The comment period to submit feedback online is now closed.

To view previously submitted comments about Developmental Origins of Health and Disease, please see below.

Developmental Origins of Health and Disease White Paper (PDF - 281KB)

Created on May 3, 2011



17 Submitted Comments on Developmental Origins of Health and Disease
The idea of the white paper is great & intriguing, dreamt of by many but not thought feasible! I want to make 2 points 1. For public support, what affects people immediately is more effective - for e.g. building on the concept "need a healthy mother for a healthy baby". Consider setting up a National center for integration of local & state statistics gathering centers, assure reliability of data collection, and then annually summarize areas for study and improvement. This will provide invaluable information to the public, scientists and policy makers. A great example is the Center for Maternal and Child Enquiries in the UK.1 Such centers can be for preterm births; school drop-outs; mental disorders etc. With rising maternal age at conception and increasing medical complications in pregnancy, the focus should also shift to multidisciplinary management of the mothers (preconceptionally, antenatally and postnatally) to optimize in utero environment. Encouragement for generation of databases of such pregnancies could yield valuable information for long-term follow-up. 2. The current system promotes growth only within the system, with researchers well established at the NICHHD / NIH and their trainees able to gain better access. Much of it has to do with the vast experience of these researchers with the system; a ?track-record? established with pre-existing data and research; and availability of good mentorship only to those associated with these researchers. Encouragement and favoring of these ?experienced? researchers to mentor and guide researchers outside their institutions (especially with wide availability of the web and its tremendous capability to disallow geography as a limitation); would foster wider acceptability of research, more co-operation between institutions and faster acquisition of data. Some ?clinically-focused? institutions can be a rich source of data & ideas for the ?research-focused? institutions. REFERENCES: 1. Centre for Maternal and Child Enquiries (CMACE). Saving Mothers? Lives: reviewing maternal deaths to make motherhood safer: 2006?08. The Eighth Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118 (Suppl. 1):1?203.

Submitted by Meena Khandelwal on May 8, 2011 at 1:11 PM
This white paper clearly identifies an exciting area. But it unfortunately ignores a lot of the work being done on this topic by social scientists. Work in the social sciences differs from some of the work surveyed in the white paper in that it tends to focus on a broader array of outcomes, including social outcomes. For example, a social scientist might ask whether insults in utero such as exposure to famine or to radiation cause people to be more likely to be on social assistance or to get less education? Another exciting question is whether there are interactions between conditions in utero and socioeconomic status in the determination of longer term outcomes? Interactions of this type would suggest that the damage done to the fetus may be reversible (at least in terms of some of its societal impact) with appropriate investments. Social scientists have focused on using natural experiments and cohort designs (e.g. comparing cohorts affected by the 1919 flu epidemic to surrounding cohorts) in order to identify the causal impact of health in utero rather than focusing on correlations alone. A recent summary of some of the research in this area is available in Janet Currie and Douglas Almond, "Killing me Softly: The Fetal Origins Hypothesis," forthcoming in the Journal of Economic Perspectives (see http://www.nber.org/~almond/p1_JEP9.pdf). The NIH should be concerned with the way that socioeconomic status is affected by health, and by the way in which socioeconomic status and health insults interact to determine developmental trajectories and long-term outcomes.

Submitted by Janet Currie on May 23, 2011 at 1:07 PM
The inclusion of social science techniques and approaches would strengthen the white paper. Examples include: 1. Understanding the causal channels through which early life events and conditions influence health and disease (p. 1). Of particular interest are quasi-experimental designs providing credible identification of causal effects with confounding factors (e.g. previous research on birth during influenza epidemics or recessions). 2. Life course approach identifying late-life factors arising from early trajectories (p. 2). Economists have developed sophisticated tools for analyzing these types of issues using longitudinal and repeated cross-sectional data. Again, the major concern is in determining causation and, also, distinguishing effects influences occurring at different life-stages as well as interacdtions across stages (e.g. dynamic complementarity). 3. Develop trans-disciplinary systems science approaches for DOHaD (p. 3). As noted, "socioeconomic and behavioral factors may contribute to adverse outcomes, but these factors are ubiquitous in nature, subtle, and difficult to study". Social scientists should therefore be brought into the analysis efforts at the initial design stages, rather than being added as a supplement to the "main" scientific effort. Disease prevention efforts should include not only the specific factors mentioned on p. 4 (e.g. less maternal smoking) but also examination of fundamental incentives and environmental factors leading to unhealthy outcomes. 4. Long-term follow-up of clinical studies and establishment of an international clearinghouse of datasets (p. 7). Major medical datasets currently usually lack sufficient socio-demographic information, while important population-based data genearlly lack sufficient information on biomarkers and relevant clinical measures. The greatest benefits will occur if both types of data are obtained simultaneously. There are also likely to be considerable benefits from linking data across sources (e.g. administrative and clinical data). 5. Interactions between health determinants at different stages of life. The paper does not adequately consider how health determinants at the beginning of life interact with subsequent reinforcing/mitigating factors. Such efforts should be central to the scientific vision, rather than being a supplement. Attention also needs to be paid to the interactions between short-term factors (e.g. economic conditions) and early life conditions in affecting health.

Submitted by Christopher Ruhm on May 25, 2011 at 9:56 AM
The Endocrine Society appreciates the opportunity to provide feedback on the white papers developed from the Vision Workshops held by the NICHD. The Society lauds the transparent and reciprocal approach taken by NICHD in the generation of its scientific vision that will inform the Institute's direction for the next decade. A significant portion of the Society's basic researchers apply for funding through NICHD, and the NICHD Scientific Vision is of particular interest to our membership. Given the tremendous overlap between Society members' research interests and those of the NICHD, The Endocrine Society considers the Institute's scientific visioning process to be an important milestone for the continued advancement of endocrine science.

It is imperative that NICHD continue to fund investigator initiated, independent research into basic physiological processes surrounding developmental origins of health and disease. Continued support of basic research will provide the basis for advances in understanding these processes and will thereby lay the essential foundation for future disease prevention and treatment strategies.

The Society supports the mission presented in the Developmental Origins of Health and Disease white paper and urges NICHD to continue to fund investigations into basic physiological processes surrounding developmental origins of health and disease. While the white paper emphasizes clinical studies, it is essential to underline the importance of relevant animal models that probe molecular mechanisms vital to the research areas presented in the white paper.

The Society is pleased with the incorporation of environmental and nutritional components, as these are important factors contributing to endocrine disruption and dysregulation. Nutritional supplements should be included in this discussion as their use is on the rise in the United States, as should studies of the effects of fetal psychological stress in utero.

Again, the Society appreciates the opportunity to comment on the white papers made available by NICHD and recognizes the efforts of the workshop participants in compiling the documents that will shape the future of the Institute. Endocrine research is a vital component of expanding our understanding of the developmental origins of health and disease, and Society members are enthusiastic about potential opportunities that may be presented by the Institute as a result of its rigorous planning.

Submitted by Kelly Mayo on June 1, 2011 at 11:30 AM
There is increasing evidence that early life influences may contribute to bone health later in adulthood. There is not mention of early fetal and neonatal influences on muscle bone development.

Submitted by Bonny Specker on June 2, 2011 at 10:47 AM
The topic of developmental origins of health and disease is extremely important and will play an increasing role in addressing chronic diseases in the future. Musculoskeletal conditions should be included in this discussion. Conditions like osteoporosis and osteoarthritis have their origins during critical periods of development similar to the other classes of disorders mentioned. Environmental factors strongly influence bone and joint health and the eventual manifestation of musculoskeletal diseases in the elderly, which are extremely common.

A significant challenge is the difficulty of studying conditions that develop over decades. To truly realize the scientific opportunity of longitudinal analyses, NICHD needs to develop mechanisms that would allow longitudinal tracking of populations for many decades from childhood through old age.

Submitted by Tishya Wren on June 2, 2011 at 5:21 PM
The white paper addresses an important issue that should be a centerpiece of the NICHD agenda. It recognizes the need to move beyond science that focuses only on proximal biological determinants of health. Integrating contributions from the demographic, behavioral, social, and economic sciences is necessary to achieve the vision laid out by the authors and produce results with real-world impact.
The terms "environment" and "exposome" need careful differentiation and definition to enable scientists working at different environmental levels to collaborate. Expertise from psychology, sociology, economics, demography, and biological anthropology is crucial if the goal is truly to establish the mechanisms linking early life experience in complex external environments to the biology of healthy development.
Given the evidence supporting the role of psychosocial stressors in conditioning the development of neuro-endocrine pathways, stress pathways should be included as priority areas in the study of developmental origins of health and disease.
Behaviors -deciding what to eat, using tobacco -play a crucial rule in environmental exposures and should be accounted for in modeling biological responses to environmental cues. Behavior-environment relationships can be integrated in biological models using tools from the behavioral and social sciences.
The call for making use of new technologies in measurement is welcome, but industry-based methods need careful evaluation. Industry's goal is to sell products and make money, ours is to improve health. Work is already underway in the social sciences (some of it funded through the NIH's GEI program) to develop appropriate technology-driven tools for measuring environmental exposures in health research.
Dissemination of research findings is important, but unlikely to improve health all by itself. We need integrated, problem-focused, approaches that depend on knowledge not only of the biological consequences of e.g., poor nutrition but also of economic, social, cultural, and built-environmental determinants.
Leveraging existing resources with complementary strengths and weaknesses is a good idea. This should include studies, such as NICHD's Add Health and NIA's Health and Retirement Survey, that obtain statistically representative samples as well as biological measures of disease pathways. These studies complement the NCS by tracing developmental pathways in adolescent, adulthood, and later life.

Submitted by Christine Bachrach on June 9, 2011 at 2:03 PM
The DOHaD position paper notes that, "The outcomes arising from chronic diseases have their roots in fetal and early childhood development". Although the concepts, questions and intervention opportunities relating to the potential importance of in utero influences on long-term health outcomes were clearly framed, the panel largely ignored the influences of neonatal and early childhood development on health outcomes. Of particular concern was the omission of references to the importance of neonatal nutrition and breastfeeding to long-term health outcomes. Breast milk is an ideal neonatal food, and breastfeeding is thought to have significant long-term health benefits, particularly for premature infants. The American Academy of Pediatrics recommends exclusive breastfeeding during the first six months of life and suggests, "adopting breastfeeding as the reference model by which other forms of infant feeding should be compared in regard to long-term developmental and health outcomes". These recommendations are in line with the DOHaD panel's call to "define normative values for fetal growth, neonatal anthropometry, and postnatal growth" and to "include new ways to evaluate "optimal" prenatal and postnatal nutrition". Surprisingly, the panel failed to recognize the value of breast milk as a possible linchpin for evaluating optimal postnatal nutrition. The science behind the nutritional and health benefits of breastfeeding is still emerging, with fundamental questions remaining about the long-term effects of breastfeeding on health outcome, the effects of breast milk on neonatal metabolic and developmental parameters, and the "nutri-omic" composition of breast milk. Breast milk is also a source of environmental cues that can impact developmental and health outcomes of offspring. Examples of such impact include effects of breast milk on infant gut development, and evidence that breastfeeding impacts offspring obesity risk. These examples also underscore the possible use of breast milk components and/or breastfeeding as interventional strategies to counter disorders originating during fetal development. By omitting references to breastfeeding and the roles of breast milk in neonatal and infant nutrition in this position paper, the authors ignore a key developmental window that deserves further research support if we hope to develop an integrated understanding of the effects of fetal and early childhood development on adult health and disease.

Submitted by James McManaman on June 9, 2011 at 3:56 PM
The Preeclampsia Foundation is submitting a longer comment under the Pregnancy Vision Paper, but we felt compelled to underscore the need to prioritize placental banking and research, as noted in this paper. We've been throwing away (literally) what history may eventually prove to be a key piece of the preeclampsia puzzle.

No where is knowledge translation more critical than in this area of "developmental origins" as we seek to not just *understand* but *practice* behaviors and lifestyle choices that science tells us will impact us and our children the rest of our lives. While it may sometimes be frustrating when the "obvious" messages are not adopted, we must be diligent in involving the patients in this effort!

Submitted by Eleni Tsigas on June 9, 2011 at 4:27 PM
The Society for Women's Health Research appreciates the opportunity to provide comments to NICHD's Vision process. SWHR strongly encourages the Developmental Origins of Health and Disease Workgroup, when exploring fetal and early childhood determinants of health, to seek out and report sex based differences in all new NICHD research. The health and environment of the mother while pregnant is an obvious starting point, but a complete analysis needs to look at both maternal and paternal health and exposures, and differential impacts on a male fetus versus a female fetus.

Already, studies are reporting differential fetal impacts from exposures as common as high fructose corn syrup and risk for metabolic disease later in life. (Vickers, et al, 2011) Additionally, exposures to endocrine disrupting compounds during critical gestation and development periods can impact carcinogenesis. (NIEHS, 2010) Additional NIEHS research is also suggesting that a low-fat diet can mitigate or eliminate the risks posed by environmental exposures.

As there are clear sex differences already noted in the incidence, diagnoses, treatment and outcomes relating to diabetes, obesity, and coronary heart disease, one would expect that these differences can be traced back to early childhood and development.

As the thought leaders in research on sex differences, it is SWHR's hope that NICHD will recognize the importance of sex as a variable, and will encourage the analysis and reporting of sex-specific differences.

Submitted by SWHR on June 10, 2011 at 9:49 AM
The White Paper tilts heavily to the early origins of diabetes and cardiovascular disease. There is little in the White Paper about the early origins of osteoporosis. However, osteoporosis has its roots in fetal life, childhood and adolescence. Pediatricians who are oriented toward maximizing children's genetic potential for attainment of peak bone mass consider osteoporosis to be a pediatric disease with grave consequences in adulthood. Fruitful lines of research include calcium balance in pregnancy, transport of calcium across the placenta, FGF and other growth factors that affect skeletal development, identifying windows of opportunity during which supplementation with calcium, vitamin D, or pharmacologic agents can increase bone mineral density in children who are not fulfilling their genetic potential.

Submitted by Gilman Grave on June 10, 2011 at 12:19 PM
agree with the message of focusing research on early childhood origins of adult outcomes but I think that a discussion of the of observational studies needed to conduct causal inference would be warranted. The paper appropriately highlights the need for following up randomizations involving early childhood health but observational studies seem to be viewed rather monolithically by the authors. Showing statistical associations is neither a necessary nor a sufficient condition for supporting causal inference. If the only two options are either random assignment or observational studies of statistical associations using "rich" datasets, I do not think we are going to learn nearly as much as we could. I think the "third way" that is neglected in the white paper would focus energy on causal identification strategies where the case for causal inference is less of a leap of faith. Obviously, this is not a substitute for gold-standard random assignment studies. But this approach to observational data has had great success in older epidemiological studies, e.g. Snow's famous study of Cholera in London, Heider's study of the lingering effects of the 1918 In.uenza, or the early work on the Dutch Famine by Stein et al. The key is uncovering situations where the treatment considered is plausibly the only thing different between treatment and control groups and the reason for that treatment di.erence known and explicit. The White Paper comes closest to discussing the identification strategy approach when talking about the Dutch Famine and the Gulf oil spill. However, missing from this discussion is a cogent reason why we would bother to focus on such events. To my mind, the central reason to look at such events is that they might provide exogenous variation in a treatment of interest, i.e. variation that is not correlated with other determinants of health. And of course, such exogenous variation need not come from an event per se, but any setting, including static ones, where variation in treatment is exogenous and driven by a known source (such as which company supplied water in Snow's study of cholera). Whether the Gulf oil spill will constitute a useful identification strategy is not clear, though it is certainly a "splashy" and topical event.

Submitted by Douglas Almond on June 10, 2011 at 1:27 PM
It is exciting to have a national vision for research into the Developmental Origins of Health and Disease. My field is breastfeeding promotion and support. Every day research and also common sense show that the human infant needs human milk to develop and grow normally. As the AAP says the fully breastfed infant should be used as the reference or normative model against which all other ways of feeding should be measured. We already have a growth grid developed by the WHO based on the growth of healthy, full term breastfeed infants. We need more research into what the purposes are for the over 400 ingredients in human milk. There is some research that suggests human milk feeding "turns on" certain genes. What happens to babies that do not get those genes "turned on" because they are fed artificial baby milk. We need to find out more about how women avoid health risks when they breastfeed. We need a good working definition of breastfeeding that all researchers use so we know what we are talking about. We also need to report on our research using breastfeeding as the norm - not report the "benefits" of breastfeeding but the risks of not breastfeeding i.e., formula feeding. Finally I encourage the NICHD to take a leadership role to bring the Surgeon General's Call to Action to Support Breastfeeding to fruition by convening a consortuim to help "overcome breastfeeding research limitations in study design, increase the generalizability of research on breastfeeding, help prioritize key reseaech areas, enable expanded and advanced research to be performed, and foster the timely translation of research into practice." Thanks you for this opportunity to comment.

Submitted by Ann Twiggs on June 10, 2011 at 1:54 PM
I have two comments I would like to make, the first of which has also been made by others. First, there is only passing reference to the role of socio-economic status (SES). SES can be both an important determinant of disease (ie, higher levels of stress among the poor during pregnancy can program the fetus) but it is also an important moderator. Not only is it important to consider issues of SES to better understand the underlying models of disease development, SES is something that is strongly influenced by public policy. Thus, the policy implications of analyses that incorporate SES would be quite large. Second, in terms of data collection and analysis, there is already a (potentially) very large source of data that can be used to study both the determinants of disease and the consequences and that is administrative data maintained by various government agencies. For example, we already have much information stored in vital statistics records (natality and mortality), school records, Medicaid records, hospital discharge records, that, if linked together, could be a very important source of data to look at the development of disease over the life course. Issues of confidentiality related to developing such a data source, while significant, can be overcome. A number of other industrial countries currently maintain linked administrative data (eg, Norway) that has been a tremendously useful source of data.

Submitted by Anna Aizer on June 10, 2011 at 2:15 PM
On behalf of the Population Association of America, I am pleased to comment on the Developmental Origins of Health and Disease paper.

Overall, there is a disconnect between the paper's purported desire to adopt a life-course developmental systems paradigm, and its recommendations. For example, use of the word "programming" implies genetic determination. Much of the language implies a passive individual who is "exposed" to good and bad environmental factors, rather than an active agent who seeks out people, environments, and experiences.

Social and behavioral predictors, correlates, trajectories and outcomes are barely mentioned. For example the paper states "mammals are immersed in a complex environment that changes over time" (p. 1) but "socioeconomic and behavioral factors may contribute to adverse outcomes, but these factors are ubiquitous in nature, subtle and difficult to study" (p. 3).

There is a key concept of "understanding complex molecular interactions," but there is no counterpart of understanding complex psychosocial interactions. This is also evident in the text on "educating the public" (Page 13). It suggests that people will respond if told how to behave. Many failed public health campaigns have proven that education alone cannot change behavior. Input from behavioral economics, psychology and sociology might be relevant here.

The paper strongly suggests NIH should support new cohorts from pre-conception. However, existing NIH-supported datasets, such as Add Health (from early adolescence) and NLSY (from birth) already do. The paper focuses heavily on pre-conception and periconception variables and influences. Add Health can address questions about these influences since it tracks women from early adolescence through, currently, to their peak childbearing ages.

Design issues are overlooked in the section about leveraging Roadmap initiatives and creating big "team science" approaches. National representation or probability sampling, the only design that can apply findings to a larger population, is not raised.

Ethical considerations and practices regarding protection of shared data, whether biological or social/survey, are not mentioned. If these different types of data cannot be linked, a systems approach is impossible.

Submitted by David Lam on June 10, 2011 at 3:46 PM
I have two comments I would like to make, the first of which has also been made by others. First, there is only passing reference to the role of socio-economic status (SES). SES can be both an important determinant of disease (ie, higher levels of stress among the poor during pregnancy can program the fetus) but it is also an important
moderator. Not only is it important to consider issues of SES to
better understand the underlying models of disease development, SES is something that is strongly influenced by public policy. Thus, the policy implications of analyses that incorporate SES would be quite large. Second, in terms of data collection and analysis, there is already a (potentially) very large source of data that can be used to study both the determinants of disease and the consequences and that
is administrative data maintained by various government agencies. For example, we already have much information stored in vital statistics records (natality and mortality), school records, Medicaid records, hospital discharge records, that, if linked together, could be a very important source of data to look at the development of disease over the life course. Issues of confidentiality related to developing such
a data source, while significant, can be overcome. A number of other industrial countries currently maintain linked administrative data (eg, Norway) that has been a tremendously useful source of data.

Submitted by Ana Aizer on June 10, 2011 at 4:36 PM
I write on behalf of Childbirth Connection, a national non-profit organization that has worked to improve maternity care quality for 93 years. We are grateful to NICHD for prioritizing Developmental Origins and appreciate the opportunity to help strengthen the excellent White Paper.

We appreciate that the concepts of environmental impact on health and disease and on development recognize the potential impact of obstetric intervention. Vaginal versus cesarean birth, preterm versus term delivery, and ART are important examples that are specifically noted. We encourage NICHD to include in this framework the full complement of obstetric exposures during the prenatal, intrapartum, and early postpartum periods. Many reach a large portion of the fetal/newborn population. In validation studies, birth certificates and discharge data undercount some of these. The national Listening to Mothers II survey found, e.g., the following rates of intervention in women who gave birth in U.S. hospitals in 2005: one or more methods of labor induction (41%), epidural/spinal analgesia (76%), narcotics (22%), augmentation (47%), c-section (32%), and mother-baby separation after birth for routine care (39%). Also, 99% had one or more ultrasounds, and 59% had 3 or more. The survey report is available at www.childbirthconnection.org/listeningtomothers/.

In addition to interventions mentioned in the report, we urgently need to better understand possible long-term impacts of such exposures as (1) infant feeding practices (see http://www.ahrq.gov/clinic/tp/brfouttp.htm), recognizing that breastfeeding is highly modifiable through supportive interventions, quality measurement, and baby-friendly hospital practices, (2) pain medications, (3) antibiotics, (4) synthetic oxytocin, and (5) ultrasound. Subtle, inadequately appreciated and understood mechanisms for early postpartum mutual maternal-newborn regulation of physiology and behavior are commonly disrupted in current hospital practice (see, e.g., Winberg J. Dev Psychobiol 2005;47(3)), with evidence of long-term effects.

It is crucial to assess from the developmental origins framework common maternity practices that are "normal" to many clinicians and women, but may have unintended long-term consequences. The growing body of evidence of undesirable effects on health and development of late preterm and early term delivery and of cesarean section suggests that these investigations must be extended to other obstetric interventions.

Submitted by Carol Sakala on June 10, 2011 at 5:41 PM

The period for commenting on NICHD scientific Vision white papers is now closed.

Last Updated Date: 06/04/2012