Nasogastric and Gastrostomy Tube Weaning in Children: An Evaluation of Outcome Data from a Systematic Multidiscliplinary Approach

Amy Shelton
Advisor: R. Leonard Vance, PhD, CIH
Preceptors: Elizabeth Clawson, PhD and
Elizabeth Eustis Turf, PhD

Objectives
The purpose of this study is to show that a multidisciplinary feeding program using applied behavioral analysis (ABA), Beckman Oral Motor Exercises, and a systematic tube weaning approach will increase the likelihood of long-term success for reducing tube feeding dependency in children with feeding disorders. Multiple independent variables were analyzed to assess which variables increase the odds of a successful weaning outcome after 12 months post-discharge.

Methods
This is a retrospective follow-up study of 62 patients admitted to the Day Feeding Program at Children’s Hospital in Richmond, VA. The outcome measure, weaned vs. non weaned, was compared to several independent variables: age, sex, race, local vs. non-local residence, prematurity vs. full term, tube type, 9 different diagnostic categories, caregiver discharge scores, outpatient therapy, location of therapy, length of time outpatient therapy was received, length of time the feeding protocol was used, telephone and email correspondence, any major illnesses or surgery since discharge, feeding clinic follow-up, nutrition follow-up, and tube feeding status at admission and discharge. The analysis included descriptive data, prevalence data, and multivariate logistic regression.

Results
Several patient variables were associated with greater odds for successful weaning. When adjusted for other factors, the variables shown to be the best predictors included patients with neurological conditions, follow-up with the feeding program dietitian after discharge, and weaned or supplemental tube feeding status at discharge. The diagnoses most likely to wean included patients with structural abnormalities, neurological conditions, and pulmonary conditions as opposed to patients with cardiovascular, cardio respiratory, and or gastrointestinal problems. Outpatient therapy specifically at Children’s Hospital and patients without significant medical problems since discharge are also most likely good predictors; however, more sample size is needed to confirm these as significant. Due to sample size, tube feeding status at discharge was the only variable with a significant confidence interval (1.09, 13.62).

Conclusions
This pilot study was beneficial to show that programs such as the feeding program at Children’s Hospital are successful at weaning children from their feeding tubes on a short-term and long-term basis. Given the limitations of a small sample size, in order to make definitive statements regarding variables associated with a positive outcome, the research should be repeated in a 2-3 years with a larger sample to strengthen the data.