One thing I keep thinking about from Monday's post is how quickly we compress a medication response into two words.
It worked.
So what did the medication actually treat?
Did it increase serotonin, which then somehow directly made the depressed mood go away?
Maybe. But when you say it that plainly, you can hear how much is being skipped over. And if that is the explanation, why did it not do the same thing for the next person with the same diagnosis on the same medication? Why did she barely respond? Why did somebody else feel worse?
That does not make serotonin irrelevant. It means "serotonin changed, therefore the depression improved" is not much of an explanation on its own.
Maybe the medication changed something that gave this particular person enough energy to start leaving the house again. Then they saw people more. They had more positive experiences. They felt less isolated. Getting out became easier. Their sleep changed. Their expectations changed. Their attention shifted.
Every one of those is itself a biological change.
Or maybe the medication did very little.
Maybe the day after they started it they got a phone call that took a major stressor off their plate. That is not biology. That is something that happened to them. But what happened next is. They had enough bandwidth to finally start the hobby they had been wanting to try. The hobby got them out of the house. They started interacting with people again. They had something to look forward to. They started sleeping better. Their mood improved.
Every one of those is a biological change too, and none of them came from the medication.
And if I see them six weeks later and they are doing much better, it would be very easy to write "responded well to sertraline."
Maybe they did.
But improvement following the medication does not tell me how much of the change came from the medication, what the medication changed if it did contribute, or how much came from everything else that started moving around the same time.
And sometimes it runs the other way. I see this a lot with OCD. Somebody notices a small or ambiguous change after starting a medication. They start monitoring for it. They start wondering what it means. The monitoring makes it more noticeable, the uncertainty makes it more distressing, and eventually they stop the medication.
At that point the clinically important effect may not be the original sensation at all. It may be everything that happened after they noticed it.
And this is where I think the way we read medication research matters. A medication response is an observation. The story we tell about what that response means is another step entirely.
Underneath all of this are two questions that are easy to run together.
Did the medication help?
And does whatever it acted on explain the disorder?
Those are different questions.
Sometimes we can be pretty sure about the first. Sometimes we cannot.
But a yes to the first is still not a yes to the second.
Think of a patient who had a meaningful response to a medication, good or bad. What changed first, and what happened after that?
Am I way off base here? Have any of you wondered about this?
When you read a study showing that changing some receptor or neurotransmitter improved symptoms, what do you actually take from it?
What about schizophrenia or bipolar disorder, where the biology seems a lot more obvious? Does this still hold?
Where do you think this way of thinking works, and where does it fall apart?