Two requirements that fight
A clinical record has to be correctable, because doctors write in a hurry and get things wrong. It also has to be trustworthy, because a record that can be silently rewritten is evidence of nothing.
Software usually resolves this badly in one of two directions. Either the note can be edited freely and the record loses its authority, or it cannot be edited at all and staff start keeping the real information somewhere else.
The paper answer was better
On paper, a correction is a struck-through line, an addendum, an initial and a date. The original stays legible. Nobody is pretending the first version did not exist, and anybody reading later can see exactly what changed and who changed it.
That is amendment, and it is a better model than either extreme. The interesting question is why so much software abandoned it in favor of an edit button.
What a good amendment records
A reason, stated before the note reopens, which is a small friction that changes behavior, because it makes the author think about what they are doing. The version as signed, preserved rather than overwritten. And what actually changed, field by field, rather than a log line saying "record edited".
Two details are worth adding. Reopening a signed note should be a separate permission from writing one, so a hospital can let every doctor write freely and still decide who may reopen. And reopening a note and changing nothing should still leave a trace, because somebody reopening a closed clinical record is worth knowing about.
What it is not
It is not version history you can roll back to. The preserved original is evidence, not a restore point, and offering a rollback would recreate exactly the problem amendment solves.
And it is not deletion. A consultation that happened, happened. Nirogix does not delete one at any permission level, and that is a design decision rather than a missing feature.