Eating and Feeding Issues in Older Adults with Dementia: Part I: Assessment

Best Practices in Nursing Care For Hospitalized Older Adults with dementia from The John A. Hartford Institute for Geriatric Nursing and The Alzheimer’s Association

Issue Number D11.1, 2007
Series Editor: Marie Boltz, PhD, APRN, BC, GNP
Managing Editor: Sherry A. Greenberg, MSN, APRN, BC, GNP
New York University College of Nursing
_________________________________________________________________________________________________________________

WHY: Hospital patients with dementia are at high risk for eating and feeding difficulties and inadequate food and fluid intake. Depending on the severity of their cognitive impairment, they may forget to eat, forget they have eaten, fail to recognize food, or eat things that are not food. They may have difficulty with specific tasks (e.g., removing plate covers and wrappings, knowing what the utensils are for and using them, moving food or fluid to their mouth, chewing, and swallowing). They may have difficulty initiating the eating process, or they may start eating, get distracted, and fail to finish (Amella, 2004). Eating difficulties may have existed before hospitalization, but they are likely to worsen in the hospital because people with dementia often become more confused in an unfamiliar place. Different mealtime routines and foods add to the problem. Factors that reduce appetite and food and fluid intake in other older patients (e.g., pain, medications, nausea, dental and oral problems, and special diets) also affect patients with dementia who are less able understand and cope with them. Because of their cognitive and related communication impairments, patients with dementia may not be able tell anyone they are hungry or that they need help eating or more time to chew and swallow. When staff members try to help, some patients with dementia resist, push the food away, refuse to open their mouths, or spit food out.

BEST TOOL: Assessment can be challenging because of patients’ cognitive and related communication impairments. The Edinburgh Feeding Evaluation in Dementia Questionnaire (EdFED-Q) (Watson & Dreary, 1997a) is a valid and reliable observational instrument that nurses can use to identify eating and feeding difficulties and determine the level of assistance needed. The EdFED-Q was developed and tested extensively with nurses in psychogeriatric units and nursing homes.

TARGET POPULATION: Hospitalized older adults with diagnosed or suspected dementia.

VALIDITY AND RELIABILITY: The EdFED-Q was developed through Mokken scaling of items and factor analysis. Factor analysis of responses in a study of 196 older persons with dementia demonstrated that items could be reduced to three groups of indicators (Watson & Deary, 1997a) and the three groupings were found to be the best fit through Structural Equation Modeling in 345 patients (Watson & Deary, 1997b). These indicators include: patient behavior – obstinacy/aversion and passivity (seven items); nursing interventions (three items); and, an indicator of feeding difficulty (one item). This instrument has been used clinically in nursing home settings and in one community-based study and among 24 pairs of raters was found to have more than acceptable inter-rater (r = .59, p = 0.013) and intra-rater (r = .95, p < .0001) reliability (Watson, McDonald, & McReady, 2001).

STRENGTHS AND LIMITATIONS: The EdFED-Q is not clinically diagnostic, but it allows the assessor to determine the level of impairment as well as evaluate the need for possible psychosocial and clinical interventions, such as referrals for speech therapy, environmental modification, dietary alterations, environmental modification, diet alterations, and communication techniques. Mokken scaling of the seven behavior items forms a hierarchy of mealtime behavior so that behavior can be predicted. The EdFED-Q can be used as both a caregiver report and observational instrument. When observing a caregiver assist a person with dementia at meals, it is possible to also assess the quality of their interaction as construct validity of the EdFED-Q was established with another instrument (Interaction Behavior Measure - IBM). Because eating and feeding is by definition a reciprocal relationship, Amella (2002) showed that both instruments (IBM and EdFED-Q) were measuring the quality of the mealtime interaction even though the EdFED-Q only measures the behavior of the person with dementia.

Permission is hereby granted to reproduce, post, download, and/or distribute, this material in its entirety only for not-for-profit educational purposes only, provided that The Hartford Institute for Geriatric Nursing, College of Nursing, New York University is cited as the source. This material may be downloaded and/or distributed in electronic format, including PDA format. Available on the internet at www.hartfordign.org and/or www.ConsultGeriRN.org. E-mail notification of usage to: hartford.ign@nyu.edu.