A model of assessment and interventions for feeding difficulties in people with dementia

Project: A - Government Institutiona - National Health Research Institutes


Nearly 80% of dementia patients have feeding difficulties. These feeding difficulties are associated with reduced food intake, dehydration, weight loss, and malnutrition. Nursing assistants (NA) in nursing homes are primary caregivers who feed patients, but they receive the least training to adequately handle feeding difficulties. NAs often fail to identify feeding difficulties and appropriate feeding strategies because of insufficient knowledge. Educating NAs to manage feeding difficulties is essential because the cognitive impairment of many residents preclude training them to manage their feeding difficulties. Using an appropriate instrument to assess the feeding difficulty among people with dementia can help nurses to identify the problems of individual patients and provide interventions such as caregiver education and supervision, and making appropriate referrals to an interdisciplinary health care team. Thus, an instrument that can measure the feeding difficulty in people with dementia and have sufficient sensitivity to detect the change in eating ability is a major concern in managing difficulties in feeding of people with dementia. It will be a four-year study. The major aim of this study will establish a model of assessment and interventions for feeding difficulties in people with dementia. The purpose of the first two-year period of this total project will develop a culturally appropriate instrument which will be called “Assessment of Feeding Difficulty in Taiwanese Dementia” (AFDTD) based on the model of feeding difficulty that has been proposed by the PI (Dr. Chang) to assess the feeding difficulty of people with dementia and establish the psychometric properties of the instrument, including its validity and reliability. A total sample of 185 people with dementia will be enrolled for completing all the measurements. Reliability of the instrument will be measured by internal consistency, inter-rater reliability and test-retest coefficients. Validity of the instrument will be measured by content validity, criterion-related validity, and construct validity. In addition, the prevalence and associated factors will also be investigated in this study period by Pearson correlation and regression analyses. The next two years will develop interventions for reducing feeding difficulties in people with dementia. A comprehensive feeding skills training program has been develop by Chang (2005). The small sample size and lack of outcome measurements for people with dementia are the limitations for the effects of this program. The proposed randomized clinical trial will assess the effects of a longer and more comprehensive feeding skills training program on NA and outcomes of people with dementia. The training will include 2 one hour in-service classes and 5 days of hands-on training. The content of the program will include information about feeding difficulties related to dementia, assessment and interventions for different types of feeding difficulties. In 78 nursing assistants randomly assigned to a feeding training program or bathing difficulties training (2 one hour in-service classes), pre and post test assessments will be made for nursing assistant outcomes (feeding skills and job satisfaction related to feeding patients with feeding difficulties) and dementia resident outcomes (food intake, feeding time, feeding difficulty). The research assistants will categorize residents into one of the three level of feeding difficulty by using AFDTD scale. Four residents with each level of feeding difficulty will be recruited in each nursing home. To allow determination of the effectiveness of NA training for each level of feeding difficulty, NAs will be randomly assigned to feed a resident with each level of feeding difficulty (NA will feed a total of 3 residents at pre and post test). NAs will feed the same residents at pre and post test. Research assistants will observe nursing assistants feeding people with dementia with 3 different level of feeding difficulty. The proposed study will establish NA and resident outcomes and provide direction for further refinement of the intervention. Once the effectiveness of the feeding training program has been established, further research will be needed to determine the effectiveness of implementing the feeding training program in real world conditions.
Effective start/end date1/1/1312/31/13


  • Feeding difficulty
  • Scale development
  • Dementia