Course talk:CPSC522/Utility Assessment for Decision Support System

I thought the descriptions incorporated in your page were good but I didn't quite understand your solution for incorporating patient preferences in a decision-support system. I found your paragraphs on calibrating low probability outcomes that patients may encounter with familiar gambles quite interesting and can see how this could be beneficial to patients. However, I didn't understand how it related to your solution and utility assessments. Perhaps adding a couple more sentences or paragraphs on your specific solution and relating the utility assessments to the familiar gambles paragraph could help.

I appreciate the thorough descriptions on your page, particularly the insights into calibrating low probability outcomes with familiar gambles, which I found intriguing. However, the connection between this concept and your proposed solution for incorporating patient preferences in a decision-support system could be clearer. Adding a few sentences to bridge the gap between the familiar gambles paragraph and your solution would enhance overall coherence. Furthermore, the page's abstract nature could be addressed by incorporating a concrete example, illustrating how the proposed technique can be practically applied and extended for future decisions. Clarifying whether your proposal is for one-off decisions or long-term elicitation would provide valuable context; an example could effectively illustrate this distinction. These enhancements would strengthen your proposal and facilitate a clearer understanding for the reader.

I think this page is too abstract to convince us to use a particular technique. A simple example would help immensely. Then explain how that example could be extended to use in practice. Is this proposal for one-off decisions (e.g., which of two procedures should be carried out for some illness), or for more long-term elicitation that can be used for future decisions? Some of the time it seems you are talking about patients in general, and sometimes elicitation of the preferences for the particular patient for who the decision is about. Which one are you proposing? An example would be useful to clear this up.

Thanks for the references to EQ-5D, SF-6D, and HUI, but are these really "methods used for utility assessment of patients"? They didn't seem to ask patients.

"The ultimate goal of this project is to measure and quantify the health-related quality of life (HRQoL) or health utility of individuals." No the aim of the project is to make or recommend a decision.

Ultimately, I'm not sure what method you are proposing. It was good to provide the landscape, but when a team has to develop a system (see the November Assignment description), they need to choose one to start with. I don't think the team would know where to start.