5

Questions from the Committee to Review the DRIs for Energy

In open session discussions held on February 24 and April 8, 2022, between the standing committee, the Federal DRI Joint U.S.–Canadian Working Group, and the DRI committee, specific questions were posed. Those questions from the DRI Committee and the standing committee’s responses follow.

Question EC1: Should a DRI Committee Approach Identifying a General or Healthy Population Based on Inclusion or Exclusion Criteria?

Response

The general population base should be as broad as possible. To ensure the consistent application of this concept, each DRI committee should consult with the standing committee regarding the rationale for exceptions being considered or proposed (see Question WG1). Once those exceptions are removed from the total population, the general population is defined for that nutrient’s DRI value. The committee’s report should specify and justify the subpopulation(s) for which the DRI values do not apply. When dietary guidance for excepted subgroups in the population is warranted, this should be discussed in the “special considerations” section of the report.

Question EC2: What Is the Standing Committee’s Recommendation for Managing Limitations in Data Needed to Support a Rationale for Recommendations?

Response

Limitations in the evidence needed to support DRI reviews are expected. For example, with limited or no data for particular life-stage groups when determining Estimated Average Requirement/Recommended Dietary Allowance/Adequate Intake (EAR/RDA/AI) and Estimated Energy Requirement (EER) values, previous DRI committees have relied on data from one age group and extrapolated to others. Tolerable Upper Intake Level (UL) for a few nutrients may have relied on animal data or on acute exposures in case studies. DRI committees need flexibility to weigh the evidence with limitations of available data. The committees must clarify the strength, source, and uncertainties of the evidence and delineate gaps in knowledge.

In the case of a relation between a nutrient and a chronic disease where there is considerable uncertainty about causality, more rigorous data should be expected (Yetley et al., 2017). In NASEM (2017), the application of GRADE (Grading of Recommendations Assessment, Development, and Evaluation) was suggested as an approach for future DRI committees to apply to evidence synthesis. This would allow them the flexibility to formulate judgments about causality between nutrient exposure, medication use, and health outcomes.

Question EC3: If Data on Some Subpopulations Are Limited, How Do You Determine When to Use the Data as a Criterion to Exclude That Group from the Defined General Population?

Response

As indicated above in Question WG6, limitations in the evidence needed to support DRI reviews are anticipated. For example, with limited or no data for particular life-stage groups when determining DRI values, previous committees have relied on data from one age group and extrapolated to others, although the extrapolation model was not consistent across applications. DRI committees need flexibility to weigh the evidence with limitations of available data. The committees must clarify the strength, source, and uncertainties of the evidence and delineate gaps in knowledge.

The goal is to have DRI values that are as inclusive of the general population as possible. When considering essential nutrient requirements

for the EAR and RDA, providing no guidance for an essential nutrient is considered to be unethical. In these cases, the review committee may derive an AI value from population-based intake data or decide to extrapolate recommendations from a similar population where more data exist (Table 5-1). However, for the UL and Chronic Disease Risk Reduction (CDRR) intake, where there is often considerable uncertainty as to the causal nature of the relationship between intake and health outcomes and completeness of documentation, the DRI review committee will need to consider the strength of the evidence in deciding whether or not to set

TABLE 5-1 Dietary Reference Intakes (DRIs) and their Descriptive Definition

SOURCES: IOM, 2002/2005; NASEM, 2017.

DRI values for some or all life-stage groups, or whether to extrapolate data from groups with evidence to those without. The DRI review committee must be transparent about the strength of the data, extrapolation models applied, and the uncertainties or limitations of the recommendation for subpopulations (see Question EC6 below).

Question EC4: Does the Standing Committee Want to Make a Statement About Using International Classification of Diseases (ICD) Codes to Exclude Population Groups in a DRI Review?

Response

The standing committee does not recommend the use of International Classification of Diseases (ICD) codes to define exceptions to definitions of the general population. While many peer reviewed studies have used or reported ICD codes as outcome measures, these codes cover a large range of health conditions and many are likely irrelevant to the DRI process and uncertainty exists in their application. While the standing committee concluded that the ICD codes are too broad and not useful for setting DRI exceptions to the definition of general population, the inclusion of human studies on the relationships between intake and health outcomes that rely on ICD codes to define outcomes are useful in many cases for better understanding DRI-related relationships (Drosler et al., 2021).

Question EC5: Are Extreme Outliers Included in the General Population? What About Normal Body Composition When Body Mass Index (BMI) is Less Than or Equal to 18.5?

Response

DRI committees should carefully examine the data sets and determine if there is clear evidence to exclude outliers beyond the defined range of normal BMI. Regarding individuals with below normal BMI but with normal body composition, unless there is evidence to exclude data for this group, they should not be considered as exceptions. If there is evidence for altered energy requirements, for example in some metabolic disorders, exceptions may be warranted. In the case where no evidence exists that energy requirements are altered, these individuals with outlying BMI values should be included in the general population.

Question EC6: Since the DRIs for Energy Are Different from Other Nutrients, How Should the Committee Manage Limitations in Data Needed to Support the Rationale for a Recommendation?

Response

It is challenging for DRI committees to set requirements for subgroups when a lack of data exists. For the DRI committee, in the absence of doubly labeled water (DLW) data for some populations the committee should consider other populations and relevant non-DLW data sources to determine the best approach for extrapolation. The report should clarify the gaps of available data. Further, the approach used to extrapolate should be clearly described in the report.

Delineation of gaps in knowledge should hopefully stimulate funding opportunities and research to providing critical information for future DRI committees. To that end, the committee notes that racial/ethnic populations are currently understudied and should be addressed in future DRI reviews.

This page intentionally left blank.