INS NYC 2024 Program

Poster	Poster Session 04 Program Schedule 02/15/2024 12:00 pm - 01:15 pm Room: Majestic Complex (Posters 61-120) Poster Session 04: Neuroimaging \| Neurostimulation/Neuromodulation \| Teleneuropsychology/Technology Final Abstract #84 Digital Cognitive Assessment in People with Neurofibromatosis 1: Reliability of Unsupervised Self-Administered Tests for Research Purposes Jennifer Nosker, Massachusetts General Hospital, Boston, United States Johanna Brait, Massachusetts General Hospital, Boston, United States Angela Chen, Chicago Medical School, Rosalind Franklin University of Science and Medicine, North Chicago, United States Justin Jordan, Massachusetts General Hospital, Boston, United States James Walker, Massachusetts General Hospital, Boston, United States Elizabeth Klerman, Massachusetts General Hospital, Boston, United States Richa Saxena, Center for Genomic Medicine, Massachusetts General Hospital, Broad Institute, Boston, United States Michael Parsons, Massachusetts General Hospital, Boston, United States Category: Assessment/Psychometrics/Methods (Adult) Keyword 1: neurofibromatosis Keyword 2: computerized neuropsychological testing Objective: The utility of self-administered digital cognitive tests has increased in recent years due to improved standardization, flexibility of test administration, and enhanced ecological validity. Digital assessments have been developed using healthy control subjects, but their reliability and validity in clinical populations such as patients with neurofibromatosis 1 (NF1), has not been evaluated. Clinical features of NF1 (i.e., peripheral nerve tumors, skeletal muscle and motor deficits, gliomas) and cognitive conditions (i.e., attention-deficit/hyperactivity disorder, learning disabilities) may interfere with test performance. The objective of this study was to investigate the reliability of a digital testing platform to assess specific cognitive abilities in research participants with NF1. Participants and Methods: These data were collected as part of a case-control observational study investigating relationships among sleep, pain, and cognition in adults with NF1. Participants completed study procedures daily for two weeks from their home, including actigraphy/sleep monitoring, patient reported outcome measures, and cognitive assessment with a battery of tests from the TestMyBrain.org (TMB) digital cognitive platform. On the initial and final days of the study period, (days 1, 14) subjects completed a ~15 minute battery including the Gradual Continuous Performance Task-150s version (CPT), Digit Symbol Matching-90s version (DSM), Choice Reaction Time-30s version (CRT), Verbal Paired Associates (VPA), and Trail Making Test (TMT). On days 2-13, a 5 minute battery including shorter versions of the CPT, DSM, and CRT was completed. Interclass correlation coefficients (ICC) were calculated to examine test-retest reliability for each task based on a mean rating (k=2), absolute-agreement, 2-way mixed effects model. Repeated measures ANOVAs were conducted to investigate learning effects on CPT, DSM, and CRT. Results: The sample included 48 people with NF1, ages 24 to 74 years old (mean age=43). Most were White (83.3%) and women (79.2%). Tests administered consecutively for 14 days revealed excellent reliability for DSM score and CRT median reaction time for correct trials (mRTC), with ICC’s greater than or equal to 0.90, and good reliability for CPT score (0.89). Reliability coefficients for tests administered only on Days 1 and 14 were lower and in the moderate range for VPA (0.63) and TMT B (0.61). Poor reliability was observed on TMT A (0.20), which was expected as the test is sensitive to practice effects. Repeated-measures ANOVAs showed significant learning effects for DSM score, F(1, 44)=13.097, p=<0.001) and CRT MRTc, F(1, 35)=8.230, p=<0.001. CPT score showed significant variability across testing days, F(1, 33)=8.121, p=<0.001, but a learning effect was not observed between the first and last administrations. Conclusions: We found moderate to excellent reliability for unsupervised cognitive outcome measures in subjects with NF1 as well as learning effects for DSM and CRT. Reliability coefficients were consistent with published TMB metrics for the general population. Overall, the results suggest that digital cognitive assessments can be reliably administered to individuals with NF1, allowing for increased integration of cognitive outcomes in NF1 research conducted remotely.

Poster

02/15/2024
12:00 pm - 01:15 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 04: Neuroimaging | Neurostimulation/Neuromodulation | Teleneuropsychology/Technology

Final Abstract #84

Digital Cognitive Assessment in People with Neurofibromatosis 1: Reliability of Unsupervised Self-Administered Tests for Research Purposes

Jennifer Nosker, Massachusetts General Hospital, Boston, United States
Johanna Brait, Massachusetts General Hospital, Boston, United States
Angela Chen, Chicago Medical School, Rosalind Franklin University of Science and Medicine, North Chicago, United States
Justin Jordan, Massachusetts General Hospital, Boston, United States
James Walker, Massachusetts General Hospital, Boston, United States
Elizabeth Klerman, Massachusetts General Hospital, Boston, United States
Richa Saxena, Center for Genomic Medicine, Massachusetts General Hospital, Broad Institute, Boston, United States
Michael Parsons, Massachusetts General Hospital, Boston, United States

Category: Assessment/Psychometrics/Methods (Adult)

Keyword 1: neurofibromatosis
Keyword 2: computerized neuropsychological testing

Objective:

The utility of self-administered digital cognitive tests has increased in recent years due to improved standardization, flexibility of test administration, and enhanced ecological validity. Digital assessments have been developed using healthy control subjects, but their reliability and validity in clinical populations such as patients with neurofibromatosis 1 (NF1), has not been evaluated. Clinical features of NF1 (i.e., peripheral nerve tumors, skeletal muscle and motor deficits, gliomas) and cognitive conditions (i.e., attention-deficit/hyperactivity disorder, learning disabilities) may interfere with test performance. The objective of this study was to investigate the reliability of a digital testing platform to assess specific cognitive abilities in research participants with NF1.

Participants and Methods:

These data were collected as part of a case-control observational study investigating relationships among sleep, pain, and cognition in adults with NF1. Participants completed study procedures daily for two weeks from their home, including actigraphy/sleep monitoring, patient reported outcome measures, and cognitive assessment with a battery of tests from the TestMyBrain.org (TMB) digital cognitive platform. On the initial and final days of the study period, (days 1, 14) subjects completed a ~15 minute battery including the Gradual Continuous Performance Task-150s version (CPT), Digit Symbol Matching-90s version (DSM), Choice Reaction Time-30s version (CRT), Verbal Paired Associates (VPA), and Trail Making Test (TMT). On days 2-13, a 5 minute battery including shorter versions of the CPT, DSM, and CRT was completed. Interclass correlation coefficients (ICC) were calculated to examine test-retest reliability for each task based on a mean rating (k=2), absolute-agreement, 2-way mixed effects model. Repeated measures ANOVAs were conducted to investigate learning effects on CPT, DSM, and CRT.

Results:

The sample included 48 people with NF1, ages 24 to 74 years old (mean age=43). Most were White (83.3%) and women (79.2%). Tests administered consecutively for 14 days revealed excellent reliability for DSM score and CRT median reaction time for correct trials (mRTC), with ICC’s greater than or equal to 0.90, and good reliability for CPT score (0.89). Reliability coefficients for tests administered only on Days 1 and 14 were lower and in the moderate range for VPA (0.63) and TMT B (0.61). Poor reliability was observed on TMT A (0.20), which was expected as the test is sensitive to practice effects. Repeated-measures ANOVAs showed significant learning effects for DSM score, F(1, 44)=13.097, p=<0.001) and CRT MRTc, F(1, 35)=8.230, p=<0.001. CPT score showed significant variability across testing days, F(1, 33)=8.121, p=<0.001, but a learning effect was not observed between the first and last administrations.

Conclusions:

We found moderate to excellent reliability for unsupervised cognitive outcome measures in subjects with NF1 as well as learning effects for DSM and CRT. Reliability coefficients were consistent with published TMB metrics for the general population. Overall, the results suggest that digital cognitive assessments can be reliably administered to individuals with NF1, allowing for increased integration of cognitive outcomes in NF1 research conducted remotely.