OBJECTIVE: Race and gender health status disparities across the lifespan may influence death outcomes in Black and White nursing home residents. Early identification of at-risk residents will facilitate timely implementation of culturally appropriate end-of-life care. This study evaluated race and gender differences in resident characteristics and death outcome during 1 year of nursing home residence.
DESIGN: As part of a larger prospective longitudinal study, residents were assessed on admission and followed at 3, 6, and 12 months after admission for death outcome.
SAMPLE: Residents (N=1011) were enrolled sequentially as admitted to two large Midsouth government-operated safety-net nursing homes. There were 74% Black and 58% male residents, with a mean age of 66.2 years and 30.9% ever married. Most residents received intermediate level of care (60.9%), and 8.3% had a prognosis less than 6 months, 2.8% had an unstable condition, and 2.6% received hospice care. Research in safety-net nursing homes controlled for decreased access to nursing homes by Black older adults, and potential effects of socioeconomic status on health status and death outcome.
VARIABLES: Death outcome (dead, alive) and resident characteristics of race (Black, White), gender, age, marital status (ever married, never married), level of care (intermediate, skilled), prognosis less than 6 months, unstable condition, and receiving hospice care were evaluated.
METHODS: Residents were initially assessed during the first 2 weeks after admission by trained interviewers using the Minimum Data Set, and followed at 3, 6, and 12 months for death outcome.
FINDINGS: ANOVA was used to analyze race, gender, and race by gender interaction effects of resident characteristics. On average, Black residents were 3.4 years younger than White residents (p < .001) and men were 11.0 years younger than women (p=.025) on admission, with no race by gender interaction effect. There was a trend for more Black residents (p=.091) and for significantly more men (p < .001) to be ever married, with no race by gender interaction effect. There was a trend for more Black residents (p=.055) to have a prognosis less than 6 months, with no significant gender or race by gender interaction effects. For level of care, unstable condition, and receiving hospice care, there were no significant race, gender, or race by gender interaction effects.
For death outcome, 12.5% of residents were dead at 3 months, 18.2% at 6 months, and 25.0% at 12 months following admission. Chi-squared tests evaluated death outcome differences between races, and between genders within race, at 3, 6, and 12 months. Race comparison of percent dead approached statistical significance at 12 months (Black residents, 26.4%; White residents, 21.9%; p=.081). Race comparisons at 3 and 6 months, and gender within race comparisons at 3, 6, and 12 months, for death outcome were not statistically significant.
CONCLUSIONS: Despite Black or male residents being significantly younger and a trend for more Black residents having a prognosis less than 6 months, there were no gender within race differences in death outcome at 3, 6, and 12 months. However, there was a trend for more Black residents to die at 12 months of nursing home residence. The non-significant race differences in death outcome at 3 and 6 months may be an artifact of the limited numbers of residents who had died, and hence decreased variance of the death outcome variable. Had residents been followed more than 12 months there may have been sufficient numbers of residents who died to document significant race differences in death outcome. Half of the residents who died during 12 months of nursing home residence had died during the first 3 months following admission.
IMPLICATIONS: Safety-net nursing homes admit disproportionately younger Black residents or younger men than the general nursing home population. Black nursing home residents are a high risk group requiring early identification for end-of-life care. Black or male residents may have different etiologies for death. End-of-life care may need to be implemented on nursing home admission and prioritized during the first 3 months after admission. Future research should develop race and gender specific prediction equations for death at 3 and 6 months using admission Minimum Data Set data, and use survival analysis of days until death as a potentially more sensitive analysis for race differences in death outcome.
Funded by NIH NINR (RO1 NR04 299-01).
Back to Posters
Back to The Advancing Nursing Practice Excellence: State of the Science