The statistical power of zero.


Today I was going through the most recent literature on sepsis. I subscribe to Medscape emails for Surgery, Critical Care, Anesthesia, and Ob/Gyn; basically all of the disciplines that actually contribute to the type of medicine I would ultimately like to be in charge of. (A combination of surgery and intensive care for emergency patients if you did not see the connections).

The Medscape articles are usually not very good compared to many journals. They are more attention grabbing headlines, but as I was reading this one, I was actually more intrigued with what it didn’t say than what it did.

The study was in all respects not worth publishing or reading. 23 patients over more than 5 years, retrospective review. That might as well be an anecdote.
The conclusion was of course the obligatory “This might help but more study is needed.”

But as I was thinking this article was moments of my life I will never get back, I realized what was not written. There was a sound theory behind the treatment. I am not sure it was complete or accurate based on the limited info in the study. It focused more on coagulation than inflammation. As a retrospective review, they tried this on individuals they thought it might help.

If you read between the lines what you come up with is: “Our study sucks, it proves nothing, adds nothing, but demonstrates that some individuals are sometimes helped by theoretical treatments.”

But instead of latching onto that idea and writing it up as an opinion piece, the authors desperately tried to use the scientific method to give credibility to that argument (very ineffectively I might add).

As I keep professing, I know a thing or two about science. I know a thing or two about publishing studies. I may even know a little about medicine!

The medical community really needs to stop trying to fit the square peg in the round hole. We need to recognize we deal with diverse biological organisms, not simply mathematical reproducible physical properties. There are hundreds if not thousands of variables on a given patient. We need to start telling non-medical people this, not trying to deceitfully disguise it is our best scientific wild-assed guess (SWAG) to help people who do not fit into epidemiological guidelines. We need to finally realize that sometimes we might not know why a treatment worked for a given patient, but at the time it seemed like a good idea and it somehow worked.

This may sound like a call for “non-scientific” medicine and I guess in a way it is. Not because I am anti-science or something crazy like that, but because of the reality that we simply cannot produce good science on critically ill patients. In many cases, such patients will die or be permanently disabled under the best circumstances. In order to not write them off as “acceptable losses” we will have to attempt “heroic” measures from time to time. Not always, but perhaps more than we are accepting?

Otherwise, we are simply accepting the joke “the operation was successful, but the patient died.”

Maybe we really should start acknowledging and accepting that there is not going to be a preponderance of evidence for every patient or condition and start having meaningful discussions on practical ways to handle those situations rather than just writing them off as “unscientific”?

Physician Gestalt is not evil. It is and has been an obvious observation for a long time. Why do we continue to treat it as a shameful topic? When has not talking about something ever produced a desired or even positive outcome relating to it? That is exactly like trying to prevent teen pregnancy by not talking about sex.

Sometimes I think medicine is the only profession that constantly devalues its own expertise. That seems self-defeating to me. Why would anyone respect or pay you for being an expert when you constantly claim it is somehow wrong in medicine to possess knowledge or abilities that cannot be replicated by anyone able to read a scientific paper? If anyone could do it, why would we need doctors? If all doctors were equal, why is there any competition among them?

P.S. I didn’t forget to link the study I was talking about, I chose not to because it is a waste of time to read.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s