Incorporating individually defined brain health priorities in clinical trial outcomes: the electronic Person-Specific Outcome Measure approach in the United States

Abstract

INTRODUCTION: While a limited number of disease-modifying treatments for Alzheimer's disease (AD) have been approved in the United States, there is caution in adopting these treatments in clinical use. The electronic Person-Specific Outcome Measure (ePSOM) tool was recently developed to establish whether, besides modifying underlying AD pathology, new treatments exerted sufficient beneficial effects in areas that matter the most to an individual. METHODS: We conducted a study to understand how findings from the ePSOM UK study applied in the United States. The ePSOM US survey (May 2023 to January 2024) collected primarily free-text responses in personally defined brain health priorities, alongside self-reported confidence in managing these priorities. We used natural language processing (k-means clustering of GloVe vectors) and chi-squared tests to examine differences between the US and UK populations' answers. We used a Mann–Whitney U test to compare the confidence ratings between participants with and without a self-reported diagnosis of neurodegenerative disease. RESULTS: Our analysis included 764 participants in the United States (68.8% female; 74.2% high educational attainment) with a total of 9010 free-text responses, of whom 53 individuals (6.90%) reported neurodegenerative disease diagnosis. The comparable sample from the UK survey included 4529 participants with a total of 38,056 responses. There were statistically significant differences in the proportion of responses between the US and UK populations. The diagnosis group showed a significant difference in average total scores of self-reported confidence compared with those without a diagnosis (median score 21, interquartile range [IQR] = 17 to 23 vs median score 24, IQR = 22 to 25, U = 8908, p