Why are we looking at self-reports instead of outcomes? If self reports remained constant since 1993 for the most privileged group and dropped for other groups

Why are we looking at self-reports instead of outcomes? If self reports remained constant since 1993 for the most privileged group and dropped for other groups, then all things being equal, outcomes should have dropped on average when you look at the entire population, but when you look at data about actual outcomes, they don't support this: https://www.healthsystemtracker.org/chart-collection/how-has-the-quality-of-the-u-s-healthcare-system-changed-over-time/ https://www.healthsystemtracker.org/chart-collection/how-has... Maybe people feel worse even though they are doing better but don't remember how things were 25+ years ago.