Mapping the History onto the Kinetic and Morphologic Axes | AI Learning Guide

A history finding only becomes useful when you translate it into a prediction about the two axes you already know. Take heavy menstrual bleeding. If the loss is ongoing and the marrow is healthy, you predict a high reticulocyte count, because the marrow is responding. If iron stores have been depleted, you predict a low MCV. Now take chronic kidney disease. Here you predict a low reticulocyte count with a normal MCV, because the marrow is not getting enough erythropoietin to respond. A family history of anemia with jaundice predicts a high reticulocyte count, since red cells are being destroyed and replaced, with a normal or slightly high MCV. A strict vegan diet with tingling predicts a low reticulocyte count and a high MCV. The reason to write these predictions down before the CBC returns is that they turn a list of facts into a hypothesis. If you predicted microcytic and the MCV comes back normal, that mismatch is itself information, and it usually means a second process is present or the history was incomplete.

0:00 / 0:00

Auto-play

History finding to predicted axis position

History finding

Heavy menstrual bleeding, ongoing
Chronic kidney disease
Family history of anemia with jaundice
Strict vegan diet with paresthesias
Black stools, weight loss, age over 50

Predicted reticulocyte count

High, if marrow is healthy
Low
High
Low
High, if loss is ongoing

Predicted MCV

Low once iron stores are depleted
Normal
Normal or slightly high
High
Low once iron stores are depleted

Why writing the prediction down matters

If you only list the history findings, you have a set of facts. If you commit to a predicted reticulocyte count and MCV for each one, you have a hypothesis that the next two tests can confirm or refute. That commitment is what makes the sequence work: when the CBC returns a normal MCV in a patient you predicted would be microcytic, the discrepancy itself is information, and it usually means a second process is present or the history was incomplete.

These predictions are provisional. They are meant to be tested by the CBC and indices, not treated as conclusions.