To most medical students, the process involved in becoming an expert problem solver seems like a big black box. I often question, and sometimes worry about, how will I go from medical student to competent physician.
So when mention of a Clinical Problem Solving Course offered by Coursera came across my twitter feed, I jumped at the opportunity to learn more about diagnostic reasoning. I enrolled in an effort to shed light on the type of problem solving that I might be faced with come clerkship but now venture that more explicit discussion of clinical reasoning is relevant to medical learners at any stage. This post describes my experience with the course.
The course is a MOOC (Massive Open Online Course), a concept explained nicely by Javier Benitez at ALIEM. It was taught by the energetic Dr. Catherine Lucey, Vice Dean of Education at UCSF. In each week of this 6-week-long course students are expected to review a series of lectures, apply skills learned in lectures in weekly assignments, and grade the work of 5 peers from the previous week. This results in a weekly time commitment of:
- 60-90 minutes total for lectures that are delivered in 15-20 minute chunks. The breakdown of material into smaller parcels made it easier to identify key themes and keep attention! You can also speed up the lectures too…I found 1.5x to be just right.
- 1-2 hours for assignments. The amount of time necessary would differ greatly based on level of expertise. We were most often comparing and contrasting 3 diseases using illness scripts (a concept that I will elaborate on in a future post). This process was quick with familiarity of a given condition but took much longer for diseases I knew nothing about. I found the assignments to be the best part of the course. They were designed in a way that really forced me to use new lessons about clinical reasoning from the lectures. Furthermore assignments allowed me to relearn information through a different contextual framework, solidifying knowledge that I had previously gained in medical school.
- 30-60 minutes to grade peer work. Using a rubric released after the deadline for submission, students were responsible for grading the work of 5 peers. The rubric was very clear and staff for the course were available to contact for any questions that came up. I found this peer assessment a useful process for a couple of reasons. First, seeing the rubric helped me understand what I did well and what I didn’t quite get. Second, applying the rubric to peers’ work helped me further understand how others were applying the same lessons that we learned and either incorporate their approach into my own thinking or appreciate where they might have gone wrong and work to avoid similar mistakes. The actual mark received from these peer assessments was far less important to me than the process.
At the end of the day, it does work out to be a fairly big commitment but not an unmanageable one. The lectures, assignments and grading can be completed at anytime during the week. Since the lectures are divided into smaller pieces, the assignments can be edited and saved as you go and you can evaluate peers one at a time you could effectively work on the course in small chunks. You do not need to find a 4 hour block of uninterrupted time each week to fulfill the requirements of this course.
Can I just watch the lectures? There is nothing preventing you from watching the lectures and ignoring the assignments. This in fact, might be a good option for experts taking the course with the purpose of informing medical education. In fact, Dr. Heather Murray who reviewed this post ‘audited’ the course by just watching the lectures and says that’s the way she would do it again. That being said, for medical students taking the course I think that going through the motions of the assignments is key. It forced me to apply lessons from lectures then provided opportunity to get structured feedback on my ability to do so.
This is too big of a topic to cover all at once. I will give a brief outline of each week here then over the coming weeks and months discuss each topic in more depth in the context of clinical situations that I come across.
- Week 1: introduced differences in novice and expert clinical problem solving, introduced illness scripts (a way of organizing information about diseases into epidemiology, time-course, clinical presentation and mechanism of disease to facilitate comparison between different conditions)
- Week 2: Introduced and compared various clinical problem solving strategies (hypothesis testing, forward thinking and pattern recognition)
- Week 3: Introduced how to create the patient’s illness script, problem processing and how to create a tiered differential diagnosis for the patient’s presentation
- Week 4: Using the tiered differential to interpret diagnostic tests (understanding sensitivity/specificity of tests then using nomograms and heuristics to understand the importance of results)
- Week 5: Introduced common causes of diagnostic error (input error, analysis error, errors in System 1 vs System 2 thinking, cognitive biases) and a model for structuring case review
- Week 6: Showed how this model is used to write test questions, how thinking about diagnostic reasoning can make you a better test taker and tips for teaching the approach
What else was done well?
- The Technology. I have never used any online lecture tool that was as user friendly as this one. I could easily speed up, slow down and pause lectures. Embedded in the lectures were questions that made the videos interactive. Lecture notes and transcripts were also easily available.The interface was very user friendly and even looked nice. Furthermore, the support for the course on the discussion board was amazing. It was easy to ask questions and have them answered by peers or course support staff.
- Continuity. We were introduced to six patients’ stories (not diagnoses) at the beginning of the course. These patients had different demographics, different acuities of presentation, different comorbidities and different worries. During each of the modules we came back to one or some of these patients to use new clinical problem solving skills to approach their problems. Dr. Lucey used these patients to illustrate how to effectively collect and process information.
What could change?
- There are two separate tracks, signature and normal. One you pay for and one you don’t. I am not clear what the difference is. It seems that for the signature track you get a certificate of completion (it is verified because you must confirm that it is you who is submitting each assignment with typing signature and webcam shot) and for the non-signature track you don’t. In the spirit of FOAM it would be nice to see this course be completely free, though given the impressive course support I do understand why that may not be feasible.
- It would be great to get qualitative as well as quantitative feedback from peers. Peer grading did not leave any room for written feedback. On a couple of occasions I had wished that I could give more detailed feedback to peers (outside of numerical rubric) and often wished that I had been the recipient of some when I didn’t nail a particular concept.
- Dr. Murray has proposed an interesting suggestion that could help deal with the feedback issue noted above. I think the idea is brilliant and certainly would have improved my learning experience. Her idea is to have local breakout groups that meet to discuss the course and assignments. Ideally the group would consist of students and professors. Both might learn something through discussion. By having an expert review my assignments there would be opportunity for descriptive feedback- helpful for me the learner. It would also offer a unique view inside the learner’s head to see where mistakes in reasoning were made and if those mistakes are predictable or surprising- helpful for the educator. I think this would be a dynamic and worthwhile extension of the course.
- Anastasia, a classmate who took the course and who has reviewed this post, wisely suggested that it would have also been beneficial to have more practice clinical cases that allowed us to work all the way through the diagnostic process. I agree! Most of the time during assignments we were charged with the task of creating illness scripts for three predetermined conditions or ranking likelihood of diagnoses from a list provided for a clinical case. Though this provides a good foundation, the next step would be to get a clinical case –> develop a differential –> tier that differential –> order tests based on the differential –> come up with a final diagnosis –> get the answer –> reflect on if and how we went wrong in this process. The opportunity to bring it all together was lacking.
The Bottom Line
Obviously completing this course does not make me an expert in clinical problem solving, in many ways the effect has been quite the opposite. It has made me more aware of how far I have to go. The course made clear that the way my brain works now as an early learner, is very different from how it will work years of experience down the road and differs from the way those teaching me think. Right now my clinical reasoning process is slow and prone to error. This realization was humbling, though not particularly surprising.
The course didn’t just leave me at this realization but offered specific tools and tips to aid in development of my clinical reasoning abilities. I mentioned at the beginning of this post that the the transition from complete newbie to expert seemed like a big black box. Big black boxes are daunting. The course seemed to make that box a little bit more transparent, not easier, but more transparent. It helped me see what will be involved in the transition and what I can do as a learner to facilitate progress.
I’ve often heard the phrase, “don’t study hard, study smart”. I’m not sure that phrase is as relevant in medical school because I find that I need to do both to stay afloat. In terms of clinical reasoning, my evolution to competent physician will require a lot of hard work but I am trying to find ways to make that transition in informed and smart ways. The bottom line is this course has helped inform my learning.
If you have any interesting resources on the topic please share! A few other related posts are:
- MOOC’s as they relate to FOAM, “What is a MOOC” by Chris Nickson (@precordialthump)
- “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)
- I would be remiss not to cite Judith Bowen’s 2006 NEJM article on “Educational Strategies to Promote Clinical Reasoning” which was used as a reference many times throughout the course and certainly worth the read for students and their educators.
There are over 300 courses offered by Coursera. A few more exciting ones on the horizon are:
- Medical Neuroscience by Duke University (starts April 8th)
- Cardiac Arrest, Hypothermia and Resuscitation Science by U Penn (starts May 20th)
- An Introduction to Global Health by University of Copenhagen (starts September 2nd)
- There are thousands of others on topics ranging from Artificial Intelligence to the Fundamentals of Rehearsing Music Ensembles
If you enjoyed this post or want to share information about Clinical Problem Solving Course with friends or colleagues please subscribe, tweet, facebook and email away!
Eve Purdy (@purdy_eve)
This post was peer reviewed by my classmate Anastasia Prokubovskaya and professor Dr. Heather Murray (@HeatherM211) who both worked through the Clinical Problem Solving Course. Many thanks to them for helping with this article and enthusiastically engaging in discussion around the topic!