# Building an unbiased pooled cohort for the study of lifecourse social and vascular determinants of Alzheimer's Disease and Related Disorders

> **NIH NIH R01** · COLUMBIA UNIVERSITY HEALTH SCIENCES · 2022 · $785,592

## Abstract

Abstract
Critical social and vascular risk factors for Alzheimer’s disease and related dementias (ADRD) occur in
childhood, early adulthood, or midlife, decades before ADRD is typically diagnosed. Most cohorts dedicated to
the study of aging are initiated in mid to late life, and are therefore not ideal for evaluating the effects of early
life risk factors. Synthetic cohorts, which pool multiple data sources that in combination span early to late life,
provide an unparalleled opportunity to rigorously evaluate lifecourse mechanisms of ADRD. Lifecourse
research, especially when based on synthetic cohorts, faces several methodological challenges related to
survival, enrollment and attrition that are differential across the pooled studies, and reverse causation from
incipient dementia. The long-term goal of our research is to pinpoint how and when we can intervene to
prevent or delay the onset of ADRD. Yet, the differential selection forces in a synthetic cohort can make it
impossible to identify protective factors, can spuriously make harmful factors appear innocuous, and can
provide incorrect guidance on prevention priorities. In this study, we propose to pool eight data sources
comprising children, young, middle-aged, and older adults to create a SYNthetic Birth cohort for research on
ADRD (SynBAD), correcting for differential survival, enrollment or attrition, and reverse causation, allowing us
to rigorously evaluate the effects of lifecourse social and vascular risk factors. SynBAD will include the
Bogalusa Heart Study, the Muscatine study, the National Longitudinal Survey of Youth 1979, The National
Longitudinal Study of Adolescent to Adult Health, the Coronary Artery Risk in Development in Young Adults,
the Health and Retirement Study, the REasons for Geographic And Racial Disparities in Stroke, and the
National Health and Nutrition Examination Studies. SynBAD will be large (N=304,171) and exceptionally
diverse, facilitating research on the drivers of ADRD among women (56%) and Black individuals (25%).
Specifically, we propose to (Aim 1) create a diverse synthetic birth cohort (age 0 to 90) for the study of social
and vascular risk factors for ADRD, incorporating corrections for differential survival, enrollment, and attrition;
(Aim 2), evaluate and correct for reverse causation -- in which incipient dementia induces changes in risk
factors -- by using a reverse Mendelian Randomization approach based on identifying the age-specific effects
of a genetic risk score for ADRD on risk factors; (Aim 3), rigorously estimate the causal effects of social and
vascular factors on ADRD risk using the synthetic cohort corrected for selection and reverse causation biases;
and (Aim 4), quantify reduction in lifetime ADRD cases and ADRD racial disparities that could be achieved with
a variety of hypothetical interventions on social or vascular risk factors at different ages. Given the role of
biological sex with social and vascular risk factors and dementia ...

## Key facts

- **NIH application ID:** 10426258
- **Project number:** 5R01AG072681-02
- **Recipient organization:** COLUMBIA UNIVERSITY HEALTH SCIENCES
- **Principal Investigator:** Medellena Maria Glymour
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2022
- **Award amount:** $785,592
- **Award type:** 5
- **Project period:** 2021-06-15 → 2026-03-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10426258

## Citation

> US National Institutes of Health, RePORTER application 10426258, Building an unbiased pooled cohort for the study of lifecourse social and vascular determinants of Alzheimer's Disease and Related Disorders (5R01AG072681-02). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10426258. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*