On the need for revising healthcare failure mode and effect analysis for assessing potential for patient harm in healthcare processes

https://doi.org/10.1016/j.ress.2016.06.011Get rights and content

Highlights

  • This article discusses the appropriateness of using HFMEA in healthcare processes.

  • We conclude that HFMEA has an important role to play in such contexts.

  • We argue for two adjustments in the traditional HFMEA.

  • One is regarding the way risk is evaluated.

  • The other is to adopt a broader evaluation of barrier performance.

Abstract

Healthcare Failure Mode and Effect Analysis is a proactive, systematic method adapted from safety-critical industries increasingly used to assess the potential for patient harm in high-risk healthcare processes. In this paper we review and discuss this method. We point to some weaknesses and finally argue for two adjustments. One adjustment is regarding the way in which risk is evaluated, and the other is to adopt a broader evaluation of barrier performance. Examples are given from prehospital critical care and from the operating room environment within hospitals to illustrate these ideas.

Introduction

Healthcare Failure Mode and Effect Analysis (HFMEA) is one of several proactive Human Failure Ergonomics (HFE) risk analyzing methods commonly used to evaluate error-prone work processes, before patient harm occurs [30], [23], [11], [16]. This method is a modified version of the traditional Failure Mode and Effect Analysis (FMEA) methodology adjusted to the healthcare setting. It was developed by the National Center for Patient Safety of the US Department of Veterans Affairs in 2002 [23], [39] and has been widely used ever since within various medical disciplines [35], [42], [20], [43]. Specifically, studies have identified the importance of applying the HFMEA method to detect previously unrecognised system errors or hazards in the operating theatre [4], [35] or to evaluate, prioritize and analyse failure modes in drug management and transfusion processes [33], [21]. Studies also suggest that approximately 50% of all adverse events in hospitals occur in the operating room [19], [34], [40]. These figures point to the operating room as a “domain in which improved safety is an urgent and significant challenge” [41, p. 2]), suited to benefit from the HFMEA method. The helicopter emergency medical service (HEMS) is another example of a complex, high-risk, sociotechnical work system particularly vulnerable to adverse events and with inherent performance obstacles in which HFMEA can be applied. As far we know HFMEA has not been used in any published studies involving HEMS. Possible applications of HFMEA in the HEMS environment could be to proactively assess the risks of care transitions and to identify and assess potential vulnerabilities that could be introduced with the acquisition and implementation of new technology.

A HFMEA analysis does not exist in isolation and should be combined with other methods, such as incident learning and Structured What If Technique (SWIFT), in order to provide a comprehensive view of risk in a system [47], [37]. This alleviates the concerns raised by Nada Atef Shebl and colleagues related to the reliability of the FMEA method [38], [27].

Like most methods, the HFMEA method has a number of strengths and weaknesses [36]. In this method we review and discuss the HFMEA methodology and argue for two adjustments. We ask to what extent it is sufficient to rank different failure modes with respect only to probabilities and consequences (severity), and to what extent it is appropriate to take no further action if an effective control measure already exists.

We believe that there is a need to rethink these issues. Firstly, the basis for categorising the different failure modes should not only be made with respect to probabilities and consequences. The knowledge or lack of knowledge (uncertainties) is then not properly reflected. A broader perspective on risk is needed to take this aspect into consideration. One way to do this is by applying the risk perspective presented in [9], which defines risk as a combination of consequences and associated uncertainties. See also [5], [7], [12], [15], [24]. Secondly, from the traditional HFMEA viewpoint, there is no need for improvements in healthcare processes if an effective control measure (barrier) already exists. Generally, we consider this practice inappropriate. Several barriers might be necessary, as existing barriers may fail. The traditional HFMEA needs to be adjusted such that the cautionary principle [13] has a stronger role to play.

An adjusted version of an HFMEA taking the above aspects into consideration is presented in this article. Other methods exist to consider uncertainty, such as Safety Cases [28], [22], but the focus of this paper is specifically on the development of HFMEA to account for uncertainty.

The article consists of five sections. In Section 2 we give a short review of the HFMEA. In Section 3 we discuss the appropriateness of using the HFMEA for assessing the potential for patient harm in healthcare processes. Then, in Section 4, an adjusted version of the HFMEA is presented. Finally, in Section 5, we draw some conclusions.

Section snippets

Review of the healthcare failure mode and effect analysis

The overall aim of the HFMEA is to reveal potential failure modes in healthcare processes, such that the process can be redesigned in order to reduce healthcare errors. The HFMEA is a five step methodology. A short review of these steps is presented below [23], [39].

  • 1.

    Define the HFMEA topic

    The topic should be a clearly defined high risk or high vulnerability area or healthcare process to be proactively evaluated. Boundaries and limitations of complex processes should be described.

  • 2.

    Assemble the team

Discussion of the healthcare failure mode and effect analysis

The HFMEA is intuitively appealing, as the analysis identifies and assesses potential risks, and proposes actions on the critical parts of healthcare processes. There are, however, some weaknesses in the HFMEA that have not yet been covered in the literature. These are:

  • (a)

    Evaluation of risk is made with respect only to probabilities and consequences/severity. The knowledge and lack of knowledge (uncertainties) are then not properly taken into consideration.

  • (b)

    Attention is given to effective control

An adjusted version of the healthcare failure mode and effect analysis

An adjusted version of the HFMEA is presented in this section. Compared to the traditional HFMEA, two adjustments are made to better cover the aspects described in Section 3. The two adjustments are clarified in 4.1 A broader evaluation of risk, 4.2 A broader evaluation of effective control measures. In Section 4.3 we show what the implications of these adjustments are for the HFMEA decision tree and the HFMEA worksheet.

An example: WHO surgical safety checklist (SSC)

To illustrate the difference between the traditional and adjusted HFMEA models, we return in more detail to the WHO Surgical Safety Checklist (SSC), mentioned in Section 3. In terms of facts, the Surgical Safety Checklist is divided into three sections or phases: “Sign-in”, “Time-out”, and “Sign-out”. The Sign-in focuses on the particular safety steps that must be performed prior to the induction of anaesthesia, such as communication with the patient (confirm identity, operating area, type of

Conclusions

In this paper we argue that there is a need for some adjustments in the traditional HFMEA. The description of risk in the traditional approach is not broad enough, since the uncertainty is not properly taken into account. A two-step procedure for evaluating risk, which takes both the probability and severity of effects into consideration in addition to the uncertainties, has been suggested. We also suggest adopting a broader evaluation with respect to effective control measures such that the

Acknowledgments

We would like to thank all members of the Norwegian Air Ambulance Foundation for the financial support that made this research possible. The authors are also grateful to three anonymous reviewers for their useful comments and suggestions.

References (48)

  • E.L. Aveling et al.

    A qualitative study comparing experiences of the surgical safety checklist in hospitals in high-income and low-income countries

    MBJ Open

    (2013)
  • T. Aven

    Risk analysis – assessing uncertainties beyond expected values and probabilities

    (2008)
  • T. Aven

    Risk, surprises and black swans – fundamental ideas and concepts in risk assessment and risk management

    (2014)
  • T. Aven

    Risk analysis

    (2015)
  • T. Aven et al.

    Risk management and governance – concepts, guidelines and applications

    (2010)
  • T. Aven et al.

    Risk management, with applications from the offshore petroleum industry

    (2007)
  • J.W. Beasley et al.

    Human factors in primary care

  • T. Bedford

    Decision making for group risk reduction: dealing with epistemic uncertainty

    Risk Anal

    (2013)
  • T. Bedford et al.

    Probabilistic risk analysis: foundations and methods

    (2001)
  • K. Catchpole et al.

    Teamwork and error in the operating room - analysis of skills and roles

    Ann Surg

    (2008)
  • C.H. Cheng et al.

    Applying HFMEA to prevent chemotherapy errors

    J Med Syst

    (2012)
  • R. Dehnavieh et al.

    Proactive risk assessment of blood transfusion process, in pediatric emergency, using the Health care failure mode and effects analysis (HFMEA)

    Glob J Health Sci

    (2015)
  • Denney E, Pai G, Habili I. Towards measurement of confidence in safety cases. In: Proceedings of the 2011 international...
  • A.W. Evans et al.

    What is wrong with criterion FN-lines for judging the tolerability of risk?

    Risk Anal

    (1997)
  • Cited by (35)

    • A practical approach for evaluating the strength of knowledge supporting risk assessment models

      2020, Safety Science
      Citation Excerpt :

      The SoK is, in turn, classified into three levels, i.e. weak, strong and medium (Askeland et al., 2016). More examples of the SoK evaluation of the risk assessment models by semi-quantitative models can be found in (Abrahamsen et al., 2016), (Aven, 2017a), (Berner and Flage, 2016a, 2016b), (Khorsandi and Aven, 2017), (Haouzi et al., 2013). Another method proposed for SoK assessment is the assumption deviation risk method, whose standpoint is that poor assumptions are main sources of weak knowledge and, hence, efforts should be made for evaluating the solidity of assumptions on which risk analysis is based (Aven, 2013b); (Berner and Flage, 2016a, 2016b).

    • Understanding railway operational accidents using network theory

      2019, Reliability Engineering and System Safety
    • Modified failure mode and effects analysis under uncertainty: A rough cloud theory-based approach

      2019, Applied Soft Computing Journal
      Citation Excerpt :

      Hence, it is a bottom-up approach and also is a step-by-step method. Due to its practicability and effectiveness, FMEA approach has been applied to various fields, such as manufacturing system [4], propulsion system of a fishing vessel [5], a street cleaning vehicle [6], and healthcare [7], etc. Generally, the potential failure modes are ranked based on combining the risk factors, i.e., the severity (S), the occurrence (O), and the Detection (D), to generate a metric named after risk priority number (RPN), which is the product of the three risk factors.

    View all citing articles on Scopus
    View full text