Pilot Projects
Through our pilot projects program we aim to facilitate the development of research in the demography and economics of aging, focusing on our themes of aging in diverse contexts, health and social/economic productivity, and measurement and methods. We provide post-award support for the development of competitive external grant proposals.
Pilots funded in Year One
Effects of health on economic well-being of older adults in the context of rapid modernization.
PIs: Linda Adair, Penny Gordon-Larsen, Jeremy Moulton (ESI), Annie Green-Howard (ESI) Elizabeth Frankenberg
The relationship between health and economic status is bi-directional. While many studies document poor health outcomes related to poverty, less attention has been paid to how health affects economic outcomes. Physical, cognitive, and mental health can have profound effects on the economic productivity and security of older adults, with consequences not only for them but also for the next generation who may benefit from their financial support or be burdened by the time and costs of caregiving and support. Health conditions which limit physical mobility and stamina may be of particular importance for productivity in occupations that require physical strength or endurance while cognitive deficits and poor mental health may be more likely to impact occupations which require cognitive skills. These limitations may develop over long periods of time. However, few longitudinal studies in low- and middle-income country (LMIC) contexts provide the long-term data that allow study of the evolution of poor health and its impact on economic outcomes across the life course. Even fewer studies have the data needed to document intergenerational effects. We plan to conduct interdisciplinary, comparative research using data from extremely rich cohort studies spanning several decades in China, Indonesia and the Philippines. The principal investigators represent ESI and senior CPC fellows (Adair, Gordon-Larsen, Frankenberg) from different disciplines and who have collaborated on other projects. The overall goal of this project is to enhance understanding of how economic productivity and security in older adults is influenced by trajectories of physical, mental, and cognitive health beginning earlier in life, and in turn, how the health and economic well-being of older adults affects the next generation.
The cohort studies for this project include: The China Health and Nutrition Survey (CHNS) is a multipurpose panel survey following all members of more than 41,000 households in 360 communities throughout China across 10 exams from 1990 to 2019. The 2015 survey includes 19,525 individuals of whom 5,173 were age 60+. The Cebu Longitudinal Health and Nutrition Study (CLHNS) recruited pregnant women from urban and rural communities of Metro Cebu in 1983-4 and followed them and an index child during bimonthly visits for the first 2 years, then in 8 subsequent surveys through 2016, when the women ranged in age from 45-78 (n=1659, 596 age 60+). The Indonesia Study of the Tsunami, Aftermath and Recovery (STAR) includes a representative population from Aceh, Sumatra, first recruited in 2004, with 6 follow-up visits through 2017. The most recent survey includes 9,707 individuals age 40 and above, of whom 2481 were age 60 and older. The three studies span time periods of rapid modernization and concomitant dramatic health and economic changes. Of note in all settings is the transition in types of work and labor markets from more to less physically-demanding jobs benefitting from modern technology, and changes in health profiles with the emergence of obesity and chronic diseases. Comparative research across these settings can take advantage of the high diversity, pace of development and level of disparities within and between countries. The aims include: 1) Harmonization and preparation of data for comparative analysis, by harnessing substantial similarities in the questionnaires and data collected in the surveys. This harmonization work will be guided by principles established by Fortier et al in their work with multiple British and European cohorts(https://academic.oup.com/ije/article/40/5/1314/658348) ; 2) Estimate the pathways through which disparities in middle adulthood nutritional status represented as underweight, overweight, and obesity affect subsequent development and patterns of cardiometabolic health, impairment in physical and cognitive functioning, and in turn, changes in type of work, work hours, and earnings. We will use longitudinal marginal structural models to address likely bidirectional associations and take advantage of repeated surveys to focus on predictors of changes in economic outcomes subsequent to changes in health status.
Social Determinants and Adverse Health Outcomes in Older Adults: The role of inappropriate medication use.
PIs: Jennifer L. Lund (ESI) and Yang Claire Yang.
Adverse drug events are costly, accounting for approximately $3.5 billion in spending in the United States (US) each year, with a large majority of these occurring among elderly individuals. Medications whose risks of harm exceed their expected clinical benefits are termed potentially inappropriate medications (PIMs). PIM use is common among older individuals, reaching a one-year prevalence of >30% among Medicare beneficiaries, and is associated with an increased risk of adverse health outcomes of paramount importance to older adults, including falls, functional and cognitive impairment, hospitalization, and mortality. Therefore, PIM use is a major public health concern and reducing PIM use will have important implications for improving quality of life and healthy life expectancy in the growing older adult population in the United States. While social disparities in health, particularly those by race/ethnicity, geography, and socioeconomic status (SES), are well documented for individuals across the life course, there is little mechanistic research describing how disparities in PIM use could potentially impact persistent and even growing health disparities in late life. It is likely that there are disparities in PIM because of the costs of medications for socioeconomically disadvantaged older adults. Our proposed study will document disparities in PIM by race/ethnicity, geography, and SES, and disentangle associations between longitudinal PIM use and important adverse health outcomes. To address our specific aims, we will use a novel linkage between the National Health and Aging Trends Study (NHATS), a nationally-representative cohort of older adults, and Medicare enrollment and claims data. Medicare data (2012-2017) will be used to measure: (1) monthly PIM dispensing (using multiple validated tools), (2) hospitalizations and ED visits during follow-up, and (3) vital status through 2017. Annual NHATS interview data (2013-2017) will be used to define: (1) detailed covariates (e.g., baseline frailty, race/ethnicity, SES measures) and (2) falls and changes in physical (gait speed, balance, lower extremity and grip strength, and peak air flow) and cognitive (memory, orientation, executive domains) function over time. Annual PIM prevalence rates will be estimated by county and trajectories will be identified using group-based trajectory modeling. We will evaluate associations between the identified trajectories and healthcare utilization and physical and cognitive function outcomes using Kaplan-Meier methods and Poisson regression to estimate one-year risk differences, risk ratios, and incidence rate ratios, accounting for measured confounding using propensity score weighting. We will test for racial/ethnic, geographic, and SES differences in PIM use as well as their interactions in models of associations between PIM use and health outcomes. Our results will help to identify high-risk medication use in socially heterogeneous older adult populations and inform the development of future interventions focused on describing harmful and inappropriate medications in disadvantaged adults most in need. We hypothesize that there are substantial social disparities in trajectories of PIM use for black individuals and those in rural areas or with low SES; and these trajectories will be differentially associated with adverse health outcomes in heterogeneous ways across groups. Our results will inform the development of interventions to promote safe and effective medication use, ultimately leading to reductions in health disparities related to adverse PIM in late life.
Life Course Cognition: Multimodal Measurement of Cognitive Functioning among Diverse Midlife Americans.
PIs: Robert Hummer, Taylor Hargrove (ESI), Allison Aiello
Understanding cognitive functioning is a critical national priority as the United States prepares for the rapid aging of its population. Recent scholarship makes it clear, though, that population-based study of cognitive functioning must begin long before old age; focused attention on midlife (i.e., ages 40-49) is essential. The National Longitudinal Study of Adolescent to Adult Health (Add Health) is poised to make a critical contribution to the understanding of midlife cognitive functioning in the United States. Our research team is in the process of developing the application to secure funding for Wave VI of Add Health, which will include in-depth attention to cognitive health in midlife. Add Health is a longitudinal study of a nationally representative sample of Americans that began following 20,000 adolescents in grades 7-12 during the 1994-95 school year. Add Health respondents are now reaching middle-age; as currently scheduled, the average age of Wave VI respondents will be 44. Add Health is thus exceptionally unique in having such a rich, multi-level, longitudinal array of data among a large nationally representative and diverse cohort of Americans as they approach midlife. These data also set the stage for rigorously studying risk and protective factors that occur earlier in the life course prior to the age-related conditions that will begin to affect this cohort, including cognitive decline and—in the future—dementia. This pilot project will assist in our preparation for cognitive data collection in Wave VI. The overall goal of this pilot is to develop, test, and evaluate potential measures of cognitive functioning among a diverse convenience sample of midlife subjects. We will work with our Add Health field contractor (RTI, International), to draw a diverse (by race/ethnicity, gender, and educational attainment) sample of North Carolina residents between the ages of 40-49. Because the plans for Wave VI of Add Health include both in-person and web-based cognitive assessments, this pilot project will develop, test, and evaluate potential measures of cognition in both the in-home and web-based data collection modes. The pilot project will address the following aims: 1) Using available tools (e.g., NIH Toolbox, Harmonized Cognitive Aging Project, Montreal Cognitive Assessment, Mini-Mental State Examination, PhenX Toolkit, and more), develop a range of in-person and web-based cognitive assessments to pilot among our convenience sample. 2) Collect pilot data on a range of cognitive functioning batteries from our convenience sample, using both the in-home and web-based modes of data collection. We will aim to collect data from as many subjects as our budget allows, with the sample balanced according to race/ethnicity (Black, Hispanic, Asian, White), gender (women, men), and educational attainment (high school degree or less, some college, college degree or more). 3) Compare the in-person with the web-based assessments and test for differences across the two modes. We will make these comparisons across both the full sample, as well as within subgroups defined by race/ethnicity, gender, and educational attainment. This pilot project stands to provide innovative and useful information for measuring cognition in middle age in large scale demographic and population-based studies, including Add Health.