The Reliability and Validity of a New Scale: Feeding Difficulty Index (FDI)

Wednesday, 1 August 2012

Chia-Chi Chang, PhD, RN
School of Geriatric Nursing and Care Management, Taipei Medical University, Taipei, Taiwan
Beverly L. Roberts, RN, PhD
College of Nursing, University of Florida, Gainesville, FL

Learning Objective 1: The learner will be able to know a new scale for assessing feeding difficulty of dementia.

Learning Objective 2: The learner will be able to know the reliability and validity of Feeding Difficulty Index (FDI).

Feeding difficulties are common problems among dementia and may lead to serious reduction of food intake, weight loss, dehydration and malnutrition and even death. A new scale called Feeding Difficulty Index (FDI) was developed by the first author.

Purpose: The purpose was to establish reliability and validity of this new scale.

Methods: The index was reviewed and revised by 6 experts and 10 clinical care providers for the content validity and face validity. Three research assistants, two nurses, and one nursing assistants were trained to observe 10 residents with dementia simultaneous in order to test the inter-rater reliability.The test-retest reliability was evaluated by calculating the Pearson product moment correlation coefficient between pretest and posttest over a two-week interval in a sample of 10 residents with dementia.In order to examine the criterion validity and construct validity, a total sample of 185 elderly with dementia were recruited for assessment feeding difficulty.

Results: The content validity showed that the average of CVI for whole scale was 92.76%. The average of face validity for whole scale was 98.68%.  The mean agreement percentage was 0.87. The test-retest reliability was 0.853 (p<0.01). Furthermore, paired-t tests indicated no difference on the FDI score between pretest and post-test (t=0.355, p~0.73).   The internal consistency was established by calculating the Cronbach α coefficient. The Cronbach α coefficient was 0.679, indicates the fair internal consistency. Concurrent validity was evaluated by correlating the FDI with EdFED. The FDI score was significantly correlated with EdFED score (r=0.653, p<0.01). Factor analysis showed that there were 7 components among the scale items, and that 7 components could explain 64.7% variance of the FDI scale.

Conclusion: The FDI has good validity and reliability. The predictive validity will be evaluated and items may need to be modified in the future.