The Quality-Cost Connection

Worst practices used in conducting FMEA projects

Part 1 of a 2-part series

By Patrice Spath, RHIT
Brown-Spath & Associates
Forest Grove, OR

Health care practitioners now have had a couple years of experience in doing failure mode and effect analysis (FMEA) projects. Now that we’ve got some know-how in how to do a FMEA, it is clear that these projects aren’t easy. Health care providers may not always know the best way to conduct a FMEA project, but it has become apparent that there are several wrong ways. This is the first of a two-part series on how FMEA projects can be made more effective by avoiding the worst practices.

Worst Practice #1: Choose a process that people aren’t passionate about.

The leaders’ visible commitment to patient safety must be reinforced through the establishment of strategic goals related to performance improvement/patient safety. The goals will be based in part on external requirements (e.g., Joint Commission on Accreditation of Healthcare Organizations’ National Patient Safety Goals, Medicare clinical quality improvement goals, etc.); however, it must be clearly evident to everyone that the organization’s commitment to achieving these goals is internally driven. Don’t force a particular project on staff members. The FMEA will be most successful if people truly have a passion for improving the process. Use information available in your organization about near misses, sentinel events, and other indicators of unsafe situations and high-risk processes to substantiate the need for the FMEA project.

Worst Practice #2: Choose a complex process.

When hospitals learned that medical mistakes occurred most often during medication administration, many initiated a FMEA project on this process. People soon learned that administering medications to a hospital patient involves many different complex processes. The scope of these projects had to be narrowed or else the project would continue on forever. Limiting the FMEA to a subprocess or selecting a particular high-risk activity or medication to evaluate can narrow the project scope. One hospital chose to conduct an FMEA just on the task of administering a second IV medication, a known problem for the nurses at that hospital. This task involves just three steps:

  1. Hang secondary IV bag.
  2. Open the clamp on the secondary bag.
  3. Infuse secondary bag.

Three critical failure modes were identified and actions taken to reduce the likelihood of future failures. It’s prudent to limit the FMEA to a narrow scope. The team should be spending the majority of their time finding and fixing problems instead of trying to articulate each of the steps in a very complex process.

Worst Practice #3: Don’t select the right team members.

The success of an FMEA is heavily reliant on the people chosen to be members of the project team. Team members should have sufficient knowledge of the process under study and be able to devote some time to the project. The team leader should be comfortable with the FMEA methodology and be skilled in leading the project. Are managers reluctant to provide release time for staff to attend project meetings? Do physician team members often fail to show up for meetings? If you have these problems, this is an indicator of worst practice #1 — people in the organization are not personally committed to making this particular process safer. It may be necessary to reaffirm everyone’s dedication to the project before proceeding, including medical staff leaders. Describe the team selection process in your PI/Patient Safety Plan. Ideally, leaders initiate the team member invitations to establish the importance of the project.

Worst Practice #4: Don’t define the FMEA team boundaries.

The FMEA project team members should know their role in the investigation. Are they just responsible for the FMEA analysis? Will they also be making recommendations for improvement? What group is responsible for approving these recommendations? Will the FMEA team continue to meet to monitor successful implementation of the actions? It can be very disheartening for a dedicated group of staff members to discover at the end of a project that their recommendations for improvement won’t get implemented because another leadership group didn’t approve them. Be as up front as possible! Let the team members know just how far their responsibilities and authority stretch and what other individuals or groups will become involved as the project progresses.

Worst Practice #5: Use an unnecessarily complex failure mode criticality-scoring scheme.

The Joint Commission standards require that the FMEA project team prioritize the failure modes for analysis and action. There is no mention in the standards of how this should be done. Some of the priority scoring schemes in FMEA models from other industries are fairly well-defined and, in some instances, too complex for use in health care. Often health care providers don’t have a sufficient amount of historical data to be precise about the expected frequency of a failure, nor can the effect be accurately predicted. The FMEA team can spend an inordinate amount of time determining if the probability that a particular failure will occur is, for example, low or moderate. Scoring the criticality of failures can be especially frustrating if the range of score choices is too wide (for example, between 1 and 10). Using a failure mode criticality scoring method that is easy to use and makes sense to health care professionals can save a lot of time.

Worst Practice #6: Identify root causes before selecting critical failures.

According to the Joint Commission, the sixth element of the required components of a proactive risk assessment project is "Determine why the prioritized breakdowns or failures could occur, which may include performing a hypothetical root-cause analysis." However, FMEA teams often identify causes for all failures — not waiting until after the high-priority failures are selected. This may be the result of the FMEA models that many organizations have adopted. The FMEA matrix has a place to record causes immediately after identifying failures. While it is not wrong to identify causes for every failure, it can be problematic. First, it takes time for the team to articulate causes — time that might be better spent in the action-planning steps. Second, when the causes for high-priority failures already are documented, the tendency is to accept these as actual root causes when corrective actions are formulated. How often does the team go back and validate its initial thoughts on what might be causing a critical failure? Without this validating step, the team is likely to be fixing causal factors, not root causes.

Worst Practice #7: Use only FMEA techniques to make the process safer.

The FMEA methodology for improving the safety of processes has some known limitations. For example, an FMEA allows you to consider only one failure at a time, and there is no logical process for considering multiple or interacting failures. To reduce the effect of this limitation, other safety improvement techniques should be used. One tool to consider is a fault-tree analysis. This is a graphical, deductive method for systematically listing various sequential or parallel events or combinations of faults that must occur for a particular undesired event to occur. A fault tree resembles a logic diagram or flowchart.

It can be used in combination with FMEA to identify interacting causes of critical failures. Another improvement technique that could come in handy during the action-planning step is inventive problem solving or TRIZ (a Russian acronym). TRIZ consists of several inventive tools that help improve the desirable features of a process and/or simultaneously eliminate the undesirable features. One of the powers of TRIZ problem solving comes from deliberately changing the question from "What might fail?" to "What are we trying to accomplish?"

Worst Practice #8: Devote too little time to process improvements.

By the time the FMEA project team gets to the process redesign step, enthusiasm for the project may have waned. This is especially true if the team spent a considerable amount of time in defining the process, selecting failure modes, and choosing the high-priority failures for action. Ideally, these steps were quickly completed, and the team has lots of energy left over for action planning. This is the phase of the FMEA project that deserves the most attention. The team must be encouraged to be innovative in redesigning processes; otherwise, the same familiar action plans will be implemented with the same less-than-desired results.

(Editor’s note: In the second part of this series, you’ll learn about four more FMEA "worst practices" and find out how to avoid these practices.)