The Development of DRIs 1994-2004 Lessons Learned and New Challenges: Workshop Summary (2008) / Chapter Skim
Currently Skimming:
3 Criteria for Scientific Decision Making: Session 2
3 Criteria for Scientific Decision Making: Session 2
Pages 63-98
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 63... ...
. General questions were asked of each participant: Can we provide more specific guidance to study committees on scientific decision making to help clarify the concepts and tasks and to promote consistency across study committees?
|
From page 64... ...
This is to be expected, given the differences in the biology and functions of essential nutrients. Throughout the experience of developing reference values, limited data have often precluded the identification of the most appropriate endpoint for any given age/gender category.
|
From page 65... ...
Selecting Endpoints In the past, a number of endpoint types have served as the basis for reference values. These have included clinical signs, measures of developmental abnormalities in children, biochemical measures, balance study outcomes, body pool measures, functional measures, and measures of chronic disease risk.
|
From page 66... ...
Although this is complicated by synthesis in the skin as a result of sun exposure, it is generally a good measure of absorption of vitamin D However, data may be emerging that relate levels of 25-hydroxyvitamin D to measures of bone density, skeletal disease risk (as in the case of osteoporotic fracture)
|
From page 67... ...
However, we must remember that one set of criteria or even an algorithm is unlikely to be "one size fits all" because there may need to be different approaches for different nutrients and types of reference values. This process will be an evolution that must be carefully planned.
|
From page 68... ...
The DRI process depends on dose– response data for both EARs and ULs. Even if there are extremely limited data on dose–response for many nutrients, DRI study committees need to establish numeric values.
|
From page 69... ...
The study committee considered several possible endpoints or biomarkers for selenium status, ranging from disease endpoints (e.g., Keshan disease and cancer) to blood or plasma selenium levels to plasma selenoprotein concentration as a biomarker of selenium status.
|
From page 70... ...
This is nutrient specific; for example, the intake data are not of good quality for vitamin E and selenium. However, in many of these same studies, we can examine plasma nutrient status as a biomarker for chronic disease risk to estimate the dose–response, which can then be related to intake data using metabolic or other relevant studies.
|
From page 71... ...
Spline models allow for the examination of nonlinear effects of continuous variables (e.g., nutrient intake or concentration) in relation to disease risk.
|
From page 72... ...
In Figure 3-1 (page 67) , from a meta-analysis looking at vitamin D supplementation and its effects on hip fracture and nonvertebral fracture, the authors performed a meta-regression to fit a linear regression to the data on the relative risk for a chronic disease endpoint as a function of achieved plasma 25-hydroxyvitamin D concentrations.
|
From page 73... ...
An example from the literature is a meta-analysis looking at observational studies on selenium intake and prostate cancer risk (Etminan et al., 2005)
|
From page 74... ...
West (6) Pooled RR 0.5 1 2 FIGURE 3-4 A meta-analysis of observational studies of selenium intake and prostate cancer risk.
|
From page 75... ...
However, human dose data supersede in vitro tests in human cell lines. Highthroughput methods are another technique where, for nutrients, different doses could be used to see if any inference or input about dose–response can be obtained, but they will not necessarily solve any problems.
|
From page 76... ...
For example, nutrients are only one of numerous factors that determine cancer risk. However, for some chronic diseases there may be a nutritional role more proximal to the disease endpoint, such as lutein and zeaxanthin for macular degeneration.
|
From page 77... ...
The paucity of specific data available based on research in infants and children is concerning. For this reason, careful consideration of extrapolation methods is needed to ensure that we are doing the best we can until that point when data are available and DRI reference values can be set without the need for extrapolation.
|
From page 78... ...
The other model for scaling is relative to body surface area, and this adjusts for metabolic differences between ages related to body surface area based on its relation to BMR. This will always result in higher nutrient reference values than those based on body weight.
|
From page 79... ...
For some nutrients, especially for the UL, extrapolation on the basis of body weight or body surface area yields a UL for children that is incompatible with known nutrient intakes. Perhaps ULs should not be set for children until we have direct experimental evidence.
|
From page 80... ...
Dr. Atkinson suggested it would be important to pursue appropriate animal models to study adverse effects as a preliminary and hypothesis-generating step for this purpose.
|
From page 81... ...
. This presentation will focus on adjusting for data uncertainty from the perspective of establishing ULs, but many of the principles may also apply to establishing reference values.
|
From page 82... ...
NOTE: EAR = Estimated Average Requirement; SD = standard deviation; NOAEL = no-observed-adverse-effect level; LOAEL = lowest-observed-adverse-effect level; UF = uncertainty factor; RDA = Recommended Dietary Allowance. External Dose External Dose Toxic Response Absorbed Dose Absorbed Dose Clearance Intracellular Intracellular Clearance Pathological Changes Concentration in Concentration in Pathological Changes General Circulation General Circulation Distribution to Distribution to Nontarget Tissues Nontarget Tissues Concentration in Concentration in Interaction with Interaction with Target Tissues Target Tissues Intracellular Intracellular Targets Targets Local Bio Local Bio activation activation Physiological Physiological Responses FIGURE 3-7 The multiple steps between intake of a nutrient and either physiological or toxic responses, depending on3-7.eps dose.
|
From page 83... ...
All the different steps can be characterized either in animals or in different age or gender groups of humans, which helps to modify or quantify the necessary adjustment or uncertainty factors. However, few data are available on these different steps, making it difficult to develop reasonable adjustment factors.
|
From page 84... ...
. Major sources and types of uncertainties in dietary exposure assessment include food consumption, body weight, and content in food.
|
From page 85... ...
. There are other uncertainty factors for use of subchronic rather than chronic studies, use of a LOAEL instead of a NOAEL, and deficiencies in the database.
|
From page 86... ...
Where a human LOAEL was used, the uncertainty factors were higher, except for magnesium, fluoride, and sodium, and the justification for the larger uncertainty factor was often the use of the LOAEL. Where animal LOAELs were used, the variation in the selection of adjustment factors was especially great.
|
From page 87... ...
An understanding of the strengths and weaknesses of the various dietary assessment methods for estimating current population intakes is important in ensuring the proper use and interpretation of these dietary estimations. Other types of data on intake relevant to DRI development include studying dose–response relationships in clinical feeding studies, evaluating DRIs in population-based epidemiological or clinical studies, and developing AIs from national dietary surveys, such as NHANES.
|
From page 88... ...
Furthermore, usual intake distributions can be estimated from as few as two dietary recalls. One weakness is that recalls rely on memory.
|
From page 89... ...
Food Frequency Questionnaires In the often self-administered FFQs, people are asked a series of questions -- usually hundreds -- about how often they usually consumed a particular food in a given time period; what preparation methods were used; and what the typical portion size was. These components vary among FFQs, as do procedures to determine the food list and the nutrient composition assigned to each food.
|
From page 90... ...
When supplement intake is included, this results in highly skewed intake distributions, which present challenges for describing usual intake distributions. Another source of error in all self-report dietary data relates to the nutrient database.
|
From page 91... ...
The NCI method builds on the NRC/ISU methods to estimate usual nutrient intake distributions. It can also handle episodically consumed dietary constituents, such as vitamin A, and it can be applied to foods and dietary supplements.
|
From page 92... ...
Therefore, it is understandable that DRIs, even when developed using the best available scientific data, might be disparate from estimated intakes from dietary surveillance data. A clear understanding of the strengths and limitations of the dietary intake estimates allows those responsible for DRI development to put the scientifically derived DRI values in the context of current estimated intakes and, in turn, advise users of DRI values about differences between values and estimated intakes and possible reasons for them; it also identifies avenues for further research.
|
From page 93... ...
Another participant suggested that statistical modeling depends on the assumptions used. The assumption that a yearly intake reflects usual intake may be appropriate in some cases but not others, specifically in developing countries.
|
From page 94... ...
All this requires some broadening in the use of our tools. There is LESS MORE MORE Complexity of Underlying Mechanisms Enhanced performance and Environmental Conditions Responsiveness to Behavior Knowledge Primary and secondary prevention of diet-related chronic diseases Avoiding classic deficiency disease MORE LESS LESS FIGURE 3-12 Nutritional health continuum.
|
From page 95... ...
4. High rates of prematurity, the health consequences of this condition, and increasing technological capabilities that enable survival at pro gressively lower gestational ages, which will bring special pressures to the DRI process The IOM undoubtedly will be faced with including one or more of these conditions in the future DRI process; there may be a need to develop an ancillary effort to consider these groups beyond the brief paragraphs that have been included in the sections of the DRI reports labeled "special considerations." In addition, metabolic and other common morbidities that
|
From page 96... ...
A second dimension of the food environment is our increasing ability to manipulate nutrient intake through fortification, genetic engineering, and supplements. The potential for adverse nutrient interaction merits continued close attention.
|
From page 97... ...
Genomics In terms of issues that fall under the broad category of genomics, current considerations for DRI development are limited to genomic variability, which specifically takes the form of including body size (to the degree that size is genetically controlled) in estimating requirements.
|
From page 98... ...
Another participant addressed the issue of environmental influences, noting that Dr. Garza had mentioned infectious diseases as pertinent to nutrient reference values for persons in developing countries.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.