Decision analysis, as a theoretical construct, has certain implications for the roles in a medical decision, as will be described below.

As I have written in this post, a decision analysis is a way to model a medical decision as something that maximizes the expected utility of the patient. In an equation, expected utility is

where is the probability of the outcome, given an action, and an initial state, and where is the patient’s reward, or utility.

The equation for expected utility implies certain roles in a medical decision.

The provider’s role is to give the probabilities of outcomes, For example, if we are considering whether to undergo a surgery for cancer, this might describe the probability of post-operative remission or complication. These probabilities are based on clinical experience or the literature.

The patient’s role is to give the utility of the outcomes, Only they know how they will feel if they have, e.g., partial remission and a complication vs. complete remission and a complication.

The final role is that of “optimization.” Given the probabilities from the provider and the utility from the patient, one uses optimization to find the decision that maximizes the expected utility. In other words, one takes an of over

Hence, decision analysis suggests that those involved in a medical decision have specific roles.

Decision analysis is a theoretical framework. In the real world, patients and providers do not always adhere to the roles described above. For example, in the real world, the provider sometimes helps with the patient’s task of defining the utility function. Also, the patient sometimes helps with the provider’s task of determining the probability of outcomes. Finally, either the patient, the provider, or both, usually perform (heuristically) the optimization.

Overall, however, decision analysis gives an interesting perspective, and this perspective can sometimes be useful, as I will try to describe in future posts.