Center on Aging & Population Sciences (CAPS)

2022. Aprile Benner (Associate Professor, Human Development & Family Sciences, UT Austin). Structural Racism, Discrimination, and Health across the Life Course .

In considering the impacts of racism on health, much is known about how individually-experienced interpersonal discrimination influences health and well-being, yet the impacts of structural racism remain vastly understudied. Moreover, this small base of extant research tends to focus on a single structural racism component, contrary to both theory and the NIMHD framework, which stress the multifaceted and interdependent nature of structural racism. This limitation is likely driven, in part, by challenges with data access, but it ultimately compromises the effectiveness of interventions targeting health and health disparities by providing an incomplete picture of the consequences of structural racism for health. The proposed project will fill this void, linking state-level structural racism data with individual survey data from the NSAL and NLAAS. Specifically, we will document the consequences of, and mediating and moderating mechanisms by which, structural racism influences health for those 60 and over. Here, we will employ multilevel SEM path analysis models to consider the interplay among levels of racism structural, individual) and their direct and indirect impacts (via personal health assets and opportunities) on mental and physical health and cognitive functioning. We will then extend these analyses to consider variation in structural racism’s pathways of influence. Multiple group and interaction analyses will investigate potential racial/ethnic variation in modeled paths. These activities and analyses will illustrate proof of concept and feasibility of the larger research aims and inform the design of the larger R01 study to be submitted to NIA.

CAPS Themes
This proposal is well-aligned with the CAPS Major Research Themes:Place, aging, and health
• Place, aging, and health
• Life course precursors of advantage and disadvantage at older ages defined.
Such exposures, however, are not uniform across geography or time in the United States, and the current study attends carefully to this variation. Specifically, we consider how structural racism permeates multiple levels of geography (i.e., Census tract, county, state) and the implications of exposure within and across levels for individuals’ physical and mental health. Moreover, the extent to which different elements of structural racism are embedded in geography is time dependent, a function of changing state-level policies, geographic mobility (e.g., migration and immigration, gentrification), and shifting national discourse around racism and discrimination, among others. This proposed project seeks to unpack how place matters for health and whether early life course exposures exact long-term impacts on health at older ages.

2022. Karen Fingerman (Professor, Human Development & Family Sciences, UT Austin). Daily Experiences and Well-being in Late Life: A Longitudinal Examination of Ecologically Valid Assessments of Older Adults’ Daily Experiences.

A vast literature documents the critical role of involvement with family, friends, and broader community for emotional, cognitive, and physical health. Yet processes underlying these associations remain poorly understood. Social engagement theory suggests that encounters with friends, family, and social groups leads to beneficial everyday activities contributing to longer term benefits. Tests of these assertions have been limited, however, due to a paucity of ecologically-valid data connecting encounters with social partners to experiences throughout the day. The proposed pilot study will lay groundwork for a second wave of data collection building on the innovative and highly successful first wave of the Daily Experiences and Well-Being Study addressing these issues.

With NIA support (R01AG046460), in 2016, we conducted the Daily Experiences and Well-being Study (DEWS) examined these propositions. DEWS provided pioneering findings identifying how encounters with different types of social partners throughout the day affect multiple outcomes: physical activity, sedentary behavior, sleep, conversational engagement, TV watching, mood and loneliness 10–14. This type of holistic and microtemporal understanding of how social ties affect older adults’ well-being was absent from the literature previously, and warrants follow up to move the field forward.

CAPS Themes
The proposed project addresses a key CAPS thematic area. The project focuses on the theme of Family Demography, Social Engagement, and Social Isolation by studying older adults’ social engagement with family, friends, and acquaintances. The initial Daily Experiences and Well-being Study (DEWS) Wave 1 provided indepth examination of physical activity, sleep, sedentary behavior, and language production associated with encounters with social partners throughout the day. This follow up study will examine the consequences of this social engagement 6 years later.

2022. Lauren Gaydosh, (Assistant Professor, Sociology, UT Austin) . Contextual Despair and Risk Behaviors in Midlife: Extending Innovative Measures to Add Health.

Life expectancy at birth in the United States (US) has declined since 2014 [1], a uniquely American phenomenon that is not observed in any other developed country [2]. Several scholars have suggested that increasing deaths due to drug overdose, alcohol-related diseases, and suicide among middle-aged adults (45-54) may explain declining life expectancy [3-5]. This cluster of causes of death – referred to as “deaths of despair” – have been attributed to community economic distress, social isolation, and other deteriorating social factors that have led individuals to feel a sense of hopelessness, which is alleviated by turning to self-destructive behaviors, such as self-harm, alcohol and drug abuse [3]. Underlying these national trends is wide variation in both levels and trends in life expectancy across place [6]. Indeed, geographically-patterned community decline increases the demand for substance use, which can ultimately lead to premature death [7, 8]. The role of community factors in the deaths of despair research indicates that the geographic context of despair should be integrated into the social etiology of recent life expectancy decline. The successful completion of this project will lead to a better understanding of the causes and consequences of rising midlife mortality in the US that can inform more effect community-level interventions.

CAPS Themes
The proposed research fits with all three of the CAPS themes. First, the project will interrogate contextual despair across the life course (from adolescence to mid adulthood) and consider how they influence health at
older ages, in line with the CAPS theme of life course precursors. Second, the pilot project will merge measures of family dissolution, social isolation, and social disengagement as aspects of contextual, in line with the CAPS theme of family demography, social engagement, and social isolation. Finally, the project investigates the role of county-level contextual factors in shaping health behaviors and outcomes, in line with the CAPS these of place, aging, and health. The proposed research fits with the UTMB Pepper Center priority
area of medical effectiveness by including measures of patient treatment facilities and providers as contextual predictors of individual health outcomes. The proposed research fits with the UTMB RCMAR priority area of health and aging in minority populations by expanding the literature on deaths of despair beyond a narrow focus on non-Hispanic White adults to consider heterogeneity in the relationship between despair and health outcomes by race/ethnicity.

2022. Mark Hayward (Professor, Sociology, UT Austin). The Dynamics of Race Differences in Life Course Pathways Associated with the Risk of Alzheimer’s Disease and Related Dementias.

The goal of the proposed pilot project is to develop a 5-year R01 NIA grant application to investigate race-specific life course pathways of Alzheimer’s disease and related Dementias (ADRD) risk. The project challenges the idea that life course risk factors are static across race groups [1] and over historical time. It considers the changing dynamics of life course pathways without homogenizing across groups, which is necessary when considering the very different lived experiences of older Blacks and Whites in a racially stratified system. The proposed research draws on our prior work showing that race differences in dementia risk are rooted in dramatic differences in the influence of early life development factors as well as adult exposures and behaviors (which we refer to as racialized life course pathways). Drawing on historical longitudinal data from Health and Retirement Study (HRS), we will evaluate cohort
changes in the influence of racialized life course pathways of ADRD risk. Over the course of the 20th century, significant shifts in life course experiences occurred in American society with the expansion of education, changes in health behaviors, and shifts in employment opportunities that may have impacted cohort changes in ADRD risk [2]. Unclear is whether this pattern of overall cohort change in ARDR risk and its ties to life course changes pertains to both Blacks and Whites, because most Black Americans were born in the Jim Crow South, where a robust and wide-reaching legal apparatus curtailed economic, social, and political rights. Our project will explore how these dynamics, along with the race stratification system in the US, may have impacted overall cognitive functioning as well as specific components (e.g., memory). We move beyond the typical global measures of ADRD risk to better understand how changes in specific cognitive health domains are linked to historical changes in cohorts’ life course exposures and ultimately tied to changes in ADRD risk for Blacks and Whites.

Additionally, this project will also leverage HRS sister studies in Brazil, South Africa, and India to develop a macro comparative approach for understanding the dynamics of the racialized pathways through a global lens. While systematic comparisons of specific macro influences are admittedly challenging, these studies allow us to develop similar life course models of ADRD risk which can be compared cross-nationally to better understand how racial caste systems, institutionalized racism, and caste systems not defined by race shape individuals’ life course exposures and ADRD risk. South Africa and the United States are similar in terms of their Black populations living in race-based caste systems. Brazil offers an important counterpoint to examine race differences in life course pathways since, while it lacks a race-based caste system, race has long defined social hierarchy and inequality. The inclusion of India in this analysis allows us to identify the similarities and differences when caste systems are not defined by race, but rather by other sociopolitical factors that also create rigid social groups.

CAPS Themes
The proposed project speaks directly to two CAPS themes.

• Life course precursors of later life health and well-being
• Place and health including neighborhoods, geographic region, and other indicators of place

The proposed study will significantly advance current understanding of the racialized pathways associated with Alzheimer’s disease and related dementias, a topic that has received little attention despite the enormous race gap in ADRD prevalence and incidence. ADRD rates are 2.5 to 3 times higher among Blacks compared to Whites and at levels that are epidemic in magnitude. The study also speaks to the often implicit assumption that the association between life course factors and ADRD is static. This study highlights the importance of considering ADRD risk in terms of dynamic changes in the life course and how these life course changes differ for Blacks and Whites. Finally, this study will shed new light on the possible ways in which macro institutional factors come into play across countries. Most research on ADRD risk is focused on the United States and Europe with much less information about the life course influences.

2022. Sae Hwang Han (Assistant Professor, Human Development & Family Sciences, UT Austin). Disability of a Family Member, Caregiving Behavior, and Caregiver Health Outcomes in Middle and Late Adulthood.

Tens of millions of middle-aged and older persons in the United States live with a disability today, with the numbers projected to grow substantially in the coming decades due to population aging. The trends of late-life disability have placed issues surrounding health of informal family caregivers at the center of political and social discourse. This is in part because many of the challenges posed by the increasing prevalence of disabilities, such fiscal concerns regarding the cost of late-life dependency and dementia care, as well as societal efforts to reduce unmet needs, are closely related to health and well-being of informal caregivers.

Despite the decades of research on this topic, however, there continues to be a lack of clear understanding of how the act of caregiving influences caregiver’s health and well-being. On the one hand, caregiving has long been studied within a chronic stress exposure paradigm (e.g., stress process model), where the act of caregiving is characterized as a chronic stressor and by extension a significant risk factor for adverse mental health outcomes, such as depression (Schulz et al., 2020). On the other hand, a stream of recent of population studies indicate that informal caregivers of people with serious debilitating conditions, including dementia, experience robust health benefits in the form of extended longevity (Roth et al. 2015), providing a picture inconsistent with the conventional view that characterizes caregiving as a chronic stressor detrimental to health.

While there have been several attempts to reconcile the seemingly contradictory set of evidence surrounding caregiving and varied health outcomes (Brown & Brown, 2014; Marino et al., 2017; Roth et al., 2013), a conceptual framework that can be used to generate testable hypotheses about when and under what circumstances caregiving behavior promotes or undermines caregiver’s health and well-being is rare. Guided by interdisciplinary research on prosocial helping behaviors and key tenets of the life course perspective, the primary goal of this pilot project is to provide a first step toward establishing a conceptual framework that can be used to test various pathways and contextual factors for explaining the linkages between caregiving and varied health outcomes for the caregiver. This project challenges the dominant research paradigm that characterizes caregiving solely as a chronic stressor detrimental to health, thereby helping to spur a needed course-correction in the public and policy discourse surrounding family caregiving (Brown & Brown, 2014; Roth et al., 2015). To this end, we take advantage of longitudinal data spanning two decades collected from a large national sample of middle-aged and older adults in the Health and Retirement Study (HRS; 1998-2018), which contains rich information on sociodemographic and health characteristics, as well as caregiving behaviors.

CAPS Themes

The central theses of the project are framed around key concepts of the life course perspective. This project investigates linked lives and of intertwined trajectories of older individuals and their close family members, where a family member’s experience of disruptions in critical life domains due to disability is also assumed to be experienced by their family member in a number of different life domains. That is, a health trajectory underlying disablement process of one family member not only influences other family members’ mental health trajectory, but it also triggers initiation of their trajectory as a caregiver (i.e., transition into a caregiver role), often described as a “unexpected career” in life course scholarship. Potential mental health outcomes of caregiving are examined through the lens of constrained agency and identity agency. As such, this project speaks to one of the key priority areas of the Center on Aging and Population Sciences, namely, investigating life course precursors of later life health and well-being.

2022. Stephanie Grasso (Assistant Professor, Speech Language and Hearing Sciences, UT Austin) . Sociodemographic and Neurocognitive Characterization of Mexican-American Bilinguals and Monolinguals Presenting with Typical and Pathological Cognitive Aging.

Society is facing a global “dementia epidemic” that will intensify with the growing aging population. Mexican-Americans are expected to demonstrate significant increases in dementia cases relative to other race and ethnic groups. The management and treatment of dementia in Mexican-Americans is hindered by the lack of comprehensive models of neural and cognitive decline that reflect the diversity of this population. Hispanics make up the largest group of bilingual speakers in the U.S. Understanding how language dominance (i.e., language preference and use), language proficiency (e.g., mastery of English comprehension and production), and age of dual language exposure (e.g., during or after childhood) influence the structure and function of the aging brain is a critical first step towards developing individualized interventions for this historically minoritized group. There is a growing body of evidence indicating that differences in bilinguals’ proficiency, dominance, and age of dual language exposure might influence executive functions (e.g., working memory, attention, and cognitive control) that can affect the risk, onset, and progression of dementia. However, the individual, family, and community factors that underly differences in dual language behavior and how these differences influence typical and pathological cognitive aging are largely unknown. To bridge this gap, we will provide an exhaustive sociodemographic (Aim 1) and neurocognitive (Aim 2) characterization of monolingual and bilingual Mexican-Americans with typical cognitive aging, MCI, and dementia. We will use these characterizations to identify the sociodemographic, linguistic, and neurobiological markers that best predict individual differences in cognitive decline. We will analyze two longitudinal datasets that include a wide variety of sociodemographic and neurocognitive measures collected from more than 3,500 aging Mexican-Americans. Our approach will combine concepts from cognitive neuroscience, linguistics, sociology, epidemiology, and demography with cutting-edge neuroimaging and machine learning tools. Machine learning models will be used to predict complex non-linear relationships in multidimensional datasets that are not captured by linear regression models. We will use the methodologies developed herein to submit NIH proposals that aim to elucidate the complex relations between the sociodemographic and neurocognitive profiles derived from this pilot project.

If our aims are realized, we expect to transform theoretical frameworks of neural and cognitive decline in Mexican-Americans. Crucially, this will lead to advances that can have immediate impacts on society and behavioral intervention approaches. For example, this proposal will serve to reduce health disparities in the historically minoritized and two underserved Mexican-American population. Additionally, our results have the potential to enhance the management and treatment of dementia in bilingual and monolingual speakers. From a methodological perspective, the machine learning approaches developed in this project have the potential to advance predictive modeling techniques used in the field.

CAPS Themes

The themes of the UTMB Pepper Center and RCMAR both emphasize the importance of conducting research on the health and aging of diverse older adult populations, with a focus on Hispanic populations in the United States. This project is directly related to these themes as we use a life course approach to investigate the role of bilingualism, English proficiency, and language dominance on changes in cognitive functioning and brain structures among older Mexican American adults. This proposal also directly addresses the three main research themes of the Center on Aging and Population Sciences. First, our proposal addresses life course precursors of advantage and disadvantage at older ages. Second, our proposal addresses the relationship between family demography and social engagement in neurotypical and pathological in underrepresented minority groups. Finally, our proposal addresses place, aging, and health in aging Hispanics in the U.S. context. The life course approach informed our decision to focus on variables in the H-EPESE and HABLE for early- (e.g., country of birth, language spoken as a child), mid-(e.g., occupation), and late- (e.g., household composition, language dominance) life characteristics that we hypothesize are associated with cognitive functioning and brain structures.

2021. Abigail Weitzman. Assessing the Feasibility of Panel Data Collection Among Adult Refugees in Costa Rica.

The proposed project aims to collect pilot longitudinal data among refugees and other migrants in need of protection (MNP) who have relocated to Costa Rica to evade an imminent threat to their survival. To do so, the project will first conduct a baseline phone survey with MNP from various different Latin American countries who are ages 18 to 95 years and currently residing in Costa Rica. Following, it will then send these same MNP a series of standardized follow-up surveys to complete online either daily, weekly, or monthly. Analyses of the resultant data will provide new information about how the retention of MNP in panel surveys varies as a function of trauma history, age, and other personal characteristics. In addition, they will reveal: (1) how much variation exists in MNPs’ health, wellbeing, and hardship on a daily, weekly, and monthly basis; (2) how trauma histories and former ecological contexts (e.g. country of origin) sort refugees and other MNP onto different health trajectories; and (3) how trauma, age, country of origin, family structure, and attributes of personal networks predict MNP health and hardship on average. The findings promise to advance research and theory on trauma, aging, and international displacement by generating pilot data for an R01 project, adapting methods to track and describe the social and health circumstances of internationally displaced populations, and examining relationships between pre- and post-flight experiences, family dynamics, social incorporation, and MNP health and wellbeing across the adult life course.

2021. Debra Umberson. How Spouses Influence Each Other’s Health in Same- and Different-Sex Marriages: A Dyadic and Longitudinal Assessment from Mid to Later Life.

Decades of population research show that married Americans are in better health and live longer than their unmarried peers and that there are gender differences in how spouses influence each other’s health. Until recently, however, this research was entirely based on marriages between women and men. In 2015, with NIA support (R21 AG0445850), we collected dyadic data to conduct the first in-depth analysis of marital processes and health comparing legally married midlife spouses in same-sex and different-sex marriages. Findings have highlighted differences in health-related marital dynamics for women and men in same and different-sex unions. These dynamics likely impact both spouses’ long-term health, but due to a lack of longitudinal and dyadic data, scientists know little about how they do so. The major aim of this pilot project is to design and launch collection of Wave 2 data to construct a longitudinal, dyadic data set that will allow us to examine how same-sex partners influence each other’s health behaviors and mental and physical health over time, and in comparison to different-sex partners.

2021. Diane Coffey. Gender Disparities in Adult Mortality and Morbidity in India: Understanding Variation and Investigating Mechanisms.

A large body of research documents health disparities between boys and girls in India.  Gender discrimination leads to skewed sex ratios at birth, and to male-female gaps in child mortality and height that are markedly different from the rest of the world.  Such research highlights the rich complexity of how biology and society interact to produce population-level child health outcomes.  For example, in most countries, girls are more likely to survive childhood and less likely to be stunted than boys (Sawyer, 2012).  Yet, in India, many child health outcomes are similar across sexes, or worse for girls.

Less research has documented, contextualized, and investigated disparities in morbidity and mortality between adult men and women.  Comparisons of gender differences in life expectancy suggest that the female life expectancy advantage that exists in almost every country is muted in India.  This proposal seeks to understand how and why women in India experience smaller, and sometimes no survival advantages compared to men.  By examining regional variation in gender discrimination, as well as regional variation in gender health gaps, we seek to build a more comprehensive model of disparities in the aging process in India, where one-sixth of people live.

This project has three aims: 1) to identify gaps in the literature about gender disparities in health in India and compare how these gaps match with NIA funding priorities; 20 to identify how data on adult health and morbidity can be used to fill gaps in the literature; 3) to prepare preliminary analyses to be used in an NIA proposal and prepare a proposal.

2021. Lauren Gaydosh. Biological Aging Among Older Sexual and Gender Minority Adults in the US South.

Older sexual minorities constitute approximately 3 million Americans, a number expected to double by 2030. Moreover, 35% of SGM adults in the US reside in the South, yet many studies focus on SGM samples in coastal metropolitan cities. Sexual and gender minority (SGM) populations experience significantly worse morbidity, mental health, physical disability, and unmet healthcare needs compared with their heterosexual and cisgender counterparts. With smaller kin networks, SGM are also more likely to experience social isolation. These health and social disparities have been exacerbated by the current COVID-19 pandemic. The Vanderbilt University Social Networks, Aging, and Policy Study (VUSNAPS) investigates the health and aging of older SGM residing in four states in the US South (Alabama, Georgia, North Carolina, and Tennessee). This pilot project will collect innovative in-home self-collected blood samples for measurement of biomarkers of aging in a subsample (n=150) of the VUSNAPS cohort. We aim to assess the feasibility of this self-collection technology both in terms of acceptability in our study population and quality of the resulting biomarker measurements. We will construct biomarker-based measures of biological aging and examine variation by sexual orientation and gender identity. Finally, we will investigate the relationship between social isolation and biological age, testing state-level variation. This project will contribute to the fields of biodemography, public health, minority health disparities, and gerontology, all of which will benefit from expanding access to biomarker data among a high-need and understudied population. The collection of biomarkers of aging among this population will be a major advancement in the measurement and study of health inequalities across gender identity and sexual orientation, especially in older age. The study will enable preliminary investigation into the disproportionate burden and lasting implications of the pandemic on health and aging in this vulnerable population.

2021. Robert Crosnoe. Health Disparities among Aging Parents of Adult Children with Serious Conditions.

This project will involve preliminary analyses and pilot data collection for a R01 proposal to NIA. The parent project—which will go through the NIH review cycle twice over the next year—is a large mixed-methods study investigating the physical and mental health toll on aging parents of children with serious conditions who have reached adulthood and how this toll can be buffered by resources in the family ecology related to marriage/partnership and by resources in the community related to the availability of health and human services. The pilot project’s goals are to conduct preliminary statistical modeling of a nationally representative data set that incorporates community-level measures from outside sources (e.g., Health Resources and Services Administration, National Center for Charitable Statistics), demonstrates feasibility, confirms power, and produces potential results supporting the proposed plan of quantitative analyses; to conduct a small number of pilot semi-structured interviews with local parents that allow refinement and expansion of the proposed qualitative sampling and data collection procedures and experimentation with the proposed textual and grounded theory analytical techniques; and to address, when possible, any suggestions and critiques that arise during the NIH review process related to data and analyses.

2020. Chandra Muller. Education and Financial Resilience During Economic Recession for Healthy Aging: Record Linkage of High School Education Longitudinal Studies to Consumer Credit Data.

2020. Elizabeth Munoz. Neighborhood Characteristics Throughout Life and Midlife Cognitive Functioning.

Aim 1: To determine whether residential neighborhoods can be successfully coded for neighborhood resources and deprivation from childhood to midlife using historical data indicative of neighborhood context.

In accordance with Aim 1, we were able to successfully code for early childhood neighborhood deprivation. In consultation with a Geographic Information Systems expert, coded age 5, 10, and 15 addresses for the Effects of Stress on Cognitive Aging, Physiology and Emotion study (ESCAPE) database.

In consultation with Dr. Gabriel Amaro, Interdisciplinary Statistician at the U.S. Census Bureau, we geocoded and linked addresses with harmonized longitudinal Census tract boundaries. Through this process, we discovered that we were only able to link participant data who had age-5 addresses from 1970 or later. Census tracts have been used since about the 1920s but only a small number of areas in the U.S. were tabulated by tracts. It was not until 1940 that tracts were officially adopted as a geographic entity by the Census Bureau and by 1970, tracts started be used in more widespread fashion. Thus, our variables of interest were only available from the 1970 Census, forward. In all, we were able to link data to 56% of the ESCAPE sample, and thus to participants who were 45 years or younger at the time of assessment. We computed an age-five neighborhood deprivation score that included the following variables: Median household income (reversed scored), percent H.S. or less, percent owner occupied units (reversed scores), % poverty rate, % unemployment rate, median contract rent (reversed scored), and percent female headed households. Our results showed that although early neighborhood deprivation was correlated with cognitive performance, this effect was not after accounting for self-reported recollections of early neighborhood environments.

Age five addresses have currently been coded for the Colorado Adoption/Twin Study of Lifespan behavior and cognitive aging (CATSLIFE). This is a younger sample between the ages of 28 and 48 and because of the dates of data collection, we were able to successfully geocode age-5 addresses for 1,236 participants and computed a composite deprivation score for age 5. Our next step will be to analyze associations with cognitive function to evaluate if results are consistent with results obtained from ESCAPE.

2020. Elliot Tucker-Drob. Large-Scale Genomic Analysis of Aging-Related Cognitive Declines Elliot Tucker-Drob.

2020. Jennifer Glass. Trajectories of Mothers’ Breadwinning Responsibilities Over the Life Course.

Our goals were to find, clean, and analyze the best available data on mothers’ financial provisioning in U.S. households to answer two questions:

(1) What is the probability that mothers today will find themselves the primary economic support for their children, and for how long a duration on average until their first child turns 18?

(2) where and why is mother’s role in financial provisioning for children growing so remarkably ?

Using longitudinal SIPP panels beginning in 2014, we first created multi-state lifetable estimates of mothers’ duration as primary earner, as well single-decrement lifetable estimates of their chance of ever being the primary earner over the first 18 years of motherhood. Using a conservative threshold of 60% or more of household earnings to determine primary earning status, around 70% of American mothers can reasonably expect to be their household’s primary financial support at some point during their first 18 years of motherhood. Mothers average 4.19 years as their families’ primary earner in the 18 years following first birth. Mothers with some college but no degree spent the most years as primary-earners, about 5.09 years on average, as did mothers with nonmarital first births, about 5.69 years

To answer the second question , we are currently using the 1996 and 2014 Survey of Income and Program Participation data to identify changes over time in the prevalence of precipitating events [a separation from a partner or other household earner, an increase in their own earnings, or a decrease in the earnings of a partner or household earner] and determine whether any changes in the effects of these events have occurred since the mid-1990s. We then disaggregate these findings by mothers’ educational attainment and race, estimating variation among mothers in the rate of transition into primary-earner status and whether the contribution of each precipitating event varies by race and class. So far, we have found that separation and divorce account for none of the growth in mothers’ financial provisioning, while the combination of ‘his earnings decline and her earnings increase’ in the same year is much more likely to result in her assuming financial provisioning than in the past. We also note that while Black mothers have seen no growth in financial provisioning (over 80% risk in both time periods), White, Hispanic, and Asian mothers have all seen risk grow over time.

2020. Marika Cabral. Sources of Variation in Private Insurance Coverage Among Medicare Beneficiaries: Decomposing Place-Based and Person-Based Factors.