shared with the Food and Drug Administration, the product manufacturers, and the Institute for Safe Medical Practices (ISMP). The ISMP is a nonprofit organization that works closely with practitioners, regulatory agencies, health care institutions, professional organizations, and the pharmaceutical industry to encourage safe use of medications through improvements in medication use systems. ISMP provides independent practitioner review of the medication errors submitted to MERP.
Although actual reports remain confidential, information learned through this review is distributed by ISMP through its biweekly Medication Safety Alert newsletter. For example, MERP data have demonstrated that many administration errors occur because there is no system of independent double-checks. Medications that bypass the pharmacy, such as floor stock items, and dose calculations without independent verification, such as flow rates for medicated fluids, are often the source of significant medication administration errors. The ISMP strongly recommends that the drug delivery process include more than one professional involved in accuracy checking.
A similar voluntary reporting program for undesirable patient care events is sponsored by the Department of Anesthesia at the University of Basel in Switzerland. Its Critical Incidents Reporting System is an international database of anesthesia-related events that represent a deviation from the expected course with the realistic potential for a negative outcome. With provider and patient identifiers deleted, the incidents are summarized and posted on the Internet at the Critical Incidents Reporting System Web site: http://www.medana.unibus.ch/eng/CIRS/vatci.htm.
Ask those involved in patient care
In addition to external resources, identifying high-risk processes in your organization should involve physicians and staff. Establish routine reporting mechanisms they can use to alert the quality management or risk management department of potentially dangerous situations. These alerts, sometimes called "Near Miss" reports, complement your routine incident reporting system. (An example of a reporting form is inserted in this issue.) The reports are individually reviewed by the quality or risk manager and appropriate action taken - for example, validate presence of situation, gather additional facts as needed, or refer to appropriate department or committee if situation represents valid quality concern. A summary of "Near Miss" situations should be regularly reported to the hospital's quality council along with a description of the corrective action plans.
Another method for obtaining information about high-risk processes is to conduct periodic "high-risk" surveys. These surveys can focus on a specific activity or a major function, concentrate on a particular unit or a single discipline, or they can range across the board. These surveys can be distributed randomly to physicians and staff. Some facilities involve committees in the survey process, asking them to take their usual meeting time to complete them. Survey participants are asked to think of the most recent situation in which they were involved where they felt the patient care activity could have been improved. Questions about the situation like the following are answered by the survey participants:
1. When and where did this happen (approximate date and place)?
2. What type of activity was involved? Was any equipment involved?
3. Briefly describe the situation at the time.
4. Exactly what occurred?
5. Would you classify this activity as being particularly complex? Why or why not?
6. What might have occurred if the situation had gotten even more out of control? For example, would the activity be more difficult to perform, would there be more chance for error, would there be more chance of an undesirable patient outcome?
Survey results are reviewed by quality or risk managers. Their first course of action is to determine whether the situation described actually exists or if it could have existed. Generally, this involves discussions with appropriate physicians, managers, and staff. Aggregate data about undesirable situations should be carefully analyzed for patterns, systemic and hidden problems, and high-risk processes. If an imminent sentinel event situation does exist or could recur, the case should be immediately referred to the appropriate committee or group for immediate preventive actions.
Once high-risk processes have been identified, ask department managers and staff to use RCA techniques to improve those processes. Their first task is to identify the critical steps or components involved in the high-risk process. At each step, managers and staff determine:
· What could possibly go wrong at this step?
· What would be the outcome of a failure at this step?
· What might cause a failure at this step?
Once the answers to these questions are as complete as possible, the team conducts a barrier analysis. This is done by determining how well existing safeguards or barriers are working and what safeguards may need to be added to be sure any errors are detected and corrected before patient harm occurs. (The steps of this analysis are depicted in the form inserted in this issue.) The process steps are listed in order in the first column of a form. Then each of the questions in the remaining columns are answered. In the form inserted in this issue, answers to the questions about the first step in the medication administration process have been added for illustration purposes. Because many processes cross departmental boundaries, be sure to include input from all relevant physicians and staff in the error-proofing discussions. The process improvement ideas generated through this exercise, listed in the last column, will require further discussion by everyone involved in the process prior to implementation.
Is your organizational climate a barrier?
Proactive RCA and error-proofing of high-risk processes will be less effective if the organizational climate does not support identification and resolution of problems. If RCAs are performed in a cursory fashion or the results are generally disregarded, the overall improvements will probably be less than impressive. Where deficiencies are found, managerial and/or system changes need to occur. If the answer to any of the questions below is "No," you may be working in an organization that is not as supportive of an error-free environment as it could be.
· Are physicians/staff provided advice regarding the hazards and risks involved in their jobs?
· Are there clear lines of communication for reporting potentially hazardous situations?
· Do managers minimize the chance of safety-related errors by requiring:
- proper task assignment?
- use of safe practices?
- briefing sessions before new high-risk processes are implemented?
· Is attention given to caregivers' personal factors that may increase the incidence of errors?
· Is there sufficient support for the training needs of physicians and staff?
· Do physicians and managers pay close attention to their departments' error rates?
· Is information provided to physicians/staff about the consequences of errors?
· Do managers have sufficient authority, staffing, and support to implement assigned responsibilities, analyses, and decisions relative to the high-risk processes?
· Do physicians, staff, and managers take appropriate actions when potential patient care hazards are recognized?
· Can physicians/staff refuse to participate in a task or activity when they perceive that unsafe conditions exist, without fear of reprisal?
· Do physicians/staff have a recognized standard reporting process for alerting leaders to potential problem areas, and can they report these situations without fear of reprisal?
· Are physicians/staff provided opportunities to actively participate in performance improvement activities?
Proactive RCA and error reduction cannot be viewed as a "program of the month." Complete commitment to error reduction involves the organization's leaders making a decisive personal choice to follow through on an agreed-upon plan of action. Physicians and staff will be committed to error reduction to the extent that leadership is committed.
[Editor's note: For more information about the USP MERP and the ISMP alert newsletters, contact the institute at (215) 956-9181 or visit its Web site at http://www.ismp.org.]
References
1. Andrews LB, et al. An alternative strategy for studying adverse events in medical care. Lancet 1997; 349:309-313.
2. Bates DW, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. JAMA 1995; 274:29-34.
3. Presentation given by Rosenberger J, MD, at the American Society of Anesthesiologists annual meeting. New Orleans: Oct. 21-23, 1996.
4. Report of the Harvard Medical Practice Study to the State of New York Patients, Doctors, and Lawyers: Medical Injury, Malpractice Litigation, and Patient Compensation in New York. Cambridge, MA: Harvard University Press; 1990.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content