The evidence-grade rubric, visualized

A letter grade is only useful if you can see how it was reached. On this site the evidence grade is not a score someone assigned, it is derived by a fixed rule from a compound's research status and study counts, so the same inputs always produce the same grade. The diagram is that rule.

How to read it

Start at the top and take the first row whose condition is true. An approved human indication earns an A, but only if indexed studies actually included female subjects; if they didn't, the grade is capped at B. Below that, Phase 2/3 human data is a B, early human data a C, animal-only data a D, and anecdotal or in-vitro reports, or no indexed studies at all, an E.

Why female evidence can cap the grade

This is the one place the rubric departs from a conventional evidence ladder. A drug can be fully approved and still have almost no data in women, because approval was often won on trials that skewed male or never analysed sex as a variable. Rather than hand a compound an A it hasn't earned for half the population, the rule caps it at B and says why. That cap is the site's thesis encoded as a rule.

Every profile shows its own one-line derivation next to the grade. For the counts that feed the rule and their known limits, see the methodology page; for the pattern across compounds, see Female vs. male evidence, compound by compound.