These tools are useful because they simplify complex problems
However, they are not 100% accurate
If a model/score gives the wrong answer, patients may be misdiagnosed or offered the wrong treatment
If models/scores are wrong more often for particular groups, this could worsen health disparities
We aim to help healthcare professionals choose the model/score which is best for a particular patient, and to ensure that future tools help as many patients as possible.
Image credit: Turing Commons. CC BY-SA 4.0 DEED