Eunice Kennedy Shriver National Institute of Child Health and
Human Development (NICHD) Scientific Vision
Developmental Origins of Health and Disease Workshop
Bethesda, MD
February 14-15, 2011

Views expressed in the workshop presentations are those of the
presenters alone and do not necessarily reflect those of the NICHD.

The overall aim of this workshop is to identify collectively the richest areas of opportunity consistent with the mission of the NICHD that will help develop a clearer understanding of early-life "environmental exposures" and genomic counterparts that affect developmental health and lead to early- and later-life disease.


Participants aimed to focus their discussions on programmable phenotypes -the area where genetic variation, environmental exposure, and epigenetic variation overlap.It is increasingly evident that adverse environmental conditions during development are associated with disease risks in childhood and adult life. Thus, the diseases and disorders that are the most taxing to developed countries’ economies and our health care delivery systems, including obesity, hypertension, coronary heart disease, type 2 diabetes, osteoporosis, and breast cancer, have well-established roots in periconceptional, embryonic, fetal, and childhood development. The NICHD identifies, as part of its mission, the promotion of health and well-being of mothers and children and is thus well placed to lead the world in discovering the early-life factors that give a child a healthy start or, alternatively, lead to a life of disease.

The Developmental Origins of Health and Disease theory states that gene-environment interactions play a key role in determining an individual’s phenotype for life. The enduring effect of this "programming" of phenotype is likely due to environment-stimulated changes in gene expression patterns that depend on an individual’s inherent genomic and epigenomic background. The "developmental origins" concept underlies and supports nearly all of the areas outlined in the NICHD visioning process and thus serves as a common link among workshop groups.

Back to Top

Where can NICHD lead NIH?

Programming of the phenotype, in its most essential form, consists of gene-environment interactions yielding either healthy phenotypes or phenotypes with high risks for disease. While the role of early-life environmental exposures in the etiology of adult cardiac, hypertensive, malignancy, brain, and other disorders is increasingly recognized at the population level, it is not yet possible to predict the risk of such disorders in an individual’s lifetime on the basis of his or her fetal or childhood health or exposures. Explanations for the variability in phenotype between and within "normal" and "disease" states resulting from gene-environment interactions remain largely unknown.

The NICHD has led the medical and scientific communities in clinical, translational, and bench top scientific approaches to tackling common etiologies of some of our most prevalent diseases, as well as investigating normative processes and protective factors.. These have included preterm birth, perinatal health and nutrition, and developmental disorders such as autism and learning disabilities. Largely clinical, research-based investigations in these areas have led to discoveries of what exposures influence risk of maternal and childhood health or disease. For example, Sudden Infant Death Syndrome research, which took into account maternal risk factors, led to the Back to Sleep campaign and a risk reduction in infant death. As other pertinent examples, we now understand that interventions such as single-course maternal glucocorticoids improve neonatal pulmonary disorders in instances of preterm birth. However, we still don’t know why such interventions work or what molecular mechanisms are involved in them.

Back to Top


7:30 a.m. Registration and Breakfast
8:00 a.m. Welcome
Alan Guttmacher, M.D.
Director, NICHD
8:15 a.m. Review of Agenda, Overview of the NICHD Vision and Charge to Participants
8:30 a.m. Keynote Presentation: Overview of Developmental Origins of Health and Disease
David Barker, M.D., Ph.D.
University of Southampton, U.K.
9:15 a.m. Plenary Presentations
  • Epidemiology and Outcomes
    Janet Rich-Edwards, D.Sc., M.P.H.
    Harvard University
  • Environment
    Mark Klebanoff, M.D., M.P.H.,
    Nationwide Children’s Hospital
10:15 a.m. Break
10:30 a.m. Plenary Presentations Continue
  • Genomic-Epigenomic Adaptations
    Sherman M. Weissman, M.D.
    Yale University
  • The Role of the Microbiome and Metagenomic Assessment in Discerning the Developmental Origins of Disease  
    James Versalovic, M.D., Ph.D.
    Baylor College of Medicine
11:30 a.m.  Panel Discussion
11:45 a.m. Charge to Breakout Groups
12:00 p.m. Working Lunch
1:00 p.m. Breakout Group Sessions

For each set of four Key Concepts, two independent breakout groups will use the listed eight key questions (KQs), which may have been modified by breakout-session co-leaders, as spurs to discussion and identification of research opportunities. Each of the eight breakout groups includes a set of co-leaders and selected scientific representatives from the four science domains (epidemiology, genomics/epigenomics, bioinformatics/biostatistics, clinical/translational). Two breakout groups address each Key Concept.

The objective for each breakout group is to identify the most compelling research opportunities in areas of science within the NICHD mission and specifically within the workshop’s overall theme. The Key Concepts are provided to indicate the scope of each breakout group’s deliberations. For each Key Concept, examples of questions are provided in order to stimulate discussion. Workshop members should not feel obligated to answer these questions, however, and may, indeed, think of other questions, bearing in mind the objective of identifying research opportunities for the coming decade.

  • Key Concept 1: Early-in-life events affect health and disease. However, the spectrum of what defines "normal variants" versus "pathologic disease" in development is highly dynamic. In other words, while human translational research is sorely needed to provide basic information and navigational cues for investigations at the molecular level, we currently lack meaningful studies that will allow us to better determine what defines “health,” a “normal variant,” and “disease.” (SIGNIFICANCE)
    • Key Concept 1 Session Leaders:
      • James Versalovic, M.D., Ph.D.
        Baylor College of Medicine
      • Aaron Caughey, M.D., Ph.D., M.P.H.
        Oregon Health Sciences University
      • Matthew W. Gillman, M.D.
        Harvard University
      • Sherman M. Weissman, M.D.
        Yale University
    • Examples of KQs that may be Important to Ask:
      • KQ1: How can we determine what constitutes "disease" versus "health" in a developmental and temporal context? Lifetime cost burden? Lifespan longevity? Disease-free intervals across the lifespan? Loss of adult productivity? How can we systematically design studies to generate consensus that will enable investigators, and ultimately clinicians, to identify meaningful thresholds in childhood for later-in-life adult disease risk?

        For example, when defining “prematurity,” what research mechanisms would enable us to define a gestational age threshold after which there might exist occasional morbidity and mortality, but generally not? Similarly, what research mechanisms would allow us to arrive at a threshold for childhood obesity whereby adult morbidity is manifested?
      • KQ2: When are evolution and adaptation maladaptive, and how does "programming" vary from "selection"? Can we use comparative biology and unique populations in our approaches to answering these questions?

        For example, can we learn whether prematurity is a disease or an adaptive advantage by studying genomic variants across primate species? What can we learn about twins by studying sheep and marmosets (obligate twinners)? Are there unique human populations with which we can optimally study the impact of maternal obesity?
      • KQ3: What constitutes normal neurodevelopment? Normal emotional development? What are the predictors? Covariates? Confounders?
      • KQ4: What constitutes normal growth and metabolism? What are the predictors? Covariates? Confounders?
      • KQ5: What constitutes normal reproduction? What defines infertility? Subfertility? Recurrent pregnancy loss?
      • KQ6: What socioeconomic factors have significant adverse outcomes? Poverty level? Access to education? Exposure to smoking, alcohol, drugs, sex, violence, neglect? Is excess wealth a risk factor? Is there similarly a threshold effect? Temporal effect?
      • KQ7: Can we use genomic variants (common or rare) to define health or disease?
      • KQ8: What common set of computational and statistical tools can we employ to define “health” and “disease” thresholds? Is there a temporal element to these definitions?
  • Key Concept 2: Environmental influences have a measurable impact on development. Extensive investigations in both human populations and animal models have established that there is a relationship between an aberrant in utero and early postnatal environment and later-in-life disease. Relevant factors include maternal nutrition, body composition, and comorbidities prenatally, alongside postnatal nutrition and environment. However, how can we engage in meaningful research in order to better define “environmental influence”? And how can we measure environmental impact? What approaches can we use to define an optimal in utero or postnatal environment? (SIGNIFICANCE AND APPROACH)
    • KC 2 Session Leaders:
      • Patrick Catalano, M.D.
        Case-Western Reserve University
      • Patricia Hunt, Ph.D.
        Washington State University
      • Rebecca A. Simmons, M.D.
        University of Pennsylvania
      • Michael G. Ross, M.D., M.P.H.
        University of California, Los Angeles
    • Examples of KQs that may be Important to Ask:
      • KQ1: What are the primary ambient exposures? How can they be measured or linked to outcome? What areas of innovation are available for their measure? Are there applicable thresholds outside of the Clean Air Act measures?
      • KQ2: What are the primary nutrient exposures? How can they be measured or linked to outcome? What areas of innovation are available for their measure?
      • KQ3: What are the primary social, economic, and behavior exposures? How can they be measured or linked to outcome? What areas of innovation are available for their measure?
      • KQ4: What are the most common teratogens? Alcohol? Cigarette smoke? Drugs for maternal or paternal comorbidities? What are the predictors? Covariates? Confounders? How can they be measured or linked to outcome? What areas of innovation are available for their measure?
      • KQ5: What constitutes "optimal" developmental nutrition? What supplements are of maximal benefit? Modest benefit? Detriment?
      • KQ6: What defines an optimal preconception and periconception environment? Gamete health? Paternal health? “Environmental health”? What are its primary impediments? How do we measure them or link them to outcome?
      • KQ7: What defines an optimal postnatal environment? Education? Nutrition? Nurture? Absence of detrimental common influences? Health and disease-free interval?
      • KQ8: How can we best leverage existing or underway studies or clinical networks (e.g., the National Children’s Study, the Human Microbiome Project) to better define environmental influence?
  • Key Concept 3: At the crux of development and the programmable phenotype is the ability to discern complex molecular interactions. How do we study these interactions in new ways and leverage available and emerging technology? What human epidemiology and animal biology studies are needed to provide basic information and navigational clues for investigation at the molecular level in areas that are now, effectively, black holes (especially physiological adaptations to pregnancy; maternal nutrition, body composition, and fetal outcomes; gamete health; periconceptional environment; mother-placenta-fetal signaling; placental gene/growth regulation; fetal organ-specific growth, environmental sensitivities, and developmental windows; embryonic/fetal nutrient requirements; regulators of developmental plasticity; growth and nutrient requirements in infancy and childhood; nutrition and Central Nervous System [CNS] and Autonomic Nervous System [ANS] development; and preterm birth)? (SIGNIFICANCE, INNOVATION, AND APPROACH)
    • KC 3 Session Leaders:
      • Tom P. Fleming, Ph.D.
        University of Southampton
      • Kelle H. Moley, M.D.
        Washington University in St. Louis
      • Claire Roberts, Ph.D.
        University of Adelaide
      • George R. Saade, M.D.
        University of Texas
    • Examples of KQs that may be Important to Ask:
      • KQ1: What genomic approaches are most optimal? “Triad” studies of maternal, paternal, and child genomics? What is the heritable genotype? Copy number polymorphisms? mtDNA variation? Ethnic and racial haplotype variants?
      • KQ2: What tools have been developed that can be readily applied? Which need to be further developed? Array-based tools? NexGen sequencing?
      • KQ3: Would we benefit in these approaches from core focused initiatives? RoadMap efforts? Core centers for genomics discovery?
      • KQ4: How do we decipher the developmental epigenome? What is actually heritable? The methylome? Hostone code? miRNA code? Epigenotype?
      • KQ5: How do we decipher mechanisms of inheritance of the epigenome? Can we take into account copy number imbalances and single nucleotide polymorphisms (SNPs)?
      • KQ6: How modifiable is the epigenome?
      • KQ7: Can we develop the computational tools to readily layer genomic discovery with transcriptomics? Proteomics? Metabolomics? What tools do we need to develop in order to do so? How do we deal with complex metadata? Is machine learning optimal and readily employable?
      • KQ8: What research, diagnostic, or therapeutic tools need to be developed or applied to plasticity in order to make substantial clinical progress?
  • Key Concept 4: Policy and social issues are important variables that influence our efforts in research and affect the potential for application of our scientific progress in each of these areas. How can we formally come to study or manage their impact? (APPLICATION AND TRANSLATION)
    • KC 4 Session Leaders:
      • Michael W. Varner, M.D.
        University of Utah
      • Larry Wallack, Ph.D., M.P.H.
        Portland State University
      • Nancy Press, Ph.D.
        Oregon Health Sciences University
      • David Washburn, Ph.D.
        Georgia State University
    • Examples of KQs that may be Important to Ask:
      • KQ1: Are there institutional and policy barriers to discovery?  How can they be overcome?
      • KQ2: Are there institutional and policy barriers to translation of discovery? 
      • KQ3: Is our workforce composed of the right balance of individuals for the execution of this field of research?
      • KQ4: Is our workforce composed of the right balance of individuals for the implementation of this field of research?
      • KQ5: Are there barriers to research in pregnancy that ought to be considered?
      • KQ6: Are there barriers to animal models that ought to be considered?
      • KQ7: Do we strike an appropriate balance in funding clinical networks and trials with discovery or hypothesis-driven translational research? Do we optimize sample collection in clinical trials?
      • KQ8: Do we effectively communicate discovery in this area to the public? Are we culturally appropriate?
2:30 p.m. Break
2:45 p.m.. Breakout Group Session Continues
5:00 p.m. Break and Extended Discussion
6:30 p.m. Dinner
7:30 p.m. Breakout Groups Prepare Statements and Presentations
9:30 p.m. Adjourn

Back to Top


7:30 a.m. Breakfast
8:00 a.m. Review of Agenda
8:15 a.m. Breakout Groups Reconvene to:
  • Summarize discussion in preparation for presentation
  • Create presentation of summary statements
9:15 a.m. Break
9:30 a.m. Breakout Group Recommendations
  • Key Concept 1
    • Breakout Group 1
    • Breakout Group 2
    • Discussion
10:40 a.m. Break
10:55 a.m. Breakout Group Recommendations Continue
  • Key Concept 2
    • Breakout Group 1
    • Breakout Group 2
    • Discussion
12:05 p.m. Lunch
1:05 p.m. Breakout Group Recommendations Continue
  • Key Concept 3
    • Breakout Group 1
    • Breakout Group 2
    • Discussion
2:15 p.m. Break
2:25 p.m. Breakout Group Recommendations Continue
  • Key Concept 4
    • Breakout Group 1
    • Breakout Group 2
    • Discussion
3:35 p.m. Closing Statements
3:50 p.m. Workshop Adjourns

Back to Top

Last Reviewed: 03/15/2011