I need help

1 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

K. Steven Brown, Kilolo Kijakazi, Charmaine Runes, and Margery Austin Turner

February 2019

Racial and ethnic disparities figure prominently into much of the analysis conducted by

policy research organizations in the US. But too often our organizations give short shrift to

the centuries of subjugation, discrimination, exclusion, and injustice that have produced

these inequities.

If, as researchers, we aim to build knowledge that helps shape and advance solutions to the

challenges of blocked mobility and widening inequality, we must do better at explicitly examining the

structural and systemic forces at work. For many established research organizations, this is more easily

said than done. It requires scholars to learn things about our history and its lasting implications that

they may not already know. It requires changes to familiar ways of working. And it requires fresh

approaches to communicating findings to our intended audiences.

Over the past several years, the Urban Institute has committed itself to making these changes. We

see this goal—to rigorously address the structures and systems of racism in the content and

communication of our research—as an essential part of our broader diversity and inclusion efforts. The

current political climate creates a heightened sensitivity and sense of urgency, but we see this evolution

as essential to our mission over the long term. Urban is by no means alone in this endeavor; many other

policy analysis and research organizations have also embarked on this undertaking and have an interest

in sharing tools, strategies, and lessons learned along the way.

In November 2018, the Urban Institute hosted a roundtable discussion with 23 organizations

representing policy research, academia, and philanthropy to share approaches, insights, and lessons

from our respective efforts to confront structural racism in our research and policy analysis. This brief

discusses the rationale for these efforts at implementing institutional change; the range of challenges

and constraints facing different types of research organizations; and our experience to date with

DI VERSI T Y, EQ UI T Y, AND I NCL USI O N AT T HE URBAN I NST I T UT E

Confronting Structural Racism in Research and Policy Analysis Charting a Course for Policy Research Institutions

2 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

specific tools and strategies. We aim to advance understanding of and attention to structural racism in

the work of our own institution and in the larger field of policy research.

Recognizing the Pervasive Legacy of Racist Policies For nearly its entire history, the United States excluded people of color from its main pathways to

economic opportunity through explicit policy decisions. In Stamped from the Beginning, Ibram X. Kendi

argues that racism does not primarily stem from hate and ignorance, but that “racist policies have driven

the history of racist ideas in America” (Kendi 2017, 9). For generations, people of African descent lived

and died in bondage. Even after the Civil War and the abolishment of slavery, black people in the United

States were subjected to legalized forms of discrimination that restricted where they lived, if and where

they could attend school, and the kinds of jobs they could hold. And even with the constitutional right to

vote (granted to men with the 15th Amendment in 1870 and to women with the 19th Amendment in

1920), barriers to exercising those rights largely prevented citizens’ ability to change the oppressive

laws that obstructed their opportunities.

For example, the federal Home Owners Loan Corporation, established in 1933 as part of the New

Deal, created maps that were color-coded to indicate the desirability of neighborhoods. Race was a

significant factor in determining the color-coding of a neighborhood (Hillier 2005), with communities of

color designated as undesirable and color-coded red. This appraisal system, called redlining, was

adopted by the Federal Housing Administration (FHA), which provided mortgage insurance enabling

many Americans to buy homes. Redlining made it much more difficult and expensive for African

Americans to obtain loans and purchase homes. Between 1930 and 1960, African Americans received

less than 1 percent of the nation’s mortgages (Conley 1999). In addition to redlining, the FHA advocated

using restrictive covenants to maintain the racial segregation of neighborhoods. The FHA’s Underwriting

Manual stated, “if a neighborhood is to retain stability, it is necessary that properties shall continue to be

occupied by the same social and racial classes” (Oliver and Shapiro 2006, 18). Because people were

unable to buy homes in the past, many families today have less wealth, 1 and schools are not much less

segregated than they were 50 years ago (Reardon and Owens 2014).

America’s history of discriminatory policies and institutional practices explains the deep disparities

in access to opportunities and in outcomes that we see today across social and economic domains. Court

cases were decided and laws were passed that outlawed these and other practices, but to paraphrase

Lyndon B. Johnson, who helped found the Urban Institute, these legal rights are the beginning of the

path to freedom, not the end. He goes on to say that “it is not enough just to open the gates of

opportunity. All our citizens must have the ability to walk through those gates.” 2 Our approach to

understanding current racial disparities is guided by an understanding that centuries of oppression,

legal discrimination, and sanctioned inequality have long tails that continue to shape where people live,

what opportunities they are exposed to, and how people engage with one another. The legacies of those

structures—if not the structures themselves—continue to have impacts today. We use the definition of

structural racism developed by the Aspen Institute Roundtable on Community Change (2004):

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 3

a system in which public policies, institutional practices, cultural representations, and other

norms work in various, often reinforcing ways to perpetuate racial group inequity. It identifies

dimensions of our history and culture that have allowed privileges associated with “whiteness” and disadvantages associated with “color” to endure and adapt over time.

Two examples highlight how structural racism operates in policy today. First, more than 50 years

after passage of the federal Fair Housing Act, America’s neighborhoods remain starkly segregated along

lines of race and ethnicity, and people of color are dramatically overrepresented in high-poverty census

tracts. By the late 20th century, after decades of discriminatory lending practices and redlining, civil

rights legislation and evolving constitutional jurisprudence prohibited overt forms of discrimination in

housing and lifted many formal barriers to residential integration. But they were quickly replaced by

subtler and ostensibly race-neutral methods to exclude people of color from predominantly white

neighborhoods. For example, exclusionary zoning policies make it difficult for lower-income residents to

live in many suburban communities. And while the incidence of housing discrimination has generally

declined, people of color looking for places to live are still told about fewer homes and apartments than

white people (Greene, Turner, and Gourevitch 2017).

A second example involves law enforcement policies that criminalize behaviors in a way that

disproportionately affects people of color. Federal guidelines impose substantially more severe

penalties for the use of crack than for powder cocaine, two forms of the same drug. Research has found

that crack is more likely to be used by socioeconomically disadvantaged members of society, among

which African Americans are disproportionately represented, and that African Americans are “at higher

risk for arrest and subject to [an] 18:1 sentencing disparity.” 3 This is an example of color-blind structural

racism, where a policy makes no reference to race but still has major disproportionate effects by race.

As Kendi argues, the differences in outcomes in these two examples, not to mention many others,

resulted from policy. Too often, however, public policy researchers ignore or overlook the structures

and systems that created and sustain inequality, focusing exclusively on individual choices and

behaviors as the main drivers of disparate outcomes. Improving public policy research requires

organizations to consider how this history of discriminatory policies affects the context, validity, and

implications of our work, and to make intentional change in how we address these racist legacies.

Navigating Institutional Choices and Constraints Policy research organizations take many institutional forms—from small, single-issue nonprofits to for-

profit firms with thousands of employees to policy centers within universities to policy research

organizations in the nation’s capital. Structural racism is undeniably relevant to the work of all these

organizations, no matter their size or type. But the challenges we face and the paths we take to more

effectively address structural racism vary widely. In particular, an organization’s primary mission, its

funding sources, and its size and internal structure are likely to shape the strategies it pursues to

explicitly address the realities of structural racism in the research and policy analysis it produces.

4 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

Organizational Mission

Every organization must give careful thought to how structural racism issues relate to its core mission

and the audiences it aims to inform and influence. Some may conclude that racial inequity and injustice

are core to their mission and that research on structural racism should take center stage. But even

research organizations with broader or less normative missions can and should find ways to accurately

and effectively analyze structural racism and its consequences.

Some organizations may explicitly focus their mission on advancing racial justice or overcoming

white supremacy. Many of these institutions target audiences that include grassroots organizations,

advocates, and social justice practitioners. These organizations can make their focus on structures and

systems explicit in their research products. They can hire people with expertise and commitment to

their mission and can expect their staff to make this work a top priority—putting them a step ahead of

organizations whose staff might not all have the same knowledge or commitment to advancing racial

justice. These organizations are also more likely to devote institutional resources to developing internal

training for staff members and to building capacity around these crucial issues.

ORGANIZATIONS WITH MISSIONS FOCUSED ON ADVANCING RACIAL JUSTICE

 PolicyLink “is a national research and action institute advancing racial and economic equity by Lifting Up What Works.”

 Kirwan Center for the Study of Race and Ethnicity works “to create a just and inclusive society where all people and communities have opportunity to succeed.”

Many other long-established research organizations have broader missions than advancing racial

justice, but they can still decide to give structural racism explicit attention within a larger frame. These

institutions’ target audiences typically include elected officials, government agencies, and business

leaders, as well as on-the-ground practitioners and advocates. They can develop a structural racism–

focused program area or an important cross-cutting initiative within the broader scope of their research

analysis, also providing institutional legitimacy in the process. They can also make it a priority to hire

staff with relevant interests and expertise to lead in this area of study. In addition, staff with this

expertise can be encouraged to contribute to other work, since structural racism is pervasive across

research areas. Organizations with broader missions can also offer training, tools, and incentives to staff

interested in engaging with the conversation around advancing racial justice in their work.

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 5

ORGANIZATIONS WITH MISSIONS BROADER THAN, BUT INCLUSIVE OF, RACIAL JUSTICE

 The Brookings Institution’s mission is “to conduct in-depth research that leads to new ideas for solving problems facing society at the local, national and global level.”

 Abt Associates aims to be “an engine for social impact, fueled by caring, curiosity and cutting- edge research that moves people from vulnerability to security.”

 Mathematica “is dedicated to improving public well-being and reimagining the way the world gathers and uses data.”

The Urban Institute, founded in 1968 to “bring power through knowledge to solve the problems

that weigh heavily on the hearts and minds of America,” has chosen to elevate issues of racial injustice

and inequity as central to our broader mission. We seek to inform and support a wide variety of

audiences, including changemakers in government, philanthropy, business, advocacy, and practice. Our

Next50 initiative, which draws on our previous 50 years of work to inform priorities for our next 50,

focuses on advancing mobility and narrowing equity gaps. One of the big questions we want to tackle in

our future work is “What would it take to eliminate the policies, programs, and institutional practices

that impede racial equity?” We are committed to devoting resources to encouraging and supporting

steps to advance racial justice in our work, but we do not mandate this focus across all staff or projects.

The Center on Budget and Policy Priorities (CBPP) State Fiscal Policy division has been exploring

ways in which structural racism has affected their staff, how they do their work, and the landscape in

which they operate. They examine how race implicitly or explicitly impacts their decisionmaking from

the policy issues they choose to the research they conduct to the partners with whom they choose to

work. They are undertaking efforts to make these systemic barriers more transparent and to develop

strategies that will help staff identify their own assumptions and biases, analyze their decisions, and

choose to use a racial equity approach to the work. An example of such work is their seminal paper

Advancing Racial and Ethnic Equity with State Tax Policy (2018), which examines structural racism in

state tax policy. In addition, CBPP administers a state policy fellowship program with a focus on

candidates who have experience with communities that are underrepresented in state policy debates.

Funding Sources

The funding sources upon which an organization relies may either constrain or accelerate its ability to

make issues of structural racism explicit in its work. Some funders find the language of structural racism

too controversial or political and are unlikely to support work that puts the issue front and center. And

policy research organizations that compete for awards with strictly defined scope and focus face

limitations on their flexibility to explore these issues.

Despite these constraints, researchers have opportunities and responsibilities to identify and

describe the structures and systems that drive disparate outcomes, when these outcomes are

6 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

addressed by their work. As discussed further below, they can avoid data and methods mistakes that

obscure key drivers, avoid language that dehumanizes people, and publish separate products targeted

to audiences other than the original funder. These additional products may leverage supplemental funds

to dive deeper into structural racism in the work and examine its effects without the constraints

imposed by the original scope of work or funding source.

In contrast, some funders, particularly in the philanthropic world, have determined that structural

racism should be a central focus or lens for the organizations and work they support. To capitalize on

these funding sources, research institutions must first prove they have the capacity to delve into this

type of work and go beyond the superficial. Developing a robust evidence base around structural racism

and its effects is critical to attracting these funders, which will in turn allow organizations to dig more

deeply into the disparate effects of structures and systems in their future work. To win support from

these funders, organizations must also actively engage with communities of color to surface questions

and gather evidence. They must have an inclusive staff with expertise in structural barriers to

opportunity. And they must identify policy and practice reforms that stretch conventional thinking.

Many other funding organizations, including those in the philanthropic, government, and corporate

spaces, are exploring how they want to tackle the structural forces that sustain inequity and injustice.

Seeking funding from these sources provides organizations with the opportunity to work together with

a partner and learn how to best address these issues through research. Institutions seeking these

funding sources can expand their research areas to ask challenging, “outside the box” research

questions they want to investigate. They can also experiment with less conventional data sources and

methods and reach out to new audiences that may be unfamiliar with their work.

Size and Structure

Organizations’ size, structure, and internal culture play a central role in how they implement efforts to

better address structural racism. Differences in these characteristics do not excuse institutions from

taking steps to improve. Rather, they offer an opportunity for organizations to take advantage of their

unique strengths and capacities.

Small organizations with a tight-knit team structure may be especially well positioned to establish a

strong shared understanding of and approach to analysis of racial inequity and injustice. In these smaller

organizations, the leader is critical to setting the tone and focus for everyone. A smaller size makes it

easier for the entire staff to take training together to improve their awareness of these issues, and it

encourages close collaboration around how to advance lessons learned and new approaches. These

advantages can also apply to small internal teams within larger, more complex organizations.

Highly centralized organizations with top-down review and approval mechanisms may be able to

mandate that everyone adopt the lens of structural racism, apply appropriate data and analytic

methods, and adhere to language guidelines. They can require all staff to complete training that centers

around the disparate effects of structural racism. These organizations can also implement a centralized

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 7

review of proposals, work plans, research designs, and research products to ensure all work takes into

account, when applicable, a racial equity lens.

Many research organizations, however, are both large and decentralized and place a premium on

researcher independence rather than top-down direction. This poses challenges for adopting and

applying new ideas and approaches. Nonetheless, signals from leadership that they see this work as a

priority can be very powerful. These larger and more decentralized organizations can prioritize racial

equity in their work by celebrating researchers who are working on these issues, making connections

between researchers who might not otherwise know each other, offering voluntary tools and training to

advance awareness and adoption around racial justice, and providing internal financial support and

incentives to researchers who commit to prioritizing racial equity in their work.

Organizations also vary in the composition and diversity of their staffs. Many research

organizations set goals and track progress for diversity in staff and leadership. Having a diverse staff is

an important goal, and research has shown that increasing diversity can bring benefits in

communication, innovation, and productivity (Ellison and Mullin 2014; Gao and Zhang 2017). 4 Just as

important, a diverse staff brings different perspectives and sensitivities, which can improve how

organizations engage with and talk about particular populations, and a diverse staff accurately reflects

and represents the world in which we live and work. However, a staff that is less representative does

not prevent an institution from advancing a structural racism lens in their work. Tools like those listed

in this report can enable organizations to take concrete steps toward a better account of structural

disadvantage in their work. Additionally, developing this lens can provide opportunities for growth for

staff of color already in the organization and may be a draw to increase the diversity of staff through

hiring.

Tools for Moving Forward Progress may look different depending upon institutional choices and constraints, but all policy

research organizations can move forward with efforts to explicitly address the structures and systems

of racism by building understanding and awareness among staff members, reexamining data sources

and analytic methods, and improving communication strategies.

Boost Awareness and Learning Among Staff

Research organizations can build their internal capacity to produce rigorous research on racial

inequities and injustice by seeking external guidance, creating intentional spaces for reflection and

education, and embedding mechanisms that raise staff consciousness at each phase of the research

process—from proposal development to product dissemination. Many organizations have few staff with

the knowledge and expertise to effectively address structural racism in their work. Institutions should

build up their staff so people with this expertise work in each of the institutions’ policy domains. Tools

being tested to boost staff awareness and learning include the following:

8 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

 Structural racism speaker series: The Urban Institute invites outside experts to spark

discussion on structural racism and advance new lines of inquiry among researchers. These

“brown bag” seminars expose researchers to established and emerging frameworks, methods,

and data sources around structural racism while providing examples of how researchers can

contextualize research results. This speaker series also helps researchers broaden their

networks and foster new partnerships for future work.

 Structural racism blog post series: The Urban Institute’s Urban Wire invites staff at all levels,

including research assistants and analysts, to write blog posts that apply a structural racism lens

to research findings and policy developments. This approach has elevated structural racism as a

topic of discussion at Urban, and it encourages collaboration among researchers across

different domains and years of experience.

 Leveraging internal funding: Several policy research organizations dedicate flexible (internal)

resources to work around structural racism, including the staff time needed to organize,

facilitate, and debrief meetings, as well as to develop public-facing products.

 READ groups: The CBPP developed learning modules about racial inequity that are designed to

spark discussion among small groups of staff. These modules include books, articles, and videos

at the intersection of public policy, research, and structural racism. They are helpful resources

for staff committed to building their knowledge and improving their research, and they

encourage engagement and discussion among staff who might not have the opportunity to talk

about these issues otherwise.

 Research project checklist: The CBPP created a checklist of questions that prompt staff to

consider structural racism at each stage of a research project. The checklist encourages

researchers to examine each decision they make throughout the project, from choosing

populations of interest to data sources to background research to participant compensation to

the structure of the analysis. These questions prompt researchers to push themselves and think

deliberately about how structural racism may play a role in their work.

Improve Data Sources and Methods

Research organizations can take concrete steps to include people and perspectives that are left out of

standard research practices by improving the data sources upon which they rely, and to develop

analytic methods for rigorously measuring the structures and systems that sustain racial inequities. As

gatekeepers for what constitutes valid research, our institutions have the obligation to develop and

elevate data sources and methods that more accurately and respectfully represent marginalized

communities and more accurately document the barriers they face. These methods and data sources

also improve the rigor of our research and the relevance of our policy analysis:

 Cultivate community-engaged research methods: Researchers can better understand the

people they study and the realities they face by actively engaging with communities and

building on their knowledge and insights. Creating a collaborative, equitable learning

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 9

partnership requires researchers to include community members in multiple phases of

research, including study design, data collection, analysis, and dissemination.

 Devote resources to translation: Language barriers among survey respondents and research

partners can result in some populations being left out of studies. Research organizations should

include in their budgets to funders resources for translation services to ensure that everyone in

a study population is included and that their responses are accurately reflected.

 Compensate survey respondents: Time and expertise are valuable assets. Researchers can

acknowledge these contributions by paying survey respondents and community partners.

Financial remuneration may be most appropriate, but if this is not possible, alternative forms of

compensation, such as providing food or securing child care, should be offered.

 Reconsider race as a dummy variable: Researchers often use dummy variables to represent

race and ethnicity in multivariate analysis, but this practice implicitly assumes there is no

relationship between race and other explanatory variables. Instead of uncritically using dummy

variables, researchers should examine what role they think race actually plays in their model

and how best to test their hypotheses about its impact on particular outcomes. Examining these

assumptions may require researchers to recognize their own biases. Failing to account for the

fact that not everyone has access to the same assets or opportunities can result in misleading

findings about differences in outcomes.

“LIKE FISH WHO DON’T SEE WATER, ECONOMISTS DON’T SEE STRUCTURAL RACISM” (EMMONS 2018) 5

A standard approach in economics is to include race in the analytic equation as a “dummy variable”—a numeric variable used to represent subgroups of a sample. This approach makes the implicit assumption that individuals are alike in every way except for their race. Emmons and Rickets (2017) demonstrated the flaw in this assumption by testing two models for explaining the racial wealth gap. The first model used standard dummy variables for race. Its results suggested that differences in wealth were almost entirely explained by differences in education, employment, and other similar independent factors. The second model expressed the independent factors as deviations from the racial group average. This approach found that, although the independent factors were important, they did not explain away most of the wealth gap. In fact, the researchers concluded that over 70 percent of the racial wealth gap stems from structural factors that lead to families of color facing greater constraints.

This study was inspired by Darrick Hamilton’s critique of the dummy variable. His research (Darity, Hamilton et al. 2018; Hamilton and Darity 2017; Hamilton et al. 2015) has demonstrated that even when African Americans have made all the “right” choices, they have substantially less wealth than white people. White high school dropouts have more wealth, on average, than African Americans with a college degree (Darity et al. 2018). Unemployed white people have more wealth than African Americans who work full time. White homeowners possess $140,000 more in net worth than African American homeowners. And the net worth of single-parent white families is more than two times that of two- parent African American families.

1 0 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

Adopt Communication Guidelines and Engage Diverse Audiences

Research institutions can better steward their reach and influence by contextualizing data on

disparities with information on historic and current inequities, committing to using respectful and

inclusive language and images in their products, and by elevating marginalized voices and perspectives

in public events and outreach. Historically, researchers have perpetuated stereotypes about people of

color by using dehumanizing language and imagery. Organizations can dismantle this harmful legacy by

favoring meaningful change over the status quo:

 Establish communication guidelines: Urban Institute staff are creating guidance documents

that share the social and historical context of various phrases, labels, and racial categories;

provide examples of labels to use and not to use; encourage researchers to use labels preferred

by the communities they study; and offer other helpful resources. These guides aim to ensure

that all products consistently use language that conveys respect for the individuals and groups

studied, and avoid language that reinforces stereotypes about groups that have been

marginalized in society.

 Employ a principled image selection process: Researchers can be more intentional about the

images they attach to their reports, presentations, and blog posts. Images should be

representative of the research and avoid perpetuating stereotypes (for example, in an

evaluation of a federal program, the image should reflect the racial breakdown of that

program). Images should also show people in marginalized groups in contexts beyond those

solely about them being marginalized (for example, researchers should include images of black

people in reports about homeownership or career advancement, rather than just research

about poverty or joblessness).

 Implement event panel guidelines: Organizations can ensure their event planning and

outreach procedures explicitly address the importance of including diverse speakers and

reaching diverse audiences. The Urban Institute’s event guidelines prompt researchers and

communications staff to ask whether the proposed speakers and audience invitation lists are

diverse and whether they include the perspectives of people with lived experience in the topic

being discussed.

 Diversify products and dissemination strategies: Researchers can make their work more

accessible by publishing a variety of products—such as technical reports, briefs, blog posts,

podcasts, and infographics—that target different audiences. A lengthy research report might

not always be the best avenue to communicate findings, as shorter and more accessible options

often reach broader audiences. Researchers can also share their work with smaller, more

specialized news outlets to reach a more diverse audience, rather than targeting only elite

media outlets.

 Partner with advocacy organizations to take the work further: Research organizations can

develop robust partnerships with advocacy organizations to ensure their products are useful in

directly or indirectly informing and creating more equitable policy and programming.

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 1 1

 Host community data walks: Researchers can share key data and findings with the people

closest to the issues through community data walks (Murray, Falkenburger, and Saxena 2015).

These data walks aim to ensure the people most affected by the research have a robust

understanding of the data, to help inform better programming and policies to address the

strengths and needs of a community or population, and to inspire individual and collective

action among community agents.

DATA WALKS: COMMUNITY-CENTERED TOOLS TO SHARE DATA

During a data walk, stakeholders split up into small groups and rotate through “stations”—each one a visual and/or textual display of data that forms a narrative about the community that participants can confirm, critique, and complicate.

Data walks provide opportunities for researchers and community members to cocreate meaning and solutions based on community data; in other words, community members are not only research participants but also equal research partners.

Assessing Progress With these considerations and strategies in mind, one big question remains: How can we, as institutions,

hold ourselves accountable and determine whether these efforts are working? As research

organizations, our assessments of progress and decisions about next steps should be rooted in evidence.

We propose five basic indicators for researchers and policy analysts to assess their progress in

confronting structural racism in our work. Each organization would need to determine which of these

indicators are most aligned with its mission and goals and develop systems to create baseline measures,

track progress, and ensure accountability.

 Increasing numbers of staff will participate in efforts to learn about structural racism and apply

this lens in their research about issues of difference and disparities.

 Wider and more diverse audiences will read our work, attend our events, and find our work

relevant and useful.

 Audiences we typically engage will recognize the value of our work on structural racism and

find it relevant and enlightening.

 More experts of color will want to work for our organization and contribute to the bodies of

work we produce.

 Independent “audits” of the language and images in our published research products will find

improvements in respect, equity, and inclusion.

1 2 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

We would argue that no research organization has fully figured out how to effectively implement a

structural racism lens in their work and that all institutions interested in making their work more

equitable and impactful still have room to grow. As we move forward with these efforts, research

organizations should continue to share ideas and strategies and to seek the expertise of others inside

and outside our field who have made more progress.

Notes 1 Signe-Mary McKernan, Caroline Ratcliffe, C. Eugene Steuerle, Caleb Quakenbush, and Emma Kalish, “Nine

Charts about Wealth Inequality in America (Updated),” Urban Institute, February 2015, http://apps.urban.org/features/wealth-inequality-charts/.

2 Lyndon B. Johnson, “To Fulfill These Rights,” commencement address at Howard University, Washington, DC, June 4, 1965, transcript, http://teachingamericanhistory.org/library/document/commencement-address-at- howard-university-to-fulfill-these-rights/.

3 New York University, “Powder vs. Crack: NYU Study Identifies Arrest Risk Disparity for Cocaine Use,” news release, February 19, 2015, https://www.nyu.edu/about/news-publications/news/2015/february/-powder-vs- crack-nyu-study-identifies-arrest-risk-disparity-for-cocaine-use.html.

4 Phillips, Katherine W., “How Diversity Makes Us Smarter,” Scientific American, 1 October 1, 2014, https://www.scientificamerican.com/article/how-diversity-makes-us-smarter/.

5 William R. Emmons, 2018, unpublished transcript from the Structural Racism Roundtable at the Urban Institute, Washington, DC, November 7, 2018.

References and Recommended Readings Aspen Institute Roundtable on Community Change. 2004. Structural Racism and Community Building. Washington,

DC: Aspen Institute Roundtable on Community Change.

Darity, William Jr., Darrick Hamilton, Mark Paul, Alan Aja, Anne Price, Antonio Moore, and Caterina Chiopris. 2018. What We Get Wrong About Closing the Racial Wealth Gap. Durham, NC: Samuel DuBois Cook Center on Social Equity; Oakland, CA: Insight Center for Community Economic Development.

“EE Framework.” Equitable Evaluation Initiative (EEI), https://www.equitableeval.org/ee-framework.

Ellison, Sara F., and Wallace P. Mullin. 2014. “Diversity, Social Goods Provision, and Performance in the Firm.” Journal of Economics and Management Strategy 23 (2): 465–481.

Emmons, William R., and Lowell R. Rickets. 2017. “College Is Not Enough: Higher Education Does Not Eliminate Racial and Ethnic Wealth Gaps.” Review (Federal Reserve Bank of St. Louis) 99 (1): 7–39. https://doi.org/10.20955/r.2017.7-39.

Flynn, Andrea, Susan R. Holmberg, Dorian T. Warren, and Felicia J. Wong. 2017. The Hidden Rules of Race: Barriers to an Inclusive Economy. New York: Cambridge University Press.

Gao, Huasheng, and Wei Zhang. 2017. “Employment Nondiscrimination Acts and Corporate Innovation.” Management Science 63 (9): 2982–2999.

Greene, Solomon, Margery Austin Turner, and Ruth Gourevitch. 2017. “Racial Residential Segregation and Neighborhood Disparities.” Washington, DC: US Partnership on Mobility from Poverty.

Hamilton, Darrick, and William A. Darity, Jr. 2017. “The Political Economy of Education, Financial Literacy, and the Racial Wealth Gap.” Review (Federal Reserve Bank of St. Louis) 99 (1): 59–76. https://doi.org/10.20955/r.2017.59-76.

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 1 3

Hamilton, Darrick, William Darity, Jr., Anne E. Price, Vishnu Sridharan, and Rebecca Tippett. 2015. Umbrellas Don’t Make It Rain: Why Studying and Working Hard Isn’t Enough for Black Americans. New York: The New School; Durham, NC: Duke Center for Social Equity; Oakland, CA: Insight Center for Community Economic Development.

Kendi, Ibram X. 2017. Stamped from the Beginning: The Definitive History of Racist Ideas in America. New York: Nation Books.

Leachman, M., Michael Mitchell, Nicholas Johnson, and Erica Williams. 2018. Advancing Racial Equity with State Tax Policy. Washington, DC: Center on Budget and Policy Priorities.

Murray, Brittany, Elsa Falkenburger, and Priya Saxena. 2015. Data Walks: An Innovative Way to Share Data with Communities. Washington, DC: Urban Institute.

Oliver, Melvin L., and Thomas M. Shapiro. 2006. Black Wealth/White Wealth: A New Perspective on Racial Inequality, 2nd ed. New York: Routledge.

Shapiro, Thomas M. 2017. Toxic Inequality: How America’s Wealth Gap Destroys Mobility, Deepens the Racial Divide, and Threatens Our Future. New York: Basic Books.

1 4 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

About the Authors K. Steven Brown is a research associate in the Center on Labor, Human Services, and

Population and the Research to Action Lab at the Urban Institute. His work covers

projects concerned with racial disparities in economic opportunity. His primary

research focuses on employment, examining racial and gender differences in career

pathways, barriers in access to work, and gaps in wages and earnings.

Kilolo Kijakazi is an Institute fellow at the Urban Institute, where she works with staff

to develop collaborative partnerships with organizations and individuals who

represent those most affected by the economic and social issues Urban addresses,

expand and strengthen Urban’s rigorous research agenda on issues affecting these

communities, effectively communicate the findings of Urban’s research to diverse

audiences, and recruit and retain more diverse research staff at all levels.

Charmaine Runes is a research analyst in the Center on Labor, Human Services, and

Population. Her work involves both quantitative and qualitative data and methods,

focusing on multigenerational antipoverty interventions that support and empower

disadvantaged youth and low-income working families. Other research interests

include immigrant integration and structural racism in public policy.

Margery Austin Turner is senior vice president for program planning and management

at the Urban Institute, where she leads efforts to frame and conduct a forward-looking

agenda of policy research. A nationally recognized expert on urban policy and

neighborhood issues, Turner has analyzed issues of residential location, racial and

ethnic discrimination and its contribution to neighborhood segregation and inequality,

and the role of housing policies in promoting residential mobility and location choice.

C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS 1 5

Acknowledgments This brief and the November convening with other research and policy organizations were supported by

a generous gift from Cheryl Cohen Effron. We are grateful to all our funders, who make it possible for

Urban to advance its mission.

The views expressed are those of the authors and should not be attributed to the Urban Institute,

its trustees, or its funders. Funders do not determine research findings or the insights and

recommendations of Urban experts. Further information on the Urban Institute’s funding principles is

available at urban.org/fundingprinciples.

We would like to thank all the roundtable participants for their thoughtful contributions to the

discussion and to this brief:

 Patrese Atine, American Indian Higher Education Consortium

 D. Crystal Byndloss, MDRC

 Martha Chavez, Goldman School of Public Policy at the University of California Berkeley

 Ngina Chiteji, NYU Wagner Graduate School of Public Service

 Aixa Cintron-Velez, Russell Sage Foundation

 Melany De La Cruz, UCLA Asian American Studies Center

 William Emmons, Federal Reserve Bank of St. Louis

 Nkechi Erondu, Jocelyn Fontaine, Lionel Foster, Leah Hendey, Nancy La Vigne, Alanna

McCargo, Marla McDaniel, Signe-Mary McKernan, Cameron Okeke, Adaeze Okoli, and Sarah

Rosen Wartell; Urban Institute

 Jessica Fulton, Joint Center for Political and Economic Studies

 Darrick Hamilton, National Economists Association

 Waldo Johnson, University of Chicago

 Crystal Loud Hawk-Hedgepeth, American Indian College Fund

 Michael Mitchell and Erica Williams, Center on Budget and Policy Priorities

 Michelle Morse, Social Medicine Consortium

 Kantahyanee Murray, Annie E. Casey Foundation

 Kathryn Newcomer, Trachtenberg School of Public Policy and Public Administration at George

Washington University

 Lynette Rawlings, The Policy Academies

 Martha Ross, Brookings Institution

 William Spriggs, Howard University

 Erik Stegman, The Aspen Institute

 Susan Sterett, University of Maryland, Baltimore County School of Public Policy

 Romie Tribble, Spelman College

 Janelle Wong, University of Maryland

1 6 C O N F R O N T IN G ST R UC T UR AL R AC ISM IN R E SE AR C H AN D PO LIC Y AN ALY SIS

ABOUT THE URBAN INSTITUTE The nonprofit Urban Institute is a leading research organization dedicated to developing evidence-based insights that improve people’s lives and strengthen communities. For 50 years, Urban has been the trusted source for rigorous analysis of complex social and economic issues; strategic advice to policymakers, philanthropists, and practitioners; and new, promising ideas that expand opportunities for all. Our work inspires effective decisions that advance fairness and enhance the well-being of people and places.

Copyright © February 2019. Urban Institute. Permission is granted for reproduction of this file, with attribution to the Urban Institute.

2100 M Street NW Washington, DC 20037

www.urban.org

M e d i c i n e a n d S o c i e t y

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 384;8 nejm.org February 25, 2021768

Debra Malina, Ph.D., Editor

How Structural Racism Works — Racist Policies as a Root Cause of U.S. Racial Health Inequities

Zinzi D. Bailey, Sc.D., M.S.P.H., Justin M. Feldman, Sc.D., and Mary T. Bassett, M.D., M.P.H.

In the 5 years since one of us published “#Black LivesMatter — A Challenge to the Medical and Public Health Communities” in the Journal,1 we have seen a sea change in the recognition of rac- ism as a durable feature of U.S. society and of its high cost in Black lives. Elected officials, corpo- rate leaders, and academics alike use the slogan “Black Lives Matter,” which has also been widely adopted by members of the public, who by the millions protested the extrajudicial killing of George Floyd.2 With this change comes growing recognition that racism has a structural basis and is embedded in long-standing social policy. This framing is captured by the term “structural racism.”

There is no “official” definition of structural racism — or of the closely related concepts of systemic and institutional racism — although multiple definitions have been offered.3-7 All defi- nitions make clear that racism is not simply the result of private prejudices held by individuals,8 but is also produced and reproduced by laws, rules, and practices, sanctioned and even implemented by various levels of government, and embedded in the economic system as well as in cultural and societal norms.3,8 Confronting racism, therefore, requires not only changing individual attitudes, but also transforming and dismantling the poli- cies and institutions that undergird the U.S. racial hierarchy.

As a legacy of African enslavement, structural racism affects both population and individual health in three interrelated domains: redlining and racialized residential segregation, mass incar- ceration and police violence, and unequal medical care. These examples, among others, share cer- tain cardinal features: harms are historically grounded, involve multiple institutions, and rely on racist cultural tropes.

Redlining and R acialized Residential Segregation

In 1933, the federal government established the Home Owners’ Loan Corporation (HOLC) to ex- pand homeownership as a part of recovery from the Great Depression.8 To guide determinations of mortgage-worthiness, HOLC created maps of at least 239 U.S. cities. Using racial composition as part of its assessment, HOLC staff literally drew red lines (hence “redlining”) around com- munities with large Black populations, flagging them as hazardous investment areas whose resi- dents would not receive HOLC loans. Redlining made mortgages less accessible, rendering pro- spective Black homebuyers vulnerable to preda- tory terms, thereby increasing lender profits, re- ducing access to home ownership, and depriving these communities of an asset that is central to intergenerational wealth transfer. Federal mort- gages were declined regardless of home loan of- ficers’ racial views; it was not personal.

This government-sanctioned practice validat- ed other racist maneuvers, such as restrictive cov- enants that barred Blacks from home ownership by means of legal agreements set up by previous owners, undervaluing of real estate in Black neighborhoods, and mob violence against Blacks who moved into White neighborhoods. Although redlining officially ended with the Fair Housing Act of 1968, its impact is seen today in the social geography of cities. Residential segregation formed a platform for broad social disinvestment, espe- cially in neighborhood infrastructure (e.g., green space, housing stock, and roads), services (e.g., transport, schools, and garbage collection), and employment.

Residential racial segregation remains a pow- erful predictor of Black disadvantage.3,5,9 There is

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

Medicine and Society

n engl j med 384;8 nejm.org February 25, 2021 769

a direct legacy of redlining in health and well- being — preterm birth, cancer, tuberculosis, ma- ternal depression, and other mental health issues occur at higher rates among residents of once- redlined areas.3-5 Plausible mechanisms for the continued health impact of redlining deserve further study, taking into account exposure to environmental toxins (teratogens, carcinogens, air pollutants, etc.) and the sustained physical impact of concentrated psychosocial stressors.5,9-11 Better HOLC neighborhood grades are associat- ed with lower levels of airborne carcinogens and higher levels of tree-canopy coverage (which miti- gates air pollutants and heat).12 Predominantly White neighborhoods generally have lower air- pollution levels,13 while higher exposures con- tribute to asthma and low-birth-weight outcomes in Black communities.14

Redlining required the cooperation of govern- ment; the banking, credit, and real estate indus- tries and private developers; as well as homeown- ers. Together, these parties helped stoke cultural beliefs that Blacks made bad neighbors whose presence would lower real-estate values and in- crease crime. Furthermore, the structural racism that enables and sustains segregation facilitates structural racism in other forms, including mass incarceration and police violence and the unjust distribution of high-quality health care.

Police Violence and the C arcer al State

The United States has the world’s highest incar- ceration rate, and U.S. police kill civilians far more often than do police in other wealthy countries.15,16 A large body of scientific research documents both racially unequal outcomes and racial bias in virtually all aspects of the criminal legal system, with Black people experiencing harsher outcomes in relation to police encounters, bail setting, sentence length, and capital punish- ment than White people.17,18 The history of courts, prisons, and police as institutions that maintain racial hierarchy is key to understanding the deeply punitive and racially unequal nature of the U.S. criminal legal system, with important and persisting implications for the health of Black communities.

Contemporary U.S. policing has roots in slave patrols, which were first established in 18th-cen- tury colonial Virginia in an effort to capture run-

aways and quell uprisings. After the abolition of slavery and the short-lived progress of the Re- construction Era, police and prisons served as key institutions for reasserting White dominance, especially in the South. Law enforcement sanc- tioned, enabled, and participated in the lynching of Black people, which White mobs typically carried out under the pretext of punishment for crime; in reality, lynching often had broader eco- nomic and political motives.19 Southern White people also used police and prisons to enforce vagrancy laws and the convict-leasing and share- cropping systems in order to compel formerly enslaved people to return to the fields — “slavery by another name,” as one author famously put it.20

By the time Congress passed the Civil Rights Acts of 1964, lynching had become rare and the convict-leasing system had been long abandoned. But just months later, President Lyndon Johnson declared a “War on Crime,” which was followed in the next decade by President Richard Nixon’s “War on Drugs,” both of which appealed to fears about supposed Black criminality. These devel- opments portended a sevenfold increase in the size of the incarcerated population, with Black people incarcerated at five times the rate for White people.21,22 As in the post-Reconstruction era, the development of mass incarceration also had economic dimensions — for example, the expansion of prisons provided employment in White, deindustrialized rural areas.23

The late 1960s also saw a massive spike in police killings of Black men,24 and it was not un- til the 1980s that the U.S. Supreme Court placed even modest restrictions on police use of force — for instance by declaring it unconstitutional for police to shoot a civilian who is fleeing a crime scene but poses no harm to others.25 Polic- ing has long been entangled in other structures that reproduce racism, such as residential segre- gation. Police once enforced racial restrictions in “sundown towns” that excluded Black people out- side working hours; they now disproportionately target Black people who enter White neighbor- hoods.26 The police activity that resulted in Bre- onna Taylor’s fatal shooting by police in Ken- tucky has been tied to an “urban revitalization” plan.27

Policing and incarceration have profound ad- verse consequences for the health of Black peo- ple. Some of these consequences are direct — police use of force kills hundreds of Black

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 384;8 nejm.org February 25, 2021770

people each year and nonfatally injures many thousands more. Incarcerated people — who are disproportionately Black — face a high risk of death after release,28 and prisons and jails have been major sites of disease transmission during the Covid-19 pandemic.29 There are also indirect effects — for instance, police violence can harm mental health for entire communities through constant surveillance and threat of violence,30 and the churn in and out of incarceration can result in community spread of sexually transmit- ted infections or other infectious diseases, such as Covid-19.31

The notion that police reform alone will solve police violence is incomplete and misleading. A structural racism lens allows us to see how polic- ing and prisons have served their intended purpose of social control of the Black population, which has long been enforced by violence. For effective change, we must determine which sectors (such as mental health and social services) should be in- volved in equitably addressing public safety with- out necessarily requiring a police response.

Unequal Health C are

Modern American medicine has historical roots in scientific racism and eugenics movements. Scientific racism reified the concept of race as an innate biologic, and later genetic, attribute using culturally influenced scientific theory and in- quiry.32 American scientists, such as Samuel Mor- ton, continued this tradition, using anatomical features such as skull size and volume to catego- rize races in ways that enshrined White superi- ority.32,33 The modern eugenics movement swept through the United States in the early 20th cen- tury, leading to laws prohibiting “miscegenation” and the forced sterilization of undesirable “races” in an effort to create a better, more intelligent, Whiter nation.32

Well-respected medical doctors cast Blacks as innately diseased and dehumanized their suffer- ing, using scientific arguments to provide the illusion of neutrality and objectivity. For instance, in 1851, Southern physician Samuel Cartwright described “drapetomania,” a “mental illness” that he claimed caused enslaved Africans to run away from their confinement; he argued that it could be prevented by keeping Black people in submis- sion and could be cured by whippings.34 Cart-

wright also “discovered” dysaesthesia aethiopica, a “disease” in Black people characterized by reduced intellectual ability, laziness, and partial insensitivity of the skin.34 Similarly, physician J. Marion Sims, who was hailed as the father of modern gynecology, owed his signal accomplish- ment of vesicovaginal fistula repair to repeated operations performed, without anesthesia, on enslaved Black women — women for whom in- formed consent had no meaning.35

Racialized conceptions of susceptibility to dis- ease persist to this day. In its 2003 report Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care, the Institute of Medicine reviewed more than 100 studies and concluded that bias, prejudice, and stereotyping contributed to wide- spread differences in health care by race and ethnicity.36 That call to action went largely un- heeded. Fifteen years later, the 2018 National Healthcare Quality and Disparities Report docu- mented that Black, American Indian and Alaska Native, and Native Hawaiian and Pacific Islander patients continued to receive poorer care than White patients on 40% of the quality measures included, with little to no improvement from decades past.37 This unequal treatment is based, at least in part, in enduring racist cultural beliefs and practices. For instance, in a 2016 study to assess racial attitudes, half of White medical stu- dents and residents held unfounded beliefs about intrinsic biologic differences between Black peo- ple and White people. These false beliefs were associated with assessments of Black patients’ pain as being less severe than that of White pa- tients and with less appropriate treatment deci- sions for Black patients.38

It would be short-sighted to think that indi- vidual prejudice and discrimination alone drive substandard care. The systematic disinvestment in public and private sectors within segregated Black neighborhoods has resulted in under- resourced facilities with fewer clinicians, which makes it more difficult to recruit experienced and well-credentialed primary care providers and spe- cialists and thereby affects access and utilization.3 Black communities became medical training grounds and a source of profit, reinforcing the American medical caste system that we have to- day. Regardless of intent, actions by parties rang- ing from medical schools to providers, insurers, health systems, legislators, and employers have

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

Medicine and Society

n engl j med 384;8 nejm.org February 25, 2021 771

ensured that racially segregated Black communi- ties have limited and substandard care.3

Acceptance of this inequitable treatment as “normal” is historically rooted in and supported by the belief that Black people are intrinsically disease-prone and, implicitly or explicitly, not deserving of high-quality care. As with policing, dismantling structural racism’s impact on health care is not an issue of “a few bad apples”; we must reflect on the ways our everyday, accepted prac- tices reify race — that is, treat the social construct of race as an intrinsic biologic difference — thereby exemplifying and contributing to a broad- er system of structural racism.

Our Role in Dismantling Struc tur al R acism

Structural racism reaches back to the beginnings of U.S. history, stretches across its institutions and economy, and dwells within our culture. Its durability contributes to the perception that Black disadvantage is intrinsic, permanent, and there- fore normal. But considering structural racism as a root cause is not a modern analogue of the theory that disease is caused by “miasmas” — something that’s “in the air,” amorphous and undifferentiated. Structural racism functions to harm health in ways that can be described, mea- sured, and dismantled. Actions to dismantle rac- ism necessarily involve the whole of society. Mov- ing beyond individual education and personal insight to change policy and social norms will require the engagement of many institutions, but the medical and public health communities can contribute directly in at least four key areas.

The first is embracing the intellectual project of documenting the health impact of racism. Despite the long and ongoing history of racism, empirical research showing its impact on health is rarely published in major medical journals. Although we find the evidence of the health ef- fects of structural racism to be convincing, and supported by more than a century of wide-rang- ing theoretical and empirical scholarship, it re- mains marginalized and eclipsed by other research priorities.3-6,39 When leading medical journals ad- dress structural racism, it is often confined to commentaries and editorials, as though these topics are suitable for discussion but not discov- ery. Broad agreement is needed — by funders,

editors, and reviewers — that racism and inequi- ties in social determinants of health more gener- ally are topics as valid for research as biologic markers (and certainly the two can be combined).

Next, the availability of data that include race and ethnicity must improve, and efforts to de- velop and improve measurement of structural racism need to be supported, particularly those using available administrative databases. Such work is under way, and we believe it should be widely encouraged.6,40-44

Third, the medical and public health com- munities need to turn a lens on themselves, both as individuals and as institutions. Faculty and students need a more complete view both of U.S. history and of the ways in which medicine and public health have participated and continue to participate in racist practices. Reflection includes recognition of harms arising from the uncritical use of racial categories, which reinforces implicit assumptions that racial differences are genetic in origin. Furthermore, it includes measuring the success of interventions in terms of how well they narrow inequitable gaps in health (here, between Black people and White people) instead of focus- ing solely on the overall population. Rigorous, clear standards for publishing research on racial health inequities have been proposed.45

Meanwhile, addressing the growing under- representation of Black students in medical school,46 and the disadvantage Black researchers face in seeking awards from the National Insti- tutes of Health47 should not wait. We should call into question claims that there is an inadequate pool of qualified Black applicants to recruit, hire, and promote.

Fourth, we should acknowledge that structural racism has been challenged, perhaps most suc- cessfully, by mass social movements. Change will require policies that restructure the chances for a healthy life for people of color, righting the wrongs done by the foundational racial hierar- chy that continue to shape everyday life. Orga- nized medicine and public health have a long history of opposing desegregation and broader access to care (e.g., Medicare), of barring Black physicians, of championing scientific racism, and of enshrining race as a biologic variable. Our fields have much to regret, and we have much still to offer to right our historical wrongs. Let’s not sit on the sidelines.

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

T h e n e w e ngl a nd j o u r na l o f m e dic i n e

n engl j med 384;8 nejm.org February 25, 2021772

Disclosure forms provided by the authors are available at NEJM.org.

From the University of Miami Miller School of Medicine, Miami (Z.D.B.); and the FXB Center for Health and Human Rights, Harvard University, Boston (J.M.F., M.T.B.).

This article was published on December 16, 2020, at NEJM.org.

1. Bassett MT. #BlackLivesMatter — a challenge to the medi- cal and public health communities. N Engl J Med 2015; 372: 1085-7. 2. Buchanan L, Bui Q, Patel JK. Black Lives Matter may be the largest movement in U.S. history. New York Times. July 3, 2020 (https://www . nytimes . com/ interactive/ 2020/ 07/ 03/ us/ george – f loyd – protests – crowd – size . html). 3. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. Lancet 2017; 389: 1453-63. 4. Gee GC, Ford CL. Structural racism and health inequities: old issues, new directions. Du Bois Rev 2011; 8: 115-32. 5. Williams DR, Lawrence JA, Davis BA. Racism and health: evidence and needed research. Annu Rev Public Health 2019; 40: 105-25. 6. Groos M, Wallace M, Hardeman R, Theall KP. Measuring inequity: a systematic review of methods used to quantify struc- tural racism. J Health Dispar Res Pract 2018; 11: 190-206. 7. Jones CP. Confronting institutionalized racism. Phylon 2003; 50(1-2): 7-22. 8. Rothstein R. The color of law: a forgotten history of how our government segregated America. New York: Liveright Publish- ing, 2017. 9. Williams DR, Collins C. Racial residential segregation: a fundamental cause of racial disparities in health. Public Health Rep 2001; 116: 404-16. 10. Simons RL, Lei M-K, Beach SRH, et al. Discrimination, seg- regation, and chronic inflammation: testing the weathering ex- planation for the poor health of Black Americans. Dev Psychol 2018; 54: 1993-2006. 11. Theall KP, Drury SS, Shirtcliff EA. Cumulative neighbor- hood risk of psychosocial stress and allostatic load in adoles- cents. Am J Epidemiol 2012; 176: Suppl 7: S164-S174. 12. Namin S, Xu W, Zhou Y, Beyer K. The legacy of the Home Owners’ Loan Corporation and the political ecology of urban trees and air pollution in the United States. Soc Sci Med 2020; 246: 112758. 13. Woo B, Kravitz-Wirtz N, Sass V, Crowder K, Teixeira S, Takeuchi DT. Residential segregation and racial/ethnic dispari- ties in ambient air pollution. Race Soc Probl 2019; 11: 60-7. 14. Alexander D, Currie J. Is it who you are or where you live? Residential segregation and racial gaps in childhood asthma. J Health Econ 2017; 55: 186-200. 15. Lartey J. By the numbers: US police kill more in days than other countries do in years. The Guardian. June 9, 2015 (https:// www . theguardian . com/ us – news/ 2015/ jun/ 09/ the – counted – police – killings – us – vs – other – countries). 16. Tsai T, Scommegna P. U.S. has world’s highest incarceration rate. Washington, DC: Population Reference Bureau, 2012 (https://www . prb . org/ us – incarceration/ ). 17. Kutateladze BL, Andiloro NR, Johnson BD, Spohn CC. Cumulative disadvantage: examining racial and ethnic disparity in prosecution and sentencing. Criminology 2014; 52: 514-51. 18. Knox D, Lowe W, Mummolo J. Administrative records mask racially biased policing. Am Polit Sci Rev 2020; 114: 619-37. 19. Ming Francis M. Ida B. Wells and the economics of racial violence. Brooklyn, NY: Social Science Research Council, Janu- ary 24, 2017 (https://items . ssrc . org/ reading – racial – conflict/ ida – b – wells – and – the – economics – of – racial – violence/ ).

20. Blackmon DA. Slavery by another name: the re-enslavement of Black Americans from the Civil War to World War II. New York: Doubleday, 2008. 21. Cullen J. The history of mass incarceration. New York: Bren- nan Center for Justice, July 20, 2018 (https://www . brennancenter . org/ our – work/ analysis – opinion/ history – mass – incarceration). 22. U.S. incarceration rates by race, 2010. Prison Policy Initia- tive, 2020 (https://www . prisonpolicy . org/ graphs/ raceinc . html). 23. Gilmore RW. Golden gulag: prisons, surplus, crisis, and op- position in globalizing California. Berkeley: University of Cali- fornia Press, 2007. 24. Krieger N, Kiang MV, Chen JT, Waterman PD. Trends in US deaths due to legal intervention among black and white men, age 15-34 years, by county income level. Harvard Pub Health Rev 2015; 3: 1-5. 25. Walker S, Fridell L. Forces of change in police policy: the impact of Tennessee v. Garner. Am J Police 1992; 11: 97-112 (https://heinonline . org/ HOL/ LandingPage?handle=hein . journals/ ajpol11&div=27&id=&page=). 26. Gaston S, Brunson RK, Grossman LS. Are minorities sub- jected to, or insulated from, racialized policing in majority– minority community contexts? Br J Criminol 2020. 27. Beck B. The role of police in gentrification. The Appeal. Au- gust 4, 2020 (https://theappeal . org/ the – role – of – police – igentrification – breonna – taylor/ ) 28. Binswanger IA, Stern MF, Deyo RA, et al. Release from pris- on — a high risk of death for former inmates. N Engl J Med 2007; 356: 157-65. 29. The COVID Prison Project tracks data and policy across the country to monitor COVID-19 in correctional facilities. The COVID Prison Project, 2020 (https://covidprisonproject . com/ ). 30. Yimgang DP, Wang Y, Paik G, Hager ER, Black MM. Civil unrest in the context of chronic community violence: impact on maternal depressive symptoms. Am J Public Health 2017; 107: 1455-62. 31. Thomas JC, Levandowski BA, Isler MR, Torrone E, Wilson G. Incarceration and sexually transmitted infections: a neighbor- hood perspective. J Urban Health 2008; 85: 90-9. 32. Jackson J, Weidman NM, Rubin G. The origins of scientific racism. J Blacks High Educ 2005; 50: 66-79. 33. Taylor HF, Hare B. Deconstructing the bell curve: racism, classism, and intelligence in America. In: Hare BR, ed. 2001 Race odyssey: African Americans and sociology. Syracuse, NY: Syracuse University Press, 2002: 60-76. 34. Willoughby CD. Running away from drapetomania: Samuel A. Cartwright, medicine, and race in the Antebellum South. J South Hist 2018; 84: 579-614. 35. Washington HA. Medical apartheid: the dark history of medical experimentation on Black Americans from colonial times to the present. New York: Doubleday Books, 2006. 36. Smedley BD, Stith AY, Nelson AR, eds. Unequal treatment: confronting racial and ethnic disparities in health care. Wash- ington, DC: National Academies Press, 2003. 37. Agency for Healthcare Research and Quality. 2018 National healthcare quality and disparities report (AHRQ publication no. 19-0070-EF). Rockville, MD: Department of Health and Human Services, 2019 (https://www . ahrq . gov/ sites/ default/ files/ wysiwyg/ research/ findings/ nhqrdr/ 2018qdr – final – es . pdf). 38. Hoffman KM, Trawalter S, Axt JR, Oliver MN. Racial bias in pain assessment and treatment recommendations, and false be- liefs about biological differences between blacks and whites. Proc Natl Acad Sci U S A 2016; 113: 4296-301. 39. Hardeman RR, Murphy KA, Karbeah J, Kozhimannil KB. Naming institutionalized racism in the public health litera- ture: a systematic literature review. Public Health Rep 2018; 133: 240-9. 40. Chambers BD, Baer RJ, McLemore MR, Jelliffe-Pawlowski LL. Using index of concentration at the extremes as indicators of

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

Medicine and Society

n engl j med 384;8 nejm.org February 25, 2021 773

structural racism to evaluate the association with preterm birth and infant mortality — California, 2011–2012. J Urban Health 2019; 96: 159-70. 41. Chambers BD, Toller Erausquin J, Tanner AE, Nichols TR, Brown-Jeffy S. Testing the association between traditional and novel indicators of county-level structural racism and birth out- comes among black and white women. J Racial Ethn Health Disparities 2018; 5: 966-77. 42. Liu SY, Fiorentini C, Bailey Z, Huynh M, McVeigh K, Kaplan D. Structural racism and severe maternal morbidity in New York State. Clinical Medicine Insights: Women’s Health. June 14, 2019 (https://journals . sagepub . com/ doi/ pdf/ 10 . 1177/ 1179562X19854778). 43. Lukachko A, Hatzenbuehler ML, Keyes KM. Structural rac- ism and myocardial infarction in the United States. Soc Sci Med 2014; 103: 42-50.

44. Wallace M, Crear-Perry J, Richardson L, Tarver M, Theall K. Separate and unequal: structural racism and infant mortality in the US. Health Place 2017; 45: 140-4. 45. Boyd RW, Lindo EG, Weeks LD, McLemore MR. On racism: a new standard for publishing on racial health inequities. Health Affairs Blog. July 2, 2020 (https://www . healthaffairs . org/ do/ 10 . 1377/ hblog20200630 . 939347/ full/ ). 46. Lett LA, Murdock HM, Orji WU, Aysola J, Sebro R. Trends in racial/ethnic representation among US medical students. JAMA Netw Open 2019; 2(9): e1910490. 47. Hoppe TA, Litovitz A, Willis KA, et al. Topic choice contrib- utes to the lower rate of NIH awards to African-American/black scientists. Sci Adv 2019; 5(10): eaaw7238.

DOI: 10.1056/NEJMms2025396 Copyright © 2020 Massachusetts Medical Society.

The New England Journal of Medicine Downloaded from nejm.org on January 18, 2022. For personal use only. No other uses without permission.

Copyright © 2021 Massachusetts Medical Society. All rights reserved.

The Major Causes of Death in Children and Adolescents in the United States

Rebecca M. Cunningham, M.D., Maureen A. Walton, M.P.H., Ph.D., and Patrick M. Carter, M.D. University of Michigan Injury Prevention Center (R.M.C., M.A.W., P.M.C.), the Firearm Safety among Children and Teens Consortium (R.M.C., M.A.W., P.M.C.), the Department of Emergency Medicine (R.M.C., P.M.C.), and the Addiction Center, Department of Psychiatry (M.A.W.), University of Michigan School of Medicine, and the Youth Violence Prevention Center (R.M.C., P.M.C.) and Department of Health Behavior and Health Education (R.M.C.), University of Michigan School of Public Health — both in Ann Arbor.

In 2016, children and adolescents (1 to 19 years of age) represented a quarter of the total

estimated U.S. population1; reflecting relatively good health, they accounted for less than

2% of all U.S. deaths.2 By 2016, death among children and adolescents had become a rare

event. Declines in deaths from infectious disease or cancer, which had resulted from early

diagnosis, vaccinations, antibiotics, and medical and surgical treatment, had given way to

increases in deaths from injuryrelated causes, including motor vehicle crashes, firearm

injuries, and the emerging problem of opioid overdoses. Although injury deaths have

traditionally been viewed as “accidents,” injuryprevention science that evolved during the

latter half of the 20th century increasingly shows that such deaths are preventable with

evidence-based approaches.

In this report, we summarize the leading causes of death in children and adolescents (1 to 19

years of age) in the United States. Unless otherwise indicated, data on deaths were obtained

from the Wide-ranging Online Data for Epidemiologic Research (WONDER) system of the

Centers for Disease Control and Prevention (CDC), known as CDC WONDER,2 in which

data are derived from U.S. death certificates compiled from 57 vital-statistics jurisdictions.2

Data are presented for 2016, the most recent year with national data available.2 Where

appropriate, rates are expressed per 100,000 children and adolescents and include the 95%

confidence interval.

LEADING CAUSES OF CHILD AND ADOLESCENT DEATH

BURDEN OF DISEASE

In 2016, there were 20,360 deaths among children and adolescents in the United States.

More than 60% resulted from injury-related causes, which included 6 of the 10 leading

causes of death (Table 1, and Table S1 in the Supplementary Appendix, available with the

Address reprint requests to Dr. Cunningham at the Department of Emergency Medicine, University of Michigan, 2800 Plymouth Rd., NCRC 10-G080, Ann Arbor, MI 48109, or at [email protected].

Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.

HHS Public Access Author manuscript N Engl J Med. Author manuscript; available in PMC 2019 July 18.

Published in final edited form as: N Engl J Med. 2018 December 20; 379(25): 2468–2475. doi:10.1056/NEJMsr1804754.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

full text of this article at NEJM.org). Injuries were classified according to underlying

mechanism (e.g., motor vehicle crash or firearm-related injury) and intent (e.g., suicide,

homicide, unintentional, or undetermined), both of which are critical to understanding risk

and protective factors and to developing effective prevention strategies. When we examined

all deaths among children and adolescents according to intent, unintentional injuries were

the most common cause of injury-related death (57%; 7047 of 12,336 deaths), and among

intentional injuries, suicide was slightly more common (21%; 2560 of 12,336) than

homicide (20%; 2469 of 12,336).

Motor vehicle crashes were the leading cause of death for children and adolescents,

representing 20% of all deaths; firearm-related injuries were the second leading cause of

death, responsible for 15% of deaths. Among firearm deaths, 59% were homicides, 35%

were suicides, and 4% were unintentional injuries (e.g., accidental discharge). (The intent

was undetermined in 2% of firearm deaths.) In contrast, among U.S. adults (≥20 years of

age), 62% of firearm deaths were from suicide and 37% were from homicide. Furthermore,

although unintentional firearm deaths were responsible for less than 2% of all U.S. firearm

deaths, 26% occurred among children and adolescents.

Despite improvements in pediatric cancer care, malignant neoplasms were the third leading

cause of death, representing 9% of overall deaths among children and adolescents. The

fourth leading cause of death was suffocation, responsible for 7% of all deaths. Suffocation

(e.g., due to bed linens, plastic bags, obstruction of the airway, hanging, or strangulation)

varies with respect to intent (e.g., homicide, suicide, or unintentional). The remaining six

leading causes of death represented less than 25% of the overall contribution to deaths in

children and adolescents in 2016.

The leading causes of death varied between younger and older children. Among children 1

to 4 years of age, drowning was the most common cause of death, followed by congenital

abnormalities and motor vehicle crashes. Children most commonly drown in swimming

pools (1 to 4 years of age) and in pools, rivers, and lakes4 (≥5 years of age). Among older,

school-aged children (5 to 9 years of age), death was relatively rare, representing only 12%

of all deaths in children and adolescents. In this age group, malignant neoplasm was the

leading cause of death, followed by motor vehicle crashes and congenital abnormalities.

Unlike in children 1 to 4 years of age, drowning was only the fourth most common cause of

death among those 5 to 9 years of age, which potentially reflects widespread swim training

among school-aged children.5

The majority (68%) of youth who died did so during adolescence. Among these adolescent

youth (10 to 19 years of age), injury deaths from motor vehicle crashes, firearms, and

suffocation were the three leading causes of death; these findings reflect social and

developmental factors that are associated with adolescence, including increased risk-taking

behavior, differential peer and parental influence, and initiation of substance use.6

There were also differences in intent for injuryrelated causes of death between children and

adolescents. Although unintentional injuries were the most common intent underlying injury

deaths among children, intentional causes (i.e., homicide and suicide) were increasingly

Cunningham et al. Page 2

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

common with injury deaths during adolescence. For example, although unintentional causes

comprised 26% of all firearm deaths among children (1 to 9 years of age), they represented

3% of firearm deaths among adolescents (10 to 19 years of age). Similarly, unintentional

causes comprised 78% of all suffocation deaths among children, whereas they comprised

7% of suffocation deaths among adolescents.

Finally, although intentional causes of death were an increasingly important factor during

adolescence, the underlying intent varied according to mechanism. For example, among

adolescents, 61% of intentional firearm deaths (1733 of 2835) resulted from homicide and

98% of intentional suffocation deaths (1103 of 1128) resulted from suicide. Such variations

highlight the need to implement public health strategies that are tailored according to age,

underlying developmental factors, and injury-related intent.

TIME TRENDS

In 1900, the leading causes of death for the entire U.S. population were pneumonia,

tuberculosis, and diarrhea or enteritis, with 40% of these deaths occurring among children

younger than 5 years of age.7 In 2016, none of these diseases were among the 10 leading

causes of child and adolescent death, with declines in mortality from infectious disease

continuing to occur.

The rate of deaths from motor vehicle crashes among children and adolescents showed the

most notable change over time (Fig. 1), with a relative decrease of 38% between 2007 and

2016. This has been attributed to the widespread adoption of seat belts and appropriate child

safety seats, the production of cars with improved safety standards, better constructed roads,

graduated driver-licensing programs,8,9 and a focus on reducing teen drinking and driving.

Such reductions in mortality occurred despite increases in the overall number of U.S.

vehicles and annual vehiclemiles traveled.10 Unfortunately, there was a reversal of this trend

in mortality, with the rate increasing annually between 2013 and 2016. Although the cause

of this reversal is not yet clear, it probably is multifactorial and includes such factors as an

increase in distracted driving by teenagers11 (e.g., because of peer passengers or cell-phone

use). Finally, although the effect of the changing landscape of marijuana legalization on

adolescent crash risk is to date unknown, decreased risk perceptions among adolescents12

arouse concern about potential drugged driving and motor vehicle crashes, with future data

needed.

Although firearm-related mortality among children and adolescents was lower in 2016 than

the most recent peak mortality observed in 1993 (8.12 per 100,000; 95% confidence interval

[CI], 7.91 to 8.23), rates remained stable between 2007 and 2016 without improvement, with

an overall rate of 3.54 per 100,000 (95% CI, 3.50 to 3.58). Between 2013 and 2016, there

was a 28% relative increase in the rate of firearm deaths. This upward trend in firearm

mortality reflected increases in rates of firearm homicide (by 32%) and firearm suicide (by

26%), whereas rates of unintentional firearm deaths remained relatively stable. The

nonfirearm suicide rate increased 15% while the nonfirearm homicide rate decreased 4%

between 2013 and 2016. Although firearm violence in school settings makes up less than 1%

of all suicides and homicides among schoolaged children and adolescents,13 a recent review

Cunningham et al. Page 3

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

noted increasing trends in school shooting incidents, with 154 between 2013 and 2015 (35,

55, and 64, respectively, per year).14

The rate of death from malignant neoplasm, the sole non–injury-related cause among the

five leading causes of death, decreased 32% between 1990 and 2016, which reflects

scientific advancements in cancer prevention, detection, and treatment.15 Drowning deaths

declined by 46% during that time period because of public health efforts, including

mandatory fencing around pools and a greater focus on pool safety (e.g., lifeguards, use of

life jackets, and swimming lessons).16 Deaths due to residential fires fell nearly 73%

between 1990 and 2016, in part owing to decreasing rates of smoking,17 increased

installation of smoke detectors, and improved building fire codes.18,19

In contrast, drug overdoses or poisonings rose to the sixth leading cause of death among

children and adolescents in 2016. This increase was largely due to an increase in opioid

overdoses,20 which account for well over half of all drug overdoses among adolescents.

GLOBAL COMPARISONS

Figure 2 shows the rates of the two leading causes of child and adolescent death in the

United States, as compared with rates in other high-income countries and in low-to-middle-

income countries with available World Health Organization (WHO) data for 2016 (see Fig.

S1 in the Supplementary Appendix for data on all countries with WHO data for 2016).21

The rate of death from motor vehicle crashes among U.S. children and adolescents was the

highest observed among high-income countries; the U.S. rate was more than triple the

overall rate observed in 12 other developed countries (5.21 per 100,000 [95% CI, 5.06 to

5.38] vs. 1.63 per 100,000 [95% CI, 1.49 to 1.77]). Although the U.S. rate of death from

motor vehicle crashes was higher than the rates in other, similar English-speaking countries,

such as Australia (2.94 per 100,000; 95% CI, 2.52 to 3.43) and England and Wales (1.04 per

100,000; 95% CI, 0.87 to 1.23), the disproportionate rate among U.S. children and

adolescents was most pronounced relative to the rate in Sweden (0.91 per 100,000; 95% CI,

0.56 to 1.45), where government investment in road-traffic safety through a Vision Zero

policy22 probably contributed to a rate that was approximately one sixth that in the United

States.

In contrast, rates of death from motor vehicle crashes among children and adolescents in

lowto-middle-income countries were more variable, probably owing to differential levels of

economic development.23 Rates of death from motor vehicle crashes are rising in developing

countries despite global initiatives such as the United Nations Sustainable Development

Goals,24 owing in large part to underinvestment in road infrastructure, underdeveloped

public health infrastructure, limited access to emergency health care services, and a lack of

widespread safety measures.25 Thus, although the rate of death from motor vehicle crashes

among children and adolescents was lower in the United States than in some low-to-middle-

income countries, there remains room for improvement in comparison with similar high-

income countries.26

The rate of firearm deaths among children and adolescents was higher in the United States

than in all other high-income countries and low to-middle-income countries with available

Cunningham et al. Page 4

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

2016 data. The rate in the United States was 36.5 times as high as the overall rate observed

in 12 other high-income countries (4.02 per 100,000 [95% CI, 3.88 to 4.16] vs. 0.11 per

100,000 [95% CI, 0.08 to 0.15]). Only 3 high-income countries (Croatia, Lithuania, and

Sweden) had rates exceeding 0.20 per 100,000. Similarly, the U.S. rate was 5 times as high

as the overall rate in 7 lowto-middle-income countries (0.80 per 100,000; 95% CI, 0.69 to

0.92). Although these comparisons use only 2016 data, the findings are similar to those of

previous analyses that used multiple years of data.27,28

One in three U.S. homes with youth under 18 years of age has a firearm, with 43% of homes

reporting that the firearm is kept unlocked and loaded, which increases the risk of firearm

injuries.29 In addition to differences in availability between the United States and other

countries, there is wide variability across countries in laws relating to the purchase of

firearms, access to them, and safe storage.30

In contrast with rates of death from motor vehicle crashes or firearms, the rate of death from

malignant neoplasm among children and adolescents in the United States (2.37 per 100,000;

95% CI, 2.27 to 2.48) was similar to the overall rate in other high-income countries (2.32 per

100,000; 95% CI, 2.16 to 2.49) (see Fig. S1 in the Supplementary Appendix for information

on all countries with available 2016 data). The U.S. rate was 36% lower than the combined

rate in low-to-middle-income countries (3.64 per 100,000; 95% CI, 3.41 to 3.89), which

probably reflects differential environmental and genetic exposures combined with early

detection and treatment from advanced diagnostics and a more developed health

infrastructure in the United States.31

HEALTH DISPARITIES — RURALITY, RACE, ETHNIC GROUP, POVERTY, AND SEX

There were disparities in patterns of mortality according to rurality, race or ethnic group, and

sex. Rural children and adolescents had higher mortality (33.4 per 100,000; 95% CI, 32.4 to

34.5) than those living in either suburban settings (27.5 per 100,000; 95% CI, 26.8 to 28.0)

or urban settings (23.5 per 100,000; 95% CI, 23.0 to 23.9). These differences were primarily

due to higher injury-related mortality in rural settings (Fig. 3, and Fig. S2 in the

Supplementary Appendix), particularly with respect to motor vehicle crashes (the rate in

rural settings was 2.7 times the rate in urban settings), fire or burn injuries (3.3 times),

drowning (1.8 times), and suffocation (1.3 times).

Several factors contribute to this disparity. First, sparsely populated rural settings are

associated with longer emergency medical service response times, which can delay available

trauma services.32,33 Second, the markedly higher rates of death from motor vehicle crashes

in rural settings persist after adjustment for the differences in vehicle-miles traveled. These

higher rates of death are probably due to environmental factors (e.g., long stretches of

uninterrupted roads, which may lead to higher speeds, and a lack of divided roads),32,34,35

behavioral factors (e.g., less use of seat belts and child safety seats and more alcohol-

impaired driving), and policy factors (e.g., lower enforcement of traffic laws).32

Deaths from residential fires were more common in rural settings than in nonrural settings,

owing to older homes, the use of more dangerous heating sources, and lower rates of smoke-

detector and fire-alarm availability.32,36–38 Children and adolescents died from firearm

Cunningham et al. Page 5

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

injuries at a similar rate in urban settings (4.05 per 100,000) and rural settings (3.84 per

100,000); however, the firearm homicide rate was 2.3 times as high among urban youth as

among rural youth, and the firearm suicide rate was 2.1 times as high among rural youth as

among urban youth. Finally, the rate of overdose death was slightly higher (1.4 times as

high) among urban youth than among rural youth. This probably reflects the mixed nature of

the opioid epidemic, with a greater availability of heroin in urban settings39 and the

disproportionate effect of prescription opioids in rural settings.40,41

For all leading causes of death, male children and adolescents died at higher rates than their

female counterparts, with the disparity widening from a ratio of 1.2 times as high among

children 1 year of age to 2.8 times as high by 19 years of age. This higher rate among male

children and adolescents was most pronounced for firearm deaths (5.1 times the rate among

female children and adolescents), drowning deaths (2.5 times), and suffocation deaths (1.8

times). Although less pronounced, disparities between boys and girls in injury-related

mortality persisted even among children 1 to 4 years of age. Such disparities probably reflect

differential socialization and normative constraints that lead to higher levels of risk-taking

behavior among boys.42

With regard to race or ethnic group, mortality was higher among blacks (38.2 per 100,000;

95% CI, 37.1 to 39.3) and American Indians or Alaska Natives (28.0 per 100,000; 95% CI,

25.4 to 30.9) than among whites (24.2 per 100,000; 95% CI, 23.8 to 24.6) and Asians or

Pacific Islanders (15.9 per 100,000; 95% CI, 14.8 to 17.0). Disparities for black youth

resulted from higher mortality for both injury-related causes (i.e., firearms, drowning, and

fire or burns) and medical causes (i.e., heart disease and respiratory disease). The disparities

were most pronounced for deaths related to firearms, which were the lead ing cause of death

among black youth and occurred at a rate 3.7 times as high as the rate among white youth.

Black youth also had higher rates of drowning deaths (1.6 times as high) and fire-related

deaths (2.3 times as high) than white youth. For medical illnesses, blacks had rates of death

from heart disease and chronic lower respiratory diseases (e.g., asthma) that were 2.1 and

6.3 times as high, respectively, as the rates among white youth. Such disparities probably

reflect underlying socioeconomic issues, including poverty, environmental exposures, and

differential access to health care services.43–45

American Indian and Alaska Native youth had the highest rates of death from motor vehicle

crashes or suffocation in comparison with other races or ethnic groups; this group also had a

higher rate of firearm deaths than white youth. These disparities probably reflect both the

rural nature of many reservation communities and higher rates of risky driving behaviors,

including drunk driving and nonuse of seat belts.46 Disproportionate rates of suicide (by

suffocation and firearm) may reflect risk factors such as alcohol misuse and untreated mental

health issues, in concert with poor access to medical and mental health care.46 In contrast,

white youth had a rate of death due to drug overdose or poisoning that was nearly twice as

high as the rates observed in other races or ethnic groups, a finding that mirrors the overdose

trends among adults, which may reflect factors related to setting (e.g., a high proportion of

whites in rural settings) as well as differential prescribing practices according to race.40,47

Cunningham et al. Page 6

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

Non-Hispanic children had higher mortality across all 10 leading causes of death than

Hispanic children, with the exception of malignant neoplasm, for which the rates were

similar. However, CDC WONDER data may underestimate rates of death among Hispanics.2

Finally, one limitation of CDC WONDER data is the lack of inclusion of poverty variables.

However, a broad literature indicates that poverty is an important risk factor for injury across

ages,48 including contributing to increased risks of motor vehicle crashes49 and firearm

injuries.50

REDUCING DEATHS IN CHILDHOOD AND ADOLESCENCE

Childhood and adolescent mortality remains overwhelmingly related to preventable injury-

related causes of death. Progress toward further reducing deaths among children and

adolescents will require a shift in public perceptions so that injury deaths are viewed not as

“accidents,” but rather as social ecologic phenomena that are amenable to prevention. The

sound application of rigorous scientific public health methods has resulted in considerable

success in some areas of injury, notably childhood deaths due to motor vehicle crashes,

drowning, and residential fires. Expanding public health approaches to encompass all the

leading causes of death could substantially reduce childhood and adolescent mortality, as

well as the disparities observed.

Supplementary Material

Refer to Web version on PubMed Central for supplementary material.

Acknowledgments

We thank Dr. Jason Goldstick for his assistance with World Health Organization and Web-based Injury Statistics Query and Reporting System data abstraction related to this manuscript, and Jessica Roche and Wendi Mohl for their assistance in the preparation of an earlier version of the manuscript.

References

1. U.S. Census Bureau, Population Division. Estimates of the U.S. population by age and sex: April 1, 2010, to July 1, 2016 (https://www.census.gov/newsroom/press-releases/2017/cb17-tps38- population-estimates-single-year-age.html).

2. Centers for Disease Control and Prevention, National Center for Health Statistics. Compressed mortality file, 1999–2016, on CDC WONDER online database. 2017 (https://wonder.cdc.gov/cmf- icd10.html).

3. ICD10Data.com home page (http://www.icd10data.com/).

4. Brenner RA, Trumble AC, Smith GS, Kessler EP, Overpeck MD. Where children drown, United States, 1995. Pediatrics 2001; 108: 85–9. [PubMed: 11433058]

5. Brenner RA. Prevention of drowning in infants, children, and adolescents. Pediatrics 2003; 112: 440–5. [PubMed: 12897306]

6. Arnett JJ. Adolescent storm and stress, reconsidered. Am Psychol 1999; 54: 317–26. [PubMed: 10354802]

7. Control of infectious diseases. MMWR Morb Mortal Wkly Rep 1999; 48: 621–9. [PubMed: 10458535]

8. Zakrajsek JS, Shope JT, Greenspan AI, Wang J, Bingham CR, Simons-Morton BG. Effectiveness of a brief parent-directed teen driver safety intervention (Checkpoints) delivered by driver education instructors. J Adolesc Health 2013; 53: 27–33. [PubMed: 23481298]

Cunningham et al. Page 7

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

9. Simons-Morton BG, Hartos JL, Leaf WA, Preusser DF. The effect on teen driving outcomes of the Checkpoints Program in a state-wide trial. Accid Anal Prev 2006; 38: 907–12. [PubMed: 16620739]

10. U.S. Department of Transportation, Federal Highway Administration, Office of Highway Policy Information. Travel monitoring: traffic volume trends. 2018 (https://www.fhwa.dot.gov/ policyinformation/travel_monitoring/tvt.cfm).

11. U.S. Department of Transportation, National Highway Traf fic Safety Administration. Traffic safety facts: distracted driving 2015. 3 2017 (https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/ documents/812_381_distracteddriving2015.pdf).

12. Department of Health and Human Services, Substance Abuse and Mental Health Service Administration, Center for Behavioral Health Statistics and Quality. Results from the 2013 National Survey on Drug Use and Health: summary of national findings. Rockville, MD: Office of Applied Studies, 2014.

13. Musu-Gillette L, Zhang A, Wang K, et al. Indicators of school crime and safety: 2017. Washington, DC: National Center for Education Statistics, Department of Education, Department of Justice Office of Justice Programs, 2018.

14. Kalesan B, Lagast K, Villarreal M, Pino E, Fagan J, Galea S. School shootings during 2013–2015 in the USA. Inj Prev 2017; 23: 321–7. [PubMed: 27923800]

15. Ward E, DeSantis C, Robbins A, Kohler B, Jemal A. Childhood and adolescent cancer statistics, 2014. CA Cancer J Clin 2014; 64: 83–103. [PubMed: 24488779]

16. Quan L, Liller KD, Bennett E. Water-related injuries of children and adolescents In: DeSafey Liller K, ed. Injury prevention for children and adolescents. Washington, DC: Alpha Press, 2012: 295– 302.

17. Leistikow BN, Martin DC, Milano CE. Fire injuries, disasters, and costs from cigarettes and cigarette lights: a global overview. Prev Med 2000; 31:91–9. [PubMed: 10938207]

18. Warda LJ, Ballesteros MF. Interventions to prevent residential fire injury In: Doll LS, Bonzo SE, Mercy JA, Sleet DA, Haas EN, eds. Handbook of injury and violence prevention. New York: Springer, 2008: 97–115.

19. Miller TR, Finkelstein AE, Zaloshnja E, Hendrie D. The cost of child and adolescent injuries and the savings from prevention In: DeSafey Liller K, ed. Injury prevention for children and adolescents. Washington, DC: Alpha Press, 2012: 50.

20. Department Of Health and Human Services, Office of Adolescent Health. Opioids and adolescents. 2017 (https://www.hhs.gov/ash/oah/adolescent-development/substance-use/drugs/opioids/ index.html).

21. WHO mortality database. Geneva: World Health Organization (http://www.who.int/healthinfo/ mortality_data/en/).

22. Johansson R Vision Zero — implementing a policy for traffic safety. Saf Sci 2009; 47: 826–31.

23. Kopits E, Cropper M. Traffic fatalities and economic growth. Accid Anal Prev 2005; 37: 169–78. [PubMed: 15607288]

24. United Nations. Transforming our world: the 2030 agenda for sustainable development — resolution adopted by the General Assembly, 2015 (https://sustainabledevelopment.un.org/ post2015/transformingourworld).

25. Staton C, Vissoci J, Gong E, et al. Road traffic injury prevention initiatives: a systematic review and metasummary of effectiveness in low and middle income countries. PLoS One 2016; 11(1): e0144971. [PubMed: 26735918]

26. Evans L Traffic fatality reductions: United States compared with 25 other countries. Am J Public Health 2014; 104: 1501–7. [PubMed: 24922136]

27. Grinshteyn E, Hemenway D. Violent death rates: the US compared with other high-income OECD countries, 2010. Am J Med 2016; 129: 266–73. [PubMed: 26551975]

28. Hemenway D, Miller M. Firearm availability and homicide rates across 26 high-income countries. J Trauma 2000; 49: 985–8. [PubMed: 11130511]

29. Schuster MA, Franke TM, Bastian AM, Sor S, Halfon N. Firearm storage patterns in US homes with children. Am J Public Health 2000; 90: 588–94. [PubMed: 10754974]

30. Carlsen A, Chinoy S. How to buy a gun in 15 countries. New York Times. 3 2, 2018 (https:// www.nytimes.com/interactive/2018/03/02/world/international-gun-laws.html).

Cunningham et al. Page 8

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

31. Howard SC, Metzger ML, Wilimas JA, et al. Childhood cancer epidemiology in low-income countries. Cancer 2008; 112: 461–72. [PubMed: 18072274]

32. Peek-Asa C, Zwerling C, Stallones L. Acute traumatic injuries in rural populations. Am J Public Health 2004; 94: 1689–93. [PubMed: 15451733]

33. Maio RF, Green PE, Becker MP, Burney RE, Compton C. Rural motor vehicle crash mortality: the role of crash severity and medical resources. Accid Anal Prev 1992; 24: 631–42. [PubMed: 1388581]

34. Karlaftis MG, Golias I. Effects of road geometry and traffic volumes on rural roadway accident rates. Accid Anal Prev 2002; 34: 357–65. [PubMed: 11939365]

35. Baker DR, Clarke SR, Brandt EN Jr. An analysis of factors associated with seat belt use: prevention opportunities for the medical community. J Okla State Med Assoc 2000; 93: 496–500. [PubMed: 11077757]

36. The rural fire problem in the United States. Washington, DC: United States Fire Administration, Federal Emergency Management Agency, 8 1997 (https://www.usfa.fema.gov/downloads/pdf/ statistics/rural.pdf).

37. Hall JR Jr. The U.S. experience with smoke detectors: who has them? How well do they work? When don’t they work? NFPA J 1994; 88: 36–46. [PubMed: 10137381]

38. Harvey PA, Sacks JJ, Ryan GW, Bender PF. Residential smoke alarms and fire escape plans. Public Health Rep 1998; 113: 459–64. [PubMed: 9769771]

39. Curtin SC, Tejada-Vera B, Warner M. Drug overdose deaths among adolescents aged 15–19 in the United States: 1999–2015 NCHS data brief no. 282. Atlanta: National Center for Health Statistics, 2017.

40. Havens JR, Young AM, Havens CE. Nonmedical prescription drug use in a nationally representative sample of adolescents: evidence of greater use among rural adolescents. Arch Pediatr Adolesc Med 2011; 165: 250–5. [PubMed: 21041587]

41. Monnat SM, Rigg KK. Examining rural/urban differences in prescription opioid misuse among US adolescents. J Rural Health 2016; 32: 204–18. [PubMed: 26344571]

42. Byrnes JP, Miller DC, Schafer WD. Gender differences in risk taking: a meta-analysis. Psychol Bull 1999; 125: 367–83.

43. Cubbin C, LeClere FB, Smith GS. Socioeconomic status and injury mortality: individual and neighbourhood determinants. J Epidemiol Community Health 2000; 54: 517–24. [PubMed: 10846194]

44. Cubbin C, LeClere FB, Smith GS. Socioeconomic status and the occurrence of fatal and nonfatal injury in the United States. Am J Public Health 2000; 90: 70–7. [PubMed: 10630140]

45. Williams DR, Wyatt R. Racial bias in health care and health: challenges and opportunities. JAMA 2015; 314: 555–6. [PubMed: 26262792]

46. Sarche M, Spicer P. Poverty and health disparities for American Indian and Alaska Native children: current knowledge and future prospects. Ann N Y Acad Sci 2008; 1136: 126–36. [PubMed: 18579879]

47. McCabe SE, West BT, Veliz P, McCabe VV, Stoddard SA, Boyd CJ. Trends in medical and nonmedical use of prescription opioids among US adolescents: 1976–2015. Pediatrics 2017; 139(4): e20162387. [PubMed: 28320868]

48. Karb RA, Subramanian SV, Fleegler EW. County poverty concentration and disparities in unintentional injury deaths: a fourteen-year analysis of 1.6 million US fatalities. PLoS One 2016; 11(5): e0153516. [PubMed: 27144919]

49. Mannocci A, Saulle R, Villari P, La Torre G. Male gender, age and low income are risk factors for road traffic injuries among adolescents: an umbrella review of systematic reviews and metaanalyses. J Public Health 2018 6 1 (Epub ahead of print).

50. Lee J, Moriarty KP, Tashjian DB, Patterson LA. Guns and states: pediatric firearm injury. J Trauma Acute Care Surg 2013; 75: 50–3. [PubMed: 23778438]

Cunningham et al. Page 9

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

Figure 1. Mortality Rates (Deaths per 100,000 Children and Adolescents) for the 10 Leading Causes of Death in the United States from 1999 to 2016. Data were obtained from the Wide-ranging Online Data for Epidemiologic Research

(WONDER) system of the Centers for Disease Control and Prevention (CDC), known as

CDC WONDER,2 according to the codes of the International Classification of Diseases, 10th Revision (ICD-10),3 for the leading causes of death among children and adolescents.

Age was restricted to children and adolescents 1 to 19 years of age.

Cunningham et al. Page 10

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

Figure 2. Global Comparison of Mortality for the Two Leading Causes of Child and Adolescent Death in the United States in 2016. Rates of death in countries other than the United States are from the World Health

Organization (WHO) Mortality Database,21 according to ICD-10 codes3 for leading causes

of death (www.who.int/healthinfo/statistics/mortality_rawdata/en/). Death counts were

tabulated with the use of the same ICD-10 codes that were specified in the CDC WONDER

query (Fig. S3 in the Supplementary Appendix). Population denominators that were used to

calculate rates were obtained from files available on the WHO Mortality Database website

for population according to country and age group, and data on the five largest high-income

countries (besides the United States) and the four largest low-to-middle-income countries are

presented here. In the case of Australia, 2016 population data were not available and 2015

data are presented. The error bars indicate 95% confidence intervals. See Figure S1 in the

Supplementary Appendix for all countries with available 2016 data.

Cunningham et al. Page 11

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

Figure 3. Mortality for the Five Leading Causes of Child and Adolescent Death in 2016, According to Rurality. Data were obtained from the CDC WONDER database,2 according to ICD-10 codes3 for the

leading causes of death among children and adolescents. The 2013 National Center for

Health Statistics Urban–Rural Classification Scheme for Counties was used to assign one of

six categories to each county in the United States. Counties were classified as urban (Large

Central Metro or Large Fringe Metro), suburban (Medium Metro or Small Metro), or rural

(Micropolitan or Noncore). The I bars indicate 95% confidence intervals. See Figure S2 in

the Supplementary Appendix for data on all 10 leading causes of death.

Cunningham et al. Page 12

N Engl J Med. Author manuscript; available in PMC 2019 July 18.

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

A uthor M

anuscript A

uthor M anuscript

Cunningham et al. Page 13

Table 1.

The 10 Leading Causes of Child and Adolescent Death in the United States in 2016, in Order of Frequency.*

Cause of Death No. of Deaths Rate per 100,000 (95% CI) Percent of Deaths

All causes 20,360 26.06 (25.70–26.42)

All injury-related causes 12,336 15.79 (15.51–16.07) 60.6

Motor vehicle crash 4,074 5.21 (5.06–5.38) 20.0

Firearm-related injury 3,143 4.02 (3.88–4.16) 15.4

Homicide 1,865 2.39 (2.28–2.50)

Suicide 1,102 1.41 (1.33–1.50)

Unintentional 126 0.16 (0.13–0.19)

Undetermined intent 50 0.06 (0.05–0.09)

Malignant neoplasm 1,853 2.37 (2.27–2.48) 9.1

Suffocation† 1,430 1.83 (1.74–1.93) 7.0

Suicide 1,110 1.42 (1.34–1.51)

Unintentional 235 0.30 (0.26–0.34)

Drowning 995 1.27 (1.20–1.36) 4.9

Drug overdose or poisoning 982 1.26 (1.18–1.34) 4.8

Suicide 123 0.16 (0.13–0.19)

Unintentional 761 0.97 (0.91–1.05)

Congenital anomalies 979 1.25 (1.18–1.33) 4.8

Heart disease 599 0.77 (0.71–0.83) 2.9

Fire or burns 340 0.44 (0.39–0.48) 1.7

Unintentional 272 0.35 (0.31–0.39)

Chronic lower respiratory disease 274 0.35 (0.31–0.40) 1.3

* Data were obtained from the Wide-ranging Online Data for Epidemiologic Research system of the Centers for Disease Control and Prevention,2

according to the codes of the International Classification of Diseases, 10th Revision (ICD-10),3 for the leading causes of death among children and adolescents. Age was restricted to children and adolescents 1 to 19 years of age. Cruderates (deaths per 100,000) were calculated with a population denominator of 78,134,923, with 95% confidence intervals (CIs) presented. All data are calculated for 2016, the most recent year with available data. See Table S1 in the Supplementary Appendix for more data regarding intent (homicide, suicide, unintentional, or undetermined).

† Suffocation includes such incidents as suffocation or strangulation due to bed linen, the mother’s body, pillows, or plastic bags. It also includes

aspiration or obstruction of the airway by a food bolus, a foreign body, or vomitus. The category also includes intentional self-harm by hanging and intentional violence by strangulation or suffocation. For a complete list of ICD codes and definitions, see Figure S3 in the Supplementary Appendix.

N Engl J Med. Author manuscript; available in PMC 2019 July 18.