Longitudinal Poverty Sequences and Cognitive Decline in Later Life

Background and Purpose
Poverty in later life is increasingly recognized as a key determinant of cognitive health, yet much of the existing aging research treats poverty as a static condition. In reality, poverty in old age is often dynamic, with individuals entering and exiting poverty over time. Guided by life course epidemiology and the theory of cumulative disadvantage, we examine how long-term poverty trajectories influence cognitive functioning among older adults in the United States. We use data from the National Health and Aging Trends Study (NHATS), a nationally representative panel survey of Medicare beneficiaries aged 65 and older, to assess whether the timing, duration, and sequencing of poverty exposure are associated with changes in cognitive performance.

Methods
We analyzed NHATS data from 2011 to 2022, focusing on respondents aged 65 and older with complete annual income records across all 12 waves. To capture the dynamic and multidimensional nature of economic hardship in later life, we employed multistate sequence analysis using 12-year income trajectories from the National Health and Aging Trends Study (NHATS). Annual poverty status was coded into four levels—above poverty, near poverty, moderate poverty, and deep poverty—based on household income relative to the federal poverty line. Using optimal matching and cluster analysis, we identified distinct poverty trajectory groups that reflect not only the timing and duration of poverty but also its depth. Cognitive functioning was assessed each year using NHATS composite scores for memory and executive functioning. We employed random coefficient regression within a multilevel growth curve framework to assess within-person cognitive change over time and to compare rates of decline across poverty trajectory groups, while controlling for sociodemographic and health-related covariates at baseline.

Results
Preliminary findings from multistate sequence analysis revealed six distinct poverty trajectories over a 12-year period: (1) persistent deep poverty, (2) fluctuating poverty with periods of deep hardship, (3) late-entry moderate poverty, (4) early-exit poverty, (5) near-poverty throughout, and (6) consistently above the poverty line. Individuals in the persistent deep poverty group experienced the most pronounced cognitive decline, particularly in memory and executive functioning. Those with fluctuating or late-entry poverty patterns also exhibited accelerated decline, though less severe than the persistent group. In contrast, individuals who remained consistently above the poverty line showed the most stable cognitive performance across time.

Conclusions and Implications
These results underscore the importance of incorporating poverty depth—not just timing and duration—into life course models of cognitive aging. A nuanced, trajectory-based approach provides richer insight into how sustained or severe economic disadvantage compounds risk and may better inform policies and interventions aimed at reducing later-life cognitive disparities.