Session 6 was moderated by Barbara Schneeman, Professor Emerita at the University of California, Davis, and chair of the Planning Committee. The session opened with a panel that provided perspectives on the workshop’s presentations from a variety of aspects. These presentations were followed by an open discussion and questions and answers with workshop participants. The session concluded with brief remarks by a speaker from each of the workshop sponsors. These speakers thanked the Planning Committee, presenters, and participants for a stimulating and thoughtful workshop and highlighted themes that complement their own organization’s mission and initiatives.
Before introducing the first panel, Schneeman briefly summarized a few of the key themes that stood out to her over the course of the presentations and discussions. One theme that repeatedly emerged was the importance of understanding the difference between approaches for treating, curing, or preventing disease, and approaches for managing a disease so that the person’s health improves by compensating for what the disease is doing to nutrient metabolism. Sometimes a drug is involved with managing disease, but sometimes the management issues are related to different nutrient requirements. Another theme for Schneeman was that as the nature of a disease is understood more fully, the ability to define the role that nutrition plays in that disease process becomes more understandable. Another issue that came up repeatedly was the importance of defining nutritional status. Finally, consideration should be given to lessons learned from instances where nutritional status is well defined
so they can be applied to other cases where nutritional status is not well defined. A related idea is the evolving understanding of what is considered a nutrient and the use of the term “essential nutrient” versus “conditionally essential nutrient.” In several sessions, particularly around discussions about intestinal diseases, the focus is shifting from specific nutrient requirements to the broader concept of special dietary requirements or special dietary patterns. In addition, although advances in the tools used to study disease and nutrients are emerging, investigators and clinicians are still working with some old paradigms. In that sense, the use of these emerging tools as part of the repertoire of approaches for looking at special nutrient requirements could be explored, Schneeman observed.
The panel on future opportunities had four members. Susan Barr, Professor Emeritus of Food, Nutrition, and Health at the University of British Columbia, reflected on the workshop presentations from a perspective of the Dietary Reference Intakes (DRIs). Kristen D’Anci, Associate Director in the ECRI Institute’s Evidence-based Practice Center and Health Technology Assessment group, provided remarks from the perspective of building clinical recommendations, evidence-based recommendations, and clinical guidelines. Timothy Morck, Founder and President of Spectrum Nutrition LLC, provided an industry perspective, and Virginia Stallings, Children’s Hospital of Philadelphia, reflected on the workshop from her experience and perspective as a clinician.
Barr posed the following key questions that the workshop presentations raised in relation to DRIs:
- Dietary requirements for what? People talk about “the requirement” for a nutrient, but in fact, Barr said, there are many requirements for any given nutrient, depending on the selected indicator of adequacy. This question has not been addressed explicitly in most of the DRIs1 to date, but one example is vitamin A. The Estimated Average Requirement (EAR) for vitamin A intake in North America is based on adequate liver stores, and was set at 500 mcg/d for women and 625 mcg/d for men. The DRI report, however, includes a statement indicating that an EAR based on correction of impaired dark adap
1 For DRI values, see http://nationalacademies.org/HMD/Activities/Nutrition/SummaryDRIs/DRI-Tables.aspx (accessed June 6, 2018).
tation can be calculated as 300 mcg/d, and might be appropriate in less developed countries where vitamin A may not be as widely available (IOM, 2001, pp. 121–122).
- Requirement for what disease or health status? Are we looking for the requirement based on the same indicator of adequacy as was used in the apparently healthy population? Or would it be a different indicator of adequacy that is somehow adversely affected by the disease? Or an indication of reduction of disease progression that acts through altered nutrient needs?
- How are requirements and recommended intakes differentiated? Nutrient requirements are for individuals, but individual nutrient requirements do differ. It is very difficult for a clinician to determine an individual’s requirement. It is possible to determine the average requirement (the EAR), but that meets the needs of only half the individuals in an age/sex group. If the goal is to cover the needs of almost everyone with a given disease, variability in requirements should be considered and recommended intakes identified.
- What happens to the requirement distribution in disease? It is possible that the presence of disease could simply shift the position of the requirement distribution by a fixed amount. However, it is much more likely that both the position and the variability of the requirement distribution may change due to factors such as disease severity, other comorbidities, inflammation, effective treatment, drugs, and surgery.
- What about upper levels of intake? A so-called safe range of intake has been established between the Recommended Dietary Allowance (RDA) and the Tolerable Upper Intake Level (UL), where both the risk of not getting enough and the potential risk of excess are low (see Figures 1-1 and 1-2). The magnitude of that safe range is not constant across nutrients, however. It is possible that the UL in a disease state might increase in parallel with the increase in the requirement, maintaining a similar “safe range” of intake. However, it is also possible that the UL in a disease state would not change if it were based on an adverse effect that was not affected by the disease state.
- How should competing risks be managed? If the disease shifts the requirement distribution to the right so that more is needed to achieve adequacy, a group of people with the disease would possibly be receiving levels above the UL, and that would place a proportion of that group of people at risk of adverse effects. How should those competing risks be managed?
- Is it possible to individualize? If sensitive, specific, non-invasively obtained, and inexpensive biomarkers were available, it might be
possible to titrate an individual’s intake of the nutrient so that he or she could meet an individual requirement.
Clinical practice guidelines are statements that include recommendations to optimize patient care. They are informed by a systematic review of the evidence and assessment of the benefits and harms of alternative care options. D’Anci identified the following challenges in translating the research:
- Nutrition interventions are complex and must be considered in the context of the diet as a whole. Unlike with drug interventions, nutrients are not being given in a vacuum. People are eating, doing other things, and they may not be compliant with the intervention.
- Many of the disease states discussed at this workshop are rare. A small patient population has the additional complication of heterogeneity due to issues such as age, disease state, other factors in the diet and environment, and genetic makeup.
- The current evidence base is extremely limited. A systematic review that is done to support a guideline must be done in a way that facilitates decision making and should use a best-evidence approach, which could potentially include using evidence from case series and case reports. Such systematic reviews differ from those that limit their inclusion criteria to randomized controlled trials (RCTs), which are done to determine the causal link between an intervention and outcome. Once the systematic review is completed, the guideline panel works with the evidence and also considers many issues besides just the evidence itself, including the balance of benefits and harms, patient values and preferences, and other factors. The panel then uses all of these factors to make a recommendation.
- Determining how to assess the quality of the evidence is challenging. Grading of Recommendations Assessments, Development and Evaluation (GRADE) is a commonly used framework in guideline development. For much of the research described in this workshop, the quality of evidence may actually be graded up based on effect sizes, generalizability of the patient population, adequate ascertainment of exposures and outcomes, dose response, alternative explanations of causality, and length of follow-up.
D’Anci observed that a number of groups are considering new paradigms for determining adequacy of evidence, such as rethinking the concept that RCTs are the best designs and considering other rigorous
study designs in supporting guidelines. Some groups are even considering ways of evaluating evidence from animal studies.
Morck reviewed the following ideas that emerged from the workshop from his perspective:
- It is important to document nutrient needs and establish therapeutic dose and composition, so that medical nutrition therapy for disease states can be standardized.
- The patient bears the burden of the disease. The more industry and researchers can accelerate the process of getting effective “therapeutic nutritional” products into the hands of patients who are already under the supervision of a doctor who has recommended them for their specific condition, the better the outcomes will be.
- The workshop has elevated the role of nutrition as a central part of cost-effective patient care. Many of the products that can be used in nutritional therapy can reduce symptoms and complications of diseases associated with nutritional imbalance, with a larger benefit–cost ratio compared to drugs that are being used to manage the disease. Such drugs, by design, may be relatively toxic, but seldom contribute to the body’s healing or restorative processes, as do nutrients.
- A clear scientific process that is able to establish distinctive nutritional requirements in a disease state will promote the rapid development of an appropriate regulatory framework to guide industry in developing safe and efficacious products that benefit patients. Faster progress in identifying and validating appropriate nutritional biomarkers is imperative.
- How do we get solutions, now? The food industry and the scientific community are both full of good ideas, but it will be important to ensure that these nutritional innovations are consistent with clinical practice guidelines to modulate the disease progression, the severity of the disease, the side effects, or the interaction with drugs and treatment therapies to improve overall nutritional status.
- Decisions about regulating medical foods should consider the fact that medical foods require initial recommendation and ongoing supervision by health care professionals. This ensures proper monitoring of outcomes and continuous modifications of a particular intervention. Therefore, the level of evidence that is needed for medical food regulations could be different from the level of evi
dence for other public health decisions that do not involve direct medical supervision.
Morck concluded his remarks by saying that the workshop highlighted the importance of efforts to increase the priority of funding for this arena of nutritional research.
Stallings rounded out the panelist remarks with the following observations:
- Clinicians have little information and feel hamstrung in trying to use evidence-based decisions to guide them in caring for their patients.
- Clinical judgment is crucial in making clinical guidelines in the absence of robust evidence. Clinicians have the responsibility to keep moving the field forward and include the new thinking about study design, data collection, and evidence standards in clinical care.
- Until recently, the Dietary Guidelines for Americans and the DRI process have remained relatively silent on chronic disease issues. However, more than 50 percent of the adult population now has a chronic disease. This new paradigm needs considerable work when thinking about nutritional requirements and the DRIs.
- Clinicians are faced with the task of identifying a variety of clinical decisions and actions that might contribute to disease prevention or that might be useful in managing the disease or preventing progression. How does nutrition fit into those actions?
- The difference in nutritional needs between an acute illness event and those for an ongoing chronic illness presents a real clinical and research opportunity. Most clinicians think about nutrition in the setting of chronic diseases, but could clinicians use nutrition-related therapy in the emergency room to manage, for example, sickle cell pain crises?
Schneeman opened the discussion by referring back to the points made in the previous session (see Chapter 5) about various kinds of tensions, including those between the nutrient focus and the dietary focus, the genetic focus and the phenotypic focus, individuals and populations, final health outcomes and biomarkers in the pathway, and acute and
chronic diseases. Such tensions can be used to advance scientific and clinical understanding by identifying the issues that need to be resolved in order to make progress.
Panelists responded that they were hopeful that the workshop would lead to action. They also concurred that much of the tension derives from a lack of understanding of what question is being asked, and time spent in dialogue would be beneficial.
Defining the Terms
Other tensions identified by the panelists related to defining terms (e.g., “nutrient” or “chronic disease”) and to differences between the scientific approach and the clinician approach. The clinical principal of “do no harm” must be central to decisions about how to apply new evidence to clinical settings. An example of tensions or questions related to terminology is the need for a term that might encompass bioactives, such as phytonutrients, and other non-nutrient compounds so that they might be considered for inclusion in medical foods.
In terms of defining special dietary requirements, situations where patients have nutrient needs within the DRI range, but perhaps with a narrower range of tolerance, were raised. In this case, a way of defining “special” or “distinct” nutrient, would not mean a shift in the DRI distribution curve of requirement. This discussion led to the needed distinction between a recommendation for a group versus a recommendation for a specific individual. In the context of the DRI, some groups that might have distinct nutrient requirements, such as smokers and vitamin C or vegetarians and iron, have been considered. In the context of a disease state, if a person’s tolerance of a nutrient is very low, that requires individual monitoring.
What Constitutes Sufficient Proof?
The panelists reflected on a question about what constitutes proof for a distinctive nutritional requirement, given the need by patients, clinicians, and industry to have a regulatory framework for nutritional therapy. Companies can create any medical foods, but clinicians need evidence of effectiveness for nutrients or other bioactive compounds. Schneeman stated that as the science has advanced because of better genetic and metabolic tools, nutrition requirements beyond those of inborn errors of metabolism are being considered.
Susan Barr was asked to comment on whether the DRI category of Adequate Intake (AI) could help address the dilemma of a limited evidence base. Barr replied that the AI has been a challenge in terms of
determining what it really means and how to interpret intakes that are below the AI. Although healthy individuals who meet or exceed the AI for a nutrient can generally be assumed to meet their requirements, no assessment can be made when intakes are below the AI. Some of the AIs were set using median intakes of a healthy population. In this setting, half the population would have intakes below the AI, but the prevalence of inadequacy (i.e., not meeting one’s requirement) could be extremely low. However, it is possible that for disease states for which the evidence base is limited, a different term could be identified.
Paul Coates, National Institutes of Health
Coates commented on three concepts repeated throughout the workshop: evidence, uncertainty, and complexity. From his perspective, gathering the evidence to clarify requirements different from those for the general population is key. The evidence that a clinician intervention with a nutritional approach may make a difference in patient-reported outcomes and clinical outcomes is still insufficient. Biomarkers may be able to help provide the evidence to estimate the impact of an intervention on an outcome.
Entirely related to that evidence question is the issue of certainty or uncertainty. Past DRI efforts have acknowledged that uncertainty is associated with those values. The use of GRADE and other systems to assess the quality of intervention trials, observational studies, and animal studies would greatly help in assessing the uncertainty of estimates.
The third issue is complexity. Inborn errors of metabolism, in all of their complexity, are a harbinger of what is in the future as principles of nutritional management are translated from specific disease states to broader chronic disease indications.
Finally, innovations in identifying people at risk, developing strategies for assessing the impact of interventions, and analyzing data are all needed to be able to make recommendations in the context of increasing disease rates in populations.
Patricia Hansen, U.S. Food and Drug Administration
Hansen noted that the workshop cuts across the U.S. Food and Drug Administration’s (FDA’s) mission areas, including medical foods, foods for special dietary use, and even some specialized infant formulas; dietary supplements; and drugs as well. This is very important as FDA carries out its role in basing sound policies and regulations on sound, strong, and consensus science.
Hansen expressed her appreciation for the thoughtful presentations on a wide spectrum of diseases and the foundational presentations, which discussed how using lessons learned from the DRIs, both positive and negative, can be applied from those experiences to considering the role of nutrition in disease. She expressed hope for the workshop themes to stimulate the research and clinical community to fill these important gaps that will be essential in efforts to move forward with a modern framework that incorporates the best of science.
Caren Heller, Crohn’s & Colitis Foundation
Heller stated that the Foundation welcomed the opportunity to support this workshop, because nutrition and diet are so important to patients with these diseases. Ways to support inflammatory bowel disease (IBD) patients within the arenas of both diet as therapy and diet to maintain nutritional health are challenging, particularly because so many are at risk of malnutrition.
In terms of diet as therapy, Heller said, it is clear that more science is needed to support interventions. In addition, regulatory and reimbursement hurdles make it difficult for patients to access enteral nutritional products. Heller noted that the Foundation is working with groups like the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition, and the American Gastrological Association, to engage on legislative acts to improve access. She highlighted other specific questions, such as what kind of study designs and scientific data would guide FDA in defining a more streamlined regulatory approval and affect reimbursement decisions and what research studies the Foundation can be engaged in to further the understanding of special nutritional requirements in IBD to facilitate the regulatory framework.
Chantal Martineau, Health Canada
Martineau explained that the Bureau of Nutritional Science at Health Canada has the responsibility to develop and update regulations related to foods for special dietary use. These include meal replacements, nutritional supplements, medical foods (e.g., enteral products), and products developed for low amino acid and low protein diets.
Gaining a better understanding of the evidence base available related to variations in nutritional requirements associated with various diseases and conditions is an important first step to potentially establish a framework for determining special nutrition requirements, she said, although much work still needs to be done.
In Canada, regulations for most foods used for special dietary use
include detailed compositional requirements for micronutrients. Canada has minimum and maximum levels for macronutrients and vitamins and minerals, along with protein quality requirements, and these compositional provisions are very strict and based on requirements for the healthy population. Canada has initiated the process to review and update the regulations, as they were all established in the 1980s and 1990s, and Health Canada intends to consider the evidence presented at this workshop as it updates the regulations.
Sarah Ohlhorst, American Society for Nutrition
Ohlhorst described that in 2015, the American Society for Nutrition (ASN) brought together a working group, the Food for Health Initiative, with members from the academic community, nutrition and health organizations, patient groups, industry stakeholders, and many others to look at nutritional requirements for those living with disease or certain medical conditions from a nutrition science perspective.
A nutritional requirement to meet or restore desired health outcomes for one individual may vary quite drastically from that of another individual living with the same disease or medical condition, and so there are many considerations at play. More research is certainly needed to help elucidate the evidence base behind these special nutritional requirements so that next steps for a path forward can be determined. ASN plans to continue the dialogue and continue having these discussions with its Food for Health working group and with any other interested parties.
Alison Steiber, Academy of Nutrition and Dietetics
Steiber commented that the Academy of Nutrition and Dietetics (AND) practitioners care for individuals with many different diseases. AND has a major interest in understanding how nutrition can improve health outcomes, particularly for at-risk individuals and to work with those patients and their caregivers to help them understand their diet as well as to help them make a diet that is palatable and will meet their chronic disease needs. Echoing other speakers, Steiber stated that biomarkers for nutrients are needed that are rapid, feasible in clinical settings, and cost-effective. Practitioners must be given practical yet evidence-based guidelines on which they can look at macro- and micronutrients and dietary patterns for individuals living with disease states.
AND has recently completed both the scoping for evidence of nutrition requirements in cystic fibrosis and a very large evidence-based practice guideline project in collaboration with the National Kidney Foundation. Using the GRADE methodology, macronutrients have some moderate and
high-level evidence recommendations. However, for micronutrient needs in the chronic kidney disease population at all stages, very few recommendations are above a low strength of evidence level.
Steiber noted that this process has helped to identify gaps in the evidence, but implementing recommendations is a challenge for practitioners. Finally, she emphasized that in lieu of good biomarkers, researchers could use (and document in the electronic medical records) nutrition-focused physical exams and composite nutrition diagnostic tools, such as subjective global assessment and malnutrition clinical characteristics, in their research across different states.
IOM. 2001. Dietary Reference Intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: National Academy Press.
This page intentionally left blank.