J Korean Acad Child Health Nurs.  2010 Jan;16(1):84-92.

A Predictive Model using Decision Tree Method on Demand for Alternative Feeding Education by Nurses

• Oh J
• Yoon CM
• Kim BS

Abstract

PURPOSE
One of the main reasons why mothers quit breast feeding is that the volume of breast milk is inadequate due to insufficiency in suckling. We believe suckling experience may be a factor affecting nipple confusion. So an alternative feeding method, namely cup, spoon, finger, or nasogastric tube feeding may be needed to prevent nipple confusion. The purpose of this study was to construct a predictive model for demand for alternative feeding education by nurses.
METHODS
A descriptive design with structured self-report questionnaires was used for this study. Data from 175 nurses working in hospitals in Busan were collected between April 1 and 15, 2009. Data were analyzed by decision tree method, one of the data mining techniques using SAS 9.1 and Enterprise Miner 4.3 program.
RESULTS
Of the nurses, 81.1% demanded alternative feeding education and 5 factors showed that most of them expressed intention to pay, desire to know about alternative feeding, age, and learning experience. From these results, the derived model is considered appropriative for explaining and predicting demand for alternative feeding education.
CONCLUSION
This confirms that knowledge and compliance in alternative breast feeding for newborn babies should be correct and any inaccuracies or insufficient information should be supplemented.

Keyword

Breast feeding; Decision trees; Linear model

MeSH Terms

Breast Feeding
Compliance
Data Mining
Decision Trees
Enteral Nutrition
Feeding Methods
Fingers
Humans
Infant, Newborn
Intention
Learning
Linear Models
Milk, Human
Mothers
Nipples
Phenothiazines
Child Health
Surveys and Questionnaires
Phenothiazines