Cognitive Impairment, Multimorbidity, and end-of-life care in older cancer patients

Case Western Reserve University, Cleveland OH

Investigators

Linked publications, trials & patents

Abstract

PROJECT SUMMARY: Over 1.5 million individuals reside in U.S. nursing homes (NH); of those, nearly 150,000 have been or will be diagnosed with cancer. The treatment of cancer among NH residents is complicated by the fact that institutionalized older adults are also likely to suffer from cognitive impairment (COG-I) and complex multimorbidity (MM; defined as co-occurrence of chronic conditions, functional limitations, and/or geriatric syndromes). Thus, NH cancer patients are particularly susceptible to severe depletion of functional and physiologic reserves, and increased morbidity and mortality. In addition, racial/ethnic disparities across these factors add further complexity to cancer care decisions among NH residents. Despite these vulnerabilities, very little is known about NH cancer patients? patterns of cancer care, including incomplete diagnostic evaluation resulting in unstaged cancers, use of cancer-directed treatment, end-of-life (EOL) care, and associated costs. Indeed, there remains a critical need to characterize this patient population and variations in cancer care, especially by gradients of COG-I and MM complexity, and when cultural factors may favor aggressive EOL care. Key questions have yet to be addressed, including: Are unstaged or late-stage cancer in institutionalized older adults observed uniformly across patients with mild or no COG-I vs. those with moderate or severe COG-I? How do these patterns vary by MM complexity? Does EOL care vary by COG-I gradients; by prognosis for poor survival; by race/ethnicity? What are the costs associated with aggressive EOL care? We aim to investigate the above research questions using the unique, newly developed data resource linking Surveillance, Epidemiology and End Results (SEER), Medicare, and nursing home Minimum Data Set (MDS). To capture cancers with varying case-fatality rates, our study population will include patients with incident or prevalent leukemias, breast, colorectal, lung, prostate, or pancreatic cancer. Our specific aims are to: 1) characterize NH cancer patients by gradients of COG-I and complex MM phenotypes (specific combinations of conditions constituting MM) by patient age, sex, cancer incident status, cancer type, and prognosis, and particularly by race/ethnicity; 2) analyze patterns of unstaged cancer, cancer-directed care, and aggressive EOL care relative to gradients of COG-I, complex MM, patient demographics, disease characteristics, and prognosis; and 3) estimate the costs associated with aggressive EOL care in patients with moderate/severe COG-I, complex MM, and poor prognosis for survival < 6 months. By addressing the above research questions, we will be able to inform targeted interventions to ensure that dying NH cancer patients receive individualized and dignified care and reduce the costs associated with unnecessary care. Thus, our findings will have a significant impact upon clinical practice and research in geriatric oncology.