Reference values for common laboratory tests were determined as part o
f a cross-sectional survey of 70 year olds in Jerusalem in 1991. Subje
cts were systematically chosen from electoral lists. They were extensi
vely interviewed regarding demographic, socioeconomic and health chara
cteristics. Subsequently, 456 underwent a thorough medical examination
including hematologic, biochemical and endocrinologic testing, Analys
is of disease-specific mortality and hospital morbidity at 3 year foll
ow-up proved these subjects to be representative of the overall Jerusa
lem Jewish population of 70 year olds. Test-specific reference subgrou
ps were determined to exclude effects of known diseases and drug consu
mption. Reference values in these subgroups differ significantly from
reference values currently used for adults without regard to age. The
determination of disease for subjects provided the opportunity to iden
tify changes that might reflect normal aging and those that might be a
ffected by subclinical pathology. Interpretation of laboratory testing
of older adults must consider the age-related differences in referenc
e values.