Understanding Information Preferences, Risk Perceptions, and Tradeoffs When Making Decisions About Multi-cancer Early Detection Tests

The test performance data presented in the vignette will focus on established evidentiary standards for the evaluation of screening tests: Sensitivity, specificity, positive and negative predictive values, and the ability of tests to find specific cancers and more early-stage cancers. The high numerical detail condition will utilize these numbers and present them according to the decision aid criteria for presenting treatment option data and addressing health literacy.

The low numerical detail condition will describe the potential for false positives and use descriptions to denote qualitatively performance (high, acceptable, moderate, low). The test performance data presented in the vignette will focus on established evidentiary standards for the evaluation of screening tests: Sensitivity, specificity, positive and negative predictive values, and the ability of tests to find specific cancers and more early-stage cancers.

The high numerical data will use diagnostic workup data from the most up-to-date findings at the time of survey fielding. Data will be presented on the number of positive tests, follow up included (type and properties of tests), likelihood of receiving each type of test upon a positive findings, and data on median length of time to diagnostic resolution for both true positives and false positives.

The low numerical detail condition will describe the possibility of further testing (more blood tests, imaging, and/or procedures) that may be required if the multi-cancer early detection (MCED) test has a positive finding, in line with prior studies communicating about MCED test workup.

Cost estimates will include insurance and out of pocket costs associated with various pathways of diagnostic workup. The investigators will use test cost estimates based on market rates and insurance coverage mandates at the time of fielding the survey. The high numerical data condition will provide precise estimates conditioned on level of workup and insurance coverage. As above, the final content will be informed by new information that may be gleaned from ongoing studies.

The low numerical data will provide broad ranges of diagnostic costs, indicating what costs are covered by insurance, and which are not.