Joint Symposium of the Roundtable on Population Health Improvement and the Interdisciplinary Association for Population Health Science
1. Identify key trends and patterns in U.S. population health and provocative challenges to our current thinking, including socio-economic and racial/ethnic disparities
2. Explore how population health science can inform policy to improve outcomes and how policies can have unintended consequences if not grounded in research and evaluation3. Showcase perspectives on how we build common ground for development and implementation of effective policy based on science.
Speaker: Sanne Magnan, HealthPartners Institute
Speaker: Bruce Link, University of California, Riverside
Moderator: Robert Hummer, University of North Carolina, Chapel Hill
Speaker: Eileen Crimmins, University of Southern California
Speaker: David Williams, Harvard University
Policy - Relevant Evidence for Population Health: Promise and Challenges Wednesday, Oct 3, 10:00-11:15am
Moderator: Allison Aiello, University of North Carolina, Chapel Hill
Speaker: Sandro Galea, Boston University
Discussant: Jennifer Doleac, University of Virgina
Discussant: Paula Lantz, University of Michigan
Effective Population Health Policy and the Science: Finding Common GroundWednesday, Oct 3, 11:30-12:45pm
Moderator: Josh Sharfstein, Johns Hopkins University
Speaker: Leana Wen, City of Baltimore
Speaker: Ellen Marie Whelan, Centers for Medicare & Medicaid Services
Speaker: Rahul Gupta, Department of Health and Human Resources Bureau for Public Health, West Virginia
Invited Plenary Sessions
This session brings together leading national experts on the integration of social and biological approaches to the study of population health. The purpose of the session is to generate updated insights and present key questions regarding contemporary scientific research that uses biosocial approaches to better understand population health in the United States.
Allison Aiello, University of North Carolina, Chapel Hill
Robert Hummer, University of North Carolina, Chapel Hill
Moderator: Bridget Goosby, University of Texas, Austin
Speaker: Kathleen Mullan Harris, University of North Carolina, Chapel Hill
Speaker: Thomas McDade, Northwestern University
Discussant: Jennifer Beam Dowd, City University of New York
Adult health and life expectancy differ markedly across U.S. states. Since the 1980s, these differences have greatly expanded. During the same period, structural trends such as deregulation, federal devolution, and state preemption laws altered the political economies of states and consolidated much decision- making authority at the state level. Are these trends related? If so, what are the driving forces? How can population health scientists intervene? The panel will discuss these provocative questions.
Organizer: Jennifer Karas Montez, Syracuse University
Advances in technology and data science research have ushered in a new era promising precision healthcare and improved population health. New models for doing science are being promulgated that emphasize machine learning, participant engagement and responsible data sharing, with researchers, health care providers, and communities working together. This panel brings together leading experts to discuss the opportunities and challenges of harnessing big data to advance health.
Allison Aiello, University of North Carolina, Chapel Hil
Lorna Thorpe, New York University Medical Center
Moderator: Lorna Thorpe, New York University Medical Center
Speaker: Joshua Denny, Vanderbilt University Medical Center
Discussant: Anthony Iton, University of California, Berkeley
Discussant: Alfredo Morabia, Barry Commoner Center, Queens College CUNY and Columbia University
Discussant: Ebony Boulware, Duke University
Poster Sessions will be held in the Lecture Room. Poster awards will be given to the top three selected posters, presented by a student. Poster Award Winners will be announced the following morning, in the opening session.
Six Concurrent Session blocks will be held October 3-5 with 5-6 sessions in each block.
Concurrent Session 1
Wednesday, October 3
2:30pm – 3:45pm
Integrating Genetics into Population Health Research
Session Organizer: Jason Fletcher, University of Wisconsin-Madison
Session Room: Auditorium
In Utero Exposure Interact with Genetic Factors to Predict Old Age Cognition
Jason Fletcher, University of Wisconsin-Madison
Genetic association of 565 variants with alcohol and tobacco use
Robbee Wedow, University of Colorado
Genetic Cognitive Epidemiology: Using Molecular Genetic Data to Explore the Link Between Cognition and Mortality
Sam Trejo, Stanford University
Modeling the genetic architecture of plasticity in health to better understand gene-environment interactions
Dalton Conley, Princeton University
The purpose of our panel is to showcase four research papers that together explore a variety of ways that population health research is being integrated with genetics data and methods. We bring together researchers from economics/policy, sociology, and education in our panel who have experience working and speaking in interdisciplinary settings. We will introduce the audience to new methods and discoveries in genetics (Genome Wide Association Studies/GWAS and polygenic scores) as well as applications of these discoveries in two datasets using alternative methodologies, including gene-environment interactions and instrumental variables. Wedow will present new findings linking over a million individuals’ alcohol and tobacco outcomes with their genetic information to uncover new genetic discoveries that may then become targets of drugs and therapies. He will provide an overview of how massive datasets are now being mined for these associations and how bioinformatics information is being leveraged to create new understandings of the biological bases for some important health behavior outcomes. He will also define and explore the use of polygenic scores—genome wide summaries—of health behavioral risk that are becoming more widely used across the social and health sciences. Fletcher will explore the life course importance of in utero exposures, with a focus on economic disruptions from the Great Depression, on old age cognition levels and trajectories using the Health and Retirement Study. He uses state and year variation in economic conditions using state and year of birth information from the HRS to link with a set of cognitive measures administered when individuals were over 60 years old. New innovations in genetics allow the use of polygenic scores (genome wide predictive measures, like those from Wedow) to be constructed in the HRS data that predict cognition. He will explore whether the Great Depression shaped the distribution of these scores in the population as well as whether time/place variation in the depth of the Great Depression interact with genetic predispositions to affect old-age cognition outcomes. Trejo will outline the growing evidence of a robust relationship between adolescent cognitive ability and health and mortality outcomes across the life course. Using newly constructed measures of genomic predispositions (polygenic scores), he will examine causal mechanisms linking cognition and health that currently remain underexplored. Using data on a longitudinal cohort of siblings born in the middle of the 19th century in the United States from the Wisconsin Longitudinal Study, he will decompose the relationship between adolescent cognitive ability and long-term health outcomes. By employing a novel instrumental variable technique using individual genetic data combined with sibling analyses, the analyses will separate and quantify the portion of the cognition-health gradient that is caused by early-life environmental stress and trauma from the portion caused by healthy behavior and selection into healthier environments. He will also explore mediators and moderators of the cognition-health pathway. Conley will present work that shows that by developing a polygenic score for plasticity for that health trait, we are better able to model gene-environment interactions than current approaches, which relies on PGSs for levels effects that are most likely robust to the range of environments being studied due to the fact that they average over meta-analyses across cohorts and have been trained on predicting levels. We demonstrate the efficacy of this approach using data from the Health and Retirement Study.
Restrictive Immigration Policy and Population Health
Session Organizer: Annie Ro, UC Irvine
Session Room: West Court
Local immigration enforcement intensity and birthweight in California
Jacqueline Torres, UCSF
A Historical Analysis of Immigrant Detentions and Birth Outcomes in California
Annie Ro, UC Irvine
Cautious Citizenship: Examining Mechanisms by Which Immigration Issue Salience Shapes Latinas/os Engagement with Public Health Institutions
Alana Lebron, UC Irvine
Perspectives on Restrictive Immigration Policies, Immigrant Families, and Health from a Practitioner
Ana Corina “Cori” Alonso-Yoder, American University, Immigrant Justice Clinic
Federal immigration policy has become a powerful tool in deporting record numbers of immigrants back to their countries of origin. A growing body of literature has established poorer health outcomes among Latino immigrants after a particular policy implementation or enforcement event. This panel expands on this existing work to identify aspects of immigration policy that are harmful to health and who is most vulnerable. Two presentations (Torres and Ro) explicitly consider the role of enforcement activities on birth outcomes in California, but distinguish between two important components: detention and deportation. LeBron and colleagues examine potential pathways to health by exploring the association between policy climates and engagement with public health institutions through a unique experimental design. Finally, Alonso-Yoder, a legal advocate from an immigrants’ rights clinic, will share her experiences on the human impact of these policies.
Note: Due to our presenters’ scheduling constraints, we request to present on 10/3/18.
#1: Local immigration enforcement intensity and birthweight in California (Torres, UCSF)
The study evaluates the intensity of Secure Communities (SCOMM) enforcement in California (2008-2014) on term birth weight and small for gestational age. SCOMM enforcement intensity was captured with Department of Homeland Security records of county deportations. These data were linked to California birth records for Latina and non-Latina women. Preliminary analyses used 1.6 million birth records from Los Angeles County (2009-2011). Among overall births, there were significant associations between higher levels of SCOMM enforcement intensity and greater term birthweight. However, infants born to Latina women had significantly lower term birthweight at all levels of prenatal SCOMM enforcement exposure. There were also significant associations between SCOMM enforcement and greater odds of SGA for Latina women at all levels of enforcement intensity.
Initial analyses from Los Angeles suggest that SCOMM exposure during pregnancy is associated with poor birthweight outcomes for Latina, but not non-Latina white women. These poor outcomes were evident at all levels of SCOMM exposure. Analyses will subsequently be expanded to more California counties and to birth records through the end of the initial SCOMM program. Future analyses will also test the robustness of findings to indicators that reflect parallel economic trends that may confound current findings.
# 2: A Historical Analysis of Immigrant Detentions and Birth Outcomes in California (Ro, UCI)
This study examines the association between immigrant detentions and the birth outcomes of Latinas and Asians in the 58 California counties during a period of rapidly increasing enforcement action. Immigrant detentions, the policy of holding undocumented immigrants until a decision is made on their deportation, are an intermediary status between apprehension and deportation. We argue that detentions may be the most consequential enforcement activity for health, as immigrants are detained relatively quickly after apprehension and may be kept indefinitely.
We have created a dataset that combines county-level information from detentions from the Department of Homeland Security and California birth records spanning 1989-2015. We test the hypothesis that the number of detentions in a county is positively associated with the county prevalence of low-birthweight and pre-term birth. We will test the association for both US-born and foreign-born Latinas. The main independent variable is the number of detentions per county in year t-1, standardized by the size of the immigrant population in the county. The outcome variable is the annual incidence of low birth weight in year t for each of the respective populations (ie-US born Latinos, foreign-born Latinos, etc). We will conduct time series analyses that model autocorrelation using the Box-Jenkins approach.
# 3: Cautious Citizenship: Examining Mechanisms by Which Immigration Issue Salience Shapes Latinas/os Engagement with Public Health Institutions (Lebron, UCI)
This study draws on data from the 2015 Latino National Health and Immigration Survey to examine mechanisms by which the politics of immigration shapes Latinas/os’ engagement with bureaucratic institutions charged with provisioning health-promoting resources in an era of heightened immigrant policing. Through an experimental design, the authors examine whether Latinas/os are more likely to engage in Cautious Citizenship, or risk-avoidance attitudes and behaviors towards public institutions in order to avoid scrutiny of legal status, particularly when being primed to think about immigration issues relative to those primed to think about health insurance concerns. The authors hypothesize that avoidance patterns are driven by those individuals with profiles that are more vulnerable to immigrant policing. Accordingly, the authors evaluate variation in these patterns by whether individuals have close ties to someone who has undocumented status, whether individuals know someone who has been deported, and perceived anti-immigrant state policy context.
Latinas/os exposed to the “immigration issues” cue express greater avoidance of daily life activities (schools, health clinics, police) relative to those exposed to the “health insurance” cue, suggesting evidence for the aversion hypothesis. The effects of the “immigration issues” cue on avoidance of health clinics is augmented for Latinas/os with family ties to someone who is undocumented. The experimental effect was sharpened for avoidance of schools and health clinics for Latinas/os who know someone who has been deported. Latinas/os who perceive their state policy context to be unwelcoming to immigrants are more likely to avoid the police.
# 4: Perspectives on Restrictive Immigration Policies, Immigrant Families, and Health from a Practitioner (Alonso-Yoder, American University)
Ms. Alonso-Yoder is the Practitioner-in-Residence with the Immigrant Justice Clinic (IJC) at the American University Washington College of Law. The legal activities of the IJC focus on four primary areas: 1) immigration deportation defense and immigration detention; 2) immigrant workers’ rights; 3) civil rights for immigrants; 4) immigration, gender, and sexual orientation. This presentation will provide an overview of the landscape of immigration policy in the past 10 years, establishing the legal and political context of the existing empirical work. Ms. Alonso-Yoder will also share her experiences in the IJC and discuss the human impact of restrictive immigration policies, highlighting pathways between such policies and social determinants of health. Finally, Ms. Alonso-Yoder will provide her perspective on the current political climate and discuss implications for immigrant communities.
State Drug Policies, the Pharmaceutical Industry, and Trends in the Opioid-Related Overdose Epidemic
Session Organizer: Magdalena Cerda, University of California at Davis
Session Room: Members
Proactive prescription drug monitoring programs are associated with the greatest reductions in fatal prescription opioid and heroin overdoses
Magdalena Cerda, University of California at Davis
Harm reduction policies and opioid-related overdose: a spatiotemporal approach to identify sources of variation in impact
Nicole Kravitz-Wirtz, University of California at Davis
Pharmaceutical industry marketing and changes in opioid-related overdoses in the United States
Scott Hadland, Department of Pediatrics, Boston University School of Medicine
Marijuana legalization (medical and recreational) and prescription opioid use, heroin use and opioid use disorder
Silvia Martins, Columbia University Mailman School of Public Health
Opioid-related overdose deaths have quadrupled since 1999, and the epidemic worsens every year. The epidemic has shifted from a prescription opioid epidemic to an epidemic driven by synthetic opioids and heroin. Its shifting nature has required a rapid response to identify effective policy solutions. Our symposium will present some of the latest applications of epidemiologic, econometric, and spatiotemporal methods to identify effective policy responses to the opioid epidemic. First, our symposium will address the impact that prescription drug monitoring programs (PDMPs) have on fatal and non-fatal prescription opioid and heroin overdose rates. While PDMPs are one of the leading state responses to the opioid epidemic, the combinations of PDMP features that may contribute to reducing overdoses are not well understood. Dr. Cerdá’s study examines how transitions across types of PDMP models (identified through latent transition analysis), contributed to change in fatal prescription opioid and heroin overdoses in counties across 49 states in 2002-2015 and to change in nonfatal prescription opioid and heroin overdoses in zipcodes across 22 states in the same period. Second, our symposium will examine the impact that policies focused on opioid harm reduction (i.e., naloxone access and Good Samaritan laws) have on fatal and non-fatal prescription opioid and heroin overdoses. Dr. Kravitz-Wirtz’s study examines how implementation of naloxone access and Good Samaritan laws contributed to changes in fatal prescription opioid and heroin overdoses in counties across 49 states, and to changes in non-fatal prescription opioid and heroin overdoses in zipcodes across 22 states, in 2002-2015. Her study also characterizes the extent to which the impact of harm reduction policies on opioid overdose varies across counties and zipcodes, and attempts to identify the sources of such variation. Third, our symposium will examine how pharmaceutical industry marketing practices affect physician opioid prescribing practices and the risk of opioid overdose. Pharmaceutical industry marketing is widespread, but its impact on prescribing and overdose is not well understood. Dr. Hadland’s study combines data on marketing practices from the Open Payments database with Centers for Disease Control restricted access data on fatal opioid overdoses to inform potential policies regarding opioid marketing to physicians. Finally, our symposium will examine whether recent changes in laws related to marijuana access (i.e., medical and recreational marijuana legalization) have affected rates of prescription opioid misuse, heroin use, and opioid use disorders. Marijuana has been proposed as a potential substitute for opioids, particularly in the context of increased restriction of access to the prescription opioid supply. Yet whether marijuana legalization has lead to change in the use of opioids has not been examined. Dr. Martins will combine data on marijuana laws with self-reported data on opioid misuse, obtained from the National Survey on Drug Use and Health restricted use files, to examine these relationships across the 50 states in 2004-2016. Dr. Katherine Keyes will moderate the panel. By identifying effective policies targeting both prescription opioids and heroin, solutions to this epidemic may emerge.
Selection and Scarring Over the Life Course
Session Organizer: Tim Bruckner, Public Health, University of California, Irvine
Session Room: NAS 120
Pregnancy Loss and the Health of U.S. Birth Cohorts
Jenna Nobles, Associate Professor of Sociology, University of Wisconsin-Madison
Adolescent deprivation in Norway during the German occupation and subsequent cohort mortality
Alison Gemmill, Assistant Professor of Family, Population and Preventive Medicine, University of Stony Brook
Early conditions, longevity and compression of morbidity in low income countries
Alberto Palloni, Professor Emeritus of Sociology, University of Wisconsin-Madison
Integrating Selection and Scarring Processes in the Context of Racial Disparities at Old Age
Elizabeth Wrigley-Field, Assistant Professor of Sociology, University of Minnesota
Much of life course research posits that societal stressors (e.g., social or economic strain, cumulative disadvantage) leave lasting adverse health consequences on the exposed population. Some of these sequelae are thought to include elevated risk of mortality and/or chronic disease morbidity later in life. Depending on the life course stage in which the presumed stressor occurs, this literature is often characterized as “fetal origins”, “developmental origins,” or simply “scarring.”
Whereas the “scarring” work enjoys widespread dissemination in public health journals, there exists a separate literature which documents a positive relation between population stressors and cohort selection. Cohort selection refers to a process in which stressors induce mortality among frailer members of the cohort, thereby leaving behind a smaller, but on average healthier, cohort over the life course. Such cohort selection occurs as early as in utero and as late as older age (i.e., >80 years). Population health scholars may be less familiar with the literature on selection because it appears more prevalent in demography, sociology, and evolutionary biology journals.
The scarring and selection fields rarely communicate or cross-cite. This circumstance may arise from the assumption that either scarring or selection―but not both―occur at particular life stages. We contend that this assumption is flawed and that both scarring and selection processes, in fact, shape cohort health.
Advances in data availability, methodology, and theory have led to important insights about the dynamics of scarring and selection over the life course. This session will highlight novel empirical advances in the selection and scarring literatures. These contributions hold direct relevance to population health because they estimate health of complete cohorts, rather than clinical samples or non-representative surveys.
In addition to presentation of four papers, the moderator (Bruckner) will put into theoretical context the empirical contributions. Bruckner will also suggest avenues by which the selection and scarring fields could reach consilience. Below we list the names of presenters as well as their title and a summary of the work. All have agreed to attend the IAPHS conference on Oct 3-5th in Washington, D.C.
Jenna Nobles, Associate Professor of Sociology at University of Wisconsin-Madison, will present “Pregnancy Loss and the Health of U.S. Birth Cohorts.” Prof. Nobles will provide results from recent work estimating the contribution of pregnancy loss to the observed health at birth (as measured by birthweight, gestational age) of cohorts from the US Natality File.
Alison Gemmill, Assistant Professor of Family, Population and Preventive Medicine, will present “Adolescent deprivation in Norway during the German occupation and subsequent cohort mortality,” During the German occupation in Norway in 1940, scarce resources, such as food and medicine, had to be strategically allocated across the population. Adolescents were given the lowest priority in the allocation schemes because they were considered the healthiest Norwegians and most able to cope with less access to nutrition and medical care. This paper will examine if the cohorts exposed to the war at adolescence exhibited late life death rates different from expected.
Alberto Palloni, Professor Emeritus of Sociology at University of Wisconsin-Madison, will present “Early conditions, longevity and compression of morbidity in low income countries.” This paper will examine scarring processes that occur early in childhood and its sequelae on chronic disease morbidity (e.g., type 2 diabetes, obesity) in adulthood.
Elizabeth Wrigley-Field, Assistant Professor of Sociology, University of Minnesota, will present “Integrating Selection and Scarring Processes in the Context of Racial Disparities at Old Age.” In the context of racial disparities in old-age mortality, selection and scarring processes are usually discussed separately. This has prevented studies of racial disparities in old-age mortality from engaging with how selection processes might interact with cumulative disadvantages and other long-term forms of inequality. In this talk, Prof. Wrigley-Field will present a framework for how the two types of process might be integrated and propose new hypotheses suggested by the synthesis.
Tim Bruckner, Associate Professor of Public Health, University of California, Irvine, will serve as the moderator and, time permitting, as discussant. Prof. Bruckner holds expertise in epidemiology and demography and has published extensively on scarring and selection in utero. His work includes examination of both historical (e.g., 1773 Swedish famine) and contemporary (e.g., terrorist attacks of 9/11) population stressors.
Landing Upon the Body: Racism as Violence in the US
Session Organizer: Maeve Wallace, Tulane University School of Public Health and Tropical Medicine
Session Room: NAS 125
Neighborhood segregation and the geographic patterning of police killings
Justin Feldman, Department of Population Health, New York University School of Medicine
Examining competing theoretical explanations for racial disparities in police violence
Jessica Judson, Department of Global Community Health and Behavioral Sciences, Tulane University School of Public Health and Tropical Medicine
Police violence and black motherhood: exploring the impact of a high profile police killing on black mothers
Rachel Hardeman, Division of Health Policy and Management, University of Minnesota School of Public Health
Invisible Women: violence as a direct and indirect cause of maternal mortality
Maeve Wallace, Department of Global Community Health and Behavioral Sciences, Tulane University School of Public Health and Tropical Medicine
“Our phrasing – race relations, racial chasm, racial justice, racial profiling, white privilege, even white supremacy – serves to obscure that racism is a visceral experience, that it dislodges brains, blocks airways, rips muscle, extracts organs, cracks bones, breaks teeth. The sociology, the history, the economics, the graphs, the charts, the regressions all land, with great violence, upon the body.”
– Ta-Nehisi Coates, Between the World and Me
Moderator: Katherine Theall, Ph.D.
Description of Panel: Growing awareness of structural racism within and outside of public health discourse has led to calls for researchers, governmental practitioners, medical providers, and policy makers to explicitly identify structural racism as a root cause of racial health inequity. Perhaps one of the most visceral examples of structural racism is the increasingly publicized killing of black citizens at the hands of police. Law-enforcement-related deaths and the ensuing trauma they engender on immediate families and the community more broadly is an important yet virtually unexplored marker of structural racism and determinant of population health.
This panel emphasizes structural racism as both a cause of violence and a form of violence in and of itself. Four presentations will include analyses of violent deaths (law-enforcement-related death and pregnancy-associated homicide) as population health outcomes resulting from structural racism, and the maternal health impacts of community exposure to racialized acts of violence (the law-enforcement-related death of unarmed black citizens). Through a shared framework of racism as violence, these presentations highlight broad-scale, policy-relevant opportunities for both anti-racist and violence prevention intervention efforts to promote population health equity.
Introduction: Katherine Theall
Presentation 1: Justin Feldman
Talk Title: Neighborhood segregation and the geographic patterning of police killings.
Killings by police are important from an epidemiologic perspective – both as a leading cause of death for young men of color and as events that can influence the well-being of communities. While prior literature has quantified racial inequalities in rates of police-related deaths, epidemiologists have not explored economic inequalities or the role of neighborhood context. This US-wide, multilevel study uses a previously validated, geocoded database of US police killings from the Guardian newspaper. It examines the association between rates of police killings and neighborhood-level measures of racial/ethnic and economic segregation.
Presentation 2: Jessica Judson
Talk Title: Examining competing theoretical explanations for racial disparities in police violence.
Sociologists and criminologists have explored the potential for minority threat and conflict theories to explain excessive police violence. While many of these studies examined these theories in relationship to disparities in arrest rate, citizen complaints, and level of police strength, lethal police violence as the outcome presented with difficulty given the lack of comprehensive data. Using combined data from The Guardian and Fatal Encounters websites, this study looks at the association of county-level structural measures of threat with the Black-white disparity in lethal police violence victimization in 2015.
Presentation 3: Rachel Hardeman
Talk Title: Police violence and black motherhood: exploring the impact of a high profile police killing on black mothers.
On November 15, 2015, Minneapolis Police Department responded to a call on the City’s North Side that resulted in the fatal shooting of Jamar Clark by a police officer. Beyond the tragic death that occurred due to police brutality, the effects of the violence permeate the very conditions of existence in communities already marginalized and vulnerable. Over-surveillance in marginalized and vulnerable communities coupled with the looming fear of death at the hands of police effect every aspect of life within said communities. The fear and anxiety that persist in the aftermath of continual loss of life of Black men in Black communities across the U.S may contribute to maternal stress and poor health for Black mothers. Based on a Reproductive Justice framework, this study leverages qualitative data to describe the stories and lived experiences of Black mothers in a North Minneapolis community which experienced a high profile incident of police brutality, combining it with geo-coded measures of structural racism (e.g. educational attainment and income inequality) to tell the story of structural racism as violence on Black motherhood.
Presentation 4: Maeve Wallace
Talk Title: Invisible Women: violence as a direct and indirect cause of maternal mortality.
Pregnancy-associated homicide is homicide of a woman while pregnant or within the first year postpartum. To date, a small number of publications attempting to bring visibility to women victims of violent death find pregnancy-associated homicide rates consistently exceed mortality rates from the leading obstetric causes of direct maternal mortality. Concurrently, maternal mortality from obstetric causes is increasing in the US despite improvements in obstetric care, and yet socio-contextual determinants remain unexplored. This study utilizes data linkage and social epidemiologic methods to examine pregnancy-associated deaths (from both violent and obstetric causes) as part of and inseparable from contexts in which they occur. Geographically-defined measures of structural racism including racial and economic residential segregation, racial inequality in educational attainment, employment, and median household income are explored as features of the social environment that place women directly and indirectly at increased risk of death during pregnancy and postpartum.
Your Trauma is Not Your Own: Research and Perspectives on Childhood Trauma and Mental Health
Session Organizer: Meghan Wolfe, The George Washington University
Session Room: Board
Biosocial Suffering: Trauma and Epigenetic Inheritance
Jessica Minor, Yale University, Department of Anthropology
The Role of Schools in Trauma-Informed Interventions
Adrian Neely, Georgie State University, Department of Middle and Secondary Education
Parental Incarceration and Children’s Mental Health Outcomes
Leslie Salas, Emory University, Department of Behavioral Sciences and Health Education
Understanding the Underlying Biological Features of Early Stress
Raven Hardy, Emory University, Department of Biological and Biomedical Sciences
The connections between childhood trauma and mental health have long been recognized and studied by public health and medical professionals. In 1998, a landmark study on Adverse Childhood Experiences (ACEs) found that chronic stress in childhood, caused by abuse or neglect, domestic violence, or parental mental illness or substance abuse, led to long-term negative health effects. Since then, increased research in the fields of medicine, psychology, social work, and public health on have shed light on issues and challenges related to ACEs.
This panel discussion seeks to expand on this conversation, bringing to bear research on some of the lesser studied contributors to and solutions for addressing childhood trauma and mental health. Panelists will discuss 1) the role of epigenetic inheritance in predisposing children of trauma survivors to adverse mental health conditions; 2) the effects of childhood stress on neurobiological systems; 3) the effects of parental incarceration on childhood trauma and mental health; and 4) the ways in which schools can address trauma and adverse mental health among students. By examining the how trauma manifests both biologically and socioculturally, researchers, practitioners, and policymakers can better understand how to provide holistic solutions to address this complex and multifaceted issue.
Biosocial Suffering: Trauma and Epigenetic Inheritance – Jessica Minor, MD/PhD Candidate, Medical Anthropology, Yale University
Communities impacted by long-standing structural violence recognize that the psychological distress of an individual is linked to the suffering of their forebears. This phenomenon has been variously described as historical, cultural, and intergenerational trauma. Psychiatric research suggests that individuals born to parents who experienced significant trauma are more likely to develop psychiatric illness, and children of parents with post-traumatic stress disorder (PTSD) are more likely to develop PTSD themselves. Epigenetic inheritance, or the transmission of molecular tags on DNA from parent to offspring, may play a role in mediating a child’s predisposition toward adverse mental health. This discussion will synthesize emerging research on the biological mechanisms underpinning the intergenerational transmission of trauma, calling attention to the overlap between biological and structural sources of suffering.
Understanding the Underlying Biological Features of Early Stress – Raven Hardy, PhD Student, Neuroscience, Emory University
Rapid brain development occurs during early childhood. Adverse early life stressors, such as violence exposure, trauma, and parental neglect/abuse, influence brain growth, connectivity and activation. Childhood trauma hijacks anatomically- and functionally-related brain networks that are involved in processing emotions, like reward and appetitive behaviors. Clinically, childhood trauma is a risk factor for the later onset of emotional disorders, psychopathology, substance use disorders, and metabolic disorders. Understanding the underlying biological features impacted by early stress is critical to identify populations vulnerable to develop severe disorders later in life and to aid in the development of therapies. This talk will highlight the most recent research on the relationship between biological features and early childhood stress, including effective interventions, prevention strategies, and treatment.
Parental Incarceration and Children’s Mental Health Outcomes – Leslie Salas, PhD Student, Behavioral Sciences and Health Education, Emory University
Adverse childhood experiences, such as parental incarceration, have been shown to have long lasting mental health effects. It is reported that one in every 28 children in the US have an incarcerated parent. Due to mass incarceration, parental incarceration is becoming an increasingly common American childhood experience but this adverse risk exposure disproportionately affects African-American and low-income children. The majority of children with incarcerated parents (CIPs) are nine years old and younger. CIPs often present anxiety, aggression, and depression and are prone to display behavioral issues at school. Various studies have used nationally representative datasets and shown the relationship between histories of parental incarceration and young adult mental health outcomes. Family disruption is often cited as a mechanism in the development of mental health issues among CIPs. Evidence-based practices that target families and communities affected by incarceration and detainment are necessary. Recent efforts to implement such practices have emphasized resiliency as a protective factor, including Sesame Street’s Little Children, Big Challenges: Incarceration toolkit. This conversation will use a public mental health lens to frame existing research and interventions highlighting CIPs and their mental health.
The Role of Schools in Trauma-Informed Interventions – Adrian Neely, PhD Student, Education, Georgia State University
More than 26 percent of children in the United States experience or witness a trauma before age four with more than 80 percent of these traumatic experiences occurring at home. Additionally, national data indicate that 1 in 4 children attending school have experienced a traumatic event and this number is higher for students living in socioeconomically disadvantaged areas and minority youth considered at risk for chronic and ongoing exposure. School-based outcomes have traditionally focused on academic achievement; however, attention to connecting academic outcomes to social, emotional, and mental health outcomes as determinants that impact academic success is increasing. As such, school systems are implementing trauma-informed service delivery as an approach to increase academic achievement. This talk will review theoretical and empirical research that focuses on the emerging field of two trauma-informed school-based interventions: teaching practices and curriculum.
Acknowledgement: All four presenters are doctoral students and Scholars in the Robert Wood Johnson Foundation-supported Health Policy Research Scholars program.
Concurrent Session 2
Thursday, October 4
10:45am – 12:00pm
How Do We Train Population Health Scientists?
Session Organizer: Christine Bachrach, University of Maryland
Session Room: Auditorium
A Brief Overview of the NAM Report on Training in Interdisciplinary Health Science
Christine Bachrach, University of Maryland
Population Health Science Training in a Medical Setting
Mark Cullen, Center for Population Health Sciences, Standford Medicine
Training Population Health Scientists at the PRC
Mark Hayward, Population Research Center, University of Texas in Austin
Harvard’s Interdisciplinary PhD program in Population Health Sciences
S.V. Subramanian, Harvard T H Chan School of Public Health
In June, 2015, the National Academy of Medicine Roundtable for Population Health Improvement hosted a meeting on Training in Interdisciplinary Population Health Science: A Vision for the Future. The goal of the meeting was to reflect on what had been learned about training interdisciplinary scientists who have the capability of producing integrative research on population health and to develop a vision for future training programs. A report, “Training in Interdisciplinary Health Science: Current Successes and Future Needs”, summarized discussion at the meeting about the essential elements needed for successful training. It also offered a potential model for future programs.
This session will consider current approaches in population health training in light of the Roundtable report and events and experience since the report was published. The goal is to describe for the audience a few specific models currently in use and to engage presenters and audience in reflecting on the state of the art, as well as gaps and opportunities for strengthening training opportunities.
The report provides a common starting point for discussion. Presenters will be encouraged to comment on what the report got right, wrong, or left out. They will also consider how the models of population health science training need to evolve in the context of developments since the report was produced, including the evolution of university and funding environments, changes in health care and public health systems, and the interdisciplinary “field” of population health science.
The focus of the session will be training at the pre- and post-doctoral levels. This necessarily omits the important and growing training in population health that occurs at the undergraduate level, and if time presenters will be asked to comment on these trends as well.
Panelists have been recruited to represent population health training in different settings – a medical school, a population center, and a school of public health – in order to ensure a diversity of perspectives.
The Diminishing Returns of Socioeconomic Status
Session Organizer: Taylor Hargrove, University of North Carolina at Chapel Hill
Session Room: Members
Racial Inequality in Health: The Roles of SES, Stressors, and Physiological Dysregulation
Tyson Brown, Duke University
Race/Ethnicity, Maternal Educational Attainment, and Infant Mortality in the United States
Samuel Fishman, University of North Carolina at Chapel Hill (UNC-CH)
Race, SES Mobility, and Allostatic Load: A Test of Diminishing Returns
Courtney Thomas, University of California, Los Angeles (UCLA)
A considerable amount of research and policy efforts indicate that socioeconomic status (SES) has a robust relationship with an array of health outcomes across the life course. Unsurprisingly, socioeconomic position has been deemed a “fundamental cause” of health given its ability to shape access to important resources that can be used to avoid health risks or ameliorate the consequences of disease after its onset (Phelan and Link 2015). Furthermore, several interventions to reduce health inequalities have focused on increasing access to socioeconomic resources among individuals of disadvantaged racial-ethnic, gender, and socioeconomic statuses (Kling et al. 2007; Ludwig et al. 2013; Thornton et al. 2016).
While the relationship between SES and health is well-documented, increasing evidence suggests that not all social groups experience the same health benefits of higher socioeconomic attainment, particularly racial/ethnic minorities. For example, several studies across sociological, public health, and demographic literatures find that African Americans receive no or less steep improvements in health at higher levels of education and income compared to their white counterparts (Brown et al. 2016; Chen et al. 2015; Cohen and Geronimus 2006; Farmer and Ferraro 2005; Walsemann et al. 2012). These patterns have generally been attributed to restricted opportunities for economic success, the non-equivalence of socioeconomic indicators across race, and increased experiences of discrimination among racial-ethnic minorities of higher SES in the context of institutional and interpersonal racism (Colen 2011; Pearson 2008; Thomas 2015; Williams 2012).
Studies documenting differential returns to SES are increasing—yet still unclear is the extent to which the patterning of SES-health relationships by race might vary by dimensions of health or life stage. Also less clear are the experiences and implications of such processes in real life. To this end, we propose a panel to discuss new research and observations on the topic of differential health returns to SES. This panel would be diverse not only by discipline, but by role in the academy as well. We envision having three academic presentations and one presentation giving an administrative and personal account of the process of socioeconomic achievement. If accepted, Cynthia Colen has agreed to moderate this panel.
Presentation 1: Racial Inequality in Health: The Roles of SES, Stressors, and Physiological Dysregulation
This study estimates the extent to which socioeconomic resources, stressors, and physiological dysregulation mediate racial inequalities in self-rated health among white, black, and Hispanic adults over age 50. The analyses simultaneously allow for racial/ethnic differences in the effects of the mediators given that inattention to conditional indirect effects may lead to over- or underestimating the extent to which these factors account for racial health inequalities. Using data from the Health and Retirement Study and structural equation models of moderated serial mediation, findings indicate that achievement, stress, and biological processes account for black-white and Hispanic-white disparities in self-rated health. While socioeconomic resources account for a substantial portion of these disparities (more than half), racial/ethnic differences in exposure to an array of stressors and physiological dysregulation play a critical role. Findings also show that the effects of several socioeconomic factors and stressors varied across race/ethnicity. Limitations and implications of these findings will be discussed.
Presentation 2: Race/Ethnicity, Maternal Educational Attainment, and Infant Mortality in the United States
This study investigates the extent to which an important indicator of population health, infant mortality, varies across racial/ethnic-education groups, and attempts to identify the potential factors producing such patterns. Two datasets are used to address these aims: 1) the 2007-10 US linked birth and infant death cohort files, and 2) the National Longitudinal Study of Adolescent to Adult Health (Add Health). Findings indicate that while education-specific infant mortality rates are similar when comparing Mexican Americans with Whites, infants of college-educated Black women experience nearly 47 percent higher mortality when compared with infants born to White women who have a high school degree or less. Further, infants of Black women with a high school degree or less exhibit more than twice the rate of mortality than their White counterparts. Ancillary results from Add Health show that both low and highly-educated Black women exhibit socioeconomic, contextual, psychosocial, and health disadvantages in both adolescence and young adulthood relative to low-educated White and Mexican American women, highlighting profound patterns of disadvantage across the life course. Limitations and implications of these findings will be discussed.
Presentation 3: Race, SES Mobility, and Allostatic Load: A Test of Diminishing Returns
Growing evidence suggests the impact of SES on health varies across racial groups—yet, left unclear is how SES across the life course may differentially shape the physiological functioning of Black and White adults. The present study aims to address this topic by examining the extent to which lifetime SES shapes allostatic load (AL) and whether these relationships are conditional on race. Using data from Nashville Stress and Health Study, findings indicate that while childhood SES is similarly linked to allostatic load for Blacks and Whites, higher levels of SES in adulthood do not benefit the health of Blacks compared to their White counterparts. Furthermore, stably high SES Blacks experience much worse health than their stably high SES White counterparts. Results provide evidence of differential returns to SES by race across the life course.
Presentation 4: Experiences of MURAP Students
Rosa Perelmuter is the Director of the Moore Undergraduate Research Apprentice Program (MURAP) at UNC-CH. MURAP is a graduate-level research experience for talented undergraduate students from diverse backgrounds who are interested in pursuing. A major aim of the program is to facilitate entrance of students from underrepresented minority groups into PhD programs and faculty positions in academia. MURAP hosts 20 students on campus each summer to participate in a ten-week long research experience. Given her role as director and her extensive interaction with the MURAP students, Dr. Perelmuter offers a unique perspective and account of the factors linking socioeconomic achievement to fewer health benefits. She will be able to speak to some of the potential strains and stressors associated with achieving higher levels of education and navigating different social spaces, particularly among racial/ethnic minorities and first-generation students.
Exploring Population Health Disparities: What are the Contributions of Genetics, Social Inequalities, and Behavior?
Session Organizer: Melinda Aldrich, Vanderbilt University Medical Center
Session Room: NAS 120
Using the electronic health record system to systematically identify health disparities and discover their underlying causes
Lea Davis, Vanderbilt University Medical Center
Getting the full picture: Social, biological, and environmental factors associated with drug response in minority children with asthma
Marquitta White, University of California San Francisco
Engaging with patients, communities, and providers for addressing disparities in precision medicine
Consuelo Wilkins, Meharry-Vanderbilt Alliance
NIMHD perspective – opportunities to address health disparities combining genetics and social factors
Nishadi Rajapakse, National Institute on Minority Health and Health Disparities
Scientific evidence points to the origin of many adult diseases as the dynamic interplay between genetic, social, behavioral, and environmental factors throughout life. Despite consensus that environmental and genetic factors contribute to disease, little is known about how to integrate these across diverse populations. This panel will address the intersection between social inequalities and human genetics necessary to address health disparities. An interdisciplinary set of experts will describe efforts underway to increase diversity in human genomics research, showcase novel strategies to model social and behavioral factors, and describe how these determinants can be incorporated into genetic studies. Understanding of these contributions has important implications for policy but also for the implementation of precision medicine, which emphasizes that prevention and treatment strategies take individual variability into account. Few health studies are integrating both genetic and social contributors when identifying risk factors for disease or assessing health disparities. This gap in knowledge can lead to increased health disparities and threatens the promise of policy changes for equitable precision medicine. A combined approach of genetic, social, and environmental factors must be considered for research initiatives and for implementation of precision medicine. This interdisciplinary panel will discuss efforts to successfully implement precision medicine and address health disparities.
Dr. Davis will describe the systematic discovery of novel health disparities associated with African ancestry in a hospital population. Dr. Davis has generated genetic data on ~30,000 individuals of diverse ancestry within the Vanderbilt biobank (BioVU), a repository of DNA extracted from discarded blood collected during routine testing and linked to de-identified electronic health records. Dr. Davis has estimated genetic ancestry and tested associations between the proportion of African ancestry and approximately 1,100 phenotypes. Efforts are underway to integrate an area deprivation index into the analysis to examine the effects of socioeconomic and genetic factors jointly. Results will be presented and the challenges of incorporating and modeling genetic as well as social and behavioral determinants of health will be described.
Dr. White will explore racial disparities in drug response to bronchodilators used to treat asthma. Genetic and epidemiologic studies have explored the independent effects of genetics, environmental exposures, and psychosocial factors on variation in drug response; however, the cumulative results of these studies explain little of the estimated heritability of bronchodilator drug response. This “missing heritability” may be captured, in part, by significant gene-environment (including social environment) interaction effects that have previously been overlooked. She will examine the combined impact of genetic, environmental, and social factors on bronchodilator drug response. These disparities have important policy implications for marketing of drugs to diverse racial/ethnic populations.
Minorities are underrepresented in precision medicine research, meaning for example, that genetic analyses are confounded by ascertainment bias and the results misapplied clinically in minority populations. Existing approaches are often not optimal for understanding health disparities and developing strategies to eliminate them. Increasing the participation of underrepresented populations in research is a mechanism to reduce health disparities. Further, community participation in health research serves as a catalyst for the development of best practices, systemic changes in healthcare delivery, and improved health outcomes. Increasing diversity in research via engaging underrepresented populations is important for reducing the unequal burden of disparities in health and healthcare. Dr. Wilkens will address opportunities to increase engagement to address disparities in precision medicine.
Benefits from medical advances are not always distributed equitably, often because structural or systemic factors limit the effectiveness of new diagnostic or therapeutic approaches in disadvantaged populations. This presentation by Dr. Rajapakse will explore perspectives of the National Institute on Minority Health and Health Disparities on (1) why greater inclusion of minorities and health disparate populations in genomic research is important, (2) examine the interplay between genetic, behavioral, social and environmental factors throughout the life course in addressing health disparities, (3) highlight ongoing studies on genomic medicine studies examining best practices and community engaged research, and (4) examine transitions, challenges and opportunities associated with precision medicine and health disparities.
Innovations in the Use of Multiple Methods in Studies of Lesbian, Gay, and Bisexual (LGB) Populations
Session Organizer: Stephen T Russell, University of Texas at Austin
Session Room: NAS 125
Generations: A study of life and health of LGB people in a changing society
Ilan Meyer, University of California Los Angeles
Identity maps: A new vantage point in studying relationships
Sara Mernitz, University of Texas at Austin
Queering quantitative and qualitative inquiry: A mixed method design to understand sexual minority alcohol use
Jessica Fish, University of Maryland
Comparing community-based and national-probability sampling methods: A comparative approach to inform LGB population health research
Evan Krueger, University of California Los Angeles
Lesbian, gay, and bisexual (LGB) youth and adults suffer compromised health outcomes compared with their heterosexual counterparts. Minority stress theory suggests that the social environment, in which LGB people experience stress due to prejudice and stigma, is influential in determining population health outcomes and disparities. Minority stress has provided an effective model for the study of health disparities in LGB populations. Despite significant sociohistorical changes, and the centrality of minority stress to the study of LGB health, we know little about how minority stress and identity development applies to contemporary LGB health and well-being.
In this session we use multiple sources of data collected as part of a program of multiple-method studies designed to understand LGB identities, relationships, and health. The Generations study consisted of two major components, one qualitative and one quantitative. The qualitative component consisted of life history interviews across three age cohorts of LGB people in 4 regions, designed for inclusion of a racially and ethnically diverse sample. The quantitative component is a national longitudinal survey, the first population-based, representative sample of LGB adults in the United States. These approaches allow for a combination of in-depth as well as broadly representative perspectives on LGB lives and health.
Linking Independent Population Data to Understand Adolescent Health
Session Organizer: Mark Hatzenbuehler, Columbia University
Session Room: Board
Geographic clustering and correlates of repeat teen childbearing in the United States, 2008-2016
Julie Maslowsky, University of Texas at Austin
Mapping the Landscape of High School Gay-Straight Alliances (GSAs)
Salvatore Loverno, University of Texas at Austin
Strategies for implementing anti-bullying policies: Are they effective?
Allen Mallory, University of Texas at Austin
Access, ethics, politics, and networks: Challenges for linking adolescent health data
Stephen T Russell, University of Texas at Austin
A number of intersecting developments are converging in population health sciences: a growing interest in and use of adolescent health and behavior data by multiple stakeholders; increasing concerns about data and identity security and protection; and dramatically changing possibilities for data collection and big data analytics. This session includes three new empirical papers that highlight examples of innovations in data matching, merging, and linking, each with the goal of understanding a distinctive dimension of adolescent health:
• Repeat teen childbirth statistics are geographically linked with demographic characteristics of counties;
• The presence of high school gay-straight alliances (GSAs) are geocoded, showing that in addition to the proximal association of GSAs to school safety and lower homophobic bullying, there is also a proximity effect on safety and bullying; and
• National survey data of school principals’ reports of school anti-bullying policies are linked with state-level student surveys, showing the efficacy of such school policies for reducing bias-based bullying.
In the final paper, a case study is provided of access, ethics, politics, and networks involved in linking multiple data sources on youth.
Panelist 1: Geographic clustering and correlates of repeat teen childbearing in the United States, 2008-2016
Julie Maslowsky, Population Research Center, University of Texas at Austin, firstname.lastname@example.org, presenting panelist
18% of the 230,000 annual US births to 15-19-year-olds are repeat teen births (births to teen mother who already has 1+ children). While overall teen births fell 51% from 1990-2010, repeat births fell only 23%. Few studies examine repeat births; evidence to understand their slow decline is lacking. This study uses linked county-level data to examine geographic clustering of repeat teen birth rates and characteristics of counties with high first versus repeat teen births. Data on N=2,795,789 births in N=3110 counties from 2008-2016 were obtained from US Natality Files. County characteristics (race/ethnicity, poverty, education, physician density) were obtained from Area Health Resources Files. Birth data were analyzed using SaTScan to generate clusters of counties with high first and/or repeat teen births. We then compared demographic and medical provider characteristics of counties with high repeat v. first birth rates. First and repeat teen births cluster in the US South and Southeast but do not completely overlap. 25.4% of counties had high rates of both repeat and first births; 8.7% repeat births only; 13.5% first births only; 52.4% neither. County clusters with high repeat versus first births had significantly higher rates of poverty, uninsured, and racial/ethnic minority residents and lower rates of high school graduation. County clusters with high repeat births versus neither had fewer primary care and OBGYN providers. The US Natality files are a source of population-level data on mothers and infants that can be linked at the county level with health datasets such as the Area Health Resource File to understand county-level determinants of health.
Panelist 2: Mapping the Landscape of High School Gay-Straight Alliances (GSAs)
Salvatore Ioverno, Population Research Center, UT Austin, email@example.com, presenting panelist
The introduction of school-based clubs such as Gay-Straight Alliances (GSAs) has emerged as one of the most effective strategies to prevent homophobic bullying. This is mostly due to the the efforts by GSAs to promote advocacy initiatives that address inequality in school and community environments. This study explored the influence of GSAs across communities using spatial statistics. Specifically, it examined whether, in addition to improving the more proximal school environment, GSAs may influence the school climate of nearby schools. Data on GSA presence at school were obtained from a census of GSA collected by a nonprofit organization (GSA Network); data from the census were geocoded, and linked to school-level data on victimization and homophobic bullying from the 2013-2015 California Healthy Kids Survey (2,378 schools). Preliminary analyses indicated spatial dependence in GSA presence (Moran’s I=.03, p=.002), victimization (Moran’s I=.34, p<.001) and homophobic bullying (Moran’s I=.18, p=.001). Spatial regression results showed that GSA presence was associated with lower levels of victimization (b = -0.28, p<.001) and homophobic bullying (b=-.01, p<.001). When spatial weights were taken into consideration, the effect of GSA presence explained a greater variance than linear regressions. Results suggest that GSAs are more effective in preventing victimization and homophobic bullying at school in communities with a higher presence of GSAs.
Panelist 3: Strategies for implementing anti-bullying policies: Are they effective?
Allen Mallory, Population Research Center, UT Austin, firstname.lastname@example.org
There is limited evidence that state anti-bullying laws are associated with lower rates of bullying. Most population-based studies examine general bullying with little attention to bias-based bullying (i.e. bullying based on a person’s perceived or actual identity) and policy implementation may vary across schools within a state. California has clear policies, definitions of bullying, and processes for addressing bullying that any school that receives state funding must follow. Data come from a representative sample of schools (N = 255) and students in California (N = 96, 765), linked at the school level. Because we were able to link at the school level data, we were able to assess if student reports of bias-based bullying and victimization (California Healthy Kids Survey) were lower in schools with anti-bullying implementation policies as reported by the schools principals (CDC School Health Profiles). Anti-bullying implementation practices were ineffective at reducing the prevalence of bias-based bullying but did reduce the frequency of bias-based bullying that students reported.
Panelist 4: Access, ethics, politics, and networks: Challenges for linking adolescent health data
Stephen T. Russell, Population Research Center, UT Austin, email@example.com
In this case study, the presenter considers a number of challenges involved in linking publically-funded, national- and state-level adolescent population education and health data. Despite the availability of numerous distinct population-level data sources, representing significant public investment, the ability to link such data remains largely thwarted. Unpacking this dilemma involves questions of ethics (e.g., which constituents are to be protected – youth, schools, systems administrators?), ultimately rooted in issues of politics and power in data and science. Implications for population health research strategies are discussed.
Concurrent Session 3
Thursday, October 4
1:30pm – 2:45pm
Funding Opportunities for Population Health Scientists
Session Organizer: Christine Bachrach, University of Maryland
Session Room: West Court
Register for a meeting with a Program Official! Click here to register. Space is limited.
National Cancer Institute, NIH
David Berrigan. Division of Cancer Control and Population Sciences, Behavioral Research Program
National Institute on Aging, NIH
Amelia Karraker, Division of Behavioral and Social Research
National Institute on Minority Health and Health Disparities, NIH
Adelaida Rosario, Community Health and Population Science Division, NIMHD
National Heart Lung and Blood Institute
Rebecca Campo, Clinical Applications and Prevention
Eunice Kennedy Shriver National Institute of Child Health and Human Development
Regina Bures and Juanita Chinn, Population Dynamics Branch
National Institute for Occupational Safety and Health, CDC
John Piacentino, Associate Director for Science
National Center for Injury Prevention and Control, CDC
Arlene Greenspan, Associate Director of Science
RWJF Culture of Health Leaders
Brian Smedley, National Collaborative for Health Equity
RWJF Evidence for Action
Matthew Trujillo, RWJF
RWJF Interdisciplinary Research Leaders
Sarah Gollust, School of Public Health, University of Minnesota
RWJF Clinical Scholars
Melissa A. Green, RWJF Clinical Scholars Program
Registration is required for this event! Click here to sign-up!
This session offers conference participants brief one-on-one meetings with program officers at organizations that fund population health science. The goal of the session is to enable participants to share their ideas for applications and to seek information about different funders’ opportunities, openness to, and criteria for population health awards.
The 75 minute session will be divided into 5 15-minute meeting slots. Program officers will meet with conference participants at tables set up in the meeting room. Individual participants will be allowed to sign up for consultations with up to three distinct funders prior to the session and to add more as space allows at the session. Individuals not able to speak to the program official of their choice may leave a business card with them and meet with them at other times during the conference, at the discretion of the program official.
The Microbiome and Population Health: Promise or Hype?
Session Organizer: Jennifer Dowd, King's College London
Session Room: Auditorium
The influence of social conditions across the life course on the human gut microbiota: A pilot project with the Wisconsin Longitudinal Study
Pamela Herd, University of Wisconsin
Sociodemographic variation in the oral microbiome; evidence from New York City Health and Nutrition Examination Survey
Audrey Renson, New York University
The population microbiome: what can we learn that we can’t from clinical samples?
Ken Krauter, University of Colorado
Jessica Faul, University of Michigan
There is growing evidence that the microbes that inhabit the human body, the microbiome, have profound implications for human health. The microbiome is now considered our “second genome” with potentially parallel importance to the genome in shaping human health. Unlike the genome, the microbiome is plastic and adaptive to the environment, opening the door to a broader research agenda to understand the microbiome’s sensitivity to the broader physical and social environments. Despite these exciting new microbial discoveries brought on by sequencing technologies, it remains to be seen what level of importance the microbiome truly has for explaining human health at the individual and population level over the life course and whether it can be modifiable public health target in the future. This panel will present new work on the microbiome in population-based data sets and debate the promise vs. pitfalls of this new area vs the danger of pursuing the microbiome as the “biomarker of the moment” in population health.
Each speaker will present some results from their own microbiome work as well as comment on the broader question of how this area can or cannot contribute to population health science. The discussant Dr Faul does not work directly on the microbiome but has worked with genetic and other biomarker data in social and population research and thus will provide an external perspective on the promise and hype of this emerging area.
Moderator: Jennifer Dowd, King’s College London
Participants, with tentative titles:
1.) Pam Herd, University of Wisconsin: “The influence of social conditions across the life course on the human gut microbiota: A pilot project with the Wisconsin Longitudinal Study”
2.) Audrey Renson, New York University (PhD student, TBD) “Sociodemographic variation in the oral microbiome; evidence from New York City Health and Nutrition Examination Survey.”
3.) Matt McQueen, University of Colorado, Boulder, “The population microbiome: what can we learn that we can’t from clinical samples?”
4.) Jessica Faul, University of Michigan: Discussant
All participants have agreed to their participation, although Dr. Faul can only participate on October 3rd or 4th, if this can be accommodated during scheduling.
Educational Attainment as a Key Social Determinant of Health Over the Life Course
Session Organizer: Elizabeth Lawrence, University of Nevada, Las Vegas
Session Room: NAS 120
Degrees of Quality: How High School Quality Shapes Educational Health Disparities.
Matthew Andersson, Baylor University
Reducing Health Inequalities for the Next Generation: Can Increases in Mothers’ Schooling Help?
Jennifer Augustine, University of South Carolina
Multigenerational Attainments and Mortality among Older Men: An Adjacent Generations Approach
Shawn Bauldry, Purdue University
Low Health Returns to Subbaccalaureate Educational Attainment: The Role of Background and Skills
Anna Zajacova, University of Western Ontario
In today’s U.S. society, educational differences in mortality are widening, and the salience of educational attainment for health and longevity appears to be growing. This session explores the role of education in population health inequalities from an interdisciplinary perspective, providing insight into pathways and mechanisms over the life course. The four papers span the life course from childhood to old age, using different data sources and diverse approaches. This session will explore how social background, intergenerational interactions, and contextual environments account for or differentially shape education’s effects on health and longevity.
The first paper, to be presented by Matt Andersson, uses national longitudinal data merged with school quality metrics to examine whether young adult health returns to high school graduation are contingent on high school quality, and if estimated effects of quality hinge on respondent socioeconomic background. Additional analyses will explore whether high school quality improves health outcomes among those who do not attain college graduation by age 25. For the second paper, Jennifer Augustine will assess whether child well-being improves subsequent to increases in maternal education, including effects of maternal education on factors theorized to vary with education (such as parenting stress). Prior research has shown that children with more highly educated mothers have better health, but many U.S. women enroll or complete their education after having a child which may have implications for child health and well-being. This paper will distinguish how the timing of maternal education might shape its import for children. Third, Shawn Bauldry will present his study that examines multiple generations of status attainment and their associations with mortality risk among men in old age. This study highlights the importance of accounting for the attainments of adjacent generations and suggest that reducing educational and occupational inequalities for children and young adults might also reduce health disparities among the growing population of older adults. The last paper, to be presented by Anna Zajacova, will look at the health for adults who attended postsecondary schooling but did not earn a BA. Despite having more education than high school graduates who never attended college, the ‘some college’ group reports equal amount or more health problems, including many chronic conditions and pain. This paper will identify the extent to which this anomaly in the education-health gradient is the due to the respondents’ background and their cognitive and noncognitive skills. Elizabeth Lawrence will moderate the panel.
These papers draw from multiple disciplines to provide insight into key questions about the education-health relationship that have yet to be answered, such as the relative importance of quantity versus quality of education, how timing shapes education’s effects, and the role of intergenerational links in the production and maintenance of health disparities. Further, these researchers employ top-notch methods including random- and fixed-effects and heterogeneous treatment effects. The conclusions of these four studies will shed light on why education is associated with better health and longevity among Americans, and through revealing how health disparities are produced, suggest ways we can reduce them.
Bringing it All Back Home: Improving Health Through Indoor Home Environment Modifications
Session Organizer: Lorna Thorpe, NYU School of Medicine Department of Population Health
Session Room: Board
Evaluating the independent impact of pest control as non-traditional medical intervention for children with asthma
Carolyn Olson, Assistant Commissioner, NYC Department of Health and Mental Hygiene, NY, NY
Subsidized Housing Renovation and Health
Ingrid Gould Ellen, Professor of Urban Policy and Planning, NYU Wagner School and Furman Center)
Evaluation of Smoke-Free Housing Policy Impacts on Tobacco Smoke Exposure and Health Outcomes
Lorna Thorpe, Professor of Epidemiology, NYU School of Medicine Department of Population Health, NY, NY
Sue Kaplan, Research Associate Professor, NYU School of Medicine Department of Population Health, NY, NY
In recent years, the number of studies showing an association between housing quality and health has grown substantially, particularly with respect to vulnerable populations such as pregnant women, children and older adults. A variety of indoor home exposures such as poor air quality, lead, mold, and pest infestations have been linked to a range of health, stress and developmental outcomes, but many gaps in evidence remain. While some studies have been longitudinal in nature and provided stronger causal evidence, most have focused on associations with less rigorous, cross-sectional methods. In this panel, presenters each evaluate a novel housing intervention or initiative, with a goal of informing policy or practice. Initiatives include: a clinical-community linkage study to evaluate improvements in childhood asthma exacerbations via integrated pest management prescriptions, an assessment of the health and social impacts of a distressed housing renovation initiative, and a policy shock analysis of the new Housing and Urban Development’s rule for public housing authorities to issue a smoke-free policy. Presenters employ rigorous longitudinal, and in some cases experimental, study designs. This panel explores how to put new knowledge into broader practice to help families live healthier lives.
(Full abstracts for each presentation are available)
Using Evidence to Inform Immigration Policy at the County, State, and National Levels
Session Organizer: Erin Hagan, Evidence for Action, a Signature Program of the Robert Wood Johnson Foundation
Session Room: Members
Immigration Enforcement and Birth Outcomes: Evidence from North Carolina
Romina Tome, Duke University
The Impact of State Immigration Enforcement and Public Health Insurance Policies on the Health Outcomes of Children of Immigrants
Heather Koball, Columbia University National Center for Children in Poverty
Successful Policy Strategies to Advance Health for Immigrant Populations
Steven Lopez, UnidosUS
The interplay between policy and population health, well-being, and equity impacts is not always well researched or understood. This is especially true for a complex political issue such as immigration, both legal and unauthorized, where the overlap and interactions of city-, county-, state-, and federal-level policies and enforcement practices may confound study measures and outcomes. The nature of immigration policy and enforcement may also impact health and other outcomes of both legal or unauthorized immigrants, as well as the broader population that includes naturalized and natural-born citizens. Layered on top of all of this are multiple biases and misconceptions around immigration that can also impact how research findings may be used or disregarded when it comes to policy development and enforcement.
This panel will feature research examining the impacts of immigration policies at county- and state- levels on health outcomes, and highlight the role of research and advocacy in influencing policy and enforcement decisions.
Dr. Rangel and his team are exploring how county-level immigration policies and enforcement activities in North Carolina affect the health and well-being of Mexican-born immigrant mothers and their newborns, and if changes in birth outcomes arise due to changes in maternal behaviors and health care access. They are specifically investigating birth outcomes and associated maternal behaviors for immigrant mothers during a time when a county enacted increased enforcement regulation.
Dr. Koball’s research team is examining the effect of state-level immigration enforcement policies and public health insurance expansions on health outcomes among children of unauthorized and legal immigrants. The research team is taking advantage of the patterns of state immigration policy variations and degree of cooperation with federal immigration enforcement to measure the health impacts of different approaches to immigration regulation. While some states have taken punitive actions for unauthorized status, many states have reduced federal restrictions on health insurance access by allowing legally present and, in some cases, unauthorized immigrants, access to state-funded public health care coverage. The analysis focuses on variation in receipt of preventive health care among children of immigrants across all 50 states and the District of Columbia from 2000-2016.
These related studies will lead to a better understanding of the interplay of overlapping county- and state-level policies and enforcement activities, and the potential consequences these policies and activities pose to the health of immigrants and the broader public, along with the day to day experiences in communities. In an ideal world, these results would be used by policymakers, public health officials, and state departments of health to inform strategies for county- and state- level policy development and enforcement activities to mitigate or remediate any negative health and well-being impacts.
Yet, a continuing challenge for researchers is whether or to what extent evidence-based findings are used to inform and shape policy—in this case, health care policy or immigration enforcement—given political and other considerations by policy-makers. While evidence alone will not likely overcome prejudices or preconceived notions about immigrants (country of origin, motivations for immigrating, etc.) and impacts of immigration (legal or otherwise) on other groups (employment, government spending, use of social services, criminal activity, etc.), it is nevertheless a crucial element of laying the groundwork for sound policies.
Steven Lopez, Associate Director of Health Policy at UnidosUS, is an expert on the intersection of health and immigration and will discuss efforts where there has been success in covering immigrant populations and thus in advancing broader public health outcomes. UnidosUS (formerly National Council of La Raza) is the nation’s largest Latino civil rights and advocacy organization, bringing a unique combination of expert research, advocacy, programs, and an Affiliate Network of nearly 300 community-based organizations across the United States and Puerto Rico, to address social, economic, and political barriers at the national and local levels. UnidosUS has been involved in every major federal debate on immigration for more than the last 30 years.
Together the panel will also discuss strategies for disseminating evidence to decision-makers, policymakers, and others, to help inform policy debates by providing a more accurate analysis of the impact of immigration enforcement on public health. Panel members will discuss the challenges to integrating evidence into policy and the need to push the boundaries for population health, well-being, and equity.
Concurrent Session 4
Friday, October 5
10:45am – 12:00pm
Looking Forward: A Population Research and Policy Agenda
Session Organizer: David Kindig, University of Wisconsin Population Health Institute
Session Room: Auditorium
Comment on Priorities
Bruce Link, University of California, Riverside
Ana Diex Roux, Drexel University
Sanne Magnan, NASEM Roundtable Co-Chair
The joint IAPHS-NASEM Symposium on Oct 3 and many of the sessions in the body of the IAPHS meeting will contain ideas for both the research and policy agenda for both organizations going forward. IAPHS will make technology available during the meeting for attendees to register opinions on what they believe are priorities going forward. Thursday night these responses will be curated and collated and made available for a session perhaps late Friday morning. David Kindig will moderate a panel of the four leaders of the two organizations: Bruce Link (firstname.lastname@example.org), Ana Diex-Roux (email@example.com), Sanne Magnan (firstname.lastname@example.org) , and Josh Sharfstein (email@example.com). Each panelist will make 5-10 minute comments on the curated lists, and the floor will be open to comments from session attendees. The list of priorities as well as commentary and discussion will then be summarized and prepared for some dissemination product, at the very least an IAPHS blog.
Income Support, Tax and Labor Policy Impacts on Health Outcomes
Session Organizer: Kerry Anne McGeary, Robert Wood Johnson Foundation
Session Room: West Court
The Impact of Income Supplements on Long-term Health
Amelie Petitclerc, Northwestern University
Credit Where It’s Due: The Impact of Federal and State EITC Expansions on the Mental Health of Mothers
Anuj Gangopadhyaya, Urban Insitute
Examining Long Term Impacts of EITC and Minimum Wage Laws on Health Outcomes
Fredric Blavin, Urban Institute
This panel will present three papers that are funded by the Robert Wood Johnson Foundation’s Policies for Action Program. Each paper focuses on a different aspect of the effects of three policies designed to supplement the income of low wage workers and families – income supplements, minimum wage, and the earned income tax credit (EITC) – on short term and long term measures of health and well-being. The research projects and the panel draw on experts in clinical psychology, economics labor and human resources policy, and health policy to investigate the effects of labor and income policies on health.
Moderator: Kerry Anne McGeary, PhD, Senior Program Officer, Robert Wood Johnson Foundation
Title: The Impact of Income Supplements on Long-term Health
Amelie Petitclerc, PhD – Assistant Professor of Medical Social Sciences
Northwestern University (presenter)
Directly improving the economic conditions of poor families with young children has not been investigated as a strategy for reducing health disparities and long-term effects of income supplements are unknown. To fill this gap, we use a microsimulation approach, drawing on existing national, longitudinal data, as well as knowledge from research in economics and psychology, to estimate for the first time the effects of income supplements on health risk behaviors into early adulthood. Our simulation approach draws from the Social Genome Project (SGP), a dynamic life cycle approach to simulate the effects of policies on social mobility, developed by Brookings, Urban Institute, and Child Trends. The Children of the National Longitudinal Survey of Youth 1979 (CNLSY) data serve as the population of reference for the simulation model. The simulations allow to predict health risk behaviors for each CNLSY participant, under simulated income supplement interventions (e.g., a $2000/year child benefit for each child
Title: Credit Where It’s Due: The Impact of Federal and State EITC Expansions on the Mental Health of Mothers
Anuj Gangopadhyaya, PhD – Research Associate, Urban Institute (presenter)
Developing effective policies targeted toward improving mental health is important in reducing health disparities of vulnerable populations and disrupting the intergenerational transmission of poverty. This study assesses whether the EITC Program, which provides refundable tax credits to low-income working families, can affect maternal mental health. This is important given the uncertain role of safety net and entitlement programs (i.e. Medicaid and SNAP benefits) in the current political environment. There are several mechanisms through which EITC program that may affect mental health. First, the it provides a direct tax credit to working families. Second, at the lowest end of the wage distribution, EITC expansions are predicted to increase labor supply. Third, the EITC program may impact the health insurance coverage of those induced to work.
Using data from the Behavioral Risk Factors Surveillance System (BRFSS) and EITC benefits across states and family size, we implement a difference-in-differences design and find that EITC expansions at both the federal and state level are associated with marked improvements in parental mental health. We use the BRFSS and Current Population Survey’s Annual Social and Economic Supplement (CPS) to assess the mechanisms through which the EITC may have affected mental health. Among our sample of non-elderly parents with a high school degree or less, we find that a $1000 increase in the maximum available credit is associated with a 4 percent reduction in the number of self-reported poor mental health days in the past month. The impact is stronger among samples that had larger employment effects including white non-Hispanics, mothers, and adults under age 45. We generally find no evidence of changes in coverage for the full sample nor for our subgroups of interest in the BRFSS and CPS.
We conclude that EITC expansions improved mental health and increased employment among our analysis sample, but did not have a significant effect on health insurance status.
Title: Examining Long Term Impacts of EITC and Minimum Wage Laws on Health Outcomes
Fredric Blavin, PhD – Senior Research Associate, Urban Institute (presenter)
While there is growing literature on the effects of labor market support policies on health outcomes, the literature does not provide adequate insight into whether labor market policies that are in effect during prenatal or in childhood affect child health in subsequent years. This study estimates long-term effects of a family being exposed to EITC and minimum wage policies during a mother’s pregnancy and early childhood on child health outcomes; i.e., the impacts of exposure when a child is 0-5, 6-12, or 13-18 on subsequent health outcomes.
We explore multiple changes in federal and state EITC and minimum wage policies over the analysis period using a difference-in-differences design, exploiting the variation of EITC and minimum wage policy changes across states, years, and eligibility groups. We control for family characteristics, economic indicators, and other changes in policies during the analysis period, such as the increasing generosity of the Medicaid program to children and their parents.
We primarily use data from Panel Study of Income Dynamics (PSID), which includes information on state residency, family income over time, and health status and health behavior measures. We will also use the National Longitudinal Survey (NLSY) to explore the contemporaneous heath impacts of EITC and minimum wage policies among children, and how they change over time.
Digital Skills and Connectivity as Social Determinants of Health
Session Organizer: Amy Sheon, Case Western Reserve University
Session Room: Members
Digital Access Disparities: Policy and Practice Overview
Michael Chris Gibbons, Senior Consultant to the Federal Communications Commission’s Connect2Health Task Force and Chief Health Innovation Advisor
A Community-Based Approach to Addressing Portal Adoption Disparities and Digital Redlining in Cleveland
Adam Perzynski, Assistant Professor of Medicine and Sociology Center for Health Care Research and Policy, The MetroHealth System, Cleveland, OH
Digital Literacy Training in San Francisco: Examples from Online Patient Portal and Home-Based Programs
Courtney Lyles, UCSF Division of General Internal Medicine at Zuckerberg San Francisco General Hospital and the UCSF Center for Vulnerable Populations
Grassroots Digital Inclusion Programs in the U.S.
Angela Siefer, Executive Director of the National Digital Inclusion Alliance
The convergence of the health information technology (HIT) revolution and health care transformation have led health care systems to seek population health improvement. Digital medicine tools, including ubiquitous patient portals to electronic health records (EHRs), hold vast potential for engaging diverse patients to manage health by, for example, messaging clinicians, scheduling appointments, reviewing laboratory results and requesting medication refills. Yet emerging data portend that digital medicine could worsen health disparities because many population groups with elevated rates of chronic illness are unable to use technology due to lack of skills or connectivity. These digital gaps also limit the ability of individuals to address their SDOH such as by using the internet to find and apply for jobs or to obtain education or government benefits.
Traditional methods of promoting patient portals neglect digital skills and access that are prerequisites to adoption. The 2018 IAPHS conference offers an ideal venue for multidisciplinary discussion of the topic. Panelists offer analyses from adult literacy, sociology, and health services research perspectives. We seek to raise awareness of digital access as an issue of social justice and stimulate partnerships between health care organizations and community advocates to measure digital exclusion, test programmatic solutions, and promote policy change. Classifying digital skills and connectivity as SDOH will help turn a growing problem into an opportunity to reduce health disparities and increase equity.
Police Contact as a Social Determinant of Health
Session Organizer: Amanda Geller, NYU Sociology
Session Room: NAS 120
Police Contact, Adolescent Anxiety, and Sleep Disruption
Amanda Geller, NYU Sociology
Aggressive Policing and Internalized Homophobia, Stigma, and Discrimination among Young Men who Have Sex with Men
Joy Scheidell, NYU School of Medicine
Aggressive Policing and Somatic Symptoms among Urban Adolescents
Michael McFarland, Florida State University
Abigail Sewell, Emory University
For the past two decades, policing strategies have shifted toward a model in which officers actively engage citizens in investigative stops in attempts to detect imminent criminal activity, and use “stop-and-frisk” activity, arrests, and citations to disrupt low-level disorder, or other circumstances interpreted as indicia that crime is afoot. A growing literature documents the ways in which contact with the police may threaten individual and population health. First, contact incidents may be physically invasive. Those stopped may be handcuffed, thrown to the ground, or slammed against walls while their bodies and belongings are searched. Second, youth stopped by the police may face psychological trauma, with the potential to cause or exacerbate mental health problems. Even when police use of force does not result in physical injury, it may be stressful, particularly when officers take out a gun, taser, or other weapons. Moreover, proactive police stops are based on low levels of suspicion, and rarely result in arrest, summons, or seizure of contraband. This failure to detect illegal activity suggests that the vast majority of individuals stopped have done nothing wrong. Repeated contacts of this nature have the potential to stigmatize those stopped, triggering stress responses and depressive symptoms. A growing literature also documents health consequences of exposure to racism; if the police invoke racial invective or people perceive they were stopped because of their race, the stresses and stigma associated with being stopped may be compounded. Particularly if the “dosage” of police contact is strong, if the police threaten future violence, or if citizens believe they could be stopped at any time, the anticipation of subsequent contact could undermine their ability to function. Stresses associated with police contact have the potential to manifest physically as well as mentally and socio-emotionally, as psychological experiences may produce physical changes within the body.
This panel includes three papers examining different dimensions of the relationship between police contact and physical and mental health:
Dr. Amanda Geller and Dr. Lauren Hale examine the relationship between police contact and anxiety in a population-based sample of urban adolescents, and the physical manifestation of police-linked anxiety in disrupted sleep patterns. Data are drawn from the Fragile Families and Child Wellbeing Study, a longitudinal cohort survey of contemporary urban families. The most recent wave of data includes interviews from over 3,400 teens, and include rich self-reports of adolescent contact with the police, self-reported mental health indicators that build upon five prior waves of caregiver-reported child health, detailed self-reports of adolescent sleep patterns, and in a subset of the teens, actigraphy data on sleep quality, timing, and duration. Preliminary findings indicate a strong and robust relationship between police and adolescent anxiety. By October, additional analyses will examine the implications of this police contact, and subsequent anxiety, for sleep patterns.
Dr. Maria Khan and colleagues examine the relationship between stop-and-frisk activity and psychosocial wellbeing among young men who have sex with men (YMSM) in New York City. Data are drawn from a two-wave survey of YMSM, and to date have been analyzed using a series of cross-sectional and longitudinal regression models. Preliminary results indicate high rates of recent stop-and-frisk experience (33%) among the analysis sample, and significant associations between stop-and-frisk experience and internalized homophobia, personalized stigma, and neighborhood discrimination based on sexual orientation. Future research will confirm these relationships using propensity score models.
Dr. Michael McFarland also uses the Fragile Families and Child Wellbeing Study and examines the relationship between aggressive policing and somatic complaints among urban adolescents. Because past work posits police maltreatment can lead to worse health through stress pathways, this project tests whether aggressive policing is linked to some of the physiological responses that would be expected to occur during the initial stages of stress dysregulation. Preliminary results indicate those that experiencing aggressive policing report more somatic complaints including trouble breathing, nausea, restlessness, and feeling keyed up and tense. Future analyses will examine whether these relationships extend to those that vicariously experience aggressive policing through family, friends, or neighbors. Additional analyses will also ascertain whether key associations vary by race, ethnicity, or gender.
The panel will be moderated by Dr. Abigail Sewell, whose work examines the political economy of racial health disparities, the social construction of racial health care disparities, and quantitative approaches for studying racial inequality and structural racism.
All panelists and the moderator have indicated availability and willingness to attend the conference in October.
What are the primary pathways through which housing vouchers influence mental health for low income families? A housing experiment.
Theresa Osypuk, Associate Professor of Epidemiology and Community Health, University of Minnesota
Then I got housing and everything changed: Transitions into subsidized housing and diabetes self-management
Danya Keene, Assistant Professor of Public Health, Yale School of Public Health
Strangers in a strange land: Achieving “home” among racialized minority newcomers to small town America
Emily Walton, Assistant Professor of Sociology, Dartmouth College
Life’s a balancing act: How positive home-to-work spillover shapes health across the life course
Katherine Lin, Assistant Professor of Sociology, Dartmouth College
A house is a building where one lives, whereas home includes subjective feelings of belonging in a place. We consider both house and home to be integral aspects of health and wellbeing. The papers in this panel are wide ranging, representing a broader literature that considers material, meaningful and spatial dimensions of one’s residence as they relate to health. The material dimension of housing refers to aspects that offer physical protection and facilitate health maintenance, such as the availability, affordability, and adequacy of housing structures. The meaningful dimension refers to the social significance of housing, such as the ability to feel accepted and connected. Finally, the spatial dimension refers to the location of housing relative to services, facilities, or environmental exposures that may benefit or damage individual health.
Taken together, the collection of papers considers how these objective and subjective features of house and home relate to health and wellbeing. In considering house and home to be multifaceted landscapes of comfort, safety, belonging, and protection, we are able to elucidate a variety of processes and mechanisms through which contextual features relate to individual health outcomes. Together, these studies provide a roadmap for improving the health of vulnerable populations by enhancing not only physical shelter, through housing vouchers and subsidized housing, but also through feelings of “home”, such as promoting inclusive public spaces, and balance between the spheres of work and home.
Contextualizing Sexual Minority Health Disparities
Session Organizer: Kasim Ortiz, University of New Mexico
Session Room: Board
Health and Access to Care Among Reproductive Age Women by Sexual Orientation and Pregnancy Status
Gilbert Gonzales, Vanderbilt University, School of Medicine
Contextualizing Barriers to Care: LGBTQ Experiences Across Health Settings
Emily Paine, University of Texas-Austin, Department of Sociology
Union Status and Overweight/Obesity among Sexual Minority Men and Women
Bridget Gorman, Rice University, Department of Sociology
Mental Health, Perceived Neighborhood Social Cohesion & Sexual Orientation: Pooled Analyses (National Health Interview Survey [NHIS], 2013—2016)
Kasim Ortiz, University of New Mexico
Since the landmark 2011 Institute of Medicine (IOM) report, “The Health of Lesbian, Gay, Bisexual & Transgender People: Building a Foundation for Better Understanding”, a proliferation of research has documented health disparities experienced by sexual minority populations. While extremely useful, the subfield of sexual minority population health appears to be a natural conjecture to begin not only documenting such disparities but also identifying the root causes of these disparities. To achieve such a new generation of sexual minority population health research, requires researchers to consider further contextualizing disparities in health by sexual orientation statuses. The proposal panel session aims to showcase research that seeks to disentangle the drivers of health disparities experienced by sexual minority populations.
1. Health and Access to Care Among Reproductive Age Women by Sexual Orientation and Pregnancy Status
Gilbert Gonzales, PhD, MHA, Vanderbilt University School of Medicine (firstname.lastname@example.org )
Nicole Quinones, Vanderbilt University School of Medicine (email@example.com )
Laura Attanasio, PhD, University of Massachusetts at Amherst (firstname.lastname@example.org )
Background. A large body of research has documented disparities in health and access to care among sexual minority populations, but very little research has focused on health care needs and health risk factors among pregnant sexual minority women.
Methods. Data for this study came from 4,020 reproductive age (18-44 years) women who identified as lesbian or bisexual and 64,188 reproductive age women who identified as heterosexual in the 2014-2016 Behavioral Risk Factor Surveillance System. Logistic regression models compared health outcomes and health care access by sexual minority and pregnancy status while controlling for demographic and socioeconomic status.
Results. Approximately 3% of reproductive age sexual minority women were pregnant. After adjusting for sociodemographic characteristics, pregnant sexual minority women were more likely to have unmet medical care needs due to cost, frequent mental distress, depression diagnoses, poor/fair health, poor physical health days, activity limitations, chronic conditions, and behavioral health risk factors, such as smoking and alcohol consumption, compared to pregnant heterosexual women.
Conclusions. This study adds new research to the limited body of evidence on health and access to care for pregnant sexual minority women, who may face stressors, discrimination, and stigma that may lead to worse health outcomes and health behaviors during pregnancy. More research and programs should focus on perinatal care that is inclusive of diverse families and sexual orientations.
2. Contextualizing Barriers to Care: LGBTQ Experiences Across Health Settings
Emily Paine, University of Texas—Austin (email@example.com)
Compared to their heterosexual and cisgender (non-transgender) peers, lesbian, gay, bisexual, transgender and queer (LGBTQ) individuals in the U.S. are at high risk of poor health and underutilization of healthcare. Among LGBTQ people, however, little is known about the two groups least likely to access care and most likely to report unmet healthcare needs: trans men, and lesbian, bisexual, gay and queer (LGBQ) cis women. Moreover, despite overwhelming evidence demonstrating that context matters for experiences of stigma and discrimination, how organizational processes shape LGBTQ perceptions of—and subsequent decisions about—healthcare has yet to be examined. In this paper, I analyze data triangulated from interviews with 50 LGBTQ patients, observation at a LGBT Health Clinic, and interviews with 23 providers and staff at this site. Analyses reveal new barriers to care for LGBQ cis women and transmasculine individuals across healthcare contexts. Ultimately, findings demonstrate that understanding the processes through which medical constructs of gender/sex/sexuality are embedded and enacted in health settings is key to understanding what leads LGBTQ individuals to seek or avoid care.
3. Union Status and Overweight/Obesity among Sexual Minority Men and Women
Zelma Oyarvide, Rice University (firstname.lastname@example.org)
Bridget K. Gorman, Rice University (email@example.com) Navya Kumar, Rice University (firstname.lastname@example.org )
In this study, we draw on data from the 2010 Social Justice Sexuality Project (SJSP) and examine how union status (married/civil union/domestic partnership, partnered, and not partnered) relates to overweight/obesity among sexual minorities. Our analytic sample includes 3,950 respondents aged 18+ who identified their sexual orientation as gay, lesbian, bisexual, or alternative and their gender as either male or female. Preliminary analyses indicate that those who are in legal unions are heavier than those who are partnered or not partnered. This pattern holds across gender, except for bisexual men (where those in legal unions have the lowest rate of overweight/obesity). Logistic regression models reveal two main findings. First, among women, that being in a legal union with either a man or a woman is positively associated with overweight/obesity – and that weight differences relative to partnered and not partnered sexual minority women are driven by demographic and socioeconomic status differences across groups. Second, among men, union status is not significantly associated with overweight/obesity, but models also show that gay men who report not having a partner are significantly less likely to be overweight/obese after we adjust for the confounding effects of outness and aspects of integration into the LGBT community.
4. Mental Health, Perceived Neighborhood Social Cohesion & Sexual Orientation: Pooled Analyses (National Health Interview Survey [NHIS], 2013—2016)
Kasim Ortiz, University of New Mexico (email@example.com )
Population health scientists have long evaluated the role of neighborhood processes in contributing to population-level health disparities. Yet very little of this research has utilized nationally representative data to evaluate how variations in mental health is shaped by perceived neighborhood social cohesion; especially when considering differences by sexual orientation and within-group racial differences among sexual minorities. To fill this gap, the current paper draws upon pooled annual data from the National Health Interview Survey (2013—2016) [N=132,893] and deploys a series of linear, logistic, and ordered logistic regression techniques. Theoretically grounding this research was intersectionality, whereby instead of modeling the intersections of social positions, I utilized perceived neighborhood social cohesion as a marker for structural inequality. Results reveal that perceptions of higher neighborhood social cohesion is positively associated with lower reports of psychological distress; and holds true for heterosexuals, lesbian/gay, and bisexual populations respectively. Furthermore, when comparing sexual minority subgroups to their heterosexual counterparts, both lesbian/gay and bisexual populations respectively exhibited lower reports of psychological distress as perceptions of neighborhood social cohesion increased. Interestingly, when evaluating within-group racial differences among sexual minority populations, results varied in terms of whether perceived neighborhood social cohesion operated in a protective dimension for deleterious psychological distress. Policy and theoretical implications are discussed in greater detail.
Concurrent Session 5
Friday, October 5
1:30pm – 2:45pm
Life Course and Contemporary Exposure to Adverse Environments and Physiological Stress During Adolescence: Evidence from the Adolescent Health and Development in Context Study
Session Organizer: Christopher R Browning, Department of Sociology, The Ohio State University
Session Room: Auditorium
Early Childhood Family Conditions and EBV Reactivation in Adolescence
Kammi K Schmeer, Department of Sociology, The Ohio State University
Contact with the Police and Salivary Cortisol Levels among Urban Adolescents
Christopher R Browning, Department of Sociology, The Ohio State University
Sex Differences in the Consequences of Adverse Spatial Exposures for Chronic Physiological Stress
Jodi L Ford, College of Nursing, The Ohio State University
The purpose of this session is to explore the link between exposure to adverse social environments and physiological stress during adolescence. We consider environments across the life course from early childhood to contemporary spatial exposures in exploring the impact of inequalities in social context on multiple stress biomarkers among urban youth. The session will be composed of three interrelated papers with commentary provided by Natalie Slopen – an expert in social disparities in stress. The first paper focuses on the impact of early childhood family conditions on immune system deregulation among adolescents. The second paper explores the link between prior interaction with police and hypothalamic-pituitary-adrenal (HPA) axis dysregulation measured by nightly salivary cortisol. The third paper examines the impact of exposure to violent contexts on a measure of chronic physiological stress (cortisol from hair).
The papers in this session leverage data from two NIH-funded studies: (1) The Adolescent Health and Development in Context (AHDC) study (Browning, 1R01DA032371) – a prospective representative cohort study that examines the impact of activity space exposures on the behavioral and health outcomes of 1400 diverse youth aged 11-17 years in Franklin County, Ohio and (2) The Linking Biological and Social Pathways to Adolescent Health and Wellbeing study (Ford, 1R21DA034960) that collected data on nightly salivary cortisol, Epstein-Barr virus salivary antibody and viral DNA, and hair cortisol among a subsample of AHDC youth (N=650-700).
Drawing on Dr. Slopen’s research expertise on the psychological and biological mechanisms through which childhood experiences become biologically embedded, we hope to generate a fruitful discussion. Topics for discussion will include: the mechanisms through which prior life course adverse exposures are manifest in contemporary physiological stress processes; the potential implications of cumulative life course adversity for stress profiles during adolescence and their long-term consequences; and methodological issues associated with employing diverse measures of physiological stress in population health research. Dr. Slopen will consider these and other issues in a brief commentary following the paper presentations and then open up the discussion for audience questions.
Mortality in the United States: Progress and Retreat
Session Organizer: Virginia W Chang, College of Global Public Health, New York University
Session Room: West Court
The Contribution of Drug Overdose to Educational Gradients in Life Expectancy in the United States
Jessica Y Ho, Davis School of Gerontology, University of Southern California
Social and Behavioral Determinants of the Narrowing Black-White Life Expectancy Gap, 1990-2015
Arun S Hendi, Center for Economic and Social Research, University of Southern California
Do the Effects of Major Sociodemographic and Behavioral Risk Factors for Mortality Rise or Fall with Age?
Neil K Mehta, School of Public Health, University of Michigan
The Excess Mortality Attributable to Living in the United States
Patrick M Kreuger, University of Colorado Denver
This panel centers on status of mortality and life expectancy in the United States, unpacking recent trends in three central axes of disparity and variation: education, race/ethnicity, and age. It also considers how the U.S. fares relative to other high income countries. While black-white racial disparities have improved, education gradients have widened, and the U.S. continues to lag behind other wealthy countries, despite large expenditures on health care, with life expectancy declining for the second year in a row.
Dr. Jessica Ho examines the contribution of drug overdose to expanding education gradients, providing in-depth analyses on this popular explanation. Since the mid-1990s, the U.S. has witnessed a dramatic rise in drug overdose mortality. Educational gradients in life expectancy widened over the same period, and drug overdose likely plays a role, particularly for non-Hispanic whites. The contemporary drug epidemic is distinctive in terms of its scope, the nature of the substances involved, and its geographic patterning, which influence how it impacts different education groups. The study uses vital statistics and National Health Interview Survey data to examine the contribution of drug overdose to educational gradients in life expectancy from 1992–2011. Over this period, years of life lost due to drug overdose increased for all education groups and for both males and females. The contribution of drug overdose to educational gradients in life expectancy has increased over time and is greatest for non-Hispanic whites. Drug overdose accounts for a sizable proportion of the increases in gradients, particularly at the prime adult ages (30–60), where it accounts for 25% to 100%. Drug overdose mortality has increased more rapidly for females than for males, leading to a gender convergence. These findings shed light on the processes driving recent changes in educational gradients in life expectancy.
Dr. Arun Hendi explores several potential explanations for dramatic improvements in the black-white life expectancy gap. Between 1990 and 2015, the life expectancy of black men and women in the United States improved dramatically, increasing by approximately 8 years for men and 5 years for women. These paces of improvement rival those of the countries with the highest-ranked life expectancy in the world, and have in part resulted in a halving of the black-white life expectancy gap. This was a period of monumental change for blacks in the United States, encompassing improvements in socioeconomic conditions, increased international migration, and changes in health behaviors and access to care. This study examines several potential explanations for these life expectancy trends, including changes in where blacks live; increased black immigration from Africa, Latin America, and the Caribbean; reductions in smoking-attributable mortality; improvements in educational attainment; and trends in causes of death amenable to health care or closely linked to health behaviors. The results also identify a striking and previously undocumented age pattern of changes in the black-white life expectancy gap: while the gap below age 50 has been narrowing rapidly since the early 1990s, the gap above age 50 did not start narrowing until the mid-2000s.
Drs. Neil Mehta and Hui Zheng consider whether the influence of specific sociodemographic and behavioral risk factors actually rise or fall with age, offering new analytic insight on a controversial area. Whether the effect of a sociodemographic or behavioral risk factor on mortality rises or falls with age has important relevancies to life-course theory and public health policy. Many studies report that the hazard ratio of dying associated with such risk factors decline over age. Risk factors found to conform to this pattern include those that are socioeconomic, behavioral, and physiological in nature. In this paper, they show that the putative declining effect of a risk factor over age is a function of interpreting an interaction between a risk factor and age on a multiplicative scale. Drawing from well-known principles on statistical interaction, they show that interpretations on the additive scale often lead to different set of conclusions about the nature of the interaction. Namely, they show that on an additive scale the excess death risks posed by many major risk factors tends to increase with age. Studies have not generally recognized this additive interpretation. The paper will discuss how the prevailing pattern of increasing susceptibilities by age has critical underpinnings for cumulative disadvantage processes and public health interventions. Data from the U.S. National Health Interview Survey are used to provide empirical support.
Lastly, Dr. Patrick Krueger and Sarah Dehry estimate the excess mortality attributable to living in the U.S. vs. other high income countries. The U.S. fares poorly in terms of survival compared to other high income countries—and continues to fall further behind—despite its affluence and high health care spending. Although many high income countries continue to experience regular improvements in longevity, life expectancy in the U.S. has declined for the second year in a row. The study uses 2014 data from the Human Mortality Database (HMD) for the U.S. and 21 other high-income countries with high life expectancies. They use age- and sex-specific counts of deaths and population size to estimate mortality rates, and estimate the number of lives we could save if the U.S. experienced the mortality rates of each of the other countries. The main results show that the U.S. would experience 682,000 fewer deaths if it had the mortality rates observed in Japan, 658,000 fewer deaths if it had the mortality rates of Australia, and 645,000 fewer deaths if it had the mortality rates observed in Switzerland. The countries with the lowest mortality differ for women and men, and separate estimates will be provided. The U.S. could eliminate hundreds of thousands of deaths each year by achieving the lower mortality rates observed in other countries. Indeed, the number of lives that could be saved would be equivalent to eliminating all heart disease or all cancer deaths. The U.S. has access to the same medical technologies as other high income countries, and spends more on health care. Thus, these excess deaths are likely a result of adverse social or behavioral factors in the U.S.
Population Health Science: NIH Goals and Challenges
Session Organizer: Christine Bachrach, University of Maryland
Session Room: Members
Yonette Thomas, International Society for Urban Health
The NIH Health Disparities Strategic Plan: The View from NIMHD
Eliseo Perez-Stable, National Institute on Minority Health and Health Disparities
Population Health Approaches to Health Disparities at the National Cancer Institute
Robert Croyle, National Cancer Institute
NHLBI, Health Disparities, and Population Health
George Mensah, National Heart, Lung and Blood Institute
NIA’s Role in Supporting Population Health Science on Health Disparities
John Haaga, National Institute on Aging
One of the hallmarks of population health science is the commitment to integrating across the multi-level influences on health – from cells to society, from genetic codes to zip codes. In 2018, NIH will be releasing (through the National Institute on Minority Health and Health Disparities) a major strategic plan for health disparities and minority health research. The framework underlying the plan adopts a multi-level perspective on health determinants, suggesting a high level of relevance for population health scientists. The panel centers on the strategic plan with an aim to inform and challenge population health scientists about goals and opportunities in minority health and health disparities research. NIMHD Director Eliseo Pérez-Stable will lead off the panel with an overview of the plan and NIMHD’s own vision for building its science, emphasizing goals with relevance for population health science. Three additional leaders from NIH institutes that support population health science will complement Dr. Perez-Stable’s presentation by speaking to the challenges and opportunities for population health science and health disparities from the perspective of their own institute. After initial brief presentations, the moderator will lead the panelists in discussion with attendees.
Advances in Understanding Health Outcomes through Network Analysis
Session Organizer: Jason Fletcher, University of Wisconsin-Madison
Session Room: NAS 120
Social Networks and Opioid Use: Social Influence and Network Dynamics among Rural Opioid Users in Western Pennsylvania
Ashton Verdery, Penn State
Incarceration and Race Disparities in STDs: An Agent Based Modeling Approach
Kate Strully, SUNY-Albany
Peer Effects of Obesity on Child Body Composition
Jason Fletcher, University of Wisconsin
The purpose of our panel is to showcase three research papers that together explore a variety of ways that social network analysis is being used to uncover new findings related to health outcomes.
We bring together researchers from economics/policy, sociology, and epidemiology/public health in our panel who have experience working and speaking in interdisciplinary settings. We will both introduce the audience to social network analysis techniques and then apply these methods to three substantive areas: childhood obesity, opioid use in rural areas, and sexually transmitted disease dynamics.
Verdery will present co-authored work that uses a mix of network data and qualitative intervies with rural opioid users in Western Pennsylvania. The data are longitudinal and trace how social influence and network dynamics predict opioid use and abuse.
Strully will present co-authored work that uses network analysis and agent based modeling approaches to explore racial disparities in STDs with a focus on incarcerated populations who are released back to their communities. The models test two effects: the first test is a “simple effect” of male removal where agents’ behavioral rules do not change in response to the imbalanced sex ratios that accompany incarceration. A second test is a “behavioral effect” of male removal where the rules governing the formation of concurrent relationship do vary in response to male incarceration rates and skewed sex ratios.
Fletcher will present co-authored work that uses weight, height, and fitness measurements from the universe of public school children (>1 million) in New York City followed longitudinally between K-12 grade. He explores how peer obesity is transmitted to classmates and whether this transmission is moderated by grade level and sociodemographic characteristics. The work also explores the impacts of school policies related to obesity as they filter through social influence dynamics.
The Role and Application of Agent-Based Modeling in Evaluating the Multi-Level Impact of Population Health Interventions: Clinical, Epidemiological and System Science Perspectives
Session Organizer: Stacy Tessler Lindau, The University of Chicago
Session Room: NAS 125
CommunityRx: A Complex Adaptive System for Population Health Improvement
Stacy Tessler Lindau, University of Chicago, Department of Ob/Gyn and Medicine-Geriatrics
The Individual-level Impact of CommunityRx on Mid-age and Older Adults: Results from a Pragmatic Clinical Trial
Elizabeth Tung, University of Chicago, Department of Medicine-General Internal Medicine
Using Agent-Based Modeling to Forecast the Multi-level Impact of the CommunityRx Intervention
Jonathan Ozik, Argonne National Laboratory and Computation Institute at University of Chicago
Parameterizing the CommunityRx Agent-Based Model: Agents, Activities, Administrative Data and Community Assets
Kelly Boyd, University of Chicago, Department of Ob/Gyn
This multidisciplinary panel aims to educate a diverse audience of population health researchers and practitioners about the (1) benefits of conceptualizing a health care system as a complex adaptive system, using the CommunityRx (CRx) intervention as a case study (Stacy Lindau, MD, MAPP); (2) CRx theories of change, based on prior studies and using new results from a pragmatic clinical trial (Elizabeth Tung, MD, MS); (3) steps, including innovative methodologies, to parameterize the CRx agent-based model (ABM) with survey, administrative, and experimental data (Kelly Boyd, BS); and (4) validation and computational aspects of operating an ABM for a population health science study (Jonathan Ozik, PhD). CRx is one of a growing number of interventions to support self-care by connecting people to community-based resources for basic and other health-related social needs (HRSNs). This panel, moderated by an expert on the applications of systems science to epidemiologic research (Ana Diez-Roux, MD, PhD, MPH), will offer scientists and practitioners tools for multi-level understanding and evaluation of this class of population health interventions.
CRx systematically matches people, using evidence-based algorithms, to nearby community resources for wellness, self-care and caregiving. We developed the intervention with NIH ARRA funding and a 2012-15 Health Care Innovation Award (HCIA) from the Center for Medicare and Medicaid Innovation (CMMI). The theory of action for CRx was informed by Ronald Andersen’s widely-applied Behavioral Model of Health Services Use which posits self-care resources as a highly mutable, individual-level enabler of health, dependent on the ability and willingness of individuals to use these resources. To promote ability and willingness to use community resources for self-care, CRx used a 3-pronged approach: (a) comprehensive, ongoing assessment (eligibility, capacity, quality) of community resources to address basic and other health-related social needs (HRSNs) using feet-on-the street and phone survey methods, (b) integration of a simple resource referral process into routine, EMR-based clinical workflows, and (c) availability of a community resource navigator.
We implemented CRx at 33 health care sites for more than 113K unique individuals. In a volunteer survey of 458 HealtheRx recipients, 71% reported finding resources on the HealtheRx they did not previously know about and 19% reported going to one or more of those places as a result of the intervention. Interestingly, 49% of participants reported sharing the resource information with another person. The scale of the intervention, the inherent heterogeneity of the patient populations, and the flow and spread of information from primary to secondary and tertiary members of the community, are features of CRx that motivate us to study it as a complex adaptive system.
CMMI contracted a third-party evaluator to assess the impact of CRx on healthcare utilization. This evaluation found that during the 3 years following implementation, Medicare beneficiaries who received at least one HealtheRx (n=7385) had an average increase of 23 primary visits per 1000 beneficiaries (90% CI: 4, 42) and an average decrease of 17 hospital admissions per 1000 beneficiaries [90% CI: -21, -13) compared to matched controls (n=7260). Over the same time period, Medicaid beneficiaries who received at least one HealtheRx (n= 2408) had an average decrease of 51 emergency department visits per 1000 beneficiaries compared to matched controls (n=2437). To our knowledge this is the first large scale effort to study the impact of an HRSN intervention on healthcare utilization. The healthcare utilization impact data are being used to parameterize the CRx ABM.
We recently completed a pragmatic trial of CRx (NCT02435511) (209 in the intervention group, 211 controls) to test the effect of delivering CRx on health-related quality of life and self-efficacy for self-care. In an intent-to-treat analysis with follow-up at 1 week and 1 and 3 months, we found a steady increase in self-efficacy for self-care at each time point and significantly higher self-efficacy at 3 months (OR: 2.08, 95% CI: 1.18, 3.63, p=0.01), but no significant impact on health-related quality of life. In this study, we found that 48% of people who recalled receiving the intervention (n=145) shared information from their HealtheRx with someone else. Outcome data from this clinical trial, including data on the flow and spread of information via patient and provider social networks, are also being used to parameterize agent communication within the CRx ABM.
The CRx ABM is being developed to run in silico experiments that will quantify the population-level impact of the intervention. The development and parameterization of the ABM has happened in parallel with many of the activities described above. In addition to data from these quasi-experimental and experimental trials, we are also using data from Synthetic Populations and Ecosystems of the World 2017 (SPEW) (as a starting point for the agent synthetic population), the American Time Use Survey (to assign activity profiles to each agent), MAPSCorps (a youth asset mapping program that generates an annual census of community resources), American Community Survey (to assign sociodemographic characteristics to each agent), NHANES (to assess average number of doctor visits per year) and other sources. Using these data, we are implementing an agent activity decision model for the CRx ABM agents. This allows for individual agents to choose between more or less self-efficacious activities, comparing each agent’s natural activation score (a), to a function of an agent’s knowledge about a resource (b), the inherent inertia that an agent needs to overcome for an activity type (c) and a resource type located at a specific distance from the agent to a resource (d), where the more self-efficacious activity is chosen if b/(c * d) > a. We are also implementing the process of information spread in the ABM through co-located activities and information “dosing” dynamics.
The scalability and sustainability of evidence-based interventions to support HRSNs depends heavily on the ability of innovators and scientists to quantify the societal value beyond individual-level effects. By integrating experimental, individual-focused intervention studies with ABM, we can appreciate beyond first order (or ripple) effects of an intervention and conduct efficient in silico experiments to understand how population health interventions will perform under varying conditions and over longer time horizons than is feasible with experimental research.
The Digital Revolution and Interdisciplinary Population Health Research: Four Case Studies on Racism, Disease Mortality, Toxic Substances, and Housing
Session Organizer: Merlin Chowkwanyun, Columbia University
Session Room: Board
Immersive Virtual Reality and Combatting Everyday Racism
Courtney Cogburn, Columbia University
Text-Mining Environmental Health Catastrophe
Merlin Chawkwanyun, Columbia University
What Unburied Vital Statistics Can Tell Us About Mortality in the United States
Elizabeth Wrigley-Field, University of Minnesota
Mapping and Renewing Inequality: Understanding the Legacy of Urban Renewal and Redlining
Brent Cebul, University of Pennsylvania
Population health research is now occurring amidst a radically new technological environment. Researchers now have access to cheap computational horsepower and storage capacity unthinkable a year ago. They collect data and analyze it in new ways, thanks to an explosion of software frameworks often managed by communities of developers scattered around the world. Data itself is also more voluminous and diverse than before, with some types of data only having come into existence in the past decade.
Discussions of these developments and their population health ramifications often occur at a highly abstract level. This panel takes a different approach. It looks at four unique projects leveraging technological innovation to shed light on a particular population health domain. Each panelist will first describe technological infrastructure he or she devised, then briefly show how it was used to answer a specific question on a specific domain. None of the projects, this panel “argues,” could have been done 10 years ago.
The four panels are:
Courtney Cogburn (firstname.lastname@example.org), Assistant Professor, Columbia University School of Social Work
Title: Immersive Virtual Reality and Combatting Everyday Racism
Panelist’s Discipline: Psychology
Domain: Everyday racism
Achieving racial justice requires understanding racism. This project uses immersive virtual reality (IVR) to create a “virtual shoes” experience through which a participant can viscerally embody an avatar who encounters various forms of racism. These experiences then allow us to examine effects of an immersive virtual environment (IVE) on changes in acute physiological functioning (e.g. heart rate, heart rate variability, blood pressure) and psychological processes, including empathy/social perspective taking, racial bias and decision making. A guiding question is to see how virtual reality can induce empathy for people different from oneself.
Merlin Chowkwanyun (email@example.com), Assistant Professor, Columbia University Mailman School of Public Health
Title: Text-Mining Environmental Health Catastrophe
Panelist’s Discipline: History
Domain: Toxic Substances
This paper briefly describes a new digital infrastructure project, ToxicDocs.org, which houses 20 million pages of once-secret documents from the vaults of multi-national corporations on asbestos; PFASs; PCBs; polyvinyl chloride, benzene, and silica. Open since January of 2018, the site allows seasoned researchers, community health advocates, and the general public to have free and open access to this once-privileged material. Rendering this many documents full-text searchable required the use of high-throughput computing, whereby thousands of servers are harnessed on a single task, in this case, optical character recognition. ToxicDocs is now developing a suite of analytics tools to help researchers attack such a large text data set. One such tool is a named entity recognition miner, which extracts names, organizations, and places from text, ranks them in order of prominence, and identifies network relationships. This presentation will show how this technique was used to ferret out hundreds of “pre-Flint” lead smelter disasters that were secretly assessed by an industrial hygienist employed by the leaded gasoline industry.
Elizabeth Wrigley-Field (firstname.lastname@example.org), Assistant Professor, Department of Sociology, University of Minnesota
Title: What Unburied Vital Statistics Can Tell Us About Mortality in the United States
Panelist’s Discipline: Sociology/Demography
Domain: Disease Mortality
In the early 20th Century, detailed vital statistics were collected annually, first in selected cities, and by 1933, in all cities in the United States. However, the data have been available only in lengthy manuscript tables that are not machine-readable. For the past couple years, a team of researchers has digitized and integrated a full historical series of cause-specific data at the city level from 1900 to 1950. These tables detail annual deaths to specific causes, a treasure trove for testing hypotheses about how and why the United States transitioned from the high, young, volatile, and heavily infectious mortality of a past historical era to the low, old, stable, and heavily chronic condition-based mortality of the present. They also allow one to explore theories about demographic and epidemiological transition within the American context. These data reveal significant regional and racial variation from infectious diseases, most prominently tuberculosis. They demonstrate the importance of analyzing local characteristics and how they explain divergent outcomes in infectious disease mortality.
Brent Cebul (email@example.com), Assistant Professor, Department of History, University of Pennsylvania
Title: Mapping and Renewing Inequality: Understanding the Legacy of Urban Renewal and Redlining
Panelist’s Discipline: History
In the mid-20th century, the United States government initiated two federal programs with a lasting impact on the urban landscape today, all with enormous population health ramifications. The first was redlining: that is, discrimination in government-backed mortgage lending that exacerbated segregation and impeded access to homeownership for a generation of African-Americans. The second was urban renewal, a set of neighborhood revitalization projects that resulted in the razing of thousands of homes and displacement of their occupants. A larger sense of precisely how many people were affected and the scale of the programs has eluded researchers, however. The Mapping and Renewing Inequality is the first attempt to systematically recover all data on displacement and red-lining previously scattered in obscure government records and repositories across the country. This presentation will discuss the projects’ genesis and how researchers are now using this unique digital resource to estimate how many people were displaced by urban renewal or blocked off from mortgage markets.
Concurrent Session 6
Friday, October 5
3:00pm – 4:15 pm
What is it Good For? Using Epigenetics to Understand Health Disparities
Session Organizer: Belinda Needham, University of Michigan
Session Room: Auditorium
Epigenetic mechanisms of emotional/behavioral health among impoverished African American youth
Darlene Kertes, University of Florida
Are stressors in pregnancy reflected in circulating miRNAs?
Carrie Breton, University of Southern California
Social adversities, epigenetics, and the obesity epidemic
Cathrine Hoyo, North Carolina State University
Epigenetic damage in women living in LA food desert zip codes
Victoria Seewaldt, City of Hope
Childhood adversity and DNA methylation in adulthood: Epigenetic mechanisms of early life stress
Chantel Martin, University of North Carolina – Chapel Hill
Disentangling genetic and environmental determinants of race/ethnic differences in DNA methylation
David Rehkopf, Stanford University
The National Institutes of Health recently launched a new research program in social epigenomics to address health disparities. To date, the National Institute on Minority Health and Health Disparities has committed more than $26 million to support nine investigator-initiated research projects under this program, and an additional award was funded by the National Cancer Institute. This panel will bring together investigators from six of these projects to discuss their work:
• Epigenetic Mechanisms of Emotional/Behavioral Health among Impoverished African American Youth (PI Kertes) – Researchers will investigate whether environmental stressors, such as racial discrimination and exposure to violence, are associated with DNA methylation and telomere length among low-income, urban minority youth, which can help inform biological mediators of stress effects on emotional/behavioral health.
• Influence of Prenatal Psychosocial Stressors on Maternal and Fetal Circulating miRNAs (MPIs Breton and Marsit) – Researchers will evaluate whether psychosocial stressors in the maternal environment impact the pattern of expression of maternal and fetal microRNA (miRNA) from low SES Hispanic women and whether the expression of these miRNA can impact critical newborn and early life health outcomes indicative of future health trajectory.
• Social Adversities, Epigenetics, and the Obesity Epidemic (MPIs Hoyo and Kaufman) – Researchers will explore mechanisms by which social adversity confers risk for obesity in youth among Blacks, Hispanics and Whites and unravel the pathways by which mothers’ prenatal stress may alter DNA methylation and influence early development, growth trajectories and childhood obesity.
• Epigenetic Damage in Women Living in LA Food Desert Zip Codes (MPIs Seewaldt, Ann, Hyslop, and Schones) – Researchers will explore in young Women-of-Color (African-American and Latina/Hispanic-American) living in food-desert zip codes in Los Angeles, whether insulin-resistance promotes epigenetic damage and triple-negative breast cancer (TNBC) risk.
• Epigenomic Predictors of PTSD and Traumatic Stress in an African American Cohort (MPIs Uddin, Aiello, and Wildman) – Researchers will characterize genome wide patterns of leukocyte DNA methylation in African American participants in the Detroit Neighborhood Health Study, a population-based study of mental disorders among adult Detroit residents. Analyzing glucocorticoid receptor regulatory network genes, they will test the effects of social adversity on DNA methylation levels.
• Race/Ethnicity, DNA Methylation, and Disparities in Cardiovascular Mortality: NHANES 1999-2002 (MPIs Needham and Rehkopf) – Researchers will study whether differences in DNA methylation between African Americans, Hispanics/Latinos, and non-Hispanic Whites helps explain why mortality rates for cardiovascular disease are higher among African Americans and how socially-patterned risk factors become physically embodied.
Panelists will provide an overview of their research projects, including preliminary results, with particular attention to how they see their results being useful for understanding how we may most effectively reduce and eliminate health disparities. Following these brief presentations, the moderators will lead a discussion of opportunities and challenges in the emerging field of social epigenomics, defined as the branch of epigenomics that focuses on the effects of socially-patterned exposures on epigenetic modifications, such as DNA methylation and histone modifications, which may increase chronic disease risk by altering gene expression. Potential topics for discussion include scientific issues, such as analytic approaches and reproducibility of effects across cohorts, as well as broader societal issues, such as media hype and explaining results to the public.
The Zika Virus: A Population Health Pespective
Session Organizer: Molly Dondero, American University
Session Room: West Court
Critical issues around the development and implementation of a Zika virus vaccine
Anna Durbin, Johns Hopkins Bloomberg School of Public Health
Live births and fertility amidst the Zika epidemic in Brazil
Leticia Marteleto, University of Texas at Austin
Exploring the social context of the Zika virus in Brazil to inform policy and practice
Deanna Kerrigan, American University
The 2015 Zika virus (ZIKV) outbreak in Brazil and other countries in the Americas constituted a global health emergency due to the severe birth defects associated with the virus. Three years after the initial outbreak, ZIKV remains an urgent public health concern due to its short- and long-term consequences, the full extent of which are still unfolding. Of particular concern is that Zika disproportionately affects the poorest and most vulnerable groups in society and thus has the potential to exacerbate existing social inequality in reproductive, child, and adult health. Understanding the scope of the effects of ZIKV on population health is complex and multifaceted, and requires interdisciplinary and multisectoral perspectives and a range of methods.
This interdisciplinary panel brings together researchers from Public Health, Medicine, Sociology, and Demography to discuss recent developments in research about the social patterning of the prevention, transmission, and consequences of ZIKV in order to inform policies about this emerging public health threat. Infectious disease specialists Anna Durbin and Stephen Whitehead will discuss social and medical challenges in the development and implementation of a ZIKV vaccine. Leticia Marteleto, Gilvan Guedes, Raquel Coutinho, and Abigail Weitzman will provide a quantitative, demographic portrait of fertility rates in the wake of the ZIKV outbreak in Brazil, paying particular attention to age, socioeconomic, and geographic differences in fertility patterns. Deanna Kerrigan, Monica Malto, and Ana Brito will offer a public health perspective on the social context of ZIKV by presenting research from their qualitative interviews with Brazilian women diagnosed with ZIKV during pregnancy (about half of whom had babies born with microcephaly) and with health officials and healthcare providers.
Critical issues around the development and implementation of a Zika virus vaccine
Anna P. Durbin, Johns Hopkins Bloomberg School of Public Health
Stephen S. Whitehead, National Institutes of Health
In response to the declaration by WHO that the Zika virus outbreak was a Public Health Emergency of International Concern (PHEIC) in February 2016, vaccine manufacturers reported interest in developing candidate vaccines for Zika. The PHEIC was ended in late 2016 with the goal of continuing Zika prevention and control efforts outside of the outbreak scenario. More than 40 candidate vaccines have been announced by different manufacturers and developers however only 5 have entered Phase 1 clinical trials. Of those, two candidates are currently in Phase 2 clinical trial but it is unclear if either one will be able to be evaluated for efficacy in a traditional Phase 3 trial. Many hurdles exist to the rapid identification and evaluation of vaccines for emerging diseases, including ZIKV. In addition, there are some features to ZIKV that make development of a vaccine challenging. These include 1.) the potential need for a vaccine to induce sterilizing immunity to ensure the protection against congenital Zika syndrome, 2.) ZIKV induces a generally mild illness make case capture difficult, and 3.) immune cross-reactivity with other related viruses that may affect ZIKV diagnosis. In addition, because ZIKV outbreaks are episodic, many manufacturers are no longer pursuing vaccine efforts. The long pathway and hurdles that are present from pre-clinical evaluation to vaccine licensure will be discussed, with emphasis on ZIKV vaccine development. Proposed strategies for introduction and use of future licensed ZIKV vaccines, including use of a ZIKV vaccine in pregnant women, will also be discussed.
Live births and fertility amidst the Zika epidemic in Brazil
Letícia J. Marteleto1
Raquel Z. Coutinho2
1Department of Sociology and Population Research Center, University of Texas at Austin
2Department of Demography, Universidade Federal de Minas Gerais
Brazil was the country most affected by the ZIKV (Zika virus) epidemic starting in 2014. In the following year, the association between ZIKV and microcephaly (the most common manifestation of congenital Zika syndrome) in babies of infected mothers was established. In late 2015, the Brazilian Ministry of Health classified the increase in congenital malformations associated with ZIKV a public health emergency, while PAHO/WHO urged women to avoid pregnancy. The threat of ZIKV-related congenital syndrome could result in declining numbers of live births (and potentially fertility) via pregnancy delays and terminations, despite restrictive abortion laws.
Using 2010-2016 microdata from the Brazilian Information System on Live Births, this paper examines live births and age specific fertility rates before and throughout the ZIKV epidemic in Brazil, and seeks evidence of key socio-economic differences (years of education and geographic location) in these trends. A potential limitation of our findings is that the ZIKV epidemic coincided with an ongoing economic and political crisis in Brazil that may have also influenced postponement or permanent decline in live births and fertility rates.
Findings suggest a decline in live births, especially nine months after the call for pregnancy postponement. While declines in total fertility were negligible, fertility trends estimated by age and socio-economic status suggest important differences on how Zika has impacted Brazil’s fertility structure. Younger, high SES-women and women in geographic areas with higher incidence of Zika were more likely to have declining fertility rates during the period analyzed.
Exploring the social context of the Zika virus in Brazil to inform policy and practice
Deanna Kerrigan, American University,
Monica Malta, Escola Nacional de Saude Publica, Rio de Janeiro, Brazil
Ana Brito, Universidade Federal de Pernambuco, Recife, Brazil
The majority of research conducted on the Zika virus (ZIKV) has been epidemiologic and quantitative in nature, aimed at documenting virus transmission patterns. To more comprehensively inform services related to ZIKV, research investigating the social dynamics of Zika essential. This study seeks to: (1) Explore the social context of Zika in Brazil, with particular attention to the perspectives and experiences of poor women of reproductive age; and (2) Examine the program and policy response to Zika in Brazil, with a focus on identifying how this response can be strengthened to meet the needs of disadvantaged women and their families.
The study will be qualitative in nature. Women will be purposively selected for inclusion in this study through referrals from public health clinics and social service centers. We will recruit up to 30 women that were diagnosed with Zika during pregnancy, with at least half that had children born with microcephaly. We will also recruit approximately 20 key informants such as health officials, providers, and those from social service agencies. A total of up to 50 semi-structured in-depth interviews will be conducted. These interviews will explore participants’ experiences with ZIKV, access to and experiences in health services, the social context surrounding reproductive health decision-making, and dynamics with relationship partners and family members.
We will first write narrative summaries of participants’ experiences in relation to Zika in order to identify trajectories of prevention, care and support and factors that participants viewed as enablers or barriers to their engagement in services. We will compare the narratives of participants against one another to identify commonalities and distinct perspectives, including across settings. These narratives will facilitate a holistic understanding of participants’ experiences negotiating reproductive decisions and access to health care for themselves and their children in relation to Zika. In addition to narrative analysis, we will also code the textual data from both the women and key informant interviews to identify salient themes and experiences. Findings will broaden the literature on Zika, from a multi-disciplinary perspective, and strengthen health and social services in Brazil for disadvantaged women and their families.
Debt and Health Disparities in the United States
Session Organizer: Justin Denney, Washington State University
Session Room: Members
Rising Household Debt and Children’s Socioemotional Well-Being Trajectories
Jason Houle, Assistant Professor, Department of Sociology, Dartmouth College
Medicaid and financial well-being
Kyle J Caswell, Senior Research Associate, Urban Institute
A new midlife crisis? The association between child-related educational debt, retirement savings, and household wealth among the late baby boom cohort
Katrina Walsemann, Associate Professor, Department of Health Promotion, Education, & Behavior, University of South Carolina
Household Debt and Children’s Risk of Food Insecurity
Mackenzie Brewer, Assistant Professor, Department of Sociology, Baylor University (starting Fall, 2018)
Over the last several decades, inflation adjusted household debt has increased dramatically and become more difficult to repay for the average American family, in part a result of rising access to credit and the expansion of credit instruments, and in part a consequence of the rising costs of medical care. For some families, access to credit (and resulting debt) can provide opportunities for investments that improve economic and social well-being, and in turn, health. For others, who take on debt to supplement stagnating wages, debt may be more burdensome. That taking on debt may improve life chances for some and diminish it for others raises important questions about the implications of debt for social inequality and well-being, as well as whether rising debt may serve to exacerbate existing health disparities in US society.
Despite the remarkable rise in debt in U.S. society, very little work has examined the linkages between debt, health, and social inequalities. In this panel, we bring together scholars from a variety of disciplines to explore debt in its many forms and the health causes and consequences of debt. Panelists will discuss a diverse array of topics, including but not limited to: a) the social, economic, and demographic distributions of debt b) the role of social policies—such as Medicaid expansion—in reducing debt; c) the impact of different types and amounts of debt on health and well-being across the life course; and d) how rising debt has (or has not) impacted health disparities by race and socioeconomic status.
Cannabis legalization in the Americas: a multidisciplinary approach to investigating the impact of legalization on drug use
Session Organizer: Silvia Martins, Columbia University
Session Room: NAS 120
Medical cannabis legalization in the United States, perceived wrongfulness, frequent cannabis use and cannabis use disorder among adults
Silvia Martins, Columbia University
Recreational cannabis legalization in the United States and cannabis use across age groups
Magdalena Cerda, University of California at Davis
The impact of cannabis legalization in Uruguay on adolescent use of alcohol, tobacco, and other psychoactive substances: a synthetic control approach
Hannah Laqueur, University of California at Davis
The economics of cannabis legalization in the Americas
Beau Kilmer, RAND Drug Policy Research Center
Changes are occurring in the regulation of access to cannabis in the Americas (i.e., United States, Uruguay, Canada). In the U.S., to date, 29 states in the United States (US) have legalized cannabis for medical purposes, 9 have legalized cannabis for recreational use (RML). In Uruguay, cannabis has been legalized for recreational consumption since 2013. Canada will begin regulating recreational cannabis sales in the second half of 2018. Our symposium will present some of the latest applications of epidemiologic and econometric, methods to identify changes in cannabis use and cannabis use disorder following these policy changes in the United States and in Uruguay. First, our symposium will address consequences of cannabis medical and recreational legalization in the US. Dr. Martins’ study will combine data on medical cannabis laws with self-reported data on perceived wrongfulness, cannabis use and cannabis use disorder using data obtained from the National Survey on Drug Use and Health restricted use files, to examine these relationships across the 50 states from 2004-2016. Then, Dr. Cerdá’s study will examine whether the prevalence, incidence and frequency of cannabis use changed following legalization of recreational marijuana use in Colorado, Washington, Alaska and Oregon from 2011 to 2016 also using data obtained from the National Survey on Drug Use and Health restricted use files. Second, our symposium will examine the impact of cannabis legalization policies in Uruguay. Legalization-driven changes in pricing, availability, and/or perceived riskiness in Uruguay may have affected adolescent marijuana use, and in turn, this may impact use of other substances, depending on whether marijuana is a substitute or complement. Dr. Laqueur’s study, using a synthetic control method approach, will examine the impact of such legalization on adolescent use of marijuana, alcohol, tobacco, and other psychoactive substances in Uruguay. Finally, our symposium will examine the economics of cannabis legalization in the Americas. Dr. Kilmer will report how the United States, Uruguay and Canada address (or propose to address) supply and taxation that could have profound implications for public health, public safety, and government budgets. Together, these studies will provide the most up-to-date information on the public health impact and economics of cannabis legalization and provide much needed data on the impact of these cannabis policies in the Americas. Finally, Dr. Galea will be the panel moderator and discuss the studies’ findings.
Biological Consequences of Social Stratification on Population Health
Session Organizer: Lindsay Fernandez-Rhodes, University of North Carolina at Chapel Hill
Session Room: NAS 125
Animal models provide powerful translational insights into social determinants of human health
Noah Snyder-Mackler, University of Washington, Department of Psychology
Pathogen Burden as a Marker of Health Disparities
Rebecca Stebbins, University of North Carolina at Chapel Hill, Department of Epidemiology
Origins of Health Disparities in Biological Aging: The Case for Studying Immune Function
Grace Noppert, Duke Universtity, Center for the Study of Aging and Human Development
A socio-epigenomic pathway to health and racial/ethnic disparities
Lindsay Fernandez-Rhodes, Universtify of North Carolina at Chapel Hill, Carolina Population Center and Department
This panel seeks to bring together interdisciplinary research examining how social stratification and biological processes interact to affect health at the population level. The research presented in this panel will examine social stratification from the lens of social position, educational attainment, income, and racial/ethnic disparities. Using population studies of molecular pathways including infection, immune function, and epigenetic profiles, this panel of research will highlight the ways in which social processes reinforce social stratification and drive population health disparities. The panel will be moderated by Allison E. Aiello, PhD of the University of North Carolina at Chapel Hill.
The first talk entitled “Animal models provide powerful translational insights into social determinants of human health” will describe how social gradients in health and survival pervade our society. In the United States, for example, low socioeconomic status individuals have poorer health and higher mortality than higher-income individuals. To date, however, such findings of social determinants of health are largely correlational, leaving large gaps in our knowledge of how the social environment alters health and survival. For instance, we still do not know how much of these associations are due to reverse causality (i.e., health selection), mediating factors (e.g., health behaviors), and/or direct effects of social adversity on health. Non-human animal models, where some complexities of the social and physical environment can be standardized, allow us to examine these possibilities – specifically how social adversity directly affects health. Here, the speak will discuss his recent work on captive and free-ranging non-human primates—our close biological relatives—that provide novel biological and translational insight into the molecular mechanisms that may underlie social determinants of health in humans.
In a second talk entitled “Pathogen Burden as a Marker of Health Disparities”, our second speaker will highlight the existing population-level disparities in persistent infections in the United States, investigating disparities seen specifically in educational attainment, income, and race/ethnicity. She will discuss several plausible mechanisms through which these social strata manifest themselves physiologically, in presence of persistent infections, and ultimately impact health. This discussion will go further and provide results from original research using data on 17,651 participants from 8 waves of continuous NHANES. The talk will include a presentation of calculated cross-sectional age-adjusted mean pathogen burden scores for each category of poverty-to-income ratio (PIR), educational attainment, and race/ethnicity. Findings include the following: across all years, the highest age-adjusted mean pathogen burden score was consistently in the lowest SES category, while the highest category had the lowest mean score; in 1999-2000, those with a PIR3.5; and persistent disparities in pathogen burden were also seen in each wave by educational attainment and racial and ethnic categories.
The third talk entitled “Origins of Health Disparities in Biological Aging: The Case for Studying Immune Function” will use data from the Health and Retirement Study (HRS) to examine population level dynamics of immune function. Critical to this pathway is the role of persistent viral infections. There is substantial evidence documenting the unequal distribution of persistent infections, such as cytomegalovirus (CMV), alone lines of socioeconomic disadvantage. Individuals of low socioeconomic status are both more likely to be infected with CMV, and also have less immune control over the infection. This talk will discuss how persistent infections, such as CMV, are inducing a process of accelerated immunosenescence evidenced by both a pro-inflammatory profile and a T-cell compartment with disproportionate amounts of aged cells. This talk will conclude by highlighting the ways in which accelerated immunosenescence may be a critical driver of many of the observed health disparities in aging.
The final talk entitled “A socio-epigenomic pathway to health and racial/ethnic disparities” will highlight data on 2,633 African American adults collected as part of population-based the Atherosclerosis Risk in Communities Study (ARIC), to describe how socioeconomic adversity gets ‘under the skin’ to influence genetic expression via epigenetic mechanisms such as DNA methylation. The speaker will present her recent epigenome-wide study of more than 450,000 methylation at cytosine proximal to guanine (CpG) sites and their association with socioeconomic adversity, as measured by highest educational attainment and income. Preliminary analyses reveal that several previously-associated obesity methylation loci also associate significantly with socioeconomic adversity in African Americans after accounting for age, sex, current smoking status, center and technical covariates. Therefore, the speaker will also describe the subset of methylation loci that robustly association with socioeconomic adversity after additional adjustments for measures of obesity—body mass index and waist hip ratio, and perhaps are unrelated to obesity or its cardiometabolic disturbances. The talk will then conclude with a discussion of the implications that such a social-epigenomic pathway to health has for our understanding of public health, policy and social justice in the United States.
Aligning Health and Social Systems to Improve Population Health: The Role of Networks, Governance and Information
Session Organizer: Glen Mays, University of Kentucky Systems for Action Research Program
Session Room: Board
Multi-sector Community Networks and their Impact on Medicare Spending
Glen Mays, University of Kentucky
Assessing the Carrying Capacity of the Local Nonprofit Sector in Supporting Population Health
Rachel Graham, Trailhead Institute
Testing a Shared Governance Model for Health and Social Service Delivery in East Harlem
Carl Letamendi, New York City Department of Health and Mental Hygiene
Using Longitudinal Information on Unmet Needs to Target Supports for Vulnerable Seniors
David Meltzer, University of Chicago
Social, economic, and environmental conditions strongly influence population health, but too often the services and supports designed to improve these conditions – such as housing, transportation, or financial assistance – are disconnected from the medical and public health services tasked with improving health. Delivery and financing systems for medical, social, and public health services operate largely in isolation from each other despite pursuing many common goals and serving overlapping populations. These systems interact in complex and often poorly understood ways through fragmented funding vehicles, information systems, governance and decision-making structures, implementation rules and strategies, and professional practices. New research is needed to untangle these interactions and test novel mechanisms for coordinating health and social services in ways that improve population health.
Systems for Action is a national research program of the Robert Wood Johnson Foundation that studies novel ways of aligning delivery and financing systems that operate across the health and social sectors. This panel highlights findings from four studies underway through the S4A program. Collectively, the studies profiled in this panel reveal how interorganizational networks, governance structures, and information flows function as mechanisms for system alignment in local communities.
Multi-sector Community Networks and their Impact on Medicare Spending.
Glen Mays, University of Kentucky
This study follows a nationally representative cohort of U.S. communities to estimate how community networks connecting medical, social, and public health organizations influence Medicare spending over time. Our retrospective cohort design follows more than 350 U.S. metropolitan communities over time using survey data collected over an 18 year span from 1998 through 2016. Each year local public health officials report on the implementation of 20 nationally-recommended health improvement activities in the community, and on the organizations that contribute to performing each activity including hospitals, primary care providers, insurers, employers, schools, public health agencies, social and community-based organizations. We classify communities into one of seven categories of multi-sector network strength based on a cluster analysis of the scope of recommended activities contributed by each type of organization, along with the density of relationships connecting these organizations. Using geographic identifiers, we link these survey data with and area-level measures of Medicare spending per beneficiary, along with measures of community demographic, socioeconomic, and health resource characteristics from other data sources. Random-effects models with instrumental-variables are used to estimate whether changes in multi-sector health activities lead to changes in Medicare spending, while controlling for both observable and unmeasured confounders. We find that communities with higher density of multi-sector health activities experienced significantly lower levels of medical care spending as measured by adjusted Medicare expenditures per person, particularly when instrumental variables-analysis is used to adjust for unobserved heterogeneity that jointly influence multi-sector activity and medical care spending. Communities achieving the highest density of activity experienced Medicare spending levels 37.2% lower than communities with the lowest density of activity (p<0.01) , after controlling for other factors. Medicare spending fell by an average of 4.1% for each year in which dense multi-sector activities were maintained over time (p<0.05). We conclude that initiatives to promote multi-sector health activities provide an effective way of containing avoidable medical costs in Medicare, alongside the gains in health status shown in prior studies. In view of these findings, Medical providers operating under Medicare’s alternative delivery and financing system models, such as accountable care organizations and value-based payments, should seek to strengthen relationships with social and public health organizations.
Assessing the Carrying Capacity of the Local Nonprofit Sector in Supporting Population Health
Danielle Varda, Trailhead Institute
Across the country, efforts are underway to improve outcomes and reduce costs within the medical care system by screening patients for unmet social needs and linking them to available community resources. These efforts present both opportunities and challenges for the nonprofit sector. On one hand, the increased demand for nonprofit community-based services brings the promise of support and capacity-building from hospital and health systems. On the other hand, the carrying capacity of the nonprofit sector (that is, the capacity of the organizations in the sector to meet increased demand) is already stretched and this increased dependency on the sector presents challenges. Currently, measures of “community carrying capacity” to address social determinants of health are not established and lack validation. This study uses concepts from network theory and analysis to develop and validate a set of indicators, generated through an exploratory, community-based participatory research (CBPR) approach, to capture the carrying capacity of community-based organizations in health care networks. Measured factors include community context, service demand, organizational capacity, input costs, collaborative processes, and funding mechanisms. The indicators are drawn from both literature and practice, and are validated through a partnership with two successful practitioner organizations in Florida and Texas who currently operate in an environment of increasing dependence on community-based organizations to provide SDOH services. Results are used to accurately assess a community’s ability to respond to increasing health system demands on the nonprofit sector.
Testing a Shared Governance Model for Health and Social Service Delivery in East Harlem
Carl Letamendi, New York City Department of Health and Mental Hygiene
This study evaluates a shared governance and decision-making model for aligning a city health department with community medical and social service providers. Employing a mixed-methods research approach, we examine data on implementing and maintaining a governance council comprised of co-located partners and explore the ways that a newly-implemented electronic referral and tracking system affects Action Center visitor satisfaction and self-reported health. These efforts lay the groundwork for evaluating the impact that cross-sectoral collaboration has on health inequities across neighborhoods.
Using Longitudinal Information on Unmet Needs to Target Supports for Vulnerable Seniors
David Meltzer, University of Chicago
This study seeks to improve health for frequently hospitalized seniors in inner-city Chicago by tracking information on the dynamics of unmet social needs experienced in the population. A randomized trial is used to examine whether community health workers and community arts and cultural programming can engage seniors in community resources that address continually changing unmet social needs.