Dietary
Risk Assessment in the WIC Program
EXECUTIVE
SUMMARY
Dietary intake
patterns of individuals are complex in
nature. However, assessing these complex
patterns has been fundamental to the
Special Supplemental Nutrition Program
for Women, Infants, and Children (WIC)
since its inception. The WIC program,
which provides nutritious supplemental
foods, nutrition education, and health
referral services to low-income pregnant
or postpartum women, infants, and
children to age 5 years, requires
applicants to meet one of several
nutrition risk categories in order to be
eligible for program services; dietary
risk is one of these categories. Others
include anthropometric risk (e.g.,
underweight, overweight), biochemical
risk (e.g., low hematocrit), medical
risk (e.g., diabetes mellitus), and
other predisposing factors (e.g.,
homelessness). Since funds are not
always available to meet the needs of
the number of applicants determined to
be eligible, a priority system is in
place in which nutrition risk criteria
are categorized based on severity of
potential effect and outcome.
The role of
dietary assessment in establishing
eligibility for WIC is a crucial one,
especially for postpartum women and
children. As stated above, although
eligibility may be based on many kinds
of nutritional risks, substantial
numbers of postpartum women and children
currently are found to be eligible only
on the basis of dietary risk. The
practice of assessing dietary intake is
widespread in part because, for those
found to be at nutritional risk, the
dietary data also influence the contents
of the food package made available,
nutrition education, and, sometimes,
referrals. For this reason, even though
many applicants are found to be at
nutritional risk for a reason other than
dietary risk, 86 percent of state
agencies assess the dietary intake of
all WIC participants. The practice
consumes considerable time resources on
the part of both WIC personnel and their
clients.
In any venue,
the assessment of dietary risk poses a
challenge. Indeed, in an earlier report,
the Institute of Medicine stated,
“Research is urgently needed to
develop practical and valid assessment
tools for the identification of
inadequate diets” (IOM, 1996).
Moreover, a joint working group of the
National Association of WIC Directors
and of the Food and Nutrition Service of
the U.S. Department of Agriculture did
not find a sufficient scientific basis
for developing standardized criteria for
two major types of dietary risk: failure
to meet Dietary Guidelines and
inadequate diet. These are the two types
of dietary risk that WIC personnel use
extensively as the sole basis for
determining that postpartum women and
children are at nutritional risk.
Failure to meet
Dietary Guidelines refers to the 10
guidelines in the Dietary Guidelines for
Americans (USDA/HHS, 2000). These
guidelines emphasize overall dietary and
lifestyle patterns that can help to
achieve favorable long-term health
outcomes. Based on current knowledge
about how dietary and physical activity
patterns may reduce the risk of major
chronic diseases and how a healthful
diet may promote health, the 10
guidelines are designed to serve as the
basis for federal policy and are used to
guide nutrition information, education,
and interventions for federal, state,
and local agencies.
Embedded in the
guidelines is the Food Guide
Pyramid—one of the major tools used
for consumer nutrition education in the
United States. The Pyramid incorporates
many of the Dietary Guidelines and gives
concrete recommendations that promote
moderation, balance, and variety in food
intake.
The
Task
Because of
concern about the quality of dietary
assessment methods and the resources in
WIC required for using them to establish
nutritional risk, the Food and Nutrition
Service asked the Institute of Medicine
(IOM) for assistance. In particular, it
contracted with the IOM’s Food and
Nutrition Board to evaluate the use of
various dietary assessment tools and to
make recommendations for the assessment
of inadequate or inappropriate dietary
patterns, especially in the category
failure to meet Dietary Guidelines. The
Food and Nutrition Service asked that an
expert committee propose a framework for
assessing dietary risk among WIC
applicants and identify and prioritize
areas of greatest concern when the
Dietary Guidelines are incorporated in
WIC. In doing so, the committee was
asked to focus on tools that could
identify dietary risk of individuals
accurately and thus be suitable for
eligibility determination. The committee
was also asked to recommend specific
cut-off points for the criteria and to
consider both food-based and
behavior-based approaches. This report
addresses those topics. However, since
the Dietary Guidelines apply only to
individuals ages 2 years and older, the
focus is on pregnant and postpartum
women and children.
Current
Practices
Since
standardized criteria have not yet been
established for failure to meet Dietary
Guidelines or inadequate diets, state
WIC agencies currently select the method
and cut-off points to be used by their
agencies. The most commonly used methods
are 24-hour diet recalls and food
frequency questionnaires. WIC personnel
generally compare dietary intake data
obtained using one or both of these
methods with specified numbers of
servings from each of the five basic
food groups of the Food Guide Pyramid.
In most cases, the methods used appear
not to have undergone studies of
accuracy or reliability. Many state WIC
agencies use the Food Guide Pyramid
servings as a standard for children ages
12 to 24 months even though the Pyramid
was designed for persons ages 2 years
and older.
A
Framework for Assessing Dietary Risk
In an interim
report (IOM, 2000c), the Committee on
Dietary Risk Assessment in the WIC
Program proposed a framework that
consists of eight characteristics
essential to a food-based and/or
behavior-based tool designed for
eligibility determination. That
framework has been modified slightly in
this report. An optimal tool should:
-
use
specific criteria that are related
to health or disease;
-
be
appropriate for age and
physiological condition (e.g.,
pregnancy or lactation);
-
serve three
purposes: screening for eligibility,
tailoring of food packages, and
nutrition education;
-
have
acceptable performance
characteristics (validity and
reliability);
-
be suitable
for the culture and language of the
population served;
-
be
responsive to operational
constraints in the WIC setting;
-
be
standardized across states/agencies;
and
-
allow
prioritization within the category
of dietary risk.
The committee
considered these characteristics as it
examined possible methods for
determining dietary risk.
FINDINGS
AND RECOMMENDATION
Findings
Basing
Risk Criteria on the Dietary Guidelines
Focusing on the
single guideline Let the Pyramid Guide
Your Food Choices was determined to be
the most feasible, comprehensive, and
objective approach to using the Dietary
Guidelines for establishing dietary risk
for those individuals 2 years of age and
older. Based on review of the Dietary
Guidelines and the scientific
underpinnings of the Food Guide Pyramid,
the committee determined that this
approach should use the recommended
number of servings based on energy needs
as the cut-off point for each of the
five basic food groups. For example, the
criterion for active, pregnant, adult
women would be at least nine servings
from the grains group. A majority of
state WIC agencies already use some
version of this approach as the basis
for setting a criterion that addresses
the dietary risk failure to meet Dietary
Guidelines.
Finding 1.
A dietary risk criterion that uses the
WIC applicant’s usual intake of the
five basic Pyramid food groups as the
indicator and the recommended numbers
of servings based on energy needs as
the cut-off points is consistent with
failure to meet Dietary Guidelines.
Prevalence
of Dietary Risk Based on the Food Guide
Pyramid Recommendations
More than 96
percent of individuals in the United
States, and an even higher percentage of
low-income individuals (such as those
served by WIC), do not usually consume
the recommended number of servings
specified by the Food Guide Pyramid
(Krebs-Smith et al., 1997; Munoz et al.,
1997). Thus, the identification of
individuals who are not at dietary risk
becomes highly problematic.
Finding 2.
Nearly all U.S. women and children
usually consume fewer than the
recommended number of servings
specified by the Food Guide Pyramid
and, therefore, would be at dietary
risk based on the criterion failure to
meet Dietary Guidelines that is
described in Finding 1.
Food-Based
Assessment of Dietary Intake
Nutritional
status and health are influenced by
usual or long-term dietary intake. For
this reason, dietary assessment for
establishing WIC eligibility should be
based on usual intake. Day-to-day
variation in food and nutrient intake by
individuals is so large in the United
States that one or two 24-hour diet
recalls or food records cannot provide
accurate information about an
individual’s usual intake. In the WIC
setting, it is impractical to obtain
more than one or two recalls or records
under standardized conditions that would
promote accurate reporting. Moreover,
most people make many errors when
reporting their food intake because of
the complex nature of the task. These
errors increase the likelihood that
eligibility status for WIC will be
misclassified in the category of dietary
risk.
Food frequency
questionnaires (FFQs) are designed to
assess usual intake and may be practical
to administer to many WIC clients.
However, they are subject to many types
of errors, and their performance
characteristics are unsatisfactory for
determining individual eligibility. For
example, when reported food or nutrient
intakes from an FFQ are compared with
the values obtained using a large number
of research-quality diet recalls or food
records, correlations generally range
between 0.3 and 0.7. Although
correlations in that range may be
considered satisfactory for making
inferences about intakes by groups of
individuals in epidemiologic research,
such data cannot accurately classify
individuals as above or below set
cut-off points—a serious problem when
the goal is determining the eligibility
of an individual. Shortening FFQs
generally makes them more responsive to
operational constraints, but further
reduces their accuracy and utility.
Few practical
methods have been developed or tested
that compare food intakes with the
Dietary Guidelines or Food Guide Pyramid
recommendations. Such methods would
require converting amounts of each type
of food consumed to Pyramid portions to
determine whether the Food Guide Pyramid
recommendations had been met. This is a
complex process, especially for mixed
dishes, and does not lend itself to
operational constraints in the WIC
setting.
Finding 3.
Even research-quality dietary
assessment methods are not
sufficiently accurate or precise to
distinguish an individual’s
eligibility status using criteria
based on the Food Guide Pyramid or on
nutrient intake.
Physical
Activity Assessment
Because the
committee was asked to identify areas of
concern when the Dietary Guidelines were
incorporated into WIC and because the
Guidelines include a quantitative
recommendation for physical activity
levels for adults and for children 2
years of age and older, the committee
considered physical activity assessment
as a part of dietary risk assessment.
Although a physical activity
recommendation appears in the Dietary
Guidelines, physical activity itself is
not currently part of dietary risk
assessment in WIC, nor is there a
separate nutritional risk criterion in
the WIC program related to physical
activity. However, given that (1)
WIC’s mandate is to focus on primary
prevention, including the primary
prevention of overweight and obesity,
(2) the increasing degree to which
overweight and obesity are now major
health concerns among those served by
WIC, and (3) proper risk assessment for
prevention or treatment must consider
both diet and physical activity, it is
likely that WIC may soon consider
assessing physical activity, even if not
for the purposes of eligibility
determination.
Physical
activity assessment relates to two of
the Dietary Guidelines (Aim For A
Healthy Weight and Be Physically Active
Each Day) and thus could potentially be
used as another way to define failure to
meet Dietary Guidelines. The physical
activity guideline specifies “Aim to
accumulate at least 30 minutes (adults)
or 60 minutes (children) of moderate
physical activity most days of the week,
preferably daily.” These
specifications could be used as WIC
eligibility criteria under the dietary
risk subgroup failure to meet Dietary
Guidelines.
A review of the
literature found no physical activity
assessment instruments that meet the
operational constraints of WIC and that
also can accurately and reliably assess
whether a woman or child is obtaining at
least the specified amount of physical
activity. Because of the inherent
cognitive challenge of accurately
recalling and characterizing the varied
activity behaviors that together
constitute an individual’s physical
activity level, it is unlikely that
there could ever be a practical
instrument to establish WIC eligibility
accurately based on the physical
activity recommendation in the Dietary
Guidelines.
Finding 4.
Physical activity assessment methods
are not sufficiently accurate or
reliable to distinguish individuals
who are ineligible from those who are
eligible for WIC services based on the
physical activity component of the
Dietary Guidelines.
Behavioral
Indicators of Diet and Physical Activity
Because certain
behaviors are correlated with dietary
intake and physical activity, interest
has arisen in the use of behavior-based
assessment as a method of identifying
those who usually fail to meet the
Dietary Guidelines. Such assessment
would require the identification of
behavioral indicators that could
distinguish individuals who meet the
Dietary Guidelines from those who do
not. The committee considered two types
of behavioral indicators: surrogate and
target. Surrogate behaviors are
behaviors that are correlated with one
or more aspects of diet or physical
activity and could be used to make
inferences about what children eat or
how much activity they engage in. For
example, the frequency of eating meals
together as a family could indicate the
adequacy of vegetable consumption.
Target behaviors are behaviors that make
good targets for change. Making changes
in a target behavior would be expected
to result in changes in dietary intake.
Target behavioral indicators are not
suitable for eligibility determination
unless they also are surrogate
indicators. Building on the example
above, if families could be encouraged
to eat meals together more frequently,
and if family meals resulted in improved
dietary intake, then frequency of eating
meals as a family would be both a
surrogate indicator and a potential
target indicator for change. By analogy,
if families could spend more time
outdoors and if this change resulted in
increased levels of physical activity,
then time spent outdoors could be both a
surrogate and target indicator for
physical activity.
A review of the
literature found few studies of
behavioral correlates of diet or
physical activity conducted among the
groups served by WIC. No strong evidence
was found that any examined behaviors
would be both adequately reliable and
accurate as surrogate or target
behavioral indicators.
Finding 5.
Behavioral indicators have weak
relationships with dietary or physical
activity outcomes of interest. As a
result, they hold no promise of
distinguishing individuals who are
ineligible for WIC from those who are
eligible in the category of dietary
risk.
RECOMMENDATION
Based on the
above findings, the following
recommendation is made:
Presume
that all women and children (ages 2 to
5 years) who meet the eligibility
requirements of income, categorical,
and residency status also meet the
requirement of nutrition risk through
the category of dietary risk based on
failure to meet Dietary Guidelines,
where failure to meet Dietary
Guidelines is defined as consuming
fewer than the recommended number of
servings from one or more of the five
basic food groups (grains, fruits,
vegetables, milk products, and meat or
beans) based on an individual’s
estimated energy needs.
Studies suggest
that nearly all women in the
childbearing years and children ages 2
years and older are at dietary risk
because they fail to meet the Dietary
Guidelines as translated by
recommendations of the Food Guide
Pyramid (Krebs-Smith et al., 1997; Munoz
et al., 1997) Tools currently used for
dietary risk assessment appear to have
very high sensitivity in that they
identify nearly everyone as failing to
meet the Dietary Guidelines, but low
specificity—poor ability to identify
persons who are not at dietary risk. No
known dietary or physical activity
assessment methods or behavioral
indicators of diet or physical activity
hold promise of accurately identifying
the small percentage of women and
children who do meet the proposed
criterion based on the Food Guide
Pyramid or the physical activity
recommendation. Even if the percentage
of individuals who meet the criterion
were to increase substantially, it
remains unlikely that methods can be
found or developed to differentiate risk
among individuals.
When WIC was
originally established in 1972, the
categorical groups that WIC serves were
selected because of their vulnerability
to nutritional insults and WIC’s
potential for preventing
nutrition-related problems. Nutritional
status and dietary intake have both
short- and long-term effects on the
health of the woman and on the growth,
development, and health of the fetus,
infant, or child. The groups served by
WIC also are at increased risk of
morbidity and mortality from virtually
every disorder listed among the leading
causes of death in the United States
(cardiovascular disease, cancer,
diabetes, and digestive diseases). The
high prevalence of overweight and
obesity and of diets that are
inconsistent with the Dietary Guidelines
(e.g., low intakes of fruits and
vegetables, high intakes of saturated
fats) may contribute to these increased
risks.
This
recommendation is not intended to affect
the current use of other nutritional
risk criteria for eligibility
determination. That is, information
should continue to be collected for the
identification of other nutrition risks
(e.g., hemoglobin or hematocrit to
identify risk of anemia, height and
weight to identify anthropometric risk,
and the presence of diabetes mellitus to
identify medical risk). Such information
is useful for nutrition education, and
it is essential to implement the
priority system. When funds are
insufficient to enroll all those
eligible for WIC, the priority system is
used to determine those at greatest
need. If dietary information is
collected in the WIC setting for food
package tailoring, nutrition education,
and/or health referrals, the methods
used should be approached with caution
given the likelihood of error and
misclassification.
Optimal
Collection and Use of Dietary and
Physical Activity Data
Although
individual-level reporting errors
greatly reduce the validity of data for
assessing diet or physical activity
levels in individuals, the errors are
less serious in group assessments.
Moreover, a variety of statistical
procedures can adjust for known sources
of error (IOM, 2000a; Traub, 1994) and
thereby provide reasonable tests of
relationships. Thus, while identified
relationships may not be true for any
specific individual, they would be true
for the group. For example, FFQs and
diet recalls can be used to identify
dietary patterns in a WIC population and
patterns needing improvement. Repeated
collection of dietary recalls or FFQs
also may be used to monitor change over
time at the group level or to assess
effects of nutrition education
interventions.
Findings from
such analyses could be used to design
nutrition education programs and monitor
their effectiveness. For example, diet
recalls can provide valid information on
the average intakes of groups, assuming
that a standardized data collection
approach is used and an adequate sample
size (50 or larger) is available. If
more than one recall is collected on at
least a subsample of the group and
appropriate adjustments are made, one
could determine the proportion of the
group with usual nutrient intakes that
are less than the Estimated Average
Requirement (IOM, 2000a). Group dietary
intake information for a WIC population
(e.g., data from a recent national
dietary survey such as the National
Health and Nutrition Examination Survey
or the Continuing Survey of Food Intakes
by Individuals or data collected in a
special WIC study) could be used to
identify areas for targeted nutrition
education services.
Likewise,
physical activity assessment tools may
be sufficiently valid to assess physical
activity levels within groups. These
data would be valuable for monitoring
groups of individuals or “target
populations” within WIC that may be at
higher risk for low physical activity
levels and/or that may benefit most from
interventions within WIC to increase
physical activity levels.
Group
assessment data would best be collected
by trained individuals on randomly
selected subsamples of the WIC
population. However, any tool used for
this purpose must still be evaluated in
terms of desired criteria (e.g., a tool
would still need to be easy to
administer, appropriate for the group,
and reasonably accurate).
CONCLUDING
REMARK
In summary,
evidence exists to conclude that nearly
all low-income women in the childbearing
years and children ages 2 to 5 years are
at dietary risk, are vulnerable to
nutrition insults, and may benefit from
WIC’s services. Further, due to the
complex nature of dietary patterns, it
is unlikely that a tool will be
developed to fulfill its intended
purpose within WIC: to classify
individuals accurately with respect to
their true dietary risk. Thus, any tools
adopted would result in
misclassification of the eligibility
status of some, potentially many,
individuals. By presuming that all who
meet the categorical and income
eligibility requirements are at dietary
risk, WIC retains its potential for
preventing and correcting
nutrition-related problems while
avoiding serious misclassification
errors that could lead to denial of
services to eligible individuals.
Last modified: 12/04/2008
|