Evaluation of Family Characteristics and Multiple Hospitalizations at the End of Life: Evidence from the Utah Population Database

Djin L. Tay, Katherine A. Ornstein, Huong Meeks, Rebecca L. Utz, Ken R. Smith, Caroline Stephens, Mia Hashibe, Lee Ellington

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


Background: Scant research has examined the relationship between family characteristics and end-of-life (EOL) outcomes despite the importance of family at the EOL. Objectives: This study examined factors associated with the size and composition of family relationships on multiple EOL hospitalizations. Design: Retrospective analysis of the Utah Population Database, a statewide population database using linked administrative records. Setting/subjects: We identified adults who died of natural causes in Utah, United States (n = 216,913) between 1998 and 2016 and identified adult first-degree family members (n = 743,874; spouses = 13.2%; parents = 3.6%; children = 51.7%; siblings = 31.5%). Measurements: We compared demographic, socioeconomic, and death characteristics of decedents with and without first-degree family. Using logistic regression models adjusting for sex, age, race/ethnicity, marital status, comorbidity, and causes of death, we examined the association of first-degree family size and composition, on multiple hospitalizations in the last six months of life. Results: Among decedents without documented first-degree family members in Utah (16.0%), 57.7% were female and 7 in 10 were older than 70 years. Nonmarried (aOR = 0.90, 95% CI = 0.88-0.92) decedents and decedents with children (aOR = 0.97, 95% CI = 0.94-0.99) were less likely to have multiple EOL hospitalizations. Family size was not associated with multiple EOL hospitalizations. Conclusions: First-degree family characteristics vary at the EOL. EOL care utilization may be influenced by family characteristics - in particular, presence of a spouse. Future studies should explore how the quality of family networks, as well as extended family, impacts other EOL characteristics such as hospice and palliative care use to better understand the EOL care experience.

Original languageEnglish
Pages (from-to)376-387
Number of pages12
JournalJournal of Palliative Medicine
Issue number3
StatePublished - Mar 2022


  • caregivers
  • end-of-life
  • population health
  • socioeconomic factors


Dive into the research topics of 'Evaluation of Family Characteristics and Multiple Hospitalizations at the End of Life: Evidence from the Utah Population Database'. Together they form a unique fingerprint.

Cite this