Learning from mistakes: Defining case descriptions to improve clinical reasoning education

Clinical reasoning is one of the core tasks in the medical profession. Complex decisions are made in an high stakes environment in which incorrect decisions can lead to severe consequences for patients. In General Practice (GP), many patients are seen each day and, by necessity, many complex decisions about diagnoses are made. Clinical reasoning is therefore an important part of the GP vocational training. A significant amount of time is dedicated to practicing with clinical cases to learn and improve clinical reasoning. Previous research has focused on how clinical reasoning takes place, how expertise develops and on testing educational methods. While some methods may be more effective than others, there is increasing evidence that the most important factor in determining educational effectiveness is the content of the cases, from which the students learn. Expert reasoning develops through focused, deliberate practice aimed at building and organizing accessible knowledge structures to make clinical decisions. Ideally, case content should reflect clinical information that trainees need to know but have not yet been exposed to sufficiently and thus do not have mastery. Given their nature, malpractice cases often reflect such knowledge gaps. An important but underutilized way to learn from malpractice claim cases is by including them in clinical reasoning education so trainees can learn from the mistakes of their peers. Learning from mistakes is considered to be important in healthcare. In most hospitals, errors must be reported through an incident reporting system so that others can learn from them and systematic problems identified and rectified. As of this year, GPs in the Netherlands are also required to report their errors. What makes learning from errors special? Is it just that they reflect domain-specific content that is generally not well understood and thus error-prone, or is there more to it? Research shows that an emotional response could contribute to better memory of a situation. It is not clear whether hearing about a case in which an error occurred contributes to remembering that specific case with its context, or whether it contributes to learning by better understanding the disease and filling existing knowledge gaps. The fundamental question is: Are errors particularly suitable for learning or can we just as well learn from successes? While trainees have educational sessions dedicated to clinical reasoning, most training occurs during their work in clinical practice. Evaluation of trainees also occurs by discussing and evaluating situations in which diagnostic errors occurred in practice. However, there are some disadvantages to relying solely on cases from clinical work to teach and assess trainees’ clinical reasoning. Specifically, the supervisors are aware of the outcome of a case and this can result in hindsight bias. Hindsight bias is the tendency to overestimate the ability to have predicted an outcome after learning about the outcome. The outcome of the diagnostic process may not always reflect the process itself. Therefore, teachers may be more effective in facilitating learning if they discuss the cases by going over the case prospectively rather than only with the benefit of hindsight. The main aim of this research proposal is to determine if learning from malpractice can be effectively used as learning material in the GP vocational training. In Study 1 we will identify important knowledge gaps of trainees by analyzing GPs’ malpractice claims. We will make use of the malpractice claim database of the VvAA, the largest liability insurance company of GPs in the Netherlands. Secondly, two experimental studies will be conducted to determine whether knowing that a clinical case resulted in a malpractice claim facilitates learning (Study 2) and to examine how the role of information processing by which information about a case (i.e. patient specific or clinical information) is better remembered (Study 3). We will select those cases from the VvAA database that reflect common knowledge gaps identified in Study 1. Lastly in Study 4, we will study the influence of hindsight bias (i.e. how knowing the outcome alters the evaluation of the process) in assessment of vocational training. Specifically, we will investigate to what extent knowing the outcome affects the supervisor’s judgment of the trainees’ diagnostic performance. With the cases that are used for the studies, we will develop a database of clinical cases that are effective learning materials for clinical reasoning education. Importantly, the database will contain a set of clinical cases that reflect important knowledge gaps of trainees which can be used in future clinical reasoning education in the GP vocational training. The proposed studies will reveal a novel evidence-based way of learning from mistakes in an educational setting and contribute to improving diagnostic accuracy in general practice.