---

Summary: decision analysis can help ensure that insurance prioritizes patient utility—at the individual and collective level.

---

I previously described decision analysis, a well-known framework for medical decision-making. As a follow up, I described the patient and provider roles in a decision analysis (specifying utilities, , and probabilities, , respectively, where is the patient’s initial state, is the treatment, and is the outcome). I also discussed the interaction between decision analysis and the pharmaceutical industry. I will now cover how decision analysis might interact with insurance.

Insurance allows people to help one another pay for healthcare services. Due to high costs of healthcare, this is important. However, from a medical decision-making standpoint, it is problematic when insurance is coordinated by a for-profit company, which might have incentive to deny claims. Doing so is not a great business strategy, but it can be done in subtle ways. For example, a company might deny claims for time-intensive lifestyle counseling and approve the quick prescription of symptom-alleviating medications. This can exert pressure on entire sub-fields of medicine.

A decision analysis makes it difficult to unjustly deny a claim, because it facilitates a structured, documented discussion between the patient, provider, and payer. For example, with a decision analysis, in order to deny a claim, an insurance company has to challenge the probability of the outcome, It’s difficult to do this, because this probability is tied to the literature. Or, an insurance company has to challenge the patient’s reward, It’s difficult to do this, because the patient defines their own reward. Without a decision analysis, it is difficult to separate probability and utility, much less determine which of the two an unjust claim is challenging. This leads to ambiguity, which often favors the insurance company.

This said, insurance is complex. Entities that coordinate insurance must cover a large number of patients who may fall ill at any time. This involves tradeoffs. For example, it might be important to deny claims for costly, after-the-fact healthcare interventions in favor of more preventive strategies. Or, it might be important to allocate more funds toward helping sick children. At a population level, the documentation from a decision analysis, which includes quantities such as and , can be transformed into a structured database, facilitating data-driven discussions between patients, providers, and payers about how we, as a society, are optimizing for (and should optimize for) collective utility.

Overall, decision analysis, in its formalism and transparency, protects the individual patient from unjust, profit-oriented denials, and also helps us better coordinate insurance for the collective good.