**2. Limitations of RCA and the traditional "Swiss Cheese" model of healthcare and hospital setting adverse events**

Attempts to reduce the incidence of adverse events and make hospitals safer have been largely unsuccessful [2, 4–6]. Like other diseases and conditions, an understanding of the underlying aetiology or "pathophysiology" of adverse events is important for the development of preventative strategies. To date the predominant theory to explain adverse events in health has been the "Swiss Cheese" model developed by James Reason from his analysis of large scale industrial and organisational accidents [9].

James Reason in his book "Managing the Risks of Organizational Accidents" states that organisational accidents, as opposed to individual accidents, are predictable events [9]. An individual accident is one in which a person or group of people makes an individual slip, lapse, or error of judgement with the net result being an adverse outcome either to the person or the people who erred, or to the person or people in the immediate vicinity. As such there is usually a relatively tight, simple explanation for cause and effect in an individual accident. On the other hand, organisational accidents have "multiple causes involving many people at different levels of an organization" [9]. These events, whilst usually infrequent, are often catastrophic. Analyses of such organisational accidents often reveal that the defences an organisation has to prevent such catastrophes are breached by a unique series of sequential hazards that play out in an environment of latent conditions, the so called "Swiss Cheese". It follows that,

*Clinical Futile Cycles: Systematic Microeconomic Reform of Health Care by Reform… DOI: http://dx.doi.org/10.5772/intechopen.106034*

one can decrease the incidence of these organisational accidents by increasing the number of defences (more cheese slices) and/or by shrinking the size of the holes in each of the defences (**Figure 1**). This is the basis for the RCA investigation of a HAE.

In 2008, Palmieri et al. published their "Health Care Error Proliferation Model" of adverse healthcare events [10]. This model takes the "Swiss Cheese Model" and specifically adapts the various factors that exist in healthcare (**Figure 2**). Most notably, they place clinician vigilance as a key defence at the sharp end of the actual adverse event, in the form of clinical improvisation and localised workarounds. This clinician vigilance repairs gaps produced by actions, changes and adjustments that are made at the blunt end of the healthcare organisation with its administrative and therefore higher level, layers of defence. A good example of this is the use of highdefinition mobile telephone devices in rural and regional settings that allow almost immediate transfer of clinical information to an appropriate clinician at a referral centre. However, this clinical workaround and improvisation is clearly at odds with most organisations' patient privacy policies that have been developed at the blunt administrative end of the organisation.

Having for the most part accepted the Reason "Swiss Cheese" model of adverse events and adapted variations, most hospitals response to adverse events has been to increase defences at the blunt end of the healthcare organisation's administration [3]. These defences, in the hospital, take the form of dedicated quality and safety units and committees, electronic event reporting systems, and the development of appropriate standards linked to hospital accreditation [3]. The aim of each of these blunt end defence layers is to continually decrease the size of the holes in each defence layer, by more audits, meetings, and RCA analysis projects combined with the use of the quality improvement cycle. Inevitably what is generated is recommendations, guidelines and more policy and procedure.

**Figure 1.** *The Reason "Swiss Cheese" model [22].*

**Figure 2.** *Healthcare error proliferation model [25].*

The "Swiss Cheese" model does explain well some types of hospital adverse events, in particular patient falls, wrong side surgery and medication errors. In the case of medication errors, root cause analysis of these events often highlights holes in the "Swiss Cheese," such as poor transcription of medication prescriptions, and failure to do appropriate checks [11]. In the case of patient falls, there is failure to identify the "at risk" patient and put in place appropriate preventative strategies. Fixing the holes or at least reducing the size of them can reduce the incidence of patient falls and medication errors. This can be done by and large with top-down policy and procedure and ensuring implementation of such [12]. The best example of this has been the reduction in incidence of wrong side surgery, with the implementation of time out, with completion of a check list before surgery [13]. The Reason "Swiss Cheese" model gives good explanation of the adverse event when there is a relatively tight temporal relationship, between the adverse event and preventative strategies. The adverse event itself in these circumstances is itself evidence that a mistake or error was made. There is usually with the "Swiss Cheese" model a series of clear errors that can be identified. This model then allows for preventative strategies to be implemented, and with the increasing move back to professional responsibility for compliance, in theory at least the Holy Grail of the perfectly safe hospital should be attainable.

However, most adverse events in hospital, particularly the more serious ones, often do not have such clear errors with a tight temporal relationship with the adverse event and the contributing errors. When the temporal relationship between the adverse event and the preventative strategies is not so tight, hospital cultural factors

start to be more significant, and the potential for policy and procedure to help is much less so, simply because it can be and often is ignored.
