Discussion about this post

User's avatar
Diana Alex's avatar

These arguments would work in a world dominated only by reason, where all patients and all doctors understand completely the evidence presented and the risk of each next step in the treatment plan. If you follow this, then all current evidence should be presented for all medical decisions - not only those for screening. It would mean every time a doctor recommends Ibuprofen for muscle aches or fever he/she should explain in detail that 1 in about 100,000 people taking ibuprofen can develop acute renal failure, and list of all the other complications that could appear with their associated risk. In the spirit of justice, it should also be included in all our daily activities - because health is a whole person and entire day thing. For example a reminder every time you start your car that the risk of death is about 1 in 100, and the risk of an accident in the next 1,000 miles is 1 in 366 people, and then all the relevant statistics for how much money the average car accident costs. All of this is evidence is reasonable and needed to make sure people make an adequate and informed decision about driving to work that day, weighing the benefits and risks involved.

The problem with a purely rational view of life and medicine is that we are not all rational beings. Even Plato recognized that there’s spirit and appetite also, apart from reason. The reality is that no matter how smart someone is, or how much knowledge they have, the perception of risk for their own person is intrinsically connected to emotion, experiences, values and beliefs. None of these are stagnant - because unlike the absolute numeric risk of an adverse event - but are changing constantly, sometimes even from one day to the next. The numbers on PSA screening might mean something for me today, but tomorrow, after my 50 year old brother in law got diagnosed with a very aggressive, thankfully still incapsulated, prostate cancer based on a random screening PSA - these numbers will mean something different, because I’ve seen and felt what being that 1 in 366 is.

The other problem is that your average patient doesn’t want to make all healthcare decisions themselves. Part of a good doctor patient relationship is the partial transference of some autonomy that comes with trusting the doctor, and allowing him/her to make some decisions, while sharing some others. This is a crucial component of medicine, that is implicitly understood by all. Me as a patient (even very well informed one as a physician myself) could not bear the weight of being responsible for the entirety of every single medical decision based on EBM, without having a trusted doctor who makes recommendations in a more qualitative way.

In the end, the core of shared decision making is not to communicate rational facts, but to do it in the context of the values and preferences of the patient. That context, the relationship with the doctor and human connection that allows SDM to occur are important and, in my opinion, is one of the reasons physicians won’t be replaced by AI, at least not for the foreseeable future.

Mary Braun Bates, MD's avatar

I'll give your script a try next time I have a PSA discussion and report back.

My usual prostate cancer spiel starts by explaining the idea of screening. I'd guess somewhere around 2/3 of my eligible patients who get physicals opt for a PSA which is quite high, I see, compared to national averages a quick search produces. Hmm. I wonder if I'm overselling it.

3 more comments...

No posts

Ready for more?