Decision Aids:The effect of context on decision making


Decision aids should include information about the risks and benefits of each possible choice. Often in decision aids these risks and benefits are not provided by themselves but are put into context. This context information could include the risks and benefits that other people face from these choices or the risk that the patient faces from other health conditions. Providing contextual information can influence the meaning people take away from risk information.

The role of context in evaluating risk. People often make different choices when they are presented each option in isolation than when all options are presented together. Similarly, providing people with information about the risk of the “average person” often changes how they interpret their own risk. We are conducting an ongoing series of studies to identify the influence of contextual information on patient decision making, including whether it influences (for better or worse) people’s perceived competence in the decision task and their ability to accurately understand the statistical meaning of critical probability information. When we identify detrimental biases, we also work to develop potential debiasing approaches.

Changes in perception of risk information for recurring versus newly developing diseases. Even when risks are quantitatively similar, people may not feel they are equally likely if they occur in different contexts. For example, people may feel more susceptible to a disease that they have already had in the past, compared to other diseases.

The effect of asking people to estimate their own risks: Many times people start conversations about health risks by asking the patient what they think their risk is. Yet, doing so provides an anchor point which can alter how people perceive subsequent risk communications. For example, our research has shown that asking women to estimate their risk of breast cancer led them to be less worried about this information than women who did not estimate their risks.

The effect of categorical / qualitative labels on reactions to test results. We have demonstrated that providing qualitative labels for a diagnostic test result (e.g., "positive” or “negative") provides additional information beyond risk probabilities and often influences how people interpret the test results.