[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Invited Commentary
Pediatrics
June 1, 2020

The Role of Neighborhood Social Characteristics on the Epigenome—Why the Lack of Investigations?

Author Affiliations
  • 1Center for Psychiatric and Neurodevelopmental Genetics Unit, Center for Genomic Medicine, Massachusetts General Hospital, Boston
  • 2Department of Psychiatry, Harvard Medical School, Boston, Massachusetts
JAMA Netw Open. 2020;3(6):e206111. doi:10.1001/jamanetworkopen.2020.6111

For centuries, scientists and clinicians alike have wondered how the contexts surrounding children and their families play a role in child health. Spanning multiple disciplines, dozens of empirical studies have investigated the association between features of place (or the macro-level social and physical environments) and numerous child health outcomes. Using a mix of observational, experimental, and quasiexperimental study designs, researchers have begun to unravel the complex ways that neighborhood features can promote or harm child health.1 Such work is important because it shows that opportunities for child health promotion and disease prevention are shaped by the settings in which children live, learn, and play. That is, such work illuminates the social determinants of child health or “the specific features of and pathways by which societal conditions affect health and that potentially can be altered by informed action.”2(p697)

For at least 20 years, phrases like cells to society and neurons to neighborhoods have been part of the scientific lexicon. Use of these phrases is intended to convey a metaphor of layers and the multiple nested factors that shape disease origin and course.3 Research to bring together multiple levels of analysis is critical because it can inform both etiological understanding and the design of interventions or policy approaches. Yet, few attempts have been made to systematically bridge these levels in empirical analyses. Frequently, researchers focus on variables within a level, without considering ways to link factors between levels.

This bridging of levels is why the study by Reuben and colleagues4 makes an important contribution to the literature. These authors analyzed data from the Environmental Risk (E-Risk) Longitudinal Twin Study to ascertain whether neighborhood characteristics throughout childhood and adolescence were associated with DNA methylation at age 18 years. They measured multiple aspects of the neighborhood, including deprivation, dilapidation, disconnection, and dangerousness. Reuben and colleagues4 also characterized the neighborhood and epigenetic associations using 3 distinct approaches. The authors found that children who grew up experiencing neighborhood social disadvantage have epigenetic profiles that are distinct from those of their peers who were raised in more advantaged settings. These findings were detected for inflammation and smoking-related DNA methylation sites as well as probes mapped to genes involved in the metabolism of cigarette smoke and ambient outdoor air pollution. These findings held true even after controlling for family-level socioeconomic status and individual-level tobacco smoking, both of which are known to affect DNA methylation values.5 The study by Reuben and colleagues4 provides a useful road map for researchers when operationalizing not just the different aspects of the neighborhood social environment but also the different dimensions of epigenetic processes. This study appears to be the largest and most comprehensive exploration of this topic to date.

The study4 implicitly prompts readers to ask, why have only a handful of studies evaluated the role of neighborhood social characteristics in the epigenome? According to the authors, only 5 studies have tested for DNA methylation differences among individuals living along neighborhood socioeconomic gradients. The lack of research in this area is curious, especially given that neighborhood-induced epigenetic processes are likely implicated in many adverse health outcomes, spanning from mental health disorders to cancer, obesity, and metabolic diseases, as suggested by the concept of multifinality. Thus, if so much could be learned through research that links neighborhoods to epigenomes, why aren’t more investigators conducting it?

Like most things, the answer is multifaceted. Perhaps it reflects the age-old problem of disciplinary silos and the tendency of researchers to not reach across the disciplinary aisle. Perhaps too few incentives exist, such as grant opportunities, or adequate infrastructures (eg, how departments are set up, how conferences are organized, and how different disciplinary groups communicate) that encourage people to work together. In addition, some aspects of nurture-focused research may make the more nature-focused investigator skittish. For example, neighborhood and other aspects of environment are unspecified and unbounded in a way that the genome and, to some degree, the epigenome are not. Such broadness produces challenges in deciding not just what is studied but also when and how such exposures are ascertained.6 Although studying the epigenome is equally complex, in my experience sometimes the perception is that nonenvironmental-focused and more biologically based work is easier to conduct. Sometimes, nurture-focused researchers talk more about the complexity of their work in ways that sound alarms to researchers outside of the field. For example, in describing the evolution of neighborhood health studies in the Foreword of the second edition of Neighborhoods and Health, Ana Diez Roux went so far as to label one of her sections “Oh my God, it’s (really) complicated….”1 The skittishness might also run the other direction, with neighborhood-focused investigators being skeptical about the possibility that genetic research is actually policy relevant. And maybe, as Diez Roux noted, another reason for the lack of research in this area is that the work is hard and complex. To make strong inferences, epigenetic studies require large samples because effect estimates are small, and the number of probes analyzed in an epigenetic study creates a multiple testing problem. Furthermore, neighborhood studies require large samples of people, who are ideally followed up prospectively and are characterized using measures that tap salient and policy-relevant qualities of neighborhoods. These challenges exist when the work is done on its own, let alone when 2 fields are integrated.

Consider these challenges. How can they be overcome, and how can researchers be encouraged to pursue more neighborhood-oriented epigenome studies? Here I think we are on the cusp of exciting days ahead, when we can begin to execute truly multilevel studies. Compared with 2 decades ago, when the field of neighborhood health research was just emerging and studies on environmentally induced epigenetic programming were just finding their footing, considerable advancements in neighborhood theory, measurement, and analyses have been made that should make the examination of place-based effects more viable in all areas of health research, including epigenetics. We now have better sources of data, methods, and ideas that enable us to piece together neighborhood social features with epigenomes. For example, strategies have been outlined to integrate multilevel thinking with multilevel analytical methods.7 A variety of measurement tools are now widely available that can be used to characterize neighborhood settings (eg, census data, multidimensional geocoded data, and smart phone–based technologies) and that can be incorporated relatively easily into existing epigenetic studies. In addition, genotyping and epigenotyping costs are decreasing, making it more affordable for neighborhood-focused researchers to collect saliva, blood, or other tissues for epigenetic analyses. Innovative analytical approaches and software are also available that allow investigators to model aspects of place in nuanced ways (eg, agent-based models, spatial analyses) and to pair those models with biological data.

I hope that studies like this by Reuben and colleagues4 will prompt researchers to explore these complex concepts and to bridge social determinants of health with epigenetic processes. I also hope the work of these authors will inspire leaders to build better infrastructure, such as funding opportunities to facilitate interdisciplinary research and strong pipelines to train the next generation of interdisciplinary scientists who can think on multiple levels. In such an environment, we can then usher in a new generation of multilevel studies that work toward the goal of understanding disease origin and course from neurons (or cells) to neighborhoods (and society).

Back to top
Article Information

Accepted for Publication: February 18, 2020.

Published: June 1, 2020. doi:10.1001/jamanetworkopen.2020.6111

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Dunn EC. JAMA Network Open.

Corresponding Author: Erin C. Dunn, ScD, MPH, Center for Psychiatric and Neurodevelopmental Genetics Unit, Center for Genomic Medicine, Massachusetts General Hospital, 185 Cambridge St, Simches Research Building 6th Floor, Boston, MA 02114 (edunn2@mgh.harvard.edu).

Conflict of Interest Disclosures: None reported.

Funding Support: This commentary was supported by grant R01MH113930 from the National Institute of Mental Health.

Role of the Funder/Sponsor: The funder had no role in the analysis and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Mariana Arcaya, ScD, Massachusetts Institute of Technology, and Natalie Slopen, ScD, University of Maryland, provided insightful feedback on this commentary. These individuals received no additional compensation, outside of their usual salary, for their contributions.

References
1.
Duncan  DT, Kawachi  I.  Neighborhoods and Health. 2nd edition. Oxford University Press; 2018.
2.
Krieger  N.  A glossary for social epidemiology.   J Epidemiol Community Health. 2001;55(10):693-700. doi:10.1136/jech.55.10.693 PubMedGoogle ScholarCrossref
3.
Kaplan  GA.  What’s wrong with social epidemiology, and how can we make it better?   Epidemiol Rev. 2004;26:124-135. doi:10.1093/epirev/mxh010 PubMedGoogle ScholarCrossref
4.
Reuben  A, Sugden  K, Arseneault  L,  et al.  Association of neighborhood disadvantage in childhood with DNA methylation in young adulthood.   JAMA Network Open. 2020;3(6):e206095. doi:10.1001/jamanetworkopen.2020.6095Google Scholar
5.
Dunn  EC, Soare  TW, Zhu  Y,  et al.  Sensitive periods for the effect of childhood adversity on DNA methylation: results from a prospective, longitudinal study.   Biol Psychiatry. 2019;85(10):838-849. doi:10.1016/j.biopsych.2018.12.023 PubMedGoogle ScholarCrossref
6.
Dunn  EC, Brown  RC, Dai  Y,  et al.  Genetic determinants of depression: recent findings and future directions.   Harv Rev Psychiatry. 2015;23(1):1-18. doi:10.1097/HRP.0000000000000054 PubMedGoogle ScholarCrossref
7.
Dunn  EC, Masyn  KE, Yudron  M, Jones  SM, Subramanian  SV.  Translating multilevel theory into multilevel research: challenges and opportunities for understanding the social determinants of psychiatric disorders.   Soc Psychiatry Psychiatr Epidemiol. 2014;49(6):859-872. doi:10.1007/s00127-013-0809-5 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    2 Comments for this article
    EXPAND ALL
    Perhaps there’s another reason
    Donna Lisi, PharmD | Healthcare system
    I read both the paper by Reuben et al that is the focus of this commentary as well as this commentary. Perhaps there is another reason to consider as to why this type of research is not more widespread. Perhaps the answer is found not on the cover of a medical journal but rather on the front page of every leading newspaper- systemic racism. The socially disadvantaged are more often people of color. We’ve seen first hand in areas like Flint, Michigan that race-based neighborhoods played a role in the safety of drinking water. There are those who try to discredit the BLM movement and try to paint people of color as less than, as people who don’t work toward their own success. Now imagine if science can once and for all debunk these racist attitudes with proof that those who grow up disadvantaged are physiologically disadvantaged their whole life no matter how hard they try. This would be a whole new narrative in discussions on race ... one that some people may not want to hear because they still want to cast fraudulent blame. The time is long past due for neighborhood-based research. There can be no health care equity unless access to quality healthcare and economic opportunity exists in all neighborhoods.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Disruptive Process
    Paul Nelson, MS, MD | Family Health Care, P.C. retired
    As applied to a community's health and its health care, each of the community's resident persons will react uniquely given the variable time-dimensions of the clustering among the disturbances that could be involved. This study would suggest that we really don't have a unified epistemology for understanding adversities.

    During the last three days, I witnessed each of my two grandsons participate in a baseball game with age-related peers. Both are pitchers, fairly good-ones at that, given their ages of 9 and 11 years (thanks to their Dad). Here they are with 4 parent coaches for
    both teams and several family members to "cheer-them-on." The play by play is reported by a parent who maintains a cell-phone app for everyone else to follow, if they wished, including lots of stats. These players and their Dad coaches really care about each other especially after every game, win or lose.

    It's not hard to understand how this will help my grandsons survive their life's struggles in the future. Putting aside sports as a factor to assure future survival, it really began with intact families that have for 9 and 11 years offered a caring relationship within a communal family as the basis to survive the disruptive processes we encounter during early childhood. Any permanent health reform must eventually be structured, community by community, to support the needs of each family. We have a long road ahead to prevent, mitigate, and ameliorate the disruptive processes affecting each person's health and their well-being beginning in early childhood. This study is a foundational contribution.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    ×