High short-term mortality in hospitalized patients with advanced dementia - Lack of benefit of tube feeding

Background: The influence of tube feeding on survival in hospitalized patie nts with advanced dementia is controversial. Objective: To assess long-term survival in an inception cohort, incident tu be feeding placement during the index hospitalization, and the influence of tube feeding on survival in this group of patients. Subjects and Methods: Ninety-nine hospitalized patients with advanced demen tia and an available surrogate decision maker were followed up through and after the index hospitalization for mortality and placement of a feeding tu be. Other variables measured included advance directive status, presence of a long-term primary care physician, level of involvement of the surrogate decision maker, admitting diagnosis, prior hospitalizations, comorbidities, and diagnosis related group diagnostic category. Results: A new feeding tube was placed in 50% (51/99) of the study patients during the index hospitalization, 31% (31/99) left the hospital without a feeding tube, and 17% (17/99) were admitted with a feeding tube already in place. By stepwise logistic regression analysis, predictors of new feeding tube placement included African American ethnicity (odds ratio, 9.43, 95% c onfidence interval, 2.1-43.2) and residence in a nursing home (odds ratio, 4.9; 95% confidence interval, 1.02-2.5). Median survival of the 99 patients was 175 days. Eighty-five (85%) survived the index hospitalization, and 28 (28%) were still alive at last follow-up, a range of 1.3 to 4.2 years afte r enrollment in the study. Tube feeding was not associated with survival (P =.90). An admitting diagnosis of infection was associated with higher morta lity (odds ratio, 1.9; 95% confidence interval, 1.01-3.6). Conclusions: In this cohort of hospitalized patients with advanced dementia , risk of receiving a new feeding tube is high, associated with African Ame rican ethnicity, and prior residence in a nursing home, and has no measurab le influence on survival. With or without a feeding tube, these patients ha ve a 50% six-month median mortality.