Guiding Principles for Developing Dietary Reference Intakes Based on Chronic Disease (2017) / Chapter Skim
Currently Skimming:
7 Intake-Response Relationships and Dietary Reference Intakes for Chronic Disease
7 Intake-Response Relationships and Dietary Reference Intakes for Chronic Disease
Pages 195-242
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 195... ...
Risks of chronic diseases will typically be expressed as relative risks (RRs) , and intakes may be expressed as daily exposures to a food substance, cumulative exposures over specified periods of time, or as a measured biomarker of intake exposure.
|
From page 196... ...
The committee also offers commentary and guidance on conceptual challenges raised in the Options Report, such as establishing risk-reductions goals for DRIs based on RR, rather than on absolute risk (the measure traditionally used for DRIs based on deficiency diseases and high intake toxicity) and developing DRIs that take into account various interactions among food substances that may influence risk of chronic disease.
|
From page 197... ...
will be one of the goals of DRI committees. The ability to quantify how much reduction in chronic disease risk would result from a given change in the intake of an NOFS will depend on the nature of the available data from randomized controlled trials (RCTs)
|
From page 198... ...
It is important to note that many RCTs are typically powered for a 10 to 20 percent difference in disease outcomes by treatment arm. The true quantifiable relationship may be at lower or higher values of the nutrient than were tested, and the true effect size in the general population also may be lower or higher, but inferences can be drawn only on what was tested in the trial so this may limit usable information for DRI committees.
|
From page 199... ...
In recent years, investigators have strongly emphasized absolute risk as sessment for a number of chronic diseases. Risk calculators, which typically project chronic disease occurrence over a period of a few years while taking into account competing risks (e.g., Gail et al., 1989)
|
From page 200... ...
Important factors that will influence, and potentially bias, the intakeresponse relationship include the nutrient intake and the chronic disease measures. Reliability and accuracy of those measures is essential when identifying chronic disease DRIs.
|
From page 201... ...
The committee assumes that the systematic review team would have evaluated whether individual studies adjusted for the appropriate confounders during the risk-of-bias assessment. DRI committees could come up with a priori lists of likely confounders of the food substance/outcome relationship.
|
From page 202... ...
More comprehensive data may be needed to guide DRI committees to better evaluate the likely long latency period between intake and chronic disease endpoints and to help determine whether they should restrict data evaluation to a defined time frame before the response or rely on data with various time exposures of intakes. Were Interactions with Other Nutrients Considered?
|
From page 203... ...
For example, some prior investigations of vitamin E and chronic disease risk may have overlooked the biological interaction of vitamin E and selenium in the design and interpretation of the data (Lippman et al., 2009)
|
From page 204... ...
The end result is a larger sample size with a greater number of disease endpoints, thus increasing statistical power. In addition, pooled relative risks can be estimated both for main effects and, importantly for effect modification by age, race/ethnicity, sex, body mass index, and other characteristics where individual studies would not have the statistical power for interaction tests.
|
From page 205... ...
These pooled data studies are subject to the same limitations, for example, concerning dietary exposure assessment and confounding, as are the individual studies being pooled. Systematic Reviews and Meta-Analyses A second approach to modeling the intake-response relationship for consideration by DRI committees is meta-analyses.
|
From page 206... ...
. Other methods have been used, such as metaregression analysis, which allows the modeling of various levels of intake and incorporates the fact that reference values are not zero in nutrition studies (Chung et al., 2016; del Gobbo, 2013; Jiang et al., 2016)
|
From page 207... ...
Therefore, although in practice questions related to causality and intake-response relationships will be included in the same systematic reviews, a separate evaluation of certainty of evidence is needed for intake-response relationships. Drawing upon the conceptual and methodologic issues described above and considering the set of options relevant to this chapter described in the Options Report, the committee has developed recommendations (and their rationale)
|
From page 208... ...
Risk of Bias of Individual Studies Risk of bias is defined as the systematic error due to limitations in the study design or execution, or inappropriate analysis. The features discussed under risk of bias for causality in Chapter 6 are applicable to assessing the evidence for the quantitative relationship between an NOFS and a chronic disease.
|
From page 209... ...
, while evaluating, assessing, and communicating uncertainties is always necessary, overly complex uncertainty analyses that provide little or no value-added in terms of the ultimate decision should not be pursued. The precision of the quantitative relationship between nutrient intake and the risk of developing a chronic disease is always limited by the available data.
|
From page 210... ...
per day for women of childbearing age were estimated to result in optimal RBC folate concentration to prevent neural tube defects in children. Investigators are already attempting to study the relationship between folic acid intake and chronic diseases (e.g., see relationship between folic acid intake and plasma folate in Yang et al., 2010, and plasma folate and FIGURE 7-3 Example of potential relationships between intake, a surrogate marker, and a disease that would allow indirect determinations of intake-response relationships.
|
From page 211... ...
As a result, qualified surrogate markers will necessarily have a significant role in the development of chronic disease DRIs. The approaches to describing intake-response relationships for qualified surrogate markers and factors such as confounding and inter-individual variability would proceed as outlined earlier in the section titled "Was the Appropriate Model of the Nutrient-Chronic Disease Relationship Selected?
|
From page 212... ...
, so the committee is reluctant to recommend these options as a general matter, as they require development of multivariate, BOX 7-2 Selecting Indicators and Specifying Intake Response Relations: Qualified Surrogate Disease Markers and Nonqualified Disease Markers Options Report Option 1: Choose a single outcome indicator on the causal pathway This option selects a single outcome indicator that is on the causal pathway, provided that it is sufficiently sensitive to quantify the relation between a food substance and a chronic disease. Option 2: Use multiple indicators of a chronic disease This option integrates information from multiple indicators of a given chronic disease that add substantially to the accuracy of the intake-response relation and the development of a reference value.
|
From page 213... ...
When such a situation arises, and the available intake-response data are adequate, there is no special difficulty with recommending reference values for each chronic disease, even when the level of confidence in the evidence for all of the chronic diseases is different; DRI committees should adhere to the approaches spelled out in the following sections for each endpoint. However, an overall DRI across multiple chronic diseases would need to consider the risks and benefits of each chronic disease across the range of intakes, which is discussed later in the section on "Balance Between Desirable and Undesirable Outcomes." Extrapolating intake-response data across populations The committee generally holds that a high degree of caution should guide extrapolation in the intake-response data across populations that differ greatly in important underlying risk factors related to the chronic disease (see Box 7-3)
|
From page 214... ...
Option 2: Allow extrapolation when sufficient evidence is available This option allows extrapolation when sufficient evidence shows that spe cific intakes of a food substance can increase or decrease the risk of a chronic disease. Committee's Recommendation 5 The committee recommends extrapolation of intake-response data for chronic disease Dietary Reference Intakes only to populations that are simi lar to studied populations in the underlying factors related to the chronic disease of interest.
|
From page 215... ...
This factor may differ across the range of intakes, which complicates interpretation. Rating the Overall Certainty in the Body of Evidence for Intake-Response In the GRADE approach to assess causality, the certainty in the evidence for a particular outcome is generally designated as the highest level among the bodies of evidence for that outcome (e.g., if both RCTs and observational studies examined blood pressure, then the body of evidence with the highest certainty should be used)
|
From page 216... ...
Therefore, selecting chronic diseases DRIs involves additional decisions related to the type of DRIs, acceptable level of confidence in the intake-response data, and balancing health risks and benefits. The factors considered in rating the certainty of evidence delineated by GRADE are still appropriate, but the committee suggests that they be used more iteratively to determine the most appropriate DRI.
|
From page 217... ...
The Options Report introduces the difficult problem of interactions among food substances that affect chronic disease risk. Approaches to address these questions are in the previous section on "Was the Appropriate Modeling of the NOFS-Chronic Disease Relationship Selected?
|
From page 218... ...
Committee's Recommendation 6 The committee recommends that Dietary Reference Intakes (DRIs) for chronic disease risk take the form of a range, rather than a single number.
|
From page 219... ...
Because the ULCD is based on changes in the relative risk of the chronic disease, intakes below the ULCD might reduce but not necessarily eliminate disease risk, reflecting the multifactorial nature of chronic diseases. Committee's Recommendation 7 The committee recommends retaining Tolerable Upper Intake Levels (ULs)
|
From page 220... ...
. These examples encompass the Options Report suggestions of a point estimate at which maximum risk reduction or a range of beneficial intakes is achieved, but additional information could be valuable for derivation and application of DRIs.
|
From page 221... ...
• Setting a DRI for chronic disease also may need to consider poten tial harm from deficiency at the lower end of intake and toxicity at the upper end of intake. Risks and benefits related to multiple chronic diseases introduce further complexity (see the later section on "Balance Between Desirable and Undesirable Consequences of a DRI")
|
From page 223... ...
Furthermore, efforts should be made to convert RR to absolute risks, which will give users of these reference values increased understanding of health benefits achieved with a given degree of intake reduction. ULs and Reduction of Chronic Disease Risk ULs were introduced in 1997 as a new type of DRI (see Figure 4 in the Options Report, Appendix B)
|
From page 224... ...
. In cases in which increased intake is associated with increased chronic disease risks at intakes less than the traditional UL (see Figure 7-4 A', B', and C and Figure 5b in the Options Report, Appendix B)
|
From page 225... ...
Relevant information is included in an Annex 7-2 to this chapter, titled "Possible New Approaches to UL Development." Recognizing the challenge described in the Options Report -- that is, the lack of data and implications for setting chronic disease DRIs higher that the traditional UL -- the committee notes the need for better information on the adverse health effects of high intake levels for many NOFSs that may become the subjects of DRIs. Certainty in Evidence for Intake-Response Relationship As already mentioned, Chapter 6 describes the use of GRADE in evaluating the existence of a causal relationship between intake of an NOFS and risk of a chronic disease.
|
From page 226... ...
. Box 7-6 shows the three potential approaches in the Options Report for accepting the necessary confidence level in the intake-response relationships for developing DRIs.
|
From page 227... ...
Several issues specific to DRIs for chronic disease need to be considered, including overlaps between benefits and harm. Overlaps Between Benefits and Harm Deficiency, toxicity, and multiple chronic diseases need to be considered when balancing benefits and harms.
|
From page 228... ...
In making their conclusions, however, DRI committees should explicitly specify the certainty in the evidence used to develop the DRI and the populations and other circumstances to which it applies. In the case of chronic disease DRIs, committees will likely find that certainty of evidence for intake-response relationships and the balance between desirable and undesirable outcomes are the most influential fac
|
From page 229... ...
For example, in addition to differences in populations, DRI committees should consider the possibility that certainty in the evidence may differ depending on the ranges of intake. Guiding Principle 14 was developed to emphasize the importance of describing all uncertainties when characterizing intakeresponse relationships (see Box 7-9)
|
From page 230... ...
here they exist, quantitative intake-response relationships should in W clude a thorough description of the scientific uncertainties associated with them.
|
From page 231... ...
However, DRI committees' charges relate only to health outcomes, and, as explained in Chapter 1, policy considerations, and values and preferences related to quality of life (e.g., taste, convenience) , equity considerations, and considerations of cost and resources that might ensue from chronic disease DRIs are outside of the scope of DRI committees.
|
From page 232... ...
2002/2005. Dietary Reference Intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids.
|
From page 233... ...
2011. Dietary Reference Intakes for calcium and vitamin D
|
From page 234... ...
2009. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement.
|
From page 235... ...
2017. Options for basing Dietary Reference Intakes (DRIs)
|
From page 236... ...
IHD death, defined as any fatal IHD, including sudden cardiac death (SCD) • Population: Adults • ll eligible studies included a multivariate-adjusted effect A estimate with a measure of uncertainty for circulating or dietary magnesium and incident CVD, IHD, or fatal IHD including SCD Exclusion criteria: • tudies reporting stroke as a distinct outcome S • tudies focused on children S • tudies that only evaluated drinking water magnesium or S water hardness, dietary patterns/food groups, intracellular free magnesium, or extracellular ionized magnesium • tudies focused on populations with disturbed mineral S homeostasis • tudies with only crude risk estimates, ecologic studies, case S reports, cross-sectional studies, retrospective case-control studies, editorials/commentaries, letters, and reviews Literature Search Earliest available online indexing year to May 2012 Dates or Year Range Number of Studies 16 prospective studies, mostly cohort (11 with estimates of dietary Included magnesium; 9 with estimates of circulating magnesium)
|
From page 237... ...
Trends toward stronger associations in cohorts with more men (P-heterogeneity=0.06) and studies evaluating fatal IHD death rather than total IHD (P-heterogeneity=0.07)
|
From page 238... ...
or IHD (P=0.26) Significant nonlinear association between dietary magnesium and fatal IHD (P-nonlinearity, 0.001)
|
From page 239... ...
Due to the multifactorial nature of chronic disease, no such bright line exists, as chronic diseases do not exhibit such "threshold"-like behavior and typically exhibit a continuous change in risk with changing intakes. However, this "threshold" interpretation of the UL is not entirely satisfactory either.
|
From page 241... ...
This may be particularly important in the scenario described in Chapter 7 in which increasing intake decreases chronic disease risk up to and perhaps beyond the UL. This type of "risk-benefit" comparison would be infeasible under the traditional "NOAEL divided by 100" approach, because there is no characterization of the gradient of the intake response over a wide enough range of doses.
|
From page 242... ...
1997. Dietary Reference Intakes for calcium, phosphorus, mag nesium, vitamin D, and fluoride.
|
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.