Commentary on Banning M (2008) A review of clinical decision making: models and current research. Journal of Clinical Nursing 17, 187–195

The clinical decision making of nurses is an increasing area of interest for researchers, educators and clinicians. Nurses are taking on a wider range of extended and professional roles and consequently more responsibility for a wider range of clinical decisions. As Banning (2008) rightly points out in the above article, a key issue for educators is how to assist student nurses (and qualified nurses taking on more specialist roles) to develop the skills they need to manage complex information. By implication this would perhaps improve their judgement and decision-making skills. The central tenet of Banning’s paper is that there are two existing ‘models’ of clinical decision making relevant to nurses: the information processing model and the intuitive-humanist model. She then goes on to discuss a ‘new’ approach to decision making, which is based on the model used by O’Neill et al. (2005) as the basis for a computerised clinical decision support tool (the N-CODES). The suggestion in this paper is that somehow the ‘new’ model is superior to existing models and has the potential to be used for education. However, there appear to be several flaws in Banning’s assessment of the current literature on clinical decision making in nursing, which may also raise doubts about the eventual conclusion to her paper. One of the most influential areas of research into healthcare decision making has been in the area of information processing theory. Based on the original work by Newell and Simon, the information processing theory has its groundings in cognitive science and suggests that how humans reason is ‘bounded’ by the limitations of our memory (Elstein et al. 1990). Humans have, therefore, developed cognitive strategies (thought processes) for dealing with complex information, which enable us to make judgements and decisions that are mostly very efficient. On the basis of this theory, a number of researchers have examined how healthcare professionals (both doctors and nurses) process information (e.g. Westfall et al. 1986, Tanner et al. 1987, Elstein et al. 1990, Grobe et al. 1991, Cioffi 1997, Elstein 2000, Greenwood et al. 2000). This work suggests that they probably use different processes depending on experience and expertise. Strategies include hypothetico-deductive reasoning (Tanner et al. 1987) (more likely in less experienced clinicians), intuition (more likely in more experienced clinicians), pattern matching, illness scripts, fast and frugal reasoning, heuristics (e.g. Elstein et al. 1990, Cioffi 1997, 2001). What is key is that all of these possible explanations for how clinicians reason lie within the ‘broad’ umbrella of information processing theory; what differs is the explanation for how individuals reason. What Banner describes as information processing theory is a mixture of the results of studies using information processing theory (focusing on hypothetico-deductive reasoning) and more ‘normative’ approaches to decision making, which focus on how clinicians should make decisions if they were logical, rational decision makers (Dowding & Thompson 2002). Specifically, the use of decision trees is associated with clinical decision analysis, an approach to individual decision making, based on subjective expected utility theory (Dowding & Thompson 2002). Although both approaches to decision making have been used in nursing, they are certainly not the same; one focuses on describing how clinicians make judgements and decisions, the other on dictating how they should make them. Another influential area of research into decision making in nursing has undoubtedly been that of Benner and her work on intuition (Benner 1982, Benner & Tanner 1987, Benner et al. 1992). This is the basis of the ‘intuitive-humanist’ model discussed by Banner. Although there are certainly several authors and researchers who subscribe to the view that in some way intuition is a more ‘spiritual’ approach to nursing, many of the researchers who have examined how nurses reason (using information processing theory approaches) have also identified it as a reasoning strategy (Cioffi 1997). It can therefore be questioned whether or not it is actually a different approach to examining nurses clinical decision making, or really just another way of explaining or describing how nurses reason. What is interesting when one reads the original O’Neill et al. paper, which Banner describes as being a ‘new’ model of decision making, is the research on which it is based. O’Neill et al. (2005) describe how they have developed the knowledge base for a clinical decision support system, which incorporates both findings from the research on how nurses reason, with up to date research evidence. In other words, the reasoning strategies that are used as the basis for manipulating the evidence incorporated into the knowledge system are those derived from studies using information processing theory as the basis for their studies. What the framework used within the decision support system is designed to do is model how novice nurses appear to reason and assist them to develop patterns in their memory to assist them with more rapid decision making (i.e. the development of pattern recognition/intuition). Hence, it is possible that the N-CODES system designed by O’Neill et al. may assist with educating nurses in clinical decision making. What it is not is a ‘new’ model of decision making. It is, in fact, a very sophisticated approach to bringing together what we know about how nurses reason, based on over 25 years of information processing research.