Center for Aging and Policy Studies (CAPS)

2023. Kelly Musick. Family Complexity and Economic Resources Within and Across Households.

Increases in childbearing outside of marriage, divorce, repartnering, and childbearing across partnerships have increased the complexity and diversity of families (McLanahan, 2004). Changes in family composition and living arrangements raise challenges for measuring family relationships that our study designs are often ill-equipped to address. This is critically important because families provide a vital safety net for their members across the life course, from childhood through older ages. In Judith Seltzer’s recent address to the Population Association of America (2019), she argued that to better understand how families meet the obligations defined by kinship, demographers need to do a better job identifying family ties that extend beyond the household. Our project answers this call.
Evidence to date shows that family complexity is common, particularly among racialized and less advantaged groups (Amorim & Tach, 2019; Cancian et al., 2011; Carlson & Furstenberg, 2006). Family complexity is associated with reduced resources for children (Brown et al., 2015; Fomby et al., 2016; Halpern-Meekin & Tach, 2008; Harknett & Knab, 2007) and weaker relationships between parents and their adult children (Cherlin & Seltzer, 2014; Furstenberg, 2014). Emerging
work in this area has shed new light on the prevalence and potential implications of family complexity for children and older adults, but our current accounting is incomplete. Prior research has focused largely on the presence and composition of parents in the household (e.g., Kennedy and Bumpass 2008; Raley and Wildsmith 2004), limiting our understanding of family ties over the life course and missing key components of kinship that make up the private social safety net. We address this gap by leveraging the genealogical design of the Panel Study of Income
Dynamics (PSID, 1985-2017). Traditional household-based surveys fail to capture relationships and resource flows between children and nonresident parents and siblings, and administrative data have limited measures of family and household characteristics. The PSID is uniquely suited to advancing our understanding of family complexity and its links to resource flows within and across households, and how these resource flows change within and across cohorts. Our project
follows a conceptual framework in which family complexity has implications for economic resources, which in turn shape the support and well-being of family members from childhood through the older adult years. It advances CAPS research on family and intergenerational supports and reflects the cross-cutting theme of distinctive circumstances of specific populations, as defined by race and social disadvantage. We use multiple waves of data from the PSID 1985-2017. The unique design of the PSID follows “gened” family members even when they leave the original household. This makes it possible to collect detailed information on nonresident parents that is typically not available in household surveys. In particular, we have information on the marriage and fertility histories of these
nonresident parents; income sources from labor, kin, public, and other sources for all members of their households; and the full set of biological, adoptive, half, and step siblings residing in their households. We follow changes in residential and nonresidential family composition from the perspective of children, as their parents’ relationships evolve and new children are brought into the family. We address four specific aims:

Aims 1 and 2: We will identify complex parent and sibling relationships within and beyond the household, and we will estimate differences in their prevalence as children age, across cohorts, and by race and socioeconomic status. To do so, we will rely on household roster data collected at each wave, which includes move-in and move-out dates for all new and departed household members; comprehensive marriage and birth histories collected from all sample members ages 15 years and older since 1985; and PSID’s family composition file (early release, 1968-2017) with identifiers for biological, adoptive, half, and step sibling relationships.

Aims 3 and 4: We will examine the income sources of households linked by nonresidential parent and sibling relationships, and we will estimate differences as children age, across cohorts, and by race and socioeconomic status in the resources potentially available to children in complex families within and beyond the household. For a subset of children (N=3,065), we have a complete
fertility history for both parents, thus enabling accurate estimates of half and step siblings on the mother’s and father’s side. We will use this subset of children as a starting sample for our analysis of resources within and across households. We will merge information on family composition to information on household income sources at multiple waves from the core PSID family files. Our project makes several innovative contributions to the literature: First, we rely on the unique
genealogical design of the PSID that follows family members as they leave the household, making it possible to map changes in both residential and nonresidential sibship composition prospectively as children age. Second, we examine change across a broader span of cohorts, including more recent cohorts of children (born 1985-2004) than those included in the 1997 NLSY (born 1980-1984). Third, we provide among the first estimates based on a national sample of
differences in patterns of change by parents’ union status, education, and race/ethnicity. The inclusion of a low-income oversample in PSID’s original design enables us to evaluate the extent to which social class differences in sibship composition emerged within and between cohorts against the backdrop of rising social inequality. Finally, we provide the first estimates of detailed income sources of households linked by nonresidential parent and sibling relationships, and our
accounting differentiates by cohort, child age, race/ethnicity, and socioeconomic status. The descriptive work proposed here will inform a larger project proposal to the National Institute on Aging to investigate associations between family complexity, economic resources, and multiple dimensions of parent and child well-being over the life course. We plan to link our detailed file on childhood family composition and economic resources to data from the Child Development
Supplement, the Transition to Adulthood Supplement, and the Rosters and Transfers module. These links will allow us to map childhood experiences onto indicators of child well-being and relationships between parents and their adult children. For the parent generation, this includes transfers of time and money with their adult children, as well as help given to children for school, a home purchase, or other financial help. This extension directly addresses critical questions
about family complexity and the functioning of the family as a safety net for their members across the life course. This project is a collaboration between CAPS affiliate Kelly Musick at Cornell University and Paula Fomby at the University of Pennsylvania. The funding will support a Research Assistantship for an emerging aging scholar with interests in family relationships and well-being over the life course. Musick and Fomby have a long record of collaboration on topics related to family investments, child well-being, and family complexity, including a recently completed R01 on the emergence and social patterning of family complexity over historical time. Fomby led PSID data collection and training initiatives for a decade and is an expert in all aspects of PSID design and content required to achieve the study aims.

Priority Research Areas: Health Trends and Disparities, Determinants of Health, Well-Being and Longevity

2023. Marc Garcia. U.S. State Policy Contexts and Disability Risk Among Older Latino Adults in the United States.

The main objective of this project is to generate insights into the role of U.S. state policy contexts on disability risks among Latino adults (as a pan-ethnic category) ages 65 and older, and among subgroups of Latino adults (i.e., by nativity status, country of origin, and gender). The project will also provide insights into which domains of state policy (e.g., labor, environment, immigration) appear to be particularly important for disability risk among Latino adults. The findings will provide new evidence on the potential role of state policy contexts on disability
among the growing and increasingly diverse population of Latino adults in the United States. This project contributes to CAPS’ signature theme of health and well-being and its cross-cutting themes of place and specific populations. It has two specific aims:

Aim 1: Estimate the associations between U.S. states’ policy contexts and disability risk among older Latino adults and subgroups of Latino adults. The analyses will focus on two measures of disability. One measure assesses difficulty with activities of daily living such as bathing and eating (ADL) or instrumental activities of daily living such as shopping and running errands (IADL). Another measure assesses cognitive difficulty. State policy contexts will be measured with a single, well-established summary score of over 120 state-level policies [5].

Aim 2: Examine how 16 specific state policy domains, such as labor and immigration, are associated with disability risk among Latino adults in midlife. This aim will analyze the same measures of disability as in Aim 1. The 16 policy domains are components of the overall state policy measure used in Aim 1.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Determinants of Health, Well-Being and Longevity

2023. Pinka Chatterji. Consolidation in the Medicare Part D market and racial disparities in access to prescription drugs.

The Medicare prescription drug program (Part D) was launched in 2006. Since then, the Part D market has experienced significant consolidation, driven by mergers and acquisitions (M&A) and regulatory changes [1]. The overarching goal of this project is to examine how this industry consolidation has affected Part D utilization and out-of-pocket spending, and racial disparities in these outcomes, through reducing plan variety and competition.

Aim 1:  Document geographic variation in Medicare Part D market characteristics from 2006-2021 using maps and descriptive tables. We will examine Part D market characteristics across geography and time, focusing on Centers for Medicare and Medicaid Services (CMS) markets with the largest shares of minority racial/ethnic elderly populations. This approach will allow us to gauge descriptively whether trends in market consolidation are different in geographic areas with vs. without high minority elderly populations.

Aim 2:  Formally test whether Part D market consolidation affects market-level plan characteristics (e.g., premiums, quality), as well as racial disparities in prescription drug use and out-of-pocket spending on drugs among elderly individuals.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Disability, Health Care and Long-Term Care

2022. Kate Strully . Exclusionary and inclusionary residential zoning laws and health disparities among metropolitan older adults.

Aim 1: Draws on data from the 2020 Medicare Disparities Database to conduct an ecological-level analysis of how DZ and IZ policies are associated with health disparities in populations 65+ in the 50 most populous core-based statistical areas (CBSAs) in the U.S., and proposes to test for endogeneity in DZ policies using an instrumental variable strategy.
Aim 2: Uses data on respondents 65+ from the 2020 Behavioral Risk Factor Surveillance Study to conduct a multi-level analysis at the individual and CBSA level of how DZ and IZ policies are associated with health outcomes across racial-ethnic and income groups.
Aim 3: Involves a review and preliminary analysis of IZ programs within large CBSAs to identify significant policy changes and comparison locations in order to design a longitudinal study of effects of changes in IZ policies on income and racial-ethnic health disparities for individuals 65+.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Effects of Policy and Place on Health

2022. Maria Brown . Biological Aging and Functional Limitations: Moderation by Discrimination.

Aim 1: Examine how accelerated biological aging, as measured by epigenetic clocks, is associated with functional limitations in older adults.
Aim 2: Evaluate how the relationship between epigenetic clocks and functional limitations is impacted by perceived discrimination in later life among Black, Hispanic, and White older adults.

Priority Research Areas: Biology, Genetics and Demography of Aging, Health and Aging of Specific Populations

2022. Samantha Friedman. Redlining Policy and Later-Life Mortality: An Exploratory Analysis of the Linked HRS-1940 Census Data.

Aim 1: To link the HRS-1940 data at the ED level to the redlining data from the digitized HOLC maps. Mapping the HOLC information onto the EDs is an arduous process because a complete set of ED shapefiles is not readily available. Once ED shapefiles are created, we will overlay the HOLC maps and create a redlining score for each neighborhood. Then, we will link these data to the HRS public use data and code the project variables.
Aim 2: Using the harmonized dataset, we will estimate a hazard model of all-cause mortality as a function of early-life-neighborhood redlined status, control variables, and mediating variables for the full sample and by race to address the following research questions: a) Does residence in a redlined neighborhood in early life contribute directly to mortality in later life, controlling for other factors? b) Is the impact of residence in a redlined neighborhood in early life on mortality in later life mediated by mid-to-late life income, wealth, and education? and c) Does the impact of redlining in early life differ by race?

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Effects of Policy and Place on Health

2022. Vida Maralani . Marital Sorting and Health Inequalities in Later Life.

Aim 1: To investigate how couples’ joint education- smoking portfolios established at the time of marriage inform their functional limitations at older ages.
Aim 2: To examine differences in smoking and exposure to second-hand smoke across the marital years as potential mechanisms linking assortative mating to inequalities in functional limitations at older ages.
Aim 3: To examine whether the association between assortative mating and health at older ages has changed across birth cohorts. This helps to illuminate one potential family-level mechanism underlying increasing educational disparities in health in the U.S.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Health Trends and Disparities

2021. Chun-Yo Ho. Elderly Health in Rural America: Impact of Hospital Closure.

Specific Aims:
Rural areas in the U.S. have been experiencing declines in access to hospital care. Since 2005, there has been 176 rural hospital closure, with considerable variation across states (North Carolina Rural Health Research Program, 2020). More than a fifth of all current rural hospitals are at high risk of closing, in which about two-third of them are considered essential to their community (Navigant 2019). Such trend echoes the access to health care is the top health priority among rural population (Bolin et al. 2015).

Aim 1: To estimate the effect of losing access to hospital care in rural U.S. counties on elderly health outcomes in response to time-sensitive conditions, and to test for disparities by race/ethnicity and by education, using Vital Statistics death certificate records from 2005-2018.

Aim 2: To estimate the magnitudes of policy-relevant factors affecting rural elder’ health outcomes directly as well as indirectly through loss of access to hospital care; these policy-relevant factors include the state’s ACA Medicaid expansion status, hospital characteristics, and local market characteristics.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Disability, Health Care and Long-Term Care

2021. Colleen Heflin. Increasing access to SNAP for older adults through the Standard Medical Deduction.

Specific Aims: One of the goals of Healthy People 2020 is to identify and address the social and physical environments that promote good health for all. In 2017, food insecurity was present for 7.7 percent of persons aged 60 and older, or 5.5 million persons (Ziliak & Gundersen, 2019). The consequences of elderly persons’ food insecurity are far-reaching, affecting their health and well-being (Gundersen,
Kreider, and Pepper 2011). The Supplemental Nutritional Assistance Program (SNAP) is the largest food and nutrition assistance program in the United States and has been shown to reduce food insecurity. States have the option to adopt the Standard Medical Deduction (SMD), a provision designed to make it easier
for older adults to deduct medical expenses from income during the eligibility process which should increases the number of older adults eligible for SNAP as well as the benefit amount for which they qualify. Despite the potential importance of this provision, currently only 21 states adopt the SMD and there are no formal evaluations of its effectiveness.

Priority Research Areas: 12. Effects of social, economic, and healthcare policy on health and aging

2021. Jun Li . Effect of Medicaid expansion on low-income, aging adults’ use of informal care.

Specific Aims:
More than 44 million Americans receive help from informal caregivers at a societal cost of $522 billion annually, which reflects a health care system that provides limited support for formal long-term services and supports. The Affordable Care Act gave states the opportunity to expand Medicaid, which covers medical care and formal long-term services and supports for low-income and disabled adults under 65, with expansion associated with greater use of health care including increased use of formal long-term services and supports. Greater use of formal medical care (including long-term services and supports) has been observed to substitute for or complement informal care depending on the health of the population and availability of family
and intergenerational supports. Thus, Medicaid expansion could be critical for how low-income, aging adults use informal care.

Aim 1: Examine the association between Medicaid expansion and informal care use among low-income adults. We will examine changes in the total intensity of care (number of weekly hours) received, the type of informal caregiving including activities of daily living (e.g., dressing, eating, bathing, and toileting) and instrumental activities of daily living (e.g., meal preparation, grocery shopping, making phone calls), and the composition of informal care (number of informal caregivers providing care). We hypothesize that expansion will be associated with reduced informal care and changes in the composition of informal care.

Aim 2: Estimate the association of Medicaid expansion with unpaid, informal care costs for low-income adults. We will assess the effects of Medicaid expansion on the monthly costs of informal care borne by care recipients and their caregivers, focusing on the indirect costs of informal care computed using the market price of hourly wages of home health care and personal care aides. We hypothesize that costs of informal care will decrease after expansion.

Aim 3: Examine whether the relationship between Medicaid expansion and informal care use and costs varied by cognitive impairment, heavy informal care use, and low-income status. We will assess whether expansion had a differential effect on individuals who are more disadvantaged and likely to benefit from changes in access to formal long-term services and supports. Specifically, we examine those with cognitive impairment, those with informal care use above the median amount, and individuals in the bottom quartile of the Medicaid expansion-eligible income distribution. We hypothesize that decreases in informal care use and costs will be greater for vulnerable individuals after expansion.

Priority Research Areas: 12. Effects of social, economic, and healthcare policy on health and aging, Disability, Health Care and Long-Term Care

2021. Adriana Reyes. Extension of Health Shocks and Changes in Spatial Proximity between Parents and Children.

Summary of existing project. Adult children provide significant amounts of caregiving to their parents, which is facilitated by relatively close geographic proximity. This project utilizes longitudinal data from the Health and Retirement Study (HRS) to understand how families mobilize in response to a health shock, such an ADRD diagnosis. Using data from 2004-2014, we estimate proximity over time between parents and their adult children using parent’s Census tract and children’s zip codes. Analyses will leverage longitudinal data to assess the time-ordering of health shocks and changes in geographic proximity to children using fixed effects models.

Rationale for broadening existing project. Alzheimer’s disease and related dementias (ADRD) are among the costliest conditions faced by individuals and society [1]. Average out of pocket costs for someone with ADRD is over $10,000, four times as much as for someone without ADRD [2]. Older adults with ADRD are at a greater risk of preventable hospitalizations than those without ADRD, and having higher quality family care may help cut down preventable hospitalizations [2, 3]. Out-of-pocket health care payments for ADRD consumed 32% of household wealth five years before death [4].  Early stage diagnosis of ADRD has been associated with declines in wealth, which may make it difficult to pay for late-stage treatment [5]. Despite the well documented financial toll of ADRD on the families of those living with ADRD, we know less about how family proximity and care may moderate this relationship.

Specific Aims. This project asks whether changes in parent-child proximity in response to ADRD is associated with wealth. In particular, does the proximity of adult children reduce health care-related spending for adults with ADRD? The overall goal of this work is to understand how family proximity and mobilization are associated with wealth trajectories and health care spending by older adults with ADRD. Having family nearby may help reduce hospitalizations and other forms of health care spending that can enable older adults to maintain wealth.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Family and Intergenerational supports

2021. Ben Shaw . Living alone and risk for the development of Alzheimer’s Disease and related dementias: When does solitary living matter and why.

Background. Existing research shows that living alone in later life is associated with greater risk for the development of ADRD than physical inactivity, hypertension, diabetes, and obesity [53]. Nevertheless, many older adults who live alone are able to maintain high levels of functioning and will not develop ADRD. While some research has identified factors that moderate the mortality risk of living alone [54], more research is needed to understand the circumstances associated with solitary living in later life that are mostly likely to confer, or buffer, ADRD risk. Social isolation is widely regarded as a key explanatory factor [55], but other demographic, behavioral, social, and environmental factors may also play important roles, either independently or jointly with social isolation. In addition, the timing of living alone (e.g., age of onset; duration in living alone status) and the circumstances precipitating a transition to solitary living may also be important in determining risk for ADRD.

Aims. This study’s aims are: (1) to determine which specific circumstances defining one’s living alone status (i.e., age of transition to living alone; duration in living alone status; event leading to living alone transition) are associated with ADRD incidence among older adults; (2) to determine the key demographic, behavioral, social, and environmental factors related to living alone during later life that are most strongly associated with ADRD incidence among older adults; and (3) to determine what demographic, behavioral, social, and environmental factors buffer the risk of ADRD among older adults living alone.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Family and Intergenerational supports

2021. Emily Wiemers . Extension of The Consequences of Disabilities in Late Middle Age for Consumption Well-Being.

Summary of existing project. The existing project uses the rich consumption and health data in Panel Study of Income Dynamics [PSID] to examine the consequences of disability in late middle age for consumption well-being and to assess whether the decline in consumption at the time an individual becomes disabled is predicted by or mediated by reports of poor health in childhood. The supplement exploits the measures of cognition, consumption, labor supply, wealth, and early childhood health in the Health and Retirement Study [HRS] to answer the same set of questions focusing on the impact of the onset of cognitive limitations and dementia on consumption well-being and its correlation with poor health in early childhood.

Specific Aim. To use the HRS to examine the impact of cognitive limitation or dementia onset on consumption well-being paying particular attention to the role of spousal labor supply and wealth as sources of consumption smoothing, and to evaluate the role of poor health in childhood in either exacerbating or mediating dementia-related consumption declines.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Disability, Health Care and Long-Term Care

2021. Kevin Heffernan. Estimated Arterial Stiffness as a Predictor of Cognitive Decline and Dementia in Middle-Aged and Older Adults.

Background. With aging, large arteries outside of the brain lose elasticity. Elevated levels of arterial stiffness expose the brain to high blood pressure that causes structural damage [23]. Numerous cross-sectional studies now support a strong association between arterial stiffening and β-amyloid plaque deposition and atrophy of key areas of the brain implicated in the pathogenesis of cognitive aging and ADRD development [24-27]. Arterial stiffness also prospectively predicts cognitive decline [28-30] and transition to dementia in older adults [31-33]. As such, arterial stiffness may be a useful vascular biomarker to predict age-related cognitive decline and transition from mild cognitive impairment (MCI) to more severe dementia in the general population.

The gold standard method for assessing arterial stiffness, carotid-femoral pulse wave velocity (cfPWV), requires specialized, relatively expensive equipment and considerable technical proficiency, which limits its widespread adoption [34]. cfPWV can be estimated from two commonly measured clinical variables – age and blood pressure – and we have shown this estimated pulse wave velocity (ePWV) is an independent predictor of cardiovascular and cerebrovascular events in the general population [35,36].This pilot project will extend our ongoing work on vascular aging to consider the relationship between ePWV, cognitive decline, and dementia.

Specific Aims. As the NIA acknowledges that “MCI research is perceived by experts as Alzheimer’s Disease prevention research,” this pilot project will leverage the strong longitudinal study design of the HRS to address two specific aims and associated hypotheses: (1) examine ePWV as a predictor of cognitive decline in middle-aged and older adults. We hypothesize that higher arterial stiffness will predict a steeper age-related decline in cognitive function. (2) Examine ePWV as a predictor of the transition from MCI to dementia in middle-aged and older adults. We hypothesize higher arterial stiffness will be associated with a higher risk of transitioning from MCI to having dementia.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Health Trends and Disparities

2021. Scott Landes . Comparing Alzheimer’s Related Mortality Among People with and without Down Syndrome.

Background and Significance. Over the past 50 years, the age of death for people with Down syndrome in the Unites States has improved markedly, from an average age at death of 25 in 1983, to 49 in 1997, to 56 in 2017 [39,40]. The increase in longevity for this population is typically ascribed to improvements in early intervention efforts, as well as social supports and medical care for people that often coincided with a transition away from institutional and toward community-based supports [41,42]. With the advent of people with Down syndrome now living into and beyond their 40s, research has documented high rates of early onset Alzheimer’s disease among this population, with some researchers claiming that a majority of individual’s with Down syndrome exhibit neuropathological characteristics of Alzheimer’s disease by age 40 [43-47].

Three studies utilize US death certificate data to document that adults with Down syndrome have higher rates of death from Alzheimer’s disease and or dementia that adults in the general population [40,48,49], with the most recent reporting adults with Down syndrome were 25 times more likely to die from these diseases [49]. While informative, there are limitations to this research that need to be addressed. The first is that all three studies rely on analysis of the underlying cause of death, as opposed to conducting broader analysis of multiple cause of death. Thus, while they accurately identify that people with Down syndrome were more likely to have Alzheimer’s/dementia identified as their underlying cause of death, they do not address whether people with Down syndrome were more/less likely to have been diagnosed with Alzheimer’s disease by their time of death. This is an important distinction. As people with Alzheimer’s disease can die from various other diseases [50], focusing solely on the underlying cause of death leads to underestimates of the prevalence of this disease at time of death that may vary between people with/without Down syndrome. In comparison, utilizing multiple cause of death files, which provide information on comorbidities present at the time of death, would provide a more accurate accounting of the prevalence of Alzheimer’s disease among both populations at the time of death. In addition, analysis of multiple cause of death files would allow for examination of whether causes of death are similar or different among people with and without Down syndrome who have Alzheimer’s disease, an important research question that has not been addressed to date.

Specific Aims. This project has two aims: (1) to analyze differences in the prevalence of Alzheimer’s disease among people with and without Down syndrome at the time of death; and (2) to document causes of death among people with and without Down syndrome who had Alzheimer’s disease reported on their death certificate.

Priority Research Areas: Cognitive Aging and the Demography of Dementia, Health and aging of specific populations

2020. Adriana Reyes. Health Shocks and Changes in Spatial Proximity between Parents and Children.

Background: Over half of caregivers who provide substantial help to older adults are adult children [1]. However, family caregiving is facilitated by relatively close geographic proximity, especially because children provide an average of 77 hours of assistance a month [2]. Living near children enables a wide range of support, and therefore older adults may migrate to be closer to children in response to or in anticipation of health declines [3]. Previous research has found that when adult children move, they are more likely to choose locations close to their parents compared to other locations, especially when parents are very old [4]. Similarly, elderly adults’ migration patterns are motivated by a desire for close geographic proximity to their children [5]. Recent estimates suggest that a majority of parents and children live near each other, especially those with fewer socioeconomic resources [6]. Yet, the number of older adults without an adult child nearby is projected to increase dramatically over the next two decades [7]. As many rural areas experience an out-migration of young people and declining economic opportunities, the ability of older adults to rely on children for caregiving may be compromised in these areas. We know that children play a significant role in caring for their aging parents; however, we know less about how health shocks associated with Alzheimer’s disease and other conditions may alter the geographic location of children and parents. Building on this context, this project will pursue two specific aims:
Aim 1: Construct a panel dataset of spatial proximity between parents and their adult children. The data will document, at each wave, the distance between parents and their adult children and changes across waves.
Aim 2: Identify the correlates of changes in proximity and co-residence. The project will examine how changes in proximity evolve relative to changes in health, such as an Alzheimer’s diagnosis.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Family and Intergenerational Dynamics

2020. Emily Wiemers. The Consequences of Disabilities in Late Middle Age for Consumption Well-Being.

Background: This project will use the rich data in the Panel Study of Income Dynamics (PSID) to examine the consequences of disability in late middle age for economic well-being using consumption-based measures (e.g., food, housing, services) rather than income-based measures. It will exploit the long health histories in the PSID to examine whether declines in consumption at the time an adult becomes disabled are predicted by or mediated by reports of poor health earlier in life, particularly in childhood. The project will address two specific aims:
Aim 1: Examine the impact of the onset of work-limiting disabilities in late middle age on health, non-durable, and non-health service consumption, focusing on spousal labor supply and wealth as sources of smoothing consumption. This aim will recognize that, for married people, there is a tension between providing care to a disabled spouse and increasing labor supply to smooth consumption after the onset of a spouse’s disability.
Aim 2: Examine whether adverse childhood health conditions explain declines in consumption for those experiencing a work-limiting disability in late middle age. Individuals with poor health in childhood may save to insure their consumption against expected future adverse health shocks. Alternatively, they may be unable to self-insure because poor health in childhood limits earnings. Adverse childhood health conditions may also predict the severity of disabilities later in life, even in the absence of insurance effects.

Priority Research Areas: Family and Intergenerational Dynamics, Disability, Health Care and Long-Term Care

2020. Pinka Chatterji. 3. Medicare Part D and Disparities in Chronic Disease among the U.S. Elderly Population .

Background: Health insurance may be a key factor underlying disparities in chronic disease among the U.S. elderly population. Notably, a study that used data from 1999 to 2006 found that racial/ethnic disparities in control of blood pressure and diabetes become smaller after age 65, suggesting that Medicare reduces disparities in chronic disease [13]. Medicare introduced optional prescription drug coverage in 2006 (Part D). The introduction of Part D may have made Medicare an even more important factor in remediating disparities given the important role of prescription drugs in controlling chronic disease. Prior work suggests that there are racial/ethnic differences in medication adherence for chronic diseases [14, 15], and Part D has increased medication usage overall among the elderly population [16, 17]. To our knowledge, however, no studies have investigated the potentially strong effects of Part D on older adults’ awareness and control of chronic disease, or the effects of Part D on racial/ethnic and education-related disparities in chronic disease. The project will address two specific aims:
Aim 1: Test whether the introduction of Medicare Part D affected disparities in awareness and control of hypertension, diabetes, and hyperlipidemia using national data sets that include biomarkers.
Aim 2: Test whether Part D affected disparities in prescription drug use and out-of-pocket spending on drugs among elderly individuals with chronic diseases.

Priority Research Areas: Determinants of Health, Well-Being and Longevity, Effects of Policy and Place on Health

2020. Shannon Monnat. Demographic, Geographic, and Temporal Trends in Co-Occurring Use Disorders and Mortality from Opioids, Other Drugs, and Alcohol among Middle-Age and Older Adults in the United States.

Background: Opioid use disorders (OUD) and overdose death rates have grown to unprecedented levels [18, 19]. The U.S. has also experienced large increases in cocaine, methamphetamine, and non-opioid prescription (especially benzodiazepines) overdoses over the past decade. Deaths from alcohol use and use disorders (AUD) are also rising, especially among older adults, women, and low-SES individuals [20, 21]. These crises are costly, getting worse, and interdependent. Excessive alcohol use is common among individuals who misuse opioids [22], with more than half of individuals with OUD also meeting criteria for AUD [23]; more than half of opioid-related deaths involve alcohol or another drug [24]. Use disorders and overdose rates are also geographically patterned [25, 26]. Moreover, while drug and alcohol mortality rates have increased across many demographic groups over the past 20 years, increases in the middle-age cohort (35-64 years) have been startling. Thus, we focus on the relationship between drug- and alcohol- use disorders and between drug- and alcohol- related mortality among middle-aged and older adults. The specific aims of this two-year study are to:
Aim 1: Characterize demographic, geographic, and temporal trends (2000-2017) in the relationships between opioids, other drugs, and alcohol—in terms of use disorders and mortality—among middle-aged and older adults.
Aim 2: Identify typologies of counties based on use disorders and mortality from opioids, other drugs, and alcohol.
Aim 3: Compare the county groups found in Aim 2 by sociodemographic, economic, and other characteristics.

Priority Research Areas: Health Trends and Disparities, Effects of Policy and Place on Health