Agitation and aggression in people living with dementia and mild cognitive impairment in shared-housing arrangements – validation of a German version of the Cohen-Mansfield Agitation Inventory-Short Form (CMAI-SF)

Abstract

Background The Cohen-Mansfield Agitation Inventory-Short Form (CMAI-SF) is a 14-item scale for assessing agitation and aggression, derived from the original 29-item CMAI, and completed by a proxy. Because the CMAI-SF has not yet been validated in German language, the aim of this study is to explore its construct validity. Methods Baseline data from a cluster-randomized trial to evaluate a non-pharmacological complex intervention for people living with dementia (PlwD) and mild cognitive impairment (MCI) were analyzed. The study sample consisted of 97 shared-housing arrangements (SHAs) in Germany, comprising N = 341 residents with mild to severe dementia and MCI. Trained nursing staff collected data by proxy-rating the CMAI-SF, Neuropsychiatric Inventory-Nursing Home Version (NPI-NH), and QUALIDEM. They also conducted the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA). Results In an exploratory factor analysis, three factors emerged: “aggressive behavior”, “verbally agitated behavior”, and “physically non-aggressive behavior”. The CMAI-SF total score showed good internal consistency (α = .85), and the factors themselves showed adequate internal consistency (α = .75/.76/.73). The CMAI-SF showed convergent validity with the NPI-NH agitation item (r = .66) and the NPI-NH “agitation & restless behavior” factor (r = .82). Discriminant validity was confirmed by a low (r = .28) correlation with the NPI-NH apathy item. Quality of life decreased significantly with agitation, as the CMAI-SF showed a moderate negative correlation with the QUALIDEM total score (r = -.35). Conclusions The 14-item CMAI-SF is a time-efficient, reliable, and valid assessment instrument. Three factors emerged that were similar to those already found in nursing home samples for the original CMAI and the CMAI-SF and in day care samples for the CMAI-SF. The findings provide preliminary evidence that the CMAI-SF can be used instead of the CMAI to reduce time, costs, and burden in future trials. Trial registration The DemWG study from which data were used to draft this manuscript was prospectively registered on 16 July 2019 at ISRCTN registry (ISRCTN89825211).

Detecting agitation and aggression in persons living with dementia: a systematic review of diagnostic accuracy

“…Thirty articles comparing agitation and/or aggression tools (i.e., no reference standard), were included as part of our secondary objective [ 16 , 22 , 25 , 31 – 58 ]. These studies determined the correlation between known agitation and/or aggression tools in PLWD.…”

“…These studies determined the correlation between known agitation and/or aggression tools in PLWD. They were conducted in North America ( n = 11) [ 25 , 31 , 33 – 35 , 40 , 48 , 49 , 51 , 55 , 57 ], Asia ( n = 7) [ 32 , 38 , 39 , 47 , 52 , 53 , 56 ] South America ( n = 1) [ 34 ], Europe ( n = 8) [ 22 , 34 , 36 , 43 – 46 , 54 , 58 ], and Australia ( n = 1) [ 41 ]. Furthermore, four studies did not report their location [ 16 , 37 , 42 , 50 ].…”

“…Furthermore, four studies did not report their location [ 16 , 37 , 42 , 50 ]. The studies were published between 1989 and 2023 [ 42 , 58 ]. Dementia severity was determined mainly with the MMSE, or variations thereof, ( n = 25) [ 22 , 25 , 31 – 34 , 36 – 41 , 43 – 48 , 50 – 56 , 58 ] with other studies using the Functional Assessment Staging Scale (FAST) ( n = 2) [ 16 , 35 ], and Global Deterioration Scale (GDS) ( n = 1) [ 49 ].…”

“…The types of dementia reported include Alzheimer’s Disease, Vascular, Lewy Body, or general dementia not otherwise specified. However, multiple articles did not report the type ( n = 11) [ 16 , 22 , 25 , 35 , 41 – 44 , 49 , 56 , 58 ] or severity ( n = 19) [ 25 , 32 , 33 , 35 , 37 , 39 – 48 , 53 , 54 , 56 , 57 ] of dementia in their population.…”

“…Most included studies demonstrated low concern that the included patients did not match the review question ( n = 29) [ 22 , 31 – 36 , 16 , 37 – 54 , 25 , 58 , 56 , 57 ], with one study demonstrating unclear concern [ 55 ] due to unclear exclusion criteria. Many studies did not indicate whether test administrators were blinded ( n = 22) [ 25 , 33 , 34 , 36 , 38 – 42 , 44 , 45 , 47 – 54 , 56 – 58 ], with ( n = 3) [ 16 , 32 , 55 ] studies indicating no blinding, thus there was varying concern regarding the conduct between the two tools (Supplemental Appendices 4 and 5). Nonetheless, there was low concern that the target condition (i.e., agitation and/or aggression) as defined by both tools did not match the review question across studies ( n = 30) [ 16 , 22 , 25 , 31 – 36 , 36 – 58 ].…”