Interventions on mealtime difficulties in older adults with dementia: A systematic review

Objectives

To evaluate the effects of interventions on mealtime difficulties in older adults with dementia.

Design

A systematic review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement.

Data sources

Pubmed, Medline (OVID), CINAHL (EBSCOHost), EBM Reviews (OVID) and PsychINFO (OVID) were searched between January 2004 and September 2012 by using keywords as dementia, Alzheimer, feed(ing), eat(ing), mealtime(s), oral intake, nutrition, intervention, experimental, quasi-experimental and any matched terms. Other sources included Google Scholar and relevant bibliographies.

Review methods

Eligibility criteria were established by defining the population, intervention, comparator, outcomes, timing and setting of interest. Studies were reviewed by title and abstract screening, and full-text assessing for eligibility. Data were abstracted from eligible studies using a self-made structured tool. Eligible studies were classified by intervention, accessed for quality using the Quality Assessment Tool for Quantitative Studies, and graded for evidence using the Grading of Recommendations, Assessment, Development and Evaluation Working Group criteria.

Results

Twenty-two intervention studies (9 RCTs), including a total of 2082 older adults with dementia and 95 professionals from more than 85 long-term care facilities, were selected, and classified into five types: nutritional supplements, training/education programs, environment/routine modification, feeding assistance and mixed interventions. Eight studies were strong, eleven moderate and three weak in quality. Limitations of body of research included lack of randomization and/or control group, small sample size without power analysis, lack of theory-based interventions and blinding, inadequate statistical analysis and plausible confounding bias. “Nutritional supplements” showed moderate evidence to increase food intake, body weight and BMI. “Training/education programs” demonstrated moderate evidence to increase eating time and decrease feeding difficulty. Both “training/education programs” and “feeding assistance” were insufficient to increase food intake. “Environment/routine modification” indicated low evidence to increase food intake, and insufficient to decrease agitation. Evidence was sparse on nutritional status, eating ability, behavior disturbance, behavioral and cognitive function, or level of dependence.

Conclusions

This review provides updated evidence for clinical practice and points out priorities for nursing research. Current evidence is based on a body of research with moderate quality and existing limitations, and needs to be further explored with more rigorous studies.