This is from the Clinical and Laboratory Standards Institute (CLSI). They provide guidelines and standards for clinical laboratories worldwide.

Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory: "A reference interval (RI) is a calculated or observed range of values that is determined from testing a reference population and between which the values of a particular analyte (e.g., a constituent of blood or urine) for a specific subpopulation of healthy individuals (e.g., adults, children, pregnant women) can be expected to fall with a given probability (usually 95%). This interval usually includes the mean ± 1.96 standard deviations (SDs) of the distribution of values in the reference population."

I found a more vague definition on the Mayo Clinic site: "Results from your hematocrit test are reported as the percentage of blood cells that are red blood cells. Typical ranges vary substantially with race, age and sex. The definition of typical red-blood cell percentage also may vary somewhat from one medical practice to another. This is because laboratories decide what is a healthy range based on the population in their area."

As for CTX and P1NP, all of the drug studies I have read use a percentage change from baseline. This is because the baseline values vary a lot from person to person. But in general, higher CTX numbers mean higher bone teardown activity from the osteoclasts and higher P1NP numbers mean higher bone building activity from the osteoblasts. Looking at them as a pair is more useful than looking at them individually.