Suzannah Alexander

Academic writing

Medical education is designed to give clinicians the tools they need to provide the best patient care possible; thus, effective clinical training is of the upmost importance to both physicians and patients. As new learning technologies have become available, medical educators have provided new ways for medical students and residents to gain as much experience with patient cases as early in the learning process as possible. Interactive case scenarios, virtual patients, and live and video simulations have been used in medical schools across the globe, but data are still lacking as to which formats are the most effective in teaching the foundational knowledge and clinical reasoning skills required for a successful, lifelong medical practice.

Medical Education and Learning Theory

Clinical training is built on the cornerstones of forming an identity that allows students to act like doctors; teaching clinical reasoning skills that enable students to correctly diagnose patients; establishing decision-making skills for proper patient management; and developing ethics, patient advocacy principles, and team-building skills to successfully interact with other clinicians. This training has traditionally taken place in the in-patient clinical setting where the Socratic Method is practiced, as attending physicians ask questions and medical students and residents answer in a dialogue meant to stimulate critical thinking (Bailey & Rutledge, 2017).

In the 1950s and early 1960s, medical education emphasized the organ system model. Basic knowledge was taught out of context before students entered the clinical setting. This framework was based on the behaviorist theory of instruction, quizzing, positive feedback, and repetition, but it was ineffective in teaching problem-solving skills (Bailey & Rutledge, 2017).

In the late 1960s and 1970s, educators realized that teaching concepts out of context was not ideal, and the problem-based learning model was initiated to enhance clinical relevance and peak student interest. Problem-based learning uses real-world, open-ended problems to teach complex concepts in small-group settings. The goal is to improve clinical reasoning skills and to encourage self-study and new knowledge acquisition (Bailey & Rutledge, 2017). These patient management problems often include vignettes that require learners to select the best management decisions. Although they are designed to encourage problem-solving skills, they have only proven to test the acquisition of knowledge, so their use has declined (Thampy et al., 2019).

In the early 1990s, the Clinical Presentation Curriculum Model built on the deficiencies of the problem-based learning model by augmenting cognitive processes, improving structural knowledge, and developing better problem-solving schema (Bailey & Rutledge, 2017). This scheme model is based on teaching typical clinical presentations for common patient complaints and abnormal examination and laboratory findings. Social support was also emphasized in training students how to think and behave like physicians.

Since the 1990s, understanding, rather than memorization, and demonstrated competencies have been emphasized. Clinical reasoning, however, has not been prioritized. It was assumed that physicians would learn clinical reasoning through the mentorship process (Guze, 2015; Richards et al., 2020). However, as the pace of medical research has accelerated, it has become nearly impossible to keep up with clinical advances, such that teaching core knowledge and skills is no longer adequate for a career in medicine (Guze, 2015). It is therefore imperative that physicians leave their formal training not only with foundational knowledge, but also with the clinical reasoning skills and commitment to lifelong learning necessary to stay up to date with changes in the field. Learning to be cognizant of the influences of biases, environmental factors, and higher-order critical thinking processes is essential to improving patient care, decreasing delays to the proper diagnosis, and reducing other medical errors.

How Cognitive Psychology Applies to Clinical Reasoning

Clinical reasoning is a critical part of medical education. It combines the domain knowledge of medicine with the structural knowledge of critical thinking, including inductive and deductive reasoning. It is the cornerstone by which physicians analyze a patient’s case, formulate differential diagnoses, determine what information is important in a patient’s history, decide what tests and imaging studies to order, weigh examination and laboratory findings, develop management strategies, and evaluate the effectiveness of treatments. To efficiently and effectively treat patients, physicians must combine their knowledge of different presentations and diagnoses with the likelihood, potential etiologies, and optimal treatment plans for each illness, using causal reasoning and systems thinking to determine how the underlying cause manifests in the signs and symptoms the patient is experiencing. The ideal management strategy treats the underlying cause, not just the resultant disease presentation (Shin, 2019).  

Because human reasoning can be faulty, physicians are prone to the same types of reasoning errors as everyone else, and the more time limited and challenging the circumstance, the more likely they are to make mistakes. Physicians must therefore be aware of different biases and common errors in judgment, and medical educators must strive to teach problem-solving skills that not only help learners treat typical case scenarios, but also manage unique encounters (Shin, 2019). Clinical reasoning is also affected by how curious, motivated, or tired the physician may be as well as the time pressures, clinical environment, number of interruptions or distractions, and dynamics within the team (Richards et al., 2020).

Learning clinical reasoning is best achieved when a community provides context through applicable situations. The in-person clinical setting is highly conducive to providing relevance because students are emersed in real patient cases and surrounded by their peers and mentors. Students bring their own experiences and scientific knowledge into a setting where they can integrate and apply new information, allowing knowledge transfer through encoding specificity. Retention is further encouraged by explaining their reasoning to their instructors and peers, and feedback can be offered in real time (Bailey & Rutledge, 2017). The clinical setting gives learners the opportunity to hone critical thinking skills by analyzing cases, synthesizing information, and reflecting on their decision-making processes (Richards et al., 2020).

Optimal teaching of clinical reasoning skills should vary for novice versus expert learners. The way in which novices and experts note and process information differs, such that different instructional methods may work well for experts but have no beneficial effect on novices (Bailey & Rutledge, 2017). Studies suggest that novice physicians often use deductive reasoning, while expert physicians use inductive reasoning for familiar cases but deductive reasoning for difficult or unusual cases. In inductive reasoning, physicians gather individual pieces of information through a means-end analysis of the patient history, physical examination, and diagnostic study findings. They eventually make a general conclusion on the proper diagnosis and patient management, but these conclusions may be drawn without considering enough data. In deductive reasoning, physicians make decisions based on general knowledge and their mental models for illnesses, looking for evidence to refute the conclusion. Each approach is useful in different circumstances, with inductive reasoning being better for diagnosis and treatment and deductive reasoning being better at finding the root cause of disease (Shin, 2019).

Traditionally, clinical reasoning has focused on a hypothetico-deductive approach, in which the physician develops a hypothesis—the differential diagnosis—that is then either confirmed or falsified by the clinical data gathered. The diagnosis that best matches the findings is the answer by which the physician bases the management strategy. This approach, however, does not always lead to the correct diagnosis and treatment because it does not account for biases, environmental influences, and abnormal findings that do not fit the clinical picture. Would it be better to use an inductive reasoning process to create and test intermediate hypotheses that focus on pathophysiological factors, or the cause of the disease process, without initially worrying about the diagnosis? The answer is probably that clinical reasoning requires a combination of hypothetico-deductive reasoning, probabilistic thinking from epidemiological studies and evidence-based research, and inductive reasoning (Richards et, 2020).

To complicate matters further, the dual systems approach suggests that two systems are at work. System 1 governs intuitive thinking that is fast, automatic, and nonconscious, while System 2 governs more complex problem-solving through slower, more reflective, conscious, and controlled thinking. Heuristics, or “rules of thumb,” also operate in medicine and influence System 1 thinking. System 1 thinking is often augmented through routine training, as physicians learn patterns or illness scripts that include typical case scenarios for various disease processes. As physicians gain experience, these mental models are reinforced, and new information is integrated into the corresponding schema. Physicians use System 1 thinking to address everyday clinical issues. By contrast, System 2 thinking must be taught, so physicians stop to think about how to solve problems (Richards et, 2020). System 2 encourages the comparison of illness scripts to arrive at more deliberate diagnoses, and more experienced physicians can better shift between these two ways of thinking (Thampy et al., 2019). Although little research shows how System 2 thinking applies to medicine, it seems that teaching more deliberate inductive reasoning that focuses on processes and intermediate hypotheses would reduce errors in judgment. Intermediate hypotheses allow physicians to incorporate new information, rather than throwing it out or ignoring it, and to change the hypothesis if needed (Richards et, 2020).

Cognitive biases must also be better addressed through medical education so that physicians become aware of the internal and external influences on their decision-making processes. Cognitive bias is a leading cause of errors in medical malpractice claims, including the misdiagnosis and delayed diagnosis of patients. Some of the major cognitive biases that affect clinical reasoning include overconfidence bias, such as being overconfident in the diagnosis; anchoring bias, such as focusing on one early aspect of a case or finding in a diagnostic workup; availability bias, such as making a diagnosis that more readily comes to mind; the framing effect, such as organizing a case in a way that influences the diagnosis; premature closure, such as determining a diagnosis without all the pertinent data; search satisficing, such as stopping the search for other diagnoses once one seems to fit; diagnostic momentum, such as not questioning previous diagnoses made by other physicians; and confirmation bias, such as looking for evidence to support rather than falsify a diagnosis. Physicians must be taught strategies to overcome this thinking, to make the effort to monitor their own thinking through metacognition. Using inductive reasoning for all cases, including typical cases, is crucial in overcoming these cognitive biases. Physicians must be encouraged to discuss uncertainty with colleagues, to consider alternatives, and to share their mental models to create a more comprehensive clinical picture (Richards et, 2020; Thampy et al., 2019).

Cognitive load theory is also relevant to how novices and experts process new material (Bailey & Rutledge, 2017). When novices are faced with challenging circumstances, they tend to abandon process-oriented thinking (Cleary et al., 2020). Learners with lower levels of prior knowledge and experience are more susceptible to cognitive overload (Kiesewetter et al., 2020).

Teaching Clinical Reasoning Skills

By nurturing clinical reasoning skills, educators can help to engender the lifelong-learning skills necessary to maintain successful medical careers. Self-directed learning becomes essential as learners leave the structured educational environment and begin their medical practices. Continuing medical education should therefore help learners to obtain and maintain critical thinking skills as well. As new technologies and standards of care influence best practices, physicians must be able to incorporate these changes into their knowledge base and skillset to become better doctors throughout their careers (Richards et, 2020).

Before clinical reasoning skills can be emphasized, novices must learn the illness scripts for typical cases. Script theory surmises that physicians create and store mental models for symptoms, findings, and treatments of various illnesses (Thampy et al., 2019). Developing pattern recognition skills is an essential first step in preventing cognitive overload. Creating concept maps that explain how information is linked between patient complaints, clinical presentations, and pathophysiological processes can help learners develop the framework needed to build effective clinical reasoning skills (Richards et, 2020). Building this base of knowledge can be achieved through studying flash cards, illness script summaries, and whole case scenarios.

Medical educators can encourage clinical reasoning skills in a variety of ways. One simple strategy is to ask “Why…” or “How…” questions that require learners to explain their thinking and justify their answers. Another strategy is the one-minute preceptor technique, the microskills model of clinical teaching, which focuses on understanding the learner’s ability to think through a case. This technique involves having the learner commit early to a diagnosis, workup, and management strategy; the learner then provides supportive evidence as to why and how they arrived at their conclusion; from this information, the educator fills in any knowledge gaps and addresses any misconceptions; and then the educator asks what was done right to reinforce those behaviors and to describe anything that could have been done better to correct any mistakes. This technique helps evaluate the learner’s understanding of the underlying disease processes and allows for prompt feedback from clinical instructors (Richards et, 2020).

Another similar strategy—the SNAPPS model—encourages learners to summarize patient cases, narrow down differential diagnoses, analyze cases, probe for more information, plan diagnostics and management, and study unclear information from the clinical interaction. The main difference between the one-minute preceptor technique and the SNAPPS model is that the SNAPPS model is primarily driven by the learner, which further promotes autonomy and responsibility. Teaching this way of thinking can result in better differential diagnoses, better justifications for decisions, and increased self-directed learning (Richards et, 2020).

Moreover, self-reflective writing and small-group discussions encourage empathy and reasoning abilities by allowing learners to identify errors in judgment and discuss biases that contribute to errors. Feedback from educators that encourages alternative behaviors is essential to avoiding these pitfalls. Clinical instructors must be willing to acknowledge their own cognitive biases, admit uncertainties, and demonstrate their own thinking processes in order to model this behavior to learners (Richards et, 2020). Educators must work to impart pattern recognition strategies with their learners and to compare and contrast pertinent information that facilitates meaningful discussion (Lisperguer Soto et al., 2020). All of these strategies, however, need to be studied further to determine their long-term effectiveness (Richards et, 2020).

Simulations, Virtual Patients, and Other Online-Learning Strategies

Studies suggest that online-learning strategies, such as interactive case-based learning, video or immersive simulations, and virtual patients, provide a modest improvement in clinical reasoning skills, but no approach has proven to be clearly superior. Most formats are at least as effective as traditional methods (e.g., lectures, textbooks) in achieving knowledge transfer (Cleary et al., 2020; Watari et al., 2020). However, online learning provides more opportunities to study unique case scenarios and to integrate knowledge into existing mental models. Essentially, technology provides a safe and effective way to explore an increased number of cases, which can remove barriers to knowledge acquisition, and to practice procedural skills (Guze, 2015). Online case scenarios provide physicians with the practice necessary to update and upgrade their illness scripts and schemas (Manesh & Dhaliwal, 2017), and because learning effective clinical reasoning requires an extensive amount of practice to achieve proficiency, online-learning strategies can effectively supplement real patient encounters (Cleary et al., 2020).

Virtual patients, which simulate real-life cases, require learners to make decisions to gain feedback. If these scenarios are well designed, they require leaners to interpret findings, generate hypotheses, and make clinically relevant decisions, and they also include information that can distract from the correct diagnosis, such as misleading findings and extraneous information. If they authentically mimic real clinical environments, then abstract knowledge can be applied to problem-solving activities. These simulations often recreate the sequence of events in patient interactions, note how other experts would handle the situation, and in turn, allow learners to compare and contrast how they would handle various diagnoses and treatment plans. They often include interactions and questions along the way (Cleary et al., 2020; Kiesewetter et al., 2020; Manesh & Dhaliwal, 2017). In addition to clinical reasoning skills, virtual patient scenarios can improve procedural skills, empathy, and team-building skills (Watari et al., 2020).

Studies of case-based learning suggest that prior knowledge and experience influence effectiveness, as learners with higher levels of prior knowledge gain more from these simulated encounters than those with lower levels of prior knowledge. Learners with lower levels of prior knowledge experience greater cognitive load, even in virtual environments. Experts may gain more from interactions at each stage of the patient encounter because they already have functional illness scripts that can help them successfully navigate cases, while novices have no such foundational knowledge to determine what information is important. For novices to gain more from virtual patient encounters, other instructional methods must first be enacted to enhance illness scripts and schema and reduce cognitive overload (Kiesewetter et al., 2020).

When comparing online video simulations to live simulations, live simulations are more effective in teaching clinical reasoning skills. Live simulations provide an opportunity for more adaptive thinking, and participants more readily demonstrate higher levels of cognitive thinking, including data analysis skills. By contrast, physicians often miss key information when watching video simulations. If post-simulation reflection exercises are included after live scenarios, they provide participants with the opportunity to consider their actions and scrutinize their assumptions (Cleary et al., 2020). While video simulations can help novices practice their clinical skills, experts may benefit more from simulated experiences. Online simulations provide more autonomy, require a more extensive knowledge base, and encourage more complex conceptualization and analysis of situations (Cleary et al., 2020). Moreover, the quality of the simulations has a large impact on effectiveness. High-fidelity simulations that use lifelike manikins to mirror real-life circumstances in live scenarios have proven to be educationally effective. Using 3D anatomy and physiology models has proven effective in enhancing learning (Guze, 2015). Virtual reality simulations are superior to online interactive case scenarios.

In general, technology use in medical education seeks to facilitate foundational knowledge acquisition, improve decision-making and critical thinking skills, enhance various perceptions, improve skills training, provide practice for unusual cases and events, and encourage team training. Technology use also has a positive impact on learner satisfaction, engagement, and knowledge acquisition. Online-learning strategies provide learners with safe, controlled environments; instruction that can be tailored to individual needs; and opportunities to repeat case scenarios until the learning goals are achieved (Guze, 2015).

Conclusions and Future Research

Other issues to be explored include how to assess the successful acquisition of clinical reasoning skills and establish best practices for gauging mastery of critical thinking processes as well as overcoming cognitive biases. Further research is also needed to determine which strategies are most effective in teaching novices versus experts so that a continuum of learning can be offered throughout undergraduate education, medical school, residencies, and fellowships, and beyond the structured learning environment. Because practicing medicine requires a commitment to lifelong learning, research into what constitutes effective continuing medical education is also warranted. Online-learning technologies are the cornerstone of continuing medical education, such that there are many opportunities to bring innovative solutions to the core goal of how to maintain cognitive excellence over a physician’s entire medical career.

Online-learning technologies have earned a lasting place in medical education, and their relevancy in improving clinical reasoning skills will only grow as technology and cognitive research advances. Immersive virtual reality environments will allow learners to explore without fear of life-threatening consequences to patients and will surely augment the traditional in-person clinical training models. As higher-quality interactions are created and various types of virtual patient cases are studied, the best educational approach may be identified, but in all likelihood, a combination of approaches is necessary to supplement traditional learning strategies.

References

Bailey, J. H., & Rutledge B. (2017, February 1). The educational psychology of clinical training. The American Journal of the Medical Sciences, 353(2): 96–100. https://doi.org/10.1016/j.amjms.2016.10.011

Cleary, T. J., Battista, A., Konopasky, A., Ramani, D., Durning, S. J., & Artino, A. R., Jr. (2020). Effects of live and video simulation on clinical reasoning performance and reflection. Advances in Simulation (London, England), 5(17):1–10. https://doi.org/10.1186/s41077-020-00133-1

Guze, P. A. (2015). Using technology to meet the challenges of medical education. Transactions of the American Clinical and Climatological Association, 126: 260–270.

Kiesewetter, J., Sailer, M., Jung, V. M., Schönberger, R., Bauer, E., Zottmann, J. M., Hege, I., Zimmermann, H., Fischer, F., & Fischer, M. R. (2020, March 14). Learning clinical reasoning: how virtual patient case format and prior knowledge interact. BMC Medical Education, 20(73): 1–10. https://doi.org/10.1186/s12909-020-1987-y

Lisperguer Soto, S. E., Soledad Calvo, M., Urrejola Contreras, G. P., Pérez Lizama, M. A. (2020). Clinical reasoning training based on the analysis of clinical case using a virtual environment. Educación Médica. https://doi.org/10.1016/j.edumed.2020.08.002

Manesh, R., & Dhaliwal, G. (2018). Digital tools to enhance clinical reasoning. The Medical Clinics of North America, 102(3): 559–565. https://doi.org/10.1016/j.mcna.2017.12.015

Richards, J. B., Hayes, M. M., & Schwartzstein, R. M. (2020, October 1). Teaching clinical reasoning and critical thinking: From cognitive theory to practical application. CHEST Journal, 158(4):1617–1628. https://doi.org/10.1016/j.chest.2020.05.525

Shin, H. S. (2019, December). Reasoning processes in clinical reasoning: From the perspective of cognitive psychology. Korean Journal of Medical Education, 31(4): 299–308. https://doi.org/10.3946/kjme.2019.140

Thampy, H., Willert, E., & Ramani, S. (2019, April 25). Assessing clinical reasoning: Targeting the higher levels of the pyramid. Journal of General Internal Medicine, 34(8):1631–1636. https://doi.org/10.1007/s11606-019-04953-4

Watari, T., Tokuda, Y., Owada, M., & Onigata, K. (2020). The utility of virtual patient simulations for clinical reasoning education. International Journal of Environmental Research and Public Health, 17(15): 5325: 1–9. https://doi.org/10.3390/ijerph17155325

---