Development and first-stage validation of a digital version of the Digit Symbol Substitution test for use in assessing cognitive function in older people with diabetes

Omri Segev*, Itamar Raz, Hertzel C. Gerstein, Hillel Aviezer, Yael Sela, Dani Cukierman, Rahul Shankar, Rachel Natovich, Tali Cukierman-Yaffe

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Aims: To describe the development and report the first-stage validation of a digital version of the digit symbol substitution test (DSST), for assessment of cognitive function in older people with diabetes. Materials and Methods: A multidisciplinary team of experts was convened to conceptualize and build a digital version of the DSST and develop a machine-learning (ML) algorithm to analyse the inputs. One hundred individuals with type 2 diabetes (aged ≥ 60 years) were invited to participate in a one-time meeting in which both the digital and the pencil-and-paper (P&P) versions of the DSST were administered. Information pertaining to demographics, laboratory measurements, and diabetes indices was collected. The correlation between the digital and P&P versions of the test was determined. Additionally, as part of the validation process, the performance of the digital version in people with and without known risk factors for cognitive impairment was analysed. Results: The ML model yielded an overall accuracy of 89.1%. A strong correlation was found between the P&P and digital versions (r = 0.76, p < 0.001) of the DSST, as well as between the ML model and the manual reading of the digital DSST (r = 0.99, p < 0.001). Conclusions: This study describes the development of and provides first-stage validation data for a newly developed digital cognitive assessment tool that may be used for screening and surveillance of cognitive function in older people with diabetes. More studies are needed to further validate this tool, especially when self-administered and in different clinical settings.

Original languageAmerican English
JournalDiabetes, Obesity and Metabolism
DOIs
StateAccepted/In press - 2024

Bibliographical note

Publisher Copyright:
© 2024 The Authors. Diabetes, Obesity and Metabolism published by John Wiley & Sons Ltd.

Keywords

  • cognition
  • cognitive-impairment
  • diabetes
  • digit-symbol-substitution test
  • elderly

Fingerprint

Dive into the research topics of 'Development and first-stage validation of a digital version of the Digit Symbol Substitution test for use in assessing cognitive function in older people with diabetes'. Together they form a unique fingerprint.

Cite this