Symptom Burden and End-of-Life Treatment Preferences in the Very Old

J Pain Symptom Manage. 2016 Sep;52(3):404-11. doi: 10.1016/j.jpainsymman.2016.03.009. Epub 2016 Jun 3.

Abstract

Context: End-of-life (EOL) treatment preferences among the very old (age 85+) may differ from preferences in younger aged populations because of high levels of symptom burden and disability and high risk of mortality. It is unclear if symptom burden or level of disability is more important for such preferences.

Objectives: To investigate whether distress from daily symptom burden was an independent correlate of EOL treatment preferences over two years of follow-up in people with median age 86 (participants) and 88 (reported by proxies) at baseline.

Methods: The End of Life in Very Old Age is an ancillary study to the Health, Aging and Body Composition study. At baseline in Year 15 of Health, Aging and Body Composition, 1038 participants and 189 proxies reported levels of symptom distress every quarter, as well as 0-8 EOL treatment preferences elicited once each year.

Results: At baseline, the mean (SD) count of EOL treatment preferences was 4.2 (2.1) in participants, and 2.9 (2.3) in proxies. EOL treatment preference was not associated with symptom distress. By contrast, black race, male gender, and reported ease walking a quarter mile were independently associated with more aggressive EOL treatment preferences.

Conclusion: Preferences for more aggressive EOL treatment were not related to daily symptom distress but were significantly more likely to be endorsed among those with better mobility, suggesting that disability is an independent predictor of EOL treatment preferences in the very old.

Keywords: Symptom burden; cohort study; old age; treatment preferences.

Publication types

  • Research Support, N.I.H., Intramural
  • Research Support, N.I.H., Extramural

MeSH terms

  • Aged, 80 and over
  • Aging / psychology
  • Cross-Sectional Studies
  • Female
  • Follow-Up Studies
  • Humans
  • Interviews as Topic
  • Logistic Models
  • Longitudinal Studies
  • Male
  • Patient Preference* / psychology
  • Severity of Illness Index
  • Terminal Care / psychology*