to allow for optimal medical decision making. One
of the first analytes for which this recognition may
come is creatine kinase. The plasma concentrations
of creatine kinase tend to be much higher in “black”
men and women than in “white” men and women,
respectively. Using the data reported by Harris
et
al.
(1991), race-based reference frequency distribu-
tions of creatine kinase are significantly different for
both men and women (Figure 6.5). The large differ-
ences between the reference frequency distributions
in “whites” and “blacks” has clear implications for
the use of creatine kinase measurements in the
diagnosis of myocardial infarction. For instance,
using the critical value determined in “white” men,
creatine kinase has a low specificity when used in
the evaluation of chest pain in “black” men. The
use of a critical value derived from a reference
frequency distribution as found in “black” men
would improve the diagnostic performance of
creatine kinase in “black” men considerably.
Practical problems in constructing race-based
reference frequency distributions
A considerable difficulty in establishing race-
based reference frequency distributions in North
America is the highly variable degree of racial
admixing that is present in our society. For
instance, many individuals descended from African
slaves brought to the American Colonies are a mix
of African and European races. A reference
frequency distribution derived from such an admixed
group would be expected to be intermediate between
a reference frequency distribution derived from
Europeans and a reference frequency distribution
derived from a group of recent West African
immigrants.
Another challenge in the establishment of refer-
ence frequency distributions based on race is deter-
mining the extent to which ethnic practices and
socioeconomic conditions rather than race itself
affect the analyte under study. As an example, it
has been suggested that the higher plasma concentra-
tions of creatine kinase in “black” men compared to
“white” men may arise not from inherent racial
differences but rather from their generally lower
socioeconomic status (Young 1979). The logic
behind this suggestion is that the members of lower
socioeconomic groups are more likely to be
employed in manual labor and to have greater
muscular development resulting in increased plasma
creatine kinase concentrations. This possibility, and
other questions relating to ethnic practices and socio-
economic conditions, can been investigated scientifi-
cally by comparing racial reference groups that are
matched for the supposed confounding variables. In
the case of creatine kinase, it would be useful to
compare office workers of African and European
extraction and to compare manual laborers of the
two races. Alternatively, the effects of muscle
development on plasma creatine kinase concentra-
tions could be investigated directly by quantifying
the degree of muscle development of each individual
in a reference population and determining the
relationship, if any, between the measure of muscle
development and the concentration of creatine
kinase.
AGE
Five distinct periods of life are usually recog-
nized when considering the effects of age upon
laboratory study results. These are (1) the fetal
period, (2) the newborn period, (3) childhood and
adolescence, (4) adulthood, and (5) old age.
Laboratory studies conducted early in the fetal
period are almost all performed with the intent of
detecting genetic disease or developmental disorders.
Direct genetic studies are not sensitive to fetal age
but indirect genetic studies and developmental
studies necessarily utilize marker substances that will
depend upon fetal age. Prenatal screening for both
trisomy 21 and for neural tube defects, for instance,
use the concentration of alpha-fetoprotein in mater-
nal plasma as one of a set of marker substances.
Biologic Variability
6-6
0.5
0.75
1
1.25
1.5
1.75
SD ratio
-1.5
-1
-0.5
0
0.5
1
1.5
Scaled difference in means
M
F
Figure 6.5
Graphical test of the heterogeneity of reference
frequency distributions for creatine kinase based on race.
M, comparison of “black” and “white” males; F, comparison
of “white” and “black” females.
