From what I can recall from years ago, we reckoned +/- 2% accuracy for tests like Hb, PCV (Packed Cell Volume - same as Haematocrit I think) and so on. We didn't do RBC counts routinely back then, as it was a dilute the blood (several times) and a person (occasionally me) counted the cells in a counting chamber using a microscope . No electronic miracle machines! PCV was probably more reliable back then than RBC counts. We did have a machine to do WBC counts (white blood cells) - lyse the RBCs and then the WBCs could be counted. Coulter Counter. PCV was always a bit of a strange one - make sure blood fully mixed (we did have a machine that rocked and rolled to keep anti-coagulated blood mixed - previous lab I used to have to try and gently mix stuff by gently rocking and upending samples by hand before sampling). Then it's take a teeny tiny narrow tube that you stick in the blood which flows up by capillary action. Then you blocked the end using something a bit like blue-tack and we had a special centrifuge to spin it down, just the right size for the teeny tiny glass tubes. Then you measured the percentage of red cells to plasma, we had a little doo-hicky measuring thing to use as all the teeny tiny tubes had slightly different amounts of blood in them. Sorry @arewenearlythereyet, been on a bit of a trip down memory lane there! I'm not sure what I've written will be any help to you, but if nothing else it might give you a laugh imagining me as a teenager working in a lab wearing a white coat and Scholls sandals (horrid to walk any distance in, but great if you needed to stand a lot). Guess it's good they are flagging up your haematocrit and keeping an eye on it.