4

Current Models for Establishing Intake Recommendations

OVERVIEW

In addition to the presentations summarized in Chapter 3, session 1 included a panel discussion, “Current Models for Establishing Intake Recommendations.” The panel was moderated by Hasan Hutchinson. Hutchinson and three other panelists from different regions of the world described their experiences with the harmonization of approaches to establishing nutrient intake recommendations. This chapter summarizes their presentations and the discussion that followed, with major points highlighted here and in Box 4-1.

Hutchinson provided an overview of the joint Canadian–U.S. effort to develop harmonized Dietary Reference Intakes (DRIs), an effort that began in 1994 and which continues to this day. By 2005, he said, a fairly complete set of DRIs had been established, but with no formal process to keep the DRIs up to date. Thus, a joint steering committee developed a structured process for deciding which nutrients to update based on changes in evidence, new methodologies, and other criteria. As King had previously discussed (summarized in the previous chapter), the sodium and potassium DRI reviews will be updated first, using the new guiding principles for chronic disease endpoints.

Clifton summarized recommendations issued in 2012 for South Australia and New Zealand, actions taken since then, and challenges faced. The 2012 recommendations included an immediate review of the chronic disease and macronutrient section, less comprehensive reviews of some micronutrients (e.g., the upper limit [UL] for fluoride), and greater transparency in the decision-making process. Among other actions taken since then, a fluoride

working group was established. Its work, all of which Clifton said is clearly documented and available online, was approved in 2016 by the Australian National Health and Medical Research Council. The fluoride work took almost 2 years, illustrating for Clifton the time-consuming nature of a review of a relatively targeted area, but one done comprehensively.

Next, Hee Young Paik described how recommended dietary allowances (RDAs) in South Korea have changed over the past several decades, beginning in 1962 when the first RDAs were published. Most recently, in 2015, a committee comprising 78 members, with subcommittees for different nutrients, reviewed the DRIs for 36 nutrients. In 2005, members of the Korean Nutrition Society decided that it was time to change the concept from RDAs to DRIs. Paik emphasized the 2015 committee’s use of Korean-specific data from the literature. For international harmonization, she suggested that core processes and methods be standardized by having an international expert group review the international literature, but be adaptable to specific countries, such as by having an expert group in each country review local literature using a standardized methodology.

Finally, Ann Prentice described past and current nutrition risk assessment work by the United Kingdom (UK) Scientific Advisory Committee on Nutrition. She emphasized the committee’s focus is on risk assessment, not risk management. The committee’s dietary recommendations are derived using the same basic estimated average requirement (EAR) framework that other speakers have been describing, according to Prentice, and have been since 1981, when the United Kingdom’s first reference nutrition intakes (RNIs) were published. The committee’s recent vitamin D recommendation, however, went beyond the traditional EAR and RNI framework, because of the risk of bone health problems below a particular intake value and the desire for everyone in the population to reach that value year-round and not just when sun exposure is greatest. Instead, they set a “population protective value.” In her experience, Prentice has found that one of the biggest problems is the acceptance, interpretation, and understanding of new terminology.

DIETARY REFERENCE INTAKES: HARMONIZED NUTRIENT STANDARDS FOR CANADA AND THE UNITED STATES¹

The Canadian–U.S. joint effort to develop DRIs began in 1994. The rationale for a common set of values, Hasan Hutchinson explained, included the expanded base of specialized scientific expertise made available through collaboration; that the science underlying nutrient requirements knows no borders; and that the two populations have similar dietary needs. Additionally, it made sense to develop similar approaches to help with trade-related matters around nutrition

According to Hutchinson, since 1994, even though the two countries have separate DRI steering committees, all of the decisions to commission DRI reviews have been made jointly, with a series of six nutrient-focused reports released between 1997 and 2005. DRIs were developed for all macronutrients as well as 35 vitamins and minerals. In 2011, updated DRIs for calcium and vitamin D were released.

Structured Process for Updating DRIs

In 2012, a decision was made with input from both steering committees to develop a structured process for deciding which nutrients needed to have updated DRIs. Rather than conducting a comprehensive review of all the nutrients one by one, it was decided to develop an open nomination process to identify priority nutrients to update. Hutchinson explained

___________________

¹ This section summarizes information presented by Hasan Hutchinson, Ph.D., N.D., director general, Office of Nutrition Policy and Promotion, Health Canada, Ottawa, Ontario.

that nominations can be received from inside or outside of the government and need to include a rationale indicating where there has been a change of evidence, a need for updating because of policy implications, a change in methodology, or some other reason to reexamine a DRI. The Canadian and American committees then examine the nominations to see if they meet a set of nomination criteria. Together, the committees prioritize nutrients for further consideration and form working groups to conduct an in-depth assessment for each candidate nutrient. Each working group includes U.S. and Canadian experts. Following in-depth assessments, joint Canadian–U.S. decisions are made regarding which candidate nutrients should be considered for further assessment. If the decision is made to update a particular nutrient, the Agency for Healthcare Research and Quality is tasked to conduct a systematic review. With sufficient evidence from the systematic review, the National Academies would then be contracted to conduct a DRI review.

Canada and the United States decide separately how the information is to be applied in their respective countries. The hope, Hutchinson said, is that the recommendations in these reports can be directly applied to dietary guidance. He added they are also used to make other decisions, such as decisions about fortification and supplementation.

Hutchinson remarked that while the 2013 nomination process received 26 nominations for a total of 16 nutrients, which resulted in a short list of nutrients to move forward, the steering committees identified a need to address the challenge of chronic disease endpoints. This led to the development of guiding principles for developing DRIs based on chronic disease endpoints (NASEM, 2017a). Hutchinson referred to the earlier presentations by MacFarlane and King (summarized in Chapter 3), noting, as they had, that the sodium/potassium DRI review, initiated in the fall of 2017, is the first to use these new guiding principles.

SOUTH AUSTRALIA AND NEW ZEALAND²

In 2012, several recommendations were put forward regarding the nutrient reference value (NRV) process in Australia and New Zealand. The first, Peter Clifton began,³ was that an immediate review should be conducted of the chronic disease and macronutrient section. The Departments of Health in Australia and New Zealand did not do this, Clifton said. A second recommendation was that a less comprehensive review should be

___________________

² This section summarizes information presented by Peter Clifton, Ph.D., professor of nutrition, University of South Australia, Adelaide, South Australia, Australia.

³ For additional details on these recommendations and the consultative process that led to them, see the Chapter 3 summary of Clifton’s earlier presentation.

conducted for B12, choline and pyridoxine, zinc (after findings are released from the International Zinc Nutrition Consultative Group [IZiNCG]),⁴ the tolerable upper intake level (UL) for fluoride and selenium, energy, protein, and chloride, as funding and time permit. Third, it was recommended that there be greater transparency in the decision-making process, including clear justification for inclusion of experts and determination of nutrient values. And finally, it was recommended that there be clear documentation of all underlying decisions, evidence, assumptions, and rounding processes.

An advisory committee was set up by a steering committee from the Departments of Health in Australia and New Zealand to decide on the priority nutrients. But they provided no justification for why they chose the nutrients they did, Clifton stated. A fluoride expert working group was set up to focus on both the UL and adequate intake (AI) for fluoride, which had been recommended. However, a sodium expert working group was also set up, which had not been recommended. It was communicated only indirectly that the reason for setting up the sodium expert working group was to model chronic disease methodology, which Clifton noted is also what Health Canada and the National Academies are currently doing. That review is still in progress. It has been 2 years, so “it’s obviously quite a thorny area,” Clifton said.

In addition to the fluoride and sodium expert working groups, an iodine expert working group was also set up. As with sodium, iodine was not on the recommended list of priority nutrients. And again, there was no published reason for the choice other than the fact that, with fortification, it is a reasonable decision to review required levels. However, in Clifton’s opinion, that decision also possibly reflects steering committee members’ interest in iodine.

Finally, although there was some suggestion that carbohydrates be reviewed, there was no interest among potential expert working group members to move that review forward. Carbohydrates are viewed as a difficult and thorny area, a view that Clifton said he shared. Nonetheless, he suspected that similar lack of interest will likely be an issue for other macronutrients, as well as micronutrients, in the future.

The fluoride working group followed the recommendations of the new framework beautifully,⁵ Clifton continued. They performed a systematic review to obtain all available evidence; they were provided administrative support and research assistants; all of their actions were clearly documented

___________________

⁴ Details about IZiNCG, an international group focused on reducing global zinc deficiency, are provided at http://www.izincg.org (accessed April 25, 2018).

⁵ See the summary of Clifton’s earlier presentation in Chapter 3 for additional details about the new framework (Australian Government Department of Health, 2015) and the consultative process that led to its development.

and available online; the public consultation and comments are available online; and the paper was reviewed by methodological experts, as well as by experts in fluoride, dentistry, and pediatric health. The National Health and Medical Research Council approved the report in March 2017. The entire process took about 2 years. Even though the focus of the work was a relatively targeted area (i.e., AI and UL for fluoride for children between 2 and 8 years of age), the review was very comprehensive. “Clearly it’s a time-consuming exercise,” Clifton said.

It remains to be seen what will happen with the sodium review and what other additional reviews will be chosen in the future, he concluded. He expressed disappointment that the funders did not follow the input provided by expert consultants.

SOUTH KOREA: 2015 DIETARY REFERENCE INTAKES FOR KOREANS⁶

The first Korean recommended dietary allowances (KRDAs) were published in 1962 by the Korean Regional Office of the United Nations’ Food and Agriculture Organization (FAO), Hee Young Paik said as she began her history of nutrient standards in Korea. That same office issued two subsequent revisions. Later, the Korea Institute for Health and Social Affairs, a government institute, published the fourth and fifth editions in 1985 and 1989, respectively. Then, the Korean Nutrition Society (KNS) took over responsibility, publishing the next two editions in 1995 and 2000. When preparing for the 2005 KRDAs, Paik said, members of the society felt it was time to change from RDAs to DRIs. The first DRI committee (i.e., the 2005 committee), of which Paik served as chair, reported values for 34 nutrients. The 2010 committee reported on 35 nutrients. Paik remarked that although it had been an unwritten rule of the KNS that the RDA and DRIs would be revised every 5 years, the 2010 National Nutrition Management Act legally mandated the Ministry of Health and Welfare to publish revised Korean DRIs (KDRIs) every 5 years.

The 2015 Korean DRIs

The Ministry of Health and Welfare set up a committee known as the General Administration Committee, composed of government officials and representatives from the nutrition community, as well as other stakeholders, to publish the 2015 KDRIs. The General Administration Committee’s job,

___________________

⁶ This section summarizes information presented by Hee Young Paik, Sc.D., director, Center for Gendered Innovations in Science and Technology Research, Korea Federation of Women’s Science and Technology Associations, Seoul, South Korea.

Paik explained, was mainly to ensure that the process was going well and to resolve any problems that arose. The KNS remained responsible for the actual scientific process of establishing the values (i.e., reviewing scientific evidence on nutrients, data on age and physical standards, and data on dietary intake). The KNS committee that conducted the evidence and data review (the 2015 KDRIs Establishment Committee) had several subcommittees, including 1 for age and physical standards, 1 for applications (e.g., food group guide), 8 for energy and macronutrients, 13 for vitamins, 14 for minerals, 1 for infants, and a review subcommittee. The total number of members was 78.

The 2015 KDRIs cover estimated energy requirements (EERs), acceptable macronutrient distribution ranges (AMDRs), and EARs, RNIs, AIs, and ULs for vitamins and minerals. The total number of nutrients covered was 36, which Paik noted was fewer than the number of U.S. and Canadian DRIs (40 in 1997–2011), but more than in most other countries (e.g., Japan had 31 in 2015, Germany/Switzerland/Austria had 27 in 2015, and Australia/New Zealand had 33 in 2005).

Changes between the 2010 and 2015 KDRIs, Paik noted, include that the upper range of the carbohydrate AMDR was lowered to 65 percent in the 2015 KDRIs, compared to 70 percent in the 2010 KDRIs, and the upper range of the total fat AMDR for adults 18 years and older was increased to 30 percent, compared to 25 percent in the 2010 KDRIs.

The strategy for setting the KDRIs is similar to the process developed by the United States and Canada, Paik continued, that is, it involves systematic evidence review of the international literature and meta-analyses when possible. Plus, it involves a rigorous review of the Korean literature, using the same level of judgment that is used for the international literature, and the use of Korean data, especially for body size and dietary intake data. While the processes are similar, Paik remarked that not all components of the DRI process used in the United States and Canada are applicable to the KDRIs. For example, while the methods and biological characteristics are adoptable, particular references to population and country characteristics are not.

Thoughts on International Harmonization

Paik concluded by offering some suggestions for international harmonization. Generally, she suggested that core processes and methods be standardized, but these should be able to be adapted by specific countries. For example, she suggested that international experts not only review the international literature and make their reviews available to each country, but also develop a standardized methodology for literature review. Expert groups in each country can then use this methodology to review local lit-

erature (i.e., literature published locally in their own language) and provide feedback to the international committee regarding their local literature review. This two-way sharing of literature reviews “will make our work more complete,” Paik said. Additionally, she suggested that international experts provide guidance on standards for national resource data, such as guidelines for national surveys in nutrition, body size, and dietary practices. Local expert groups could also use guidelines on food composition tables and database methodology. Finally, she suggested periodic updating of DRIs.

UNITED KINGDOM⁷

Ann Prentice described the UK Scientific Advisory Committee on Nutrition, of which she was currently chair, as a committee of independent experts who are appointed and who are not employed by the government. In many ways, she said, they are accountable to the whole of the UK government, across a multitude of departments and agencies. Their focus is specifically in risk assessment in diet, nutrition, and health. Risk management, in contrast, is the role of government. Although the committee is sometimes invited to help government authorities understand information being communicated by the committee, it is the responsibility of those authorities, not the committee, to make a decision and consider how to implement their decision.

Nor is the committee involved with aspects related to adverse effects, Prentice clarified further, although there are joint working groups that collaborate on issues that overlap. As an example, she mentioned a recent sodium and potassium joint working group.

The committee’s dietary recommendations are very much based on the framework that previous speakers had been discussing, as they have been since 1991 when the original set of RNIs were first developed, Prentice remarked. (An RNI is the amount of nutrient that is enough to ensure that the needs of 97.5 percent of a population are being met.) In addition to RNIs, the UK dietary reference values (DRVs) also include EARs and lower reference nutrient intakes (LRNIs; an LRNI is the amount of nutrient that is enough for a small percentage of people in a population who have low requirements). She noted that LRNIs are a useful monitoring tool for the UK National Diet and Nutrition Survey.

When evaluating and synthesizing the evidence, the committee is required to consider the entirety of the data, but it uses what Prentice

___________________

⁷ This section summarizes information presented by Ann Prentice, OBE, Ph.D., DUS, director, Medical Research Council (UK) Elsie Widdowson Laboratory, University of Cambridge, England.

described as a formalized and transparent hierarchy framework. At the beginning of each work package, the committee is obliged to consider and decide exactly how they will move forward. Then, depending on the nutrient under consideration, the committee may decide, for example, to use only systematic reviews or to also include animal data. Often, the committee uses systematic reviews from Canada and the United States as a basis for its evaluation and then looks for additional, more recent data. It also examines the assumptions of the existing systematic reviews to determine whether they are appropriate for the particular piece of work the committee is addressing. As an example of its use of a different set of assumptions, in a recent review of weight loss and carbohydrates, the committee decided to consider only those studies with 12-month outcomes, not 3-month outcomes.

With its recent vitamin D review, the UK committee felt the need to go beyond the traditional RNI/EAR framework. Specifically, Prentice explained, based on the 25(OH)D vitamin D biomarker, the working group concluded that evidence indicated a risk of bone health problems below a particular level. Recognizing that most people would exceed that level, certainly in summer, the committee also recognized that there are ethnic minorities in the United Kingdom with no seasonal variation in 25(OH)D. To address the risk of bone health problems across the entire population, the committee set what it called a “population protective level” of 25(OH)D for the entire population to be above year-round. The committee used experimental intake–response data from studies commissioned by the UK government to develop this value. Its abandonment of the traditional framework in the case of vitamin D raises questions around terminology, particularly the acceptance of new terminology, that Prentice thought worth considering during this meeting.

In conclusion, Prentice mentioned that in addition to their recent vitamin D publication and forthcoming sodium/potassium publication, the committee is also currently working on saturated fat, folate, iodine, complementary infant feeding and cognitive development, and health in old age. Additionally, she expressed curiosity about how consideration of chronic disease endpoints will move forward and opined that although current DRVs are largely based on avoidance of deficiency, chronic disease is not a separate challenge. Finally, she emphasized, “I have found the biggest problem has been, in many ways, trying to interpret the terminology that we as nutrition scientists and nutritionists use to my clinical colleagues and, indeed, to my toxicological colleagues.” She observed, in addition, how various agencies are hesitant to accept not just new terminology, but also new models.

DISCUSSION

Following the panel presentations, the panel was opened to discussion and questions from the audience, as summarized here.

New Terminology: Acceptance and Understanding

In response to Prentice’s remarks about the acceptance of new terminology, Clifton commented that when there are huge differences in terminology among Europe, the United Kingdom, and the United States, it is difficult to know which terminology is the best one to accept. Most people choose to continue using what they have been using, he observed. However, if there is agreement on a specific terminology, he said, “Australia would certainly follow that.”

Paik added that most people want very clear communication around dietary standards. When South Korea changed from RDA to DRI terminology, it took a lot of work to convince people, including nutritionists and dietitians, that the change was needed. An added challenge in non-English speaking countries, she pointed out, is the development of new words to accompany changed English terminology. But, she said, “I think we have to all realize that things evolve, our knowledge evolves, our thinking evolves.” Communicating these changes is challenging, but necessary.

Change in UK Vitamin D Recommendation: Real-World Impact

Clifton wondered whether, since the United Kingdom’s new vitamin D guidelines were published, recommending 10 micrograms per day across essentially the whole population, there has been any conspicuous change in food fortification. Or, have pharmacies noticed an increased uptake of vitamin D supplements? He said that he was asking, essentially, “Does what the scientists and the nutritional community advocate have any effect in the real world?”

Prentice replied that it has been too short a time to see whether there has been a change in intake. Those data will come from the United Kingdom’s continuous National Diet and Nutrition Survey, and the most recent data are not available yet. However, she said, there has been a change in the way that government recommendations are being published. But there has been no change in thinking about vitamin D fortification, she added. She mentioned the folic acid recommendations that were made 20–25 years ago. Today, she said, “that one is still rumbling on. I’m afraid nothing happens.”

Clifton commented on the very positive press coverage of the new

vitamin D guidelines. He said, “You would think people reading that might think, ‘I’d better go and get some vitamin D from the supermarket.’”

Prentice agreed that vitamin D will be more available in the United Kingdom than it has been in the past. In the past, it was available only through prescription. Now it can be bought off the shelf. However, she added, based on past UK survey data, it is what she referred to as the “worried well” who take supplements even though they do not need them, and the people who really need them who do not take them.

Hutchinson commented that, in Canada, based on 2015 data, they should know within the next year whether there have been any changes in vitamin D intake patterns. They are ready to move forward on fortification, he said, if necessary, but they want to have an idea of what the consumption patterns are before they make any changes.

The Value of a Regional Approach to Setting NIVs

John Muyonga pointed out that one of the benefits of harmonization for developing countries is the sharing of resources and the opportunity for regional, as opposed to national, nutrition intake values (NIVs). Currently, many countries are using FAO and the World Health Organization guidelines because they do not have the resources to put together national NIV committees. Referring to Paik’s mention of the 78 members on the committee that developed the 2015 KNRVs, Muyonga said that some countries do not even have 78 nutrition experts.

Paik agreed that the development of some kind of regional cooperation would be very helpful, given that diets and nutrition-related health problems are often similar across a region. For example, when South Korea developed its first set of KDRIs in 2005, at that time Japan was similarly adopting the DRI terminology. The two countries did not share the output of their work, partly because the KDRI work was done by the Korean Nutrition Society, a nongovernmental organization of scientists, whereas Japan’s DRIs were set up by the governmental National Institute of Nutrition and Health. However, she said, “We had some communication.” When asked by Clifton whether there had been any collaboration with China, given all of the nutrition intake data being collected in China (e.g., the PURE Study), Paik replied that, at the time, China had not yet established DRIs. She predicted that, in the future, there would be some communication.

Additionally, acknowledging that she was unsure who would be responsible for running such resources, Paik suggested that each country contribute to an international research literature repository so the international community has at its disposal not just English literature, but local literature as well.

Hutchinson reiterated that, while Canada and the United States obviously take a bit of a regional approach, entering usual intake data from both countries into the report analyses, risk management is very much handled by the individual countries. “We can sometimes have different approaches using the same data,” he said.

Anna Lartey observed that while it is the developed countries that are really working in this area, every country has its own national academy of science. She called for efforts to work with these national academies so each country can develop its own original nutrient intake recommendations and suggested that the U.S. National Academies could play a big role in this.

South Korean Change in Carbohydrate Recommendation

During her presentation, Paik had described how the upper range of the carbohydrate AMDR was lowered to 65 percent in the 2015 KDRIs, compared to 70 percent in the 2010 KDRIs. She also said the upper range of the lipid total fat AMDR for adults 18 years and older was increased to 30 percent, compared to 25 percent in the 2010 KDRIs. Muyonga asked why the carbohydrate and fat AMDRs were changed in the 2015 KDRIs. Paik replied that the former UL of 70 percent for carbohydrates was very high, given many recent reports of the associated risks for chronic disease. Plus, there has been a trend toward decreased carbohydrate consumption. In addition to lowering the UL of the recommended carbohydrate distribution, they also felt that it would be appropriate to modify the ratio of fat to carbohydrates for adults.

Models for Recommendations for Infants and Children

When asked by Inga Thorsdottir how the UK Scientific Advisory Committee on Nutrition established intake recommendations for infants and children, Prentice replied that because there is relatively little evidentiary data for children those recommendations are made generally by extrapolation and by weight. But many of the committee’s recent deliberations have not even required that, she added. For example, while carbohydrates and their energy report certainly did, there is no evidence to suggest that there is any difference in the requirement for vitamin D. For Prentice, the question raised yet another question, one regarding physiological variance in general. There is so little information on variation in physiology, she lamented, with most available data being based on means.

Paik added that, in Korea, one of the 2015 KDRI subcommittees was for infants, because infants’ diets differ than those of other age groups.

Developing Food-Based Dietary Guidelines

Rune Blomhoff remarked that different food-based dietary guidelines (e.g., those developed by the American Institute for Cancer Research or the World Cancer Research Fund) are developed using different methodologies. He asked how nutrient recommendations can be integrated with these different food-based dietary guidelines. Hildegard Przyrembel agreed that NRVs are a prerequisite to food-based dietary guidelines. She referred to a 2006 FAO paper and European Food Safety Authority (EFSA) opinion piece on the principles of formulating food-based dietary guidelines. The EFSA scientific opinion, she noted, provides a stepwise approach. She emphasized, however, “You cannot globalize it.” It can only be done on a national basis. Hutchinson added that many countries are in fact following the FAO model, which, he explained, incorporates the EFSA stepwise approach, for developing food-based dietary guidelines.

This page intentionally left blank.