
vitamins and riboflavin analysis. For the enzymatic
assays of erythrocyte thiamin, riboflavin, and vitamin
B
6
heparinized blood should be stabilized with proper
ACD solution (ACD solution is a stabilizer made of
citric acid, sodium citrate and d-glucose) for storage
at 4–6
C for 10 days. For the erythrocyte thiamin
pyrophosphate (TPP) assay the determination should
be carried out within 2 h of drawing blood because of
TPP instability. For vitamin C determination, hepar-
inized blood, after stabilization with metaphosphoric
acid, can be stored frozen, preferably at 70
C, for
no more than 3 weeks. For vitamin C determination
by high-performance liquid chromatography (HPLC),
heparinized blood specimens can be stored for a
few weeks at 80
C after addition of reduced
glutathione within 20 min of collection. The addition
of dithiothreitol increases stability of plasma samples
to more than 1 year at 70
C.
0023 Acidified urine specimens for thiamin, riboflavin,
and vitamin B
6
determinations are stable at 20
C
for 3 months. The specimens should be protected
from light. Longer storage is possible for biotin and
pantothenic acid.
0024 For mineral and trace element determination serum,
plasma, whole blood, or blood cells are used with no
great problems for storage, but glass containers
should be avoided.
0025 For cellular studies of immunocompetence the
blood must be rapidly processed; in some cases the
testing of frozen cells was successful.
Recommended Methods for Nutritional
Status Assessment
0026 Table 1 is a synopsis of some recommended methods.
For a detailed description of them as well as for other
methods, see the Further Reading section.
0027 These methods can be used for the whole popula-
tion as well as for single patients. In the clinical
setting, nutritional assessment and support are best
performed by a nutritional support team, including
clinician, nutritionist, and clinical chemist. If needed,
advanced methods for the assessment of body com-
position, nitrogen balance, and some prognostic in-
dices are available to detect patient malnutrition and
to predict outcome (sepsis, wound dehiscence, death).
Data Processing and Calculation
0028 The most important stages of data processing are:
coding (including a preliminary quality control);
data input; quality control data; data bank; data
analysis (using parametric and nonparametric tech-
niques); and reporting. Specialized books describe in
detail the above stages. For each type of variable,
specific calculation methods may be used. Only
general rules will be provided here.
0029For variables not normally distributed (some an-
thropometric data, ferritin, etc.), a logarithmic trans-
formation can be applied. In some cases the use of
nonparametric tests is preferred.
0030For the use and interpretation of anthropometric
measurements an Expert Committee Report of the
World Health Organization is highly recommended.
Technical framework and detailed guidance on the
use and interpretation of anthropometric measure-
ments in pregnant and lactating women, newborn
infants, infants and children, adolescents, overweight
and thin adults, and adults aged 60 years and over are
provided. An extensive series of reference data is
included in an annex.
0031Computer programs are now commonly used to
test specific hypotheses. Because of the highly special-
ized skills required for proper processing and analysis
of data, it is recommended that specialists in these
fields take part in the planning and execution of
evaluation. To improve the quality of this work it is
advisable that these specialists are also involved in the
data recording – an essential preliminary step of data
processing. This means that they must be involved in
the design of the study and in the preparation and
testing of various forms and/or questionnaires needed
for the study.
0032For data analysis, a data stratification by age
groups and sex or any other meaningful stratification
like altitude (for hemoglobin), socioeconomic status,
and health/disease data, must often be used. Results
may be expressed in terms of means and standard
deviations or standard errors or as percentages of
values (prevalence) above or below an arbitrary cut-
off point defining adequacy. Anthropometric indices
can be expressed in terms of Z-score (or standard
deviation score). If the sample size is large enough,
the percentiles or the frequency distributions are very
useful. This data presentation facilitates comparison
between surveys in different countries; it permits at
any time the selection of the percentile or the range of
percentiles of reference, i.e., acceptability, and it can
represent a combination of the descriptive and inter-
pretative ways for analyzing and presenting collected
data.
0033The final step is the interpretation of data. In gen-
eral the prevalence of malnutrition or deficiency is
obtained using cut-off from a reference population.
This reference population and the population under
study are often from different areas or the criteria
used to specify the reference population are not care-
fully defined. However, since most reference values
are population-specific, it is recommended that data
from a clinically healthy reference population of the
NUTRITIONAL ASSESSMENT/Importance of Measuring Nutritional Status 4177