ABSTRACT
There are >6 million individuals with Alzheimer’s disease and related dementias (ADRD) in the US; this is
predicted to triple in the next 30 years. Research indicates that those with dementia have substantially greater
healthcare utilization. Yet, dementia represents only the extreme end of a chronic progressive disorder. Fully
characterizing patterns of healthcare utilization across the continuum of cognitive impairment is
critical to (i) planning resource use in our aging population, and (ii) identifying areas where care might
be modified/improved. In particular, better understanding care patterns in early stages of cognitive
impairment could be crucial since modifications may be easier for those with less disease burden.
Indeed, growing research suggests changes in healthcare utilization may begin even with mild cognitive
impairment (MCI). However, the limited studies generally lack either rigorous data on cognitive status, integrate
health services data from 10-15+ years in the past, or include very small samples. Thus, we propose to
evaluate the spectrum of cognitive health in relation to healthcare utilization, including ADRD drugs and end of
life (EOL) care, in 4 ongoing cohorts at the Rush Alzheimer’s Disease Center (RADC). All participants receive
annual, harmonized cognitive assessments and neurologic exams, with ~90% followup; this includes 2,317
older individuals with linked Medicare fee-for-service claims from 2010-19 (11,000PY of followup: 2000PY for
MCI, 1200PY for dementia). Further, 20% of participants are African American (AA), enabling initial work on
cognition and health services focused in AAs, who have high risk of MCI and dementia; this will also establish
a platform for future research as the RADC AA cohort continues to expand. Briefly, we will compare
participants diagnosed as no cognitive impairment (NCI), MCI, and dementia at each annual cohort evaluation,
considering differences in the subsequent year in their utilization of outpatient and inpatient services (Aim 1),
and ADRD drug prescription fills (Aim 2). Aim 2 will focus on examining participant and disease characteristics
that may be related to prescription fills. In Aim 3, we will compare EOL care across decedents diagnosed with
NCI, MCI or dementia. Finally, while cohort data are essential to validly studying the continuum of cognition,
cohorts often target specific participants; in Aim 4, we will enhance larger translation of cohort findings by
assessing any differences in health services, ADRD drugs, and EOL care between RADC participants versus
the general Medicare population. We will then apply statistical methods, as needed, to “mimic” our results in
larger groups of the general Medicare population. Impact: By comprehensively evaluating differences in
healthcare utilization across the continuum of cognition, we can begin to detect stages where care is accessed
differentially, including types of care which may be affected. This will both inform targeted interventions to
optimize health services, and improve future resource planning.