Three approaches to determining clinically meaningful benefit on the Cohen‐Mansfield Agitation Inventory in dementia clinical trials for agitation

INTRODUCTION: There is a need to understand the clinical meaningfulness of symptom score changes in treatment trials of dementia-related agitation. We estimated minimal clinically important differences (MCIDs) for commonly employed agitation scales and contextualized their clinical application. METHODS: We employed anchor- and distribution-based approaches to determine changes in scores corresponding to minimal symptom improvement. An opinion-based approach assessed expert clinicians' agreement on the meaningfulness of score changes through three clinical vignettes. RESULTS: Minimal symptom improvement for Cohen-Mansfield Agitation Inventory total score ranged from -4 (over <1 month) to -11 (over 1 to 3 months) points. Greater symptom severity correlated with higher MCID estimates. The clinical importance of score changes was influenced by treatment duration, pharmacological side effects, and impacts on caregiver distress/time resources. DISCUSSION: The clinical meaningfulness of agitation scale MCIDs is influenced by trial-specific and clinical factors. Shorter trial durations and measuring caregiver distress/time resources enhance the clinical interpretation of agitation treatment outcomes. Highlights: For the CMAI total score, the MCID was -4 points over shorter time scales and -11 points for longer time scales.Worse agitation severity was associated with higher MCID estimates.There was high expert consensus that a noticeable treatment benefit was not worthwhile if it occurred after 12 weeks or had no impact on caregiver/staff distress/time resources.