Skip to main content
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: J Am Geriatr Soc. 2017 Sep 12;66(1):41–47. doi: 10.1111/jgs.15041

Table 2.

Associations Between Time-Varying Exposures and Admission to Hospice in Last Year of Life

Time-Varying Exposure Unadjusted Hazard Ratio (95% CI) Adjusted a Hazard Ratio (95% CI)
Specific month b
  Any restricting symptoms 1.62 (1.27–2.07) 1.66 (1.30–2.12)
  Per no. restricting symptoms c 1.08 (1.04–1.12) 1.09 (1.05–1.12)
  Per no. disabilities d 1.10 (1.05–1.14) 1.10 (1.05–1.14)
Cumulative number of months e
Any restricting symptoms 1.07 (1.01–1.12) 1.07 (1.01–1.13)

Abbreviation: CI, confidence interval.

a

Covariates included age, sex, race/ethnicity, years of education, living situation, number of chronic conditions, cognitive impairment, depressive symptoms, low social support and months since prior comprehensive assessment.

b

Results reflect the marginal (i.e. average) effect of each exposure at month t-1 on hospice admission at month t.

c

Possible range: 0 to 15

d

Possible range: 0 to 13

e

Results reflect the effect of each additional month with any restricting symptoms as a time-varying exposure (up to month t-1) on hospice admission at month t.