Mealtime caregiving knowledge, attitudes, and behaviors for persons living with dementia: A systematic review of psychometric properties of instruments

Background: Using valid instruments to assess caregiving knowledge, attitudes, skills, and behaviors in mealtime care for people living with dementia is critical to evaluate the process and effects of mealtime assistance interventions. Yet, the quantity and psychometric quality of such instruments are unknown. Objectives: This systematic review described and evaluated psychometric properties of instruments that were developed and used to assess mealtime caregiving knowledge, attitudes, skills, and behaviors for people with dementia. Methods: We searched Pubmed, CINAHL, AgeLine, PsychINFO, and Cochrane Library for records published between January 1st, 1980 and June 31st, 2019, with follow-up searches by December 20th, 2019. Records were eligible if they included any instrument developed, tested, and/or used to measure the concepts of interest, including mealtime caregiving knowledge, attitudes, skills, and/or behaviors. After eligible records were identified, instruments that were reported in the eligible records were identified and extracted. Instruments were eligible if they were originally developed to measure the concepts of interest or developed in non-mealtime activities and later used or tested to measure the concepts of interest. From eligible records, eight characteristics of eligible instruments were extracted: (1) development process, (2) the concept/construct the instrument operationalizes, (3) sample and setting the instrument was used/tested in, (4) administration method, (5) description of items, (6) scoring format/interpretation, (7) reliability, and (8) validity. The psychometric quality of eligible instruments was evaluated using a newly developed psychometric quality assessment tool. Results: A total of 9438 records were retrieved and 19 eligible instruments were identified. Ten instruments assessed mealtime caregiving skills or behaviors; 5 assessed attitudes, intention, self-efficacy, empathy; and 4 assessed knowledge. All instruments were scored as having low psychometric quality, except for Mealtime Engagement Scale with moderate psychometric quality in assessing mealtime engagement toward people with dementia. Reasons for low psychometric quality included limited psychometric testing, inadequate estimates of psychometric properties, and use of small sample size. Conclusions: While all instruments warrant further testing, Mealtime Engagement Scale demonstrated moderate psychometric quality with preliminary evidence of reliability and validity to assess mealtime engagement toward people with dementia. Future testing of Mealtime Engagement Scale is needed in larger diverse samples in different care settings to accumulate psychometric evidence and expand the use.