Social determinants, multimorbidity, and patterns of end-of-life care in older adults dying from cancer.

TitleSocial determinants, multimorbidity, and patterns of end-of-life care in older adults dying from cancer.
Publication TypeJournal Article
Year of Publication2017
AuthorsKoroukian, SM, Schiltz, NK, Warner, DF, Given, CW, Schluchter, M, Owusu, C, Berger, NA
JournalJ Geriatr Oncol
Date Published2017 03
ISSN Number1879-4076
KeywordsAge Factors, Aged, Aged, 80 and over, Emergency Service, Hospital, Female, Health Surveys, Hospices, Hospital Mortality, Humans, Logistic Models, Male, multimorbidity, Neoplasms, Population Surveillance, Quality of Health Care, Risk Factors, Socioeconomic factors, Terminal Care

OBJECTIVE: Most prior studies on aggressive end-of-life care in older patients with cancer have accounted for social determinants of health (e.g., race, income, and education), but rarely for multimoribidity (MM). In this study, we examine the association between end-of-life care and each of the social determinants of health and MM, hypothesizing that higher MM is associated with less aggressive care.

METHODS: From the linked 1991-2008 Health and Retirement Study, Medicare data, and the National Death Index, we identified fee-for-service patients age ≥66years who died from cancer (n=835). MM was defined as the occurrence or co-occurrence of chronic conditions, functional limitations, and/or geriatric syndromes. Aggressive care was based on claims-derived measures of receipt of cancer-directed treatment in the last two weeks of life; admission to the hospital and/or emergency department (ED) within the last month; and in-hospital death. We also identified patients enrolled in hospice. In multivariable logistic regression models, we analyzed the associations of interest, adjusting for potential confounders.

RESULTS: While 61.2% of the patients enrolled in hospice, 24.6% underwent cancer-directed treatment; 55.1% were admitted to the hospital and/or ED; and 21.7% died in the hospital. We observed a U-shaped distribution between income and in-hospital death. Chronic conditions and geriatric syndromes were associated with some outcomes, but not with others.

CONCLUSIONS: To improve quality end-of-life care and curtail costs incurred by dying patients, relevant interventions need to account for social determinants of health and MM in a nuanced fashion.

User Guide Notes

Short TitleJournal of Geriatric Oncology
Alternate JournalJ Geriatr Oncol
Citation Key8750
PubMed ID28029586
PubMed Central IDPMC5373955
Grant ListR21 HS023113 / HS / AHRQ HHS / United States
KL2 TR000440 / TR / NCATS NIH HHS / United States
U48 DP005030 / DP / NCCDPHP CDC HHS / United States
P30 CA043703 / CA / NCI NIH HHS / United States
UL1 TR000439 / TR / NCATS NIH HHS / United States
R01 MD009699 / MD / NIMHD NIH HHS / United States