Cognition Over Time: Comparing Single Timepoint and Time Series Based Evaluations of Cognitive Functioning in the Glucog Study

Lanee Jung, BA

McLean Hospital – Research Assistant
Jung_Lanee poster

Scientific Abstract

Background: Variables that impact cognitive functioning in Type I Diabetes (T1D). The integration of findings based on temporal dynamic assessments is an ambitious undertaking, and thus research on fluctuations on cognitive functioning over time have typically only included a few timepoints. The objective of this study is to characterize different patterns of cognitive variability in T1D and determine mediators and moderators of this relationship. Although the overall study will look at the specific relationship between glycemic excursions, here we use initial data to characterize distinct patterns of cognitive functioning that are only apparent in time series data. 

Methods: For this pilot study, adults (N=20) with T1D (>1 year) completed daily ecological momentary assessments (EMA) of cognition, sleep, mood and anxiety. EMA permits modeling the dynamic interactions of short-term variability (over hours-days) in cognitive performance, glycemia, and patient reported data. Individuals completed up to 60 assessments over 10 days. All cognitive EMAs were developed through TestMyBrain.org using an iterative test development procedure to validate ultra-brief assessments of cognition on mobile devices (Germine et al., 2020).

Results: We demonstrate multiple approaches to quantify cognitive function based on these data: (1) in terms of mean performance, averaged across time points to reduce temporal sampling error, (2) variability in performance, expressed as the coefficient of variability in test scores, and (3) differences in the slopes (change over time) estimated over cognitive time series, reflecting differential learning and practice effects. These different metrics are thought to reflect different aspects of cognitive performance, and potentially distinct mechanisms.

Conclusions: The utilization of such multi-time-point approaches can provide comprehensive information regarding daily cognitive functioning of individuals with T1D and increases opportunities to guide treatment planning in a more evidence-based manner. It can also lead to an empirically supported self-management tool for tracking cognitive status that can be rapidly translated to clinical care.

Live Zoom Session – March 9th

research Areas

Authors

Laneé Jung, Eliza Passell, Roger Strong Ph.D., Elizabeth Grinspoon Ph.D., Shifali Singh Ph.D., Luke Scheuer, Martin J Sliwinski Ph.D., Naomi S Chaytor Ph.D., and Laura Germine Ph.D.

Principal Investigator

Laura Germine Ph.D.