Heterogeneous Health Trajectories and Healthcare Utilization and Expenditure Among Low-Income Older Adults

Purpose & Background: The relationship between health and healthcare utilization and expenditure has been extensively explored in aging research over recent decades. However, understanding how distinct health trajectories are associated with various types of healthcare service use and spending remains limited, partly due to the predominant use of cross-sectional approaches in previous studies. Additionally, the analytical capabilities of past research have been constrained in several ways: (1) previous research has rarely explored the multidimensionality of health (e.g., morbidities, cognitive functioning, depression); (2) analytical techniques sensitive to intraindividual changes in health among older adults are still underutilized; and (3) few studies have investigated the unobserved heterogeneity of health trajectories among older adults. Considering these gaps and acknowledging the impact of social determinants of health, this study aims to explore the heterogeneity of multidimensional health trajectories in older adults. This is crucial as older adults may display significant variation in the timing, duration, and severity of health domain declines, along with the rate of change in their health over time. Subsequently, we aim to investigate the associations of these trajectories with healthcare utilization and expenditure.

Methods: Secondary data analysis was conducted from twelve waves of the National Health and Aging Trends Study, which is a nationally representative panel of Medicare beneficiaries aged 65 and older in the United States. The study participants were individuals: (1) aged 65 or older who responded to the surveys in more than six waves; and (2) with incomes lower than 400% of the federal poverty level. Multi-channel sequence analysis was used to simultaneously consider the status of six health domains (cardiometabolic, somatic conditions, cognitive, mobility, depressive symptoms, self-care) and classify respondents with similar histories. Then, a regression model was used to examine the association between the health trajectories and healthcare service use/expenditure (doctor visits, inpatient care, outpatient surgery, medication).

Results: Six distinct clusters were identified: (1) Severe Cardiometabolic & Decreasing mobility; (2) Moderate Cardiometabolic; (3) Moderate Somatic, Steep decline in mobility; (4) Severe Somatic; (5) Severe chronic, Depressed, & Decline in mobility & Self-Care; and (6) Severe chronic, Depressed, Decline in cognitive, mobility and Self-care. Clusters 5 and 6 were characterized by low socioeconomic status and showed a higher tendency for using inpatient care and having medical bills that are being paid off over time compared to the other clusters.

Conclusions and Implications: Policy implications are crucial for older adults in clusters 5 and 6, who encounter socio-demographic challenges and multiple health declines, resulting in heightened healthcare needs and costs. Prioritizing investments in primary preventive care services could effectively deter these health declines. Additionally, there is a need to explore strategies aimed at enhancing healthcare coordination and support services. It is also essential to analyze the factors contributing to higher utilization rates in specific clusters and identify potential interventions to optimize healthcare utilization and reduce costs.