Adina Zeki Al Hazzouri

Social Epidemiologist and Assistant Professor at Columbia University Mailman School of Public Health

Recent Publications 

Research Program

Critical Determinants of Cognitive Aging and Dementia for Minority and Migrant Populations

I have extensively evaluated life course socioeconomic determinants of cognitive aging and dementia among minority populations, with a focus on minority populations. I have evaluated the role of a childhood socioeconomic construct, determined by variables such as sibling mortality and parental education and occupation, along with adult and late-life socioeconomic constructs on cognitive function and dementia incidence. I also evaluated how socioeconomic mobility across the life course (upward or downward mobility) shaped these older age outcomes

Poverty, Financial Security and Cognitive Aging

Owing to persistent socioeconomic inequalities many working age Americans will enter retirement in worse financial standing than in prior decades. While financial well-being and security is acknowledged as a social determinant of health, rarely are its potential far-reaching effects on cognitive and aging-related outcomes like ADRD mentioned or considered. My team and I have been leading work addressing the role of poverty and financial hardship in cognitive aging and other related outcomes.

Causal Inference Methods in Aging and Health Disparities

I have been leading innovative  causal inference work to better understand biases and methodological challenges in observational studies of cognitive aging. I have investigated analytical approaches used within the same population–whether prevalent user design or new user design, with conventional regression adjustment, inverse probability weighting or propensity score matching. I have additional work validating Regression Discontinuity Design (RDD) models for treatment effects (now extended to ADRD outcomes). I contribute to work addressing methodological challenges of observational data, including work using honest causal forests and other machine learning methods for heterogeneous treatment effects

Aging Research in Low- to Middle- Income Countries

I began my career with work at AUB focused on the functional health among older adults in Lebanon and have had continued collaborations there including a current R01. I currently lead efforts to strengthen research capacity building in Lebanon and the region—bolstering outreach in current collaborative networks and increasing training and opportunities—for the next cohort of health researchers to address uniquely present factors of LMICs beneficial to learning about ADRD risk and disparities, as well as how to conduct successful research in low-resourced settings. 

Current Funded Projects

  • This study will combine eight data sources on lifecourse social and vascular risk factors for Alzheimer’s Disease and Related Dementias (ADRD) to create a large (N=304,171) and diverse (25% Black) synthetic birth cohort (age 0 to 90) for research on ADRD. We will apply rigorous methods to estimate and incorporate corrections for selection forces, including survival, enrollment, attrition, and reverse causation from incipient dementia to risk factor changes. The synthetic cohort will then be used to rigorously estimate how lifecourse social and vascular factors and hypothetical age-specific interventions on those risk factors would affect ADRD risk and ADRD disparities.

  • To build a prospective cohort that will enable us to evaluate the relationships of late life learning (LLL) with cognitive function, Alzheimer’s Disease and Related Dementias (ADRD), and other related outcomes in older age, whether these relationships are mediated by educational and social engagement, and whether LLL would modify the influence of earlier and lifecourse psychosocial exposures.

  • the goal of this grant is to use electronic health record data to create a and diverse sample for evaluating the relationship between statin medications and Alzheimer’s disease and Alzheimer’s related dementias (AD/ADRD) risk using a quasi-experimental approach and machine learning methods.

  • To rigorously examine the relationships between hypertension, its treatment, and ADRD risk among U.S. veterans – an increasingly racially/ethnically and geographically diverse population.

  • In this study, we leverage our previously created pooled binational life course dataset to estimate the association of migration and ADRD risk using the newly improved HCAP data, assess how the socio-political context influences his relationship, and examine underlying biosocialpathways