OBJECTIVES : It is unclear whether modest changes in standardized outcome measures from dementia drug trials are clinically meaningful to patients and their caregivers. In contrast, tracking personalized symptom severity provides a robust method for interpreting relevant and non-arbitrary treatment effects. Here, we outline the development and validation of a tool to facilitate symptom tracking in dementia drug trials by caregivers.

METHODS : To develop the dementia SymptomGuide™ (SG-D) we qualitatively analyzed goals set using Goal Attainment Scaling (GAS) in two clinical trials and a memory clinic. From this the dementia symptom menu, SG-D, was elaborated. The SG-D content next was reviewed by Alzheimer’s Society group leaders, memory clinic nurses, geriatricians, and geriatric psychiatrists (demonstrating content validity). SG-D was used as a secondary outcome measure in VASPECT, a phase IV trial of donepezil in mild-moderate Vascular or mixed Alzheimer’s disease/Vascular dementia (face validity). Other outcomes measures included the Mini-Mental State Examination (MMSE), Clinical Global Impression (CGI), and Neuropsychiatric Inventory Questionnaire (NPI-Q).

RESULTS : A menu of 61 dementia symptoms was created, each with 8-12 descriptions, to form the SG-D. From this, a Canadian panel selected 32/61 symptoms as appropriate for VASPECT. Of the 108 subjects with SG-D scores who completed the trial, 102 completed SG-D (completion rate 94.4%; feasibility). For other measures, the average completion rate was 97.5%; range 94.0-100%. The average SG-D symptom rating correlated with CGI-Improvement scores (rho=-0.64; construct validity). SG-D scores correlated notionally with change in other outcome measures (MMSE r=0.37, NPI-Q r=-0.38; construct validity). The SG-D showed similar responsiveness (Standardized Response Mean: SG-D 0.30, NPI-Q -0.29, MMSE 0.22; sensitivity to change).

CONCLUSIONS : SG-D is feasible for dementia symptom tracking and rating change. SG-D provided an inherently clinically meaningful outcome measure that is valid and responsive, suitable for evaluating the clinical meaningfulness of other clinical trial measures.