Is failure mode and effect analysis reliable?

To test the reliability of failure mode and effect analysis (FMEA) within a hospital setting in the United Kingdom.

Two multidisciplinary groups were recruited, within 2 hospitals from the same National Health Services (NHS) Trust, to conduct separate FMEAs in parallel on the same topic. Each group conducted an FMEA for the use of vancomycin and gentamicin. The groups followed the basic FMEA steps, which included mapping the process of care; identifying potential failures within the process; determining the severity, probability, and detectability scores for these failures; and finally making recommendations to decrease these failures.

Both groups described the process with 5 major steps: starting vancomycin or gentamicin, prescribing the antibiotics, administering the antibiotics, monitoring the antibiotics, and finally stopping or continuing the treatment. Although each group identified 50 failures, only 17 (17%) of them were common to both. Furthermore, the severity, detectability, and risk priority number scores for both groups differed markedly resulting in their failures being prioritized differently.

Failure mode and effect analysis is a useful tool to aid multidisciplinary groups in understanding a process of care and identifying errors that may occur. However, the results of this study call into question the reliability of the FMEA process that was tested. The 2 groups identified similar steps in the process of care but different potential failures with very different risk priority numbers. Such discrepancies make it impossible to identify reliably those failures that should be prioritized and thus where money, time, and effort should be allocated to avoid these failures. Health care organizations should not solely depend on FMEA findings to improve patient safety.