Incorporated into the preclerkship medical school curriculum is a weekly Facilitated Small Group Learning (FSGL) session. I imagine there are similar curricular components at other medical schools, so I hope this post might be relevant to more than just the Queen’s Med community.
Every Wednesday, groups of six students meet with a physician tutor to work through a case. More often than not, the case is related to topics we will learn about in lecture and through other routes of education delivery. In addition to the important task of bonding over food (previously discussed here), I understand the idea is to help us develop clinical reasoning skills and to begin considering information in a more clinical context. Though we always seem to make our way through cases and even manage to reach the right conclusions, we rarely (if ever) explicitly discuss our approach to clinical reasoning. I know the reasoning is happening because we get from Point A to Point B, but I often leave the room not quite knowing how we got there and not quite sure that I could do it again.
Perhaps I am hyperaware about a lack of explicit discussion around the process because of recent time spent completing an online clinical problem solving course (reviewed here) or maybe it is because I did my undergrad degree at Mac. It might also be that I really only noticed it for the first time this week because I struggle with neurology. That struggle more clearly highlighted a few of the process issues.
This post aims to do two things: 1) Provide a satirical, yet honest, overview of how many FSGL meetings end up 2) Offer a few of my ideas for ways to optimize FSGL time based on experiences with the Clinical Problem Solving course. In a future post I will demonstrate with a real FSGL case how these suggestions could be easily incorporated.
Disclaimer: I enjoy FSGL, I have learned a lot in FSGL and I have had very positive experiences with each of my FSGL groups and all of my FSGL tutors. This post simply presents my ideas about how I think we could use the time to learn more efficiently and effectively. More than anything it is meant to open the conversation about how medical students at Queen’s are and could be learning. The post is based on my three terms of experience with FSGL. Each students’ experience is different and is shaped by specific groups and tutors. I hope that students, tutors and educators can use this as a jumping off point to further build on an already strong aspect of our curriculum.
1) How many FSGL meetings play out (admittedly exaggerated so that you catch the drift):
- We read the case and underline what we think are important details. There is lots of underlining because we don’t want to miss something that might be important. The fact that she is wearing pink shoes might just be important.
- We are then prompted to generate a broad differential diagnosis. This is usually an extensive list of diseases, some of which we know and understand and some of which we just know the names of. Thanks to Dr. House, the list inevitably includes lupus. We often compare diseases on the list we have generated to the patient’s presentation but the process of doing so is rarely structured.
- We are asked to generate a list of tests to order. This is usually an extensive list of tests, some of which we know and understand and some of which we just know the names of. Thanks to a certain hematologist, the list inevitably includes a CBC. If we are lucky, the list is loosely driven by our differential. More often than not, it isn’t. The process of relating our investigations to our differential diagnosis is rarely structured.
- In the next session we receive more information about the case which acts to confirm or deny our suspicions. If it it looks like we were right, we move on quickly; if it looks like we were wrong, we move on quickly.
2) My thoughts on how we could optimize each of those steps to use FSGL as an even better tool for fostering the development of clinical reasoning?
- After reading the case create a patient illness script: spend time processing the case. Underlining alone does not equal processing. Go through the motions of generating a problem list for the patient. Translate that
problem list into medical terminology, this is so called problem processing (i.e. temp 38.5 with sweats and chills = febrile or symptoms for 2 days = acute). The purpose of problem processing is to put presentations that are described by patients into the same language that we store information about diseases, which makes a comparison of the two much easier. After generating the processed problem list it should be easy to come up with the patient illness script that consists of epidemiology (who is the patient and what her risk factors i.e. age, exposures), time course (what is the time course of her presentation i.e. hyperacute, acute, subacute, chronic), syndrome statement (what are the features of her presentation on history and physical) and other history (what other medical history does the patient have). This becomes a very useful and tidy comparator once you have generated a differential diagnosis . I imagine having a group member present this patient illness script as an oral report would be a good way to see if everyone agrees on the picture.
- Just as before, generate a broad differential diagnosis but we should ensure that we know information about each of the diseases that ends up on the differential. Organizing this information into disease illness scripts, which include information about epidemiology (who gets the disease, what important risk factors are there), time course (over what period of time and in what patterns does the disease usually present), syndrome statement (what are the key features of the disease on history, physical and testing), mechanism (
pathophysiology of the disease) will come in handy. If we don’t know this information about one of the diseases on the differential, we need to find it. Organizing information in this way makes it easy to compare with the patient illness script.
- Compare these disease illness scripts with the patient illness script to generate a tiered differential diagnosis. This requires making a commitment to the diseases we think are most likely and those we think are least likely. The system I am familiar with is Tier 1 (very likely), Tier 2 (less likely), Tier 3 (unlikely) and Tier 1e (unlikely but could kill you by tomorrow if missed).
- Use this tiered differential to drive investigations. What tests will help us differentiate between the Tier 1 diagnoses on the list. What investigations can help us rule out the big, bad Tier 1e diagnosis. Why are we ordering each test? How will it inform our team’s decision making? These are questions that we could be asking more often and I expect would be easier with a more explicit visual representation of where each diagnosis sits in our mind. If necessary we could assign someone from the team to be the ‘devil’s advocate’/question asker each session to ensure that these questions do in fact get addressed.
- When presented with new information take time to appreciate where we went wrong and where we went right. Was our thought process sound? Did we get lucky? What biases affected our reasoning? What did we miss? What would we do differently next time?
The bottom line
So what does this all really mean? This means spending a bit of extra time and effort in being more explicit with our reasoning. For some cases this will seem unnecessary. For those cases the answers seem to stick out like a sore thumb but that is only because we recognize the pattern. As a newbie to medicine, I don’t think that pattern recognition is a safe bet. I know that I have not seen enough cases to feel good using it as my go to tool in the clinical reasoning toolbox. This week’s neurology case highlighted that to me. Our group didn’t recognize the pattern and we were at a bit of a loss for what to do next. This approach might have helped us structure our approach to an otherwise overwhelming case. I’ll explore exactly how, with a real FSGL case, in a future post.
So moving forward, I challenge my peers to try this approach whether it be in an FSGL case, a small group learning case or when seeing a patient on the wards. I also encourage tutors and students to challenge each other more openly about their reasoning processes. Let’s make the black box less of a black box. The question “why?” goes a long way in exploring what and how we are thinking. As early learners we think very differently than experts so I imagine that tutors may be surprised at some of the answers that we generate to that simple question.
In some awesome cross-country communication and in a reply to Dr. Sanfillipo’s recent blog post, U of S medical student Danica Kindrachuck (@want2BeMD), challenged medical schools to include a better background in cognitive science. FSGL seems like the natural place to start.
Ashley Thomson (one of my FSGL group members who reviewed this post), Dr. Michelle Gibson (FSGL coordinator who also reviewed this post) and Dr. Heather Murray have already started to generate a few interesting ways to incorporate more explicit reasoning into the course. Their ideas seem to centre around educating students and tutors about reasoning processes from the get go. What do you think would be effective? Do you think more focus on explicit reasoning would be helpful?
For more information on clinical reasoning, see these great resources- many of which were referenced in my previous review of the Clinical Problem Solving Course.
- “Teaching Clinical Reasoning” by Michelle Lin (@M_Lin) at ALIEM
- Blogging to promote clinical reasoning and metacognition in physiotherapy fieldwork programs by Melissa Tan, Richard Ladyshewsky and Peter Gardner
- Lauren Westafer’s (@LWestafer) great medical student thoughts on “Thinking About Thinking”and “Metacognition for the Pragmatist”
- “Teaching Clinical Reasoning” by Nadim Lalani (@ERMentor)
- “Thinking about teaching thinking” by Robert Centor (@medrants)
- Judith Bowen’s 2006 NEJM article on “Educational Strategies to Promote Clinical Reasoning”
If you enjoyed this post or want to share thoughts about explicit clinical reasoning with friends or colleagues please subscribe, tweet, facebook and email away! More than anything I am interested in what you think on the topic. Please leave your comments! An awesome prize will go to the first student who leaves a comment about their experience with efforts to explicitly explore clinical reasoning (alliteration was unintentional). Don’t think the prize will be worth it? Just ask Rebecca who won the previous logo contest about her delicious Vietnamese lunch!!!
This post was peer reviewed by my classmate and FSGL group member Ashley Thomson and professor/FSGL coordinator Dr. Michelle Gibson (@MCG_MedEd). Many thanks to them for helping with this post.
Latest posts by Eve Purdy (see all)
- Early Goal-Directed Dating: applied lessons from evidence in sepsis - December 18, 2014
- Evidence-Based Dating: the Canadian ITAD (is this a date?) decision tool - December 1, 2014
- SHPOS…never heard of it. - November 24, 2014