Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Reading around (e.g. [0]) it sounds like variance isn't something researchers care for, probably with the assumption that screening with a higher value is better than missing an issue from a lower value.

We see the same for body temperature (speed and convenience is usually prioritized over accuracy) and weight (2% variation is largely accepted). Afterall guidelines are already off as by definition, as they don't account for individual circumstances, so perhaps aiming for accuracy is useless in most settings.

[0] https://www.mayoclinic.org/diseases-conditions/high-blood-pr...



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: