Importance: As older adults live longer in the community, many, particularly those with dementia, may become bedbound at the end of life and require extensive support from caregivers.
Objective: To estimate the prevalence and associated characteristics of bedbound status during the last year of life among community-dwelling decedents, and to examine differences by dementia status.
Design, setting, and participants: This cross-sectional study used data from rounds 1 through 13 from the National Health and Aging Trends Study to identify individuals aged 65 years or older who were community-dwelling 12 months before death between the years 2011 and 2023. Data were analyzed between December 2024 and August 2025.
Main outcomes and measures: The primary outcome was bedbound status, defined by self or proxy report incorporating frequency, help required, and level of difficulty leaving bed and/or bedroom. Probable dementia was defined using an algorithm that includes cognitive testing and diagnosis history. The number of months between each participant's date of death and their last interview prior to death was calculated. Bedbound prevalence was estimated by dementia status in monthly intervals up to 1 year before death, along with the amount of caregiver help received. Survey-weighted multivariable regression was used to estimate the likelihood of being bedbound during the last year of life by sociodemographic, clinical, and health characteristics.
Results: Among 3168 decedents (mean [SD] age, 83.0 [0.2] years, 51.9% female; 36.1% [95% CI, 34.7%-39.0%] with dementia), 16.6% (95% CI, 15.0%-18.3%) were bedbound during their last year of life, representing nearly 2.6 million decedents. As individuals with dementia neared death, bedbound prevalence increased from 28.6% (95% CI, 13.9%-47.4%) in the 12 months prior to death to 77% (95% CI, 56.4%-91.6%) in the last month of life. Probable dementia was associated with nearly 5 times greater odds of being bedbound in the last year of life compared with having possible or no dementia (odds ratio, 4.58; 95% CI, 3.09-6.79; P < .001). Bedbound decedents received nearly 3 times as many hours of care per week compared with individuals who were not bedbound (98.00 [95% CI, 88.86-107.14] vs 34.03 [95% CI, 31.74-36.32] h/wk; P < .001).
Conclusions and relevance: In this cross-sectional study of decedents, being bedbound during the last year of life was common among individuals with dementia living in the community and imposed substantial demand on caregivers. These findings underscore the urgent need for policies that ensure access to comprehensive home-based care and services to support patients and families.