Aim: After completion of treatment, breast cancer survivors experience cancer-related cognitive impairments such as problems with memory and attention. Cognition is described as varying from day-to-day yet is typically measured using one-time assessments which do not capture these fluctuations. The present study aims to assess the feasibility, validity, and usability of our novel ecological momentary assessment (EMA) app of cognition.
Methods: Nineteen breast cancer survivors 6–36-month post-chemotherapy and 26 healthy controls participated. Participants completed the NIH Toolbox Cognition Battery in person followed by the EMA. The EMA app contains four cognitive tasks which assess processing speed, executive working memory, spatial working memory, and inhibition/attention. Participants completed the tasks once a day for 30 days on their smart phones, thereafter, completing an app usability questionnaire. Pearson’s whole group correlations were conducted between the NIH scores and the dependent variables of the EMA app to assess construct validity. Inductive qualitative analysis was also conducted on the open-ended usability questions.
Results: Our EMA response rates were 78.8% across all participants demonstrating the app and study design were feasible. Overall, correlations were significant between expected NIH tasks and EMA cognitive tasks. For example, EMA assessed processing speed was negatively correlated with the NIH pattern-comparison task (r[45]= -0.573, p < 0.001), which assesses processing speed. Further, the EMA processing speed task was also negatively correlated with NIH Toolbox card sorting task (r[45]= -0.557, p < 0.001), a measure of executive function. Qualitative analysis highlighted four themes contributing to app usability, namely self-development, altruism, engagement, and functionality.
Conclusions: Our correlation analysis suggests that our novel app tasks have good construct validity. Further we present valuable insights into what participants find important for app usability. Our results show that objective ambulatory measures of cognition are feasible, valid and show acceptable usability in breast cancer survivors and healthy comparison groups alike.