Reference intervals for parathyroid hormone for 70-year-old males and females: exclusion of individuals from the reference interval based on sex, calcium, diabetes, cardiovascular diseases or reduced kidney function has limited effects on the interval

Background A problem when producing reference intervals for elderly individuals is that they often suffer from a number of diseases and they are most often on medication. If all such persons are excluded, there is a risk that the residual subgroup may not be representative of the population, we therefore wanted to compare the effects different exclusion criteria has on the reference intervals. Methods We measured parathyroid hormone (PTH), calcium, albumin and cystatin C in a cohort of 70-year-old males and females (n=1003). Reference intervals for PTH for males and females were calculated for the entire population and after exclusion of persons with calcium >2.60mmol/L, calcium >2.51mmol/L, diabetes, reduced glomerular filtration rate (GFR), and cardiovascular diseases. Results The calculated PTH reference interval 16 (CI 14-17) to 94 (CI 87-101)ng/L. Exclusion of study subjects resulted in smaller reference sample groups, but the reference limits remained within the 90% confidence intervals of the original reference limits. The selections thus had a very limited effect on the calculated reference interval for PTH. Conclusions Exclusion of elderly individuals with high calcium concentrations, diabetes, reduced GFR or cardiovascular disease has little effect on the reference interval for PTH. It is better not to exclude these individuals, as it will provide a broader base for the reference interval.