Madrid - Sep 26, 2022. Cognitive biases contribute to diagnostic errors in more than 70% of cases, based on an analysis of malpractice claims of the US National Practitioner Data Bank developed between 1986-2010. Acknowledging this and other foundational considerations in cognitive processing could benefit medical educators to teach clinical reasoning. That’s the conclusion expressed by Prof Jeremy B Richards, Director of the Medical Education Research Laboratory at Beth Israel Deaconess Medical Center (Boston, the US) and member of the panel of experts of Practicum Script in Internal Medicine, in a paper he published in 2020. Today, he states that “critical thinking can be taught and it is necessary to make appropriate clinical decisions.”

Teaching clinical reasoning and critical thinking: from cognitive theory to practical application defines clinical reasoning as “a complex process that involves identifying pertinent clinical data, prioritising these data to develop a hypothesis or conclusion, and developing a plan to confirm or refute the clinical hypothesis”. It comprises both the biomedical knowledge and a range of thinking skills, but there is an inclination of schools to focus on content transfer and not to address thinking abilities. Prof Richards explains in his article, which he has written along with Margaret M Hayes and Richard M Schwartzstein, that clinical reasoning has traditionally been assumed “to be acquired as part of the apprenticeship phase of training”.

According to the authors, “clinical reasoning is related to and informed by critical thinking skills, which describe one’s ability to use higher cognitive processes such as analysis, synthesis, and self-reflection, skills necessary to demonstrate effective clinical reasoning.” They believe that “clinical reasoning is best represented as a combination of hypothetico-deductive, inductive, and probabilistic approaches”. Of these, medical educators have taught both evidence-based medicine and Bayesian analysis most explicitly, but this approach is limited in solving non-statistical cases, so their approach underlines how to teach hypothetico-deductive and inductive thinking.

When a practicing physician is faced with a case, the brain quickly scans the situation to see if it recognises a pattern in an exercise of routine expertise –probably just heuristics working–. System 1 thinking is not taught; it evolves as the brain learns more, and, for Prof. Richards, its use can be facilitated by illness scripts. System 2 thinking, in contrast, is analytical, deliberative, and conscious, and it must be developed by the learner in sessions designed to support specific thinking skills. This type of thinking is described as “adaptive expertise” and it is, at its best, represented by inductive reasoning. When used to evaluate a new problem, system 2 tends to consider intermediate hypotheses rather than jumping to a diagnosis straight away.

Expert clinicians will often move quickly back and forth between systems 1 and 2. Simpler forms of system 2 thinking include a forced pause when evaluating a patient to ask “what else could this be?” Learning to consistently pursue inductive reasoning, even when a case seems clear is crucial and can avoid a typical cognitive bias such as closing one’s mind to other possible diagnoses before having all the necessary data. Actually, one of the most useful de-biasing strategies is to acknowledge one’s uncertainty about a given clinical scenario and to consider alternative diagnoses. By thinking out loud and articulating one’s own uncertainties and thought process, attending physicians can demonstrate a good attitude.

Clinical reasoning teaching techniques 

To foster reasoning, faculty must help learners develop analytic reasoning skills and habits of life-long self-directed learning. Prioritising questions that begin with “why” or “how” will force learners to justify their answers. In fact, a common strategy is the “one-minute preceptor”. With probing questions aimed at evaluating the learner’s understanding of underlying mechanisms of disease, this technique teaches higher level concepts and prompts provision of feedback. Likewise, the SNAPPS model, which is an acronym for Summarize, Narrow, Analyse, Probe, Plan, and Self-study, makes the learner responsabile for the patient-centred discussion; but, if not conducted properly, these techniques may reinforce cognitive biases.

Storytelling can be a very powerful education tool. Reflective exercises, and reflective writing specifically, can improve medical learners’ metacognitive and clinical reasoning skills. The act of writing requires one to process events with the intent of relating them to others. One study of 41 internal medicine residents who participated in reflective writing exercises and small group discussions found that the residents were able to use writing to identify personal experiences with diagnostic errors and cognitive biases that may have contributed to them.

Visual representations of knowledge can also help. Concept maps relate knowledge and principles to organise information. Additionally, mechanism maps may develop inductive reasoning skills by explicitly linking elements of the history, physical examination, and laboratory data in a graphic manner. Rather than ask the learner for a diagnosis, instructors push students to explain how pieces of the case fit together to account for all that is being observed. By placing the data on paper or a screen and then creating links, knowledge is retrieved and reinforced, and a solution can be achieved, even one not seen before.

In any case, internal and external factors in the clinical setting can affect students’ clinical reasoning abilities, as well as their capacity to incorporate the above mentioned strategies. According to Prof Richards, intellectual curiosity, measured by different questionnaire-based tools, decreases during undergraduate medical education, potentially affecting learners’ clinical reasoning abilities, and burnout and fatigue are negatively associated with reasoning skills. From outside, distractions and interruptions may predispose learners to engage in cognitive shortcuts and result in increased multitasking and decreased attention to clinical tasks. 

Future panorama

In conclusion, Prof Richards, who found his collaboration at Practicum Script “very engaging and intellectually challenging”,  claims for multi-institutional research to assess the effects of educational interventions on learners’ clinical reasoning and critical thinking skills, as well as confirmatory studies reviewing the previous work. In his opinion, “clinical reasoning abilities are directly informed by and dependent on critical thinking skills, and the cognitive processes of clinical reasoning and critical thinking are inexorably linked”. For that reason, “understanding foundational considerations in cognitive processing, from the role of system 1 and system 2 thinking to the impact of cognitive biases on critical thinking, can allow medical educators to effectively teach clinical reasoning skills”.

Reference
Richards JB, Hayes MM, Schwartzstein RM. Teaching Clinical Reasoning and Critical Thinking: From Cognitive Theory to Practical Application. Chest. 2020 Oct;158(4):1617-1628.