Generally, when we are considering a new medical treatment, our first question is, “Does it work?” But that question is usually not enough to get a good sense of the potential benefits and harms of an intervention. In a perspective piece in the The Washington Post, Dr. Daniel Morgan, professor of epidemiology, public health and infectious diseases at the University of Maryland School of Medicine, explains why shared decision making requires us to ask, “How likely is this treatment to work?” and why the answer to that question can be tricky.
“If doctors don’t understand how likely — or unlikely — it is that a treatment will help, they can’t give patients the best advice for their care.”
Dr. Daniel Morgan
What is the difference between these two questions? The example of remdesivir can lend some insight. The antiviral drug recently approved to treat Covid-19 has had mixed evidence, but in one study was found to reduce the risk of death in hospitalized Covid-19 patients from 15.2% to 11.4% after 30 days. According to that study, the likelihood that it works for any one patient is small. This means that clinicians and patients or family members have to discuss the potential benefits in context of side effects and cost.
As Morgan writes, most medical treatments fall into this gray area of having a known but unlikely benefit for patients. This is especially true when it comes to preventive treatments for patients without symptoms.
For example, for patients without existing heart disease, taking a statin reduces the risk of having a heart attack or stroke, but only by 0.4%. On a population level, that adds up to a lot of heart attacks prevented over the years. But on an individual level — when patients want to know if this medication is “right for me”– the small benefit may not be seen as worth it, especially for people who are already taking many medications, or those who have experienced disruptive side effects from statins.
“Our job isn’t only to cure people; it’s also to help them make it through when there is no easy cure.”
Dr. Daniel Morgan
It might make sense that patients who aren’t trained in medicine may overestimate the benefits of medications. But surprisingly, doctors are also overly optimistic when it comes to the potential benefits of treatments — why? Morgan cites several potential reasons: There is very little training for doctors (if any) in statistics and probability, which are skills needed to evaluate benefits and harms of medications; promotion from drug and device companies may influence clinician thinking; clinicians may fear getting sued for not doing enough; and performance metrics often use “one size fits all” guidelines, even when a certain treatment doesn’t benefit all patients.
For better patient-centered care and less overuse, we have to reverse some of these trends. Providing better clinician training in basic probability, removing performance metrics that incentivize overuse, and prioritizing spending time with patients and having shared decision making conversations would go a long way, Morgan writes.
He also calls on doctors to embrace the gray area rather than avoiding tough conversations. Acknowledging the limits of medical treatments can be difficult for clinicians that view themselves as needing to be “invincible.” But doing this will help “foster a more holistic and meaningful sense of the doctor’s role,” Morgan writes. “Our job isn’t only to cure people; it’s also to help them make it through when there is no easy cure.”