Does the concept of safety culture help or hinder systems thinking in safety?

[Note: This is an update and repost of an old summary.]

ABSTRACT

The concept of safety culture has become established in safety management applications in all major safety-critical domains. The idea that safety culture somehow represents a “systemic view” on safety is seldom explicitly spoken out, but nevertheless seem to linger behind many safety culture discourses. However, in this paper we argue that the “new” contribution to safety management from safety culture never really became integrated with classical engineering principles and concepts. This integration would have been necessary for the development of a more genuine systems-oriented view on safety; e.g. a conception of safety in which human, technological, organisational and cultural factors are understood as mutually interacting elements.

Without of this integration, researchers and the users of the various tools and methods associated with safety culture have sometimes fostered a belief that “safety culture” in fact represents such a systemic view about safety. This belief is, however, not backed up by theoretical or empirical evidence. It is true that safety culture, at least in some sense, represents a holistic term—a totality of factors that include human, organisational and technological aspects. However, the departure for such safety culture models is still human and organisational factors rather than technology (or safety) itself.

The aim of this paper is to critically review the various uses of the concept of safety culture as representing a systemic view on safety. The article will take a look at the concepts of culture and safety culture based on previous studies, and outlines in more detail the theoretical challenges in safety culture as a systems concept. The paper also presents recommendations on how to make safety culture more systemic.

***********************************

From the full-text paper:

I feel like I’ve done a really poor job summarising this one – it’s absolutely worth reading the full paper.

Some points:

·????????Although many definitions exist for culture, they can broadly be divided into interpretive and functionalist approaches. Interpretive approaches to culture “share an interest in the meanings and beliefs that the members of an organisation assign to organisational elements (structures, systems and tools) and how these assigned meanings influence behaviour … Culture in these approaches can be considered a (research) framework for conceptualising the organisation and inspecting various phenomena perceived in the context of the organisation. Culture acts as a metaphor for the organisation; organisation as a culture” (p6).

·????????For functionalists, “organisational culture includes aspects that are shared by all members and that contribute to the social integration and equilibrium of the system. While interpretative approaches treat an organisation as a culture, functionalist approaches view culture as a variable, i.e. that an organisation has a culture” (p6).

·????????Subcultures are also another element described in cultural research. Subcultures operate within organisational cultures and may include: 1) occupational or professional subcultures based on educational background or other factors, 2) departmental subcultures of the work unit, c) age or tenure related subcultures. The general notion of a subculture is said to be ambiguous.

·????????The differences between systems and culture are covered. They note that while all cultures can be perceived as systems (in a sense), not all systems can be perceived as cultures. E.g. while we could have a technical system, and this could be influenced by a technical culture, it’s not a culture.

·????????Further, while a lingering belief may be that safety culture “somehow represents a “systemic view” on safety is seldom explicitly spoken out …safety culture never really became integrated with classical engineering principles and concepts. This integration would have been necessary for the development of a more genuine systems-oriented view on safety; e.g. a conception of safety in which human, technological, organisational and cultural factors are understood as mutually interacting elements” (p5). They note that this belief isn’t matched by evidence (as of 2014).

·????????Safety culture may be misused in research and practice, e.g. safety culture “sometimes tends to be used as an “excuse” for not dealing with some more fundamental safety problems involving technological design” (p8).

·????????On the above, it’s argued just like some approaches which try to adapt individuals to fit the environment/context, safety culture may also be treated similarly where it’s believed that culture “should be adapted to the current techno-logical conditions and not the other way around. We have seen this trend in some consultancy projects where the company has tried to compensate its problems with bad/weak technical solutions by emphasising only the improvement of the human component (often conceptualised as a problem of attitude)” (p8).

·????????Limitations of both functionalist and interpretive perspectives are covered. Functionalist perspectives may treat culture as more of a variable, where the key elements can be listed; although how they interact as a sociotechnical system is less understood. Interpretive perspectives may take the view of culture as a root metaphor too literally, where “culture [is used as] a synonym for organisation”.

·????????On the above limitations of any conceptualisation, safety culture, which may have normative tones, may obscure risks that are not cultural in nature. Thus, the “holistic view of culture as everything may seriously obscure other perspectives of safety” (p10).

·????????Safety culture theories may also have neglected the emerging nature of culture – like how cultures are created and the interaction within sociotechnical systems of structure, processes, behaviour and technology.

·????????“if an organisation has been found to exhibit signs of a weak (or bad) safety culture, this might implicitly be interpreted as a sign of deficient morality associated with the organisation’s managers and/or employees … However, the attribution of blame is not necessarily the best way to proceed ... Looking for “bad apples” has seldom been a fruitful strategy for coping with individual human errors, and … the concept of safety culture may be misused in the sense of attributing moralistic statements towards whole organisations ... To blame an organisation for having a weak safety culture has become almost the equivalent easy response to system problems as was blaming individuals for human errors a few decades ago” (pp. 8-9).

·????????Others, like Andrew Hopkins, has also discussed how safety culture has been misused to target individual attitudes over organisation-specific phenomenon.

·????????Referring to the work of Schein, Schein was “very critical of the whole concept of safety culture. He also emphasized the importance of technology in creating culture because it is the technology that creates the hazards that need to be controlled” (pg. 13).

·????????Furthermore, “Schein further pointed out that “culture is a property of a group not a concept” and thus “safety culture” should not have academic conceptual status (Schein, 2013b). Otherwise we would have to grant the same conceptual status to all kinds of “cultures”, from “team culture” to “service culture” (pg. 13).

·????????The authors do however see merit in safety culture as a concept and provide an example: “[Vogus et al., 2010] propose a framework of how patient safety is produced and sustained through safety culture. They define three main processes through which practices are gathered into a coherent safety culture … “Enabling” means singling out and drawing attention to safety-relevant aspects of the larger organisational culture, and creating contexts that make it possible for people to translate these aspects into meaningful activities in their local work context. These enabling conditions such as psychological safety to speak up or safety climate where the importance of safety is acknowledged must be turned into practices by enactment. Enacting a safety culture requires highlighting and accurately representing latent and manifest threats to safety and acting collaboratively to reduce them. Finally, elaborating a safety culture means to enlarge and refine practices. It includes rigorously reflecting on safety outcomes and using feedback to modify enabling practices and enacting processes. In their view safer practices need to be embedded in a coherent safety culture that sustains the salience and further development of these practices. (pg. 13).

·????????The authors assert that, “Embedding safety culture into a framework for conceptualizing system safety could provide a much needed counterforce to the prevalent focus in safety science on accidents” (pg. 14). In a closing paragraph, they state that blaming a poor safety culture should be viewed with as equal scepticism as blaming humans for errors and that similarly to human error, “safety culture” should be the start of the investigation and not its end (pg. 14).

Link in comments.

Authors: Teemu Reiman, Carl Rollenhagen, 2014, Safety Science, 68