Effectiveness of Feeding Tubes Among Persons with Advanced Cognitive Impairment This application addresses broad challenge area (05) Comparative Effectiveness Research, Challenge Topic 05-AG-101: Data Infrastructure for Post-Marketing Comparative Effectiveness Studies. Advanced dementia is a common life-defining illness with a disease trajectory of eating problems, weight loss, and recurrent infections. The current use of percutaneous endoscopic gastronomy (PEG) feeding tubes in persons with advanced cognitive impairment raises two concerns. First, there is striking variation in use by states, hospital referral regions, and hospitals. This variation is in contrast to the widely cited interpretation of the existing observational data that questions the effectiveness of feeding tubes in persons with dementia. Second, this evidence base itself is imperfect. These studies rely on observational data, yet none have attempted to control for selection bias, a potential problem in any non- randomized controlled trial. Furthermore, with only two exceptions, this work has not adequately adjusted for confounders. Two recent studies have found dementia not to be an independent predictor of survival with a corresponding editorial calling for better outcomes data to judge the efficacy of feeding tubes in persons with dementia. Our previous research and preliminary studies have found that hospitals are a key locus of care and final site of decision making for feeding tube insertions among the elderly with advanced dementia. We have found that the majority of PEG feeding tube decisions are made during an acute care hospitalization for an infection, and the rate at which hospitals insert PEG feeding tubes varies from 0% to 31% of hospital admissions of nursing home residents with advanced cognitive impairment. The proposed research has two overarching goals. First, we aim to use advanced methods to account for selection bias to provide a more definitive study of the potential benefits (survival, weight change, prevention and healing of pressure ulcers), risks (use of restraints, prescription of antipsychotics or benzodiazepines, and transitions to the ER and acute care hospital), and health care expenditures. This goal will be accomplished through the linkage of four files (Medicare Claims, Minimum Data Set, American Hospital Association data, and Dartmouth Atlas of Health Care) from 1999-2008 and Medicare Part D data from 2006-2008. This analysis will examine individual-level results among persons with advanced cognitive impairment, as well as persons who receive their medical care at institutions with varying rates of PEG feeding tube insertion. A second goal is to understand the variation in hospital insertion rates of PEG feeding tubes. In this goal, we will examine potentially modifiable hospital characteristics, market factors, and state policies associated with this variation in hospitals'rate of PEG feeding tube insertion. Knowledge gained will be used to inform public policy and formulate hospital-based interventions to ensure that PEG tube insertion in persons with advance cognitive impairment is ethically appropriate and consistent with the best available data on their effectiveness. Eating problems are common issues with persons afflicted with dementia. There are two concerns with the use of feeding tubes. First, hospitals vary in their rate of feeding tube insertion despite two widely cited reviews of the evidence that question the effectiveness of feeding tubes. Second, there is controversy about the evidence of effectiveness of feeding tubes. The proposed research will provide authoritative evidence of the effectiveness of feeding tubes and identify modifiable hospital characteristics and state policies that account for the observed striking variation in feeding tube use.

Public Health Relevance

Eating problems are common issues with persons afflicted with dementia. There are two concerns with the use of feeding tubes. First, hospitals vary in their rate of feeding tube insertion despite two widely cited reviews of the evidence that question the effectiveness of feeding tubes. Second, there is controversy about the evidence of effectiveness of feeding tubes. The proposed research will provide authoritative evidence of the effectiveness of feeding tubes and identify modifiable hospital characteristics and state policies that account for the observed striking variation in feeding tube use.

Agency
National Institute of Health (NIH)
Institute
National Institute on Aging (NIA)
Type
NIH Challenge Grants and Partnerships Program (RC1)
Project #
5RC1AG036418-02
Application #
7938007
Study Section
Special Emphasis Panel (ZRG1-HDM-G (58))
Program Officer
Silverberg, Nina B
Project Start
2009-09-30
Project End
2012-08-31
Budget Start
2010-09-01
Budget End
2012-08-31
Support Year
2
Fiscal Year
2010
Total Cost
$325,414
Indirect Cost
Name
Brown University
Department
Public Health & Prev Medicine
Type
Schools of Public Health
DUNS #
001785542
City
Providence
State
RI
Country
United States
Zip Code
02912
Hwang, Deborah; Teno, Joan M; Gozalo, Pedro et al. (2014) Feeding tubes and health costs postinsertion in nursing home residents with advanced dementia. J Pain Symptom Manage 47:1116-20
Givens, Jane L; Mitchell, Susan L; Kuo, Sylvia et al. (2013) Skilled nursing facility admissions of nursing home residents with advanced dementia. J Am Geriatr Soc 61:1645-50
Teno, Joan M; Gozalo, Pedro L; Mitchell, Susan L et al. (2012) Does feeding tube insertion and its timing improve survival? J Am Geriatr Soc 60:1918-21
Cai, Shubing; Gozalo, Pedro L; Mitchell, Susan L et al. (2012) Do Patients With Advanced Cognitive Impairment Admitted to Hospitals With Higher Rates of Feeding Tube Insertion HaveýýImproved Survival? J Pain Symptom Manage :
Teno, Joan M; Gozalo, Pedro; Mitchell, Susan L et al. (2012) Feeding tubes and the prevention or healing of pressure ulcers. Arch Intern Med 172:697-701
Teno, Joan M; Mitchell, Susan L; Gozalo, Pedro L et al. (2010) Hospital characteristics associated with feeding tube placement in nursing home residents with advanced cognitive impairment. JAMA 303:544-50