Clinical Experience: The Neglected Leg of the Three-Legged Stool of Evidence-Based Medicine

(and yes, I am still thinking about masking)

The definition of evidence-based medicine that I carry with me is “the process of integrating clinical experience and expertise with the best available evidence from systematic research.”

Mariana Barosa recently offered a nuanced update of this definition on Sensible Medicine that looked like this:

The messages are similar, there are three components that need to be considered and integrated when making evidence-based decisions. All three are important.

•       An understanding of biomedical science is critical for hypothesis generation, for weighing importance of unexpected research results, and for understanding clinical experiences.

•       Clinical experience is also necessary for hypothesis generation and for making decisions about the effectiveness of interventions suggested by research. Most importantly, we need to use our clinical experience because most of our practice decisions must be made in the absence of relevant data.

•       Systematic research, especially RCTs, are critical because we only truly know that an intervention works, and how well it works, when studies show it works, not because it should work.

Three recent situations highlighted what a reliable guide clinical experience can be.

Gabapentin for Cannabinoid Hyperemesis Syndrome

We are currently subjects of an enormous national observational trial of cannabis legalization. I have no moral qualms about this. However, from a strictly medical perspective, I do not think the “exposed group” is thriving. One endpoint is the number of cases of cannabinoid hyperemesis syndrome (CHS) that I care for on the inpatient wards these days. I only learned about this diagnosis five or so years ago.

If you look on UpToDate you will find that there are many potential treatments for CHS – a sure sign that none of them are perfect. A couple of years ago my team and I used gabapentin for this indication. I am not a huge gabapentin fan. I often use its history as a great example of pharma malfeasance. I have to say, however, that a recent personal experience made me a bit of a believer.

Our decision to use gabapentin for a case of CHS was based on desperation and a case report.  The patient got better. I know that experience does not count as evidence. Although I don’t know if the placebo effect was at work – the patient didn’t even know we had changed her regimen – her flare may just of been ending. Believe me I know this. VP and I wrote a whole chapter on this in Ending Medical Reversal.¹

But, over the last few years I have used gabapentin again and again for CHS, with different people, with different timing. It is not perfect, but it does seem to work.

I am not sure if there is bioplausibility here. I am still waiting for a good treatment trial. My patients have benefited from my clinical experience.

Human Metapneumovirus

I have been doing my job long enough that I have a good sense of the cause of people’s respiratory infections. I know when it is pneumonia, I know when it is influenza or RSV, I kind of know when it is COVID though that tends to be a bit of a diagnostic moving target. I also know when it is human metapneumovirus (hMPV). We often see this in the spring, when the other viruses are waning. People come in a bit later. They look miserable but have no findings suspicious for pneumonia. After an RVP or two, I know hMPV is circulating and I can use my clinical experience to give people a more accurate prognosis, though usually not the one they were hoping for. Symptoms sometimes last 2-4 weeks.

I think my diagnostic reasoning is pretty plausible. I can’t imagine we will ever have a study of this. My patients benefit from my clinical experience.

Masking in Clinic and COVID

I could absolutely quote something Dr. Rind wrote in a post a few weeks ago.

With regard to “the strain of Covid that circulated in the spring of 2020, having the provider wear a surgical mask and eye protection and the patient wear a surgical mask was highly effective in an outpatient setting. I started seeing patients in a dedicated Covid clinic in April 2020 and we had no access to N95 masks. My very first morning in the clinic, I saw eight patients. Seven of them tested positive for Covid and I suspect the eighth test was a false negative. Over seven months in that clinic, surrounded by coughing, Covid-infected patients,” I did not develop COVID.

I know the mask literature, believe me, I KNOW THE MASK LITERATURE. I know about viral particle size; I know the dynamics of mask filtering. I absolutely do not think that “mandatory masking” (which is really “suggested masking” since we don’t live in Winston Smith’s Oceania) works anywhere. That includes Bangladesh, UCSF, red states, or blue states.

I do know from my clinical experience that if you are seeing patients with a highly contagious, novel respiratory virus, and you are in the room alone with the patient, and you can assure that both doctor and patient are wearing a surgical mask properly, that you are protected, not perfectly protected, but protected.²

I do not think that masking should be part of the new “Universal Precautions.” I think that masks do a little good and a little harm, and I am willing sacrifice the little good for avoiding the little harm.³

There is some, though imperfect bioplausibility here. I hope we never have the opportunity to run a trial. Even if the opportunity presents itself, we would not be able to because no doctor would accept being randomized to no masking. My patients (and I) benefited from my clinical experience.

1. Chapter 2, for those running to your bookshelf to reread it.

2. I should add that me and my colleagues did this for nine months and many of us were doing weekly asymptomatic testing. Oh, and no, I don’t think that throwing children into a volcano guarantees a good harvest as someone suggested after Dr. Rind’s post. Though thank you for that comment. It totally cracked me up.

3. I was tempted to reference a major wedge issue in our national debate here but I’m not that stupid. I think the two sides of this debate are between those who think it is wrong to abandon the “little good” and those who think the little good doesn’t exist.