While the last years of life have an
intense focus on medical care, the same emphasis
has not been placed on social well-being. Consequently, the epidemiology of key indicators of social
health is unknown.
240 Abstracts Vol. 60 No. 1 July 2020
Objective(s). To determine the prevalence and correlates of loneliness and social isolation among older
adults during the last four years of life.
Method(s). We used nationally-representative health
and retirement study data to examine adults age >50
who died while enrolled (N¼3,540). Subjects were interviewed once in the last four years of life and classified into one of eight 6-month cohorts based on the
number of months between the interview and death.
We used validated measures of loneliness (3-item
UCLA Loneliness Scale) and social isolation (13-item
scale describing whether an individual lived alone,
was unmarried/unpartnered, had minimal interaction
with children, family, or friends, and had minimal
community engagement). We modeled the relationship between loneliness or social isolation and time
before death adjusting for sociodemographic and
key health covariates.
Results. The mean age at death was 76 (SD¼11.4) and
50% were female. The prevalence of loneliness and social isolation in the last 4 years of life was 51% and
9%, respectively, and these rates were constant for four
years prior to death. After adjustment, there were
distinct risk factors for loneliness and social isolation
(p<0.01); cognitive impairment (Normal: 44%, CIND:
52%, Dementia: 56%), vision impairment (54% vs
46%), and incontinence (54% vs 46%) were risk factors
for loneliness, whereas race (White: 11% vs Non-White:
5%), low income (20% vs 8%), and inability to walk a
block (12% vs 8%) were risk factors for social isolation.
Conclusion(s). Loneliness occurs in half of older
adults and social isolation occurs in nearly 10% of older adults, with a consistent prevalence for the four
years prior to death.