Printed intervals vary by laboratory and by method, so they are not what decides whether a reading is inside range. Each marker we interpret carries our own range from a named published source, by sex and age band. Your lab's printed interval is stored verbatim and shown next to ours — partly so you can compare against your paper, and partly because a sharp disagreement between the two is itself a signal that something is off.
Suggestions are labelled established, limited, or uncertain, and the source is named. Where we have nothing worth saying about a marker, we say nothing about it rather than filling the space. Some markers are far enough outside range that no self-directed action is appropriate; those are marked as one for your doctor and carry no suggestions at all.
A model is used for one job: turning a document into fields. It reads names, values, units and dates out of your report, and it is checked by you before any of it counts. It does not decide which markers matter, does not rank them, and does not write the suggestions you read. That is ordinary code working from a reviewed catalogue, so the same report produces the same result every time.
Powermetrica is educational. It does not diagnose, treat, or replace your doctor, it does not answer open-ended health questions, and it will not score your health or estimate a biological age. A single blood test is a narrow view of one moment, and nothing here should be read as a verdict on your health.