Successful delivery of CANTAB cognitive assessments across diverse health systems

At the virtual AAIC 2020 conference Dr Elizabeth Baker presented data on the successful delivery of CANTAB^T^Massessments across diverse health systems and ages: 10 – 90 years old.

The study captures normative performance across the lifespan to support a cognitive screening tool for characterising a range of neuropsychological disorders in clinics across India.

Background

The objective assessment of cognitive function is an important component for the early detection and differential diagnosis of Alzheimer’s disease and dementia. Near-patient assessments in the clinic, community or at home using digital health tools can meet requirements of patient characterisation in dementia and facilitate assessment in a range of resource settings.

These tools, including cognitive screening tools, have been deployed in the community, primary and secondary health care settings in relatively resource-rich healthcare settings (primarily UK and US). In partnership with Eris Lifesciences Limited, we are now investigating this approach in India.

To support the development of a cognitive screening tool, a first step in this partnership, is the assessment of healthy participants. Here we present the performance of over 4000 healthy participants assessed on CANTAB^TM in various clinical settings across India.

Methods

Participants were recruited across clinical centres in India, with voice-over instructions translated into 9 Indian languages to support assessment delivery. To best capture normative performance healthy participants were targeted for recruitment across ages 10 – 90 years old and across a range of educational attainments.

CANTAB^TM tests including Paired Associates Learning (PAL), Spatial Working Memory (SWM) and Matching to Sample (MTS) were chosen as they could be delivered as a short battery in 20 minutes and capture a range of domains including episodic memory, working memory and executive function, psychomotor speeds and attention.

Participants were assessed using Geriatric Depression Scale (GDS-15) to measure participant mood and asked to self-report their memory performance.

To understand the consistency of our assessments across platforms of delivery, we compared performance to that of our online normative data collection in 2017.

Result

4970 participants aged 10-90 years with average age 41.2 years were assessed on CANTAB^TM between September 2019 and May 2020; with a 97.8% completion rate across all tests; only 1% did not complete any CANTAB^TM assessments.
The participants were aged 10-90 with average age 41.2 years (SD 18.0 years); highly educated individuals made up a high proportion of the normative sample (46.1%), although participants also left school aged 18 or before (32.0%) or we unable to read (21.9%).
Age-related decline in episodic memory, working memory, psychomotor speeds and attention were as expected for adults.
Not all domains showed education to be important. Those unable to read (Education: No) and leaving school before the age of 18 (Education: Low) showed poorer performance on episodic memory, working memory and attention, relative to those with undergraduate or higher degrees and equivalent qualifications (Education: High).
Executive function performance did not change with age, and showed small differences between males and females.
When comparing the performances trends between this normative collection and the online normative collection, similar trends in performance occurred by age, gender and education attainment.
Differences in performance existed between the two collections; participants within the online assessment performed with lower errors on episodic memory (Figure 1) and working memory and showed better executive function. This may be attributed to the selection bias due to recruitment through Prolific academic and will be explored as part of the next phase of development.

Figure 1. Episodic Memory performance in the Indian Cognition Study and Online normative collection

Conclusions

This study presents the normative database to support the development of a cognitive screening tool in clinics across India.
This study supports that our assessments can be consistently delivered across a range of platform and delivery settings.
For effective normative evaluations, the recruitment setting is an important consideration and supports the requirement to develop a project-specific normative database.

View poster

Contact us

Dr Elizabeth Baker, Director of Statistical Sciences, Cambridge Cognition.