Decisions and inequalities

I have two ideas I’ve been thinking a lot about recently which I have trouble reconciling. First, I’ve come to think that scientific evidence should always be produced and kept on a continuous scale with the only exception being when a categorical decision is required. This means no use of language that describes categorizes the outcome of a study the way statistical significance does. We should only categorize the outcome of a study when we are forced to such as when we are forced to make decisions. Through this line of thinking, I’ve come to think that, in ideal world, we should view all research through the lens of decision-making and cost-effectiveness. And by ideal world, I mean where we had access to certain pieces of information about costs and benefits that we don’t always have access to in our non-ideal world. This way of thinking shifts research away from answering questions like, “is this true?” toward answering questions like, “what decision should I make?” There are a whole host of issues with this that I’m glossing over here but I think there’s a lot of benefit for be gained from recognizing that at the end of the day, our research serves to make decisions.

But I’m also interested in studying how health inequalities can be reduced. And it doesn’t take much imagination to end up in places where you start thinking about reduction in gini coefficient or some other measure of inequality per euro. Or taking this even further, a measure that combines the effectiveness in health inequality reduction with the general health benefit and dividing all that by the cost. Health inequalities are almost always the result of some injustice and to put a price on fixing it seems absurd. Would we prioritize health inequalities that were the cheapest to address? That also seems absurd. Surely we’d have to incorporate some measure of how unjust the origin of different inequalities is because we’d like to reduce the most unjust inequalities first. If we had a measure of injustice, you could even get into territory where you’re thinking about the most cost-effective injustice-reducing interventions. I’m squirming just thinking about this.

Hopefully I’ll come back to this post in the future because I realize my ideas are not well laid out. And also, I’ve written some ideas that make me uncomfortable.

Jeremy A. Labrecque
Postdoctoral fellow in epidemiology and causal inference

My research is on how we know what we know.