The study, carried out by Dr Chloe Fawns-Ritchie and Professor Ian Deary of the University of Edinburgh, aimed to assess whether researchers could have confidence in the quality of the UK Biobank cognitive test data.
Dr Fawns-Ritchie, questionnaire officer for the Generation Scotland cohort, based at the University of Edinburgh, said: ‘UK Biobank is a health resource with data from over half a million adults. At each study visit, UK Biobank participants completed a short and unsupervised cognitive assessment on a touchscreen computer. Many of the cognitive tests were bespoke measures designed specifically for UK Biobank. Although the cognitive test data is widely used, psychometric information on the UK Biobank tests is limited.
‘We set out to investigate the reliability and validity of the UK Biobank cognitive tests – in other words, their consistency and accuracy. We recruited 160 participants who completed the UK Biobank cognitive tests and a range of well-validated comparable neuropsychological tests that are thought to assess the same underlying cognitive abilities.’
The participants were recruited from the University of Edinburgh’s volunteer panel and Join Dementia Research, and had not previously taken the UK Biobank cognitive tests.
Dr Fawns-Ritchie added: ‘We found that some of the UK Biobank cognitive tests showed good construct validity (that is, they correlated highly with comparable neuropsychological tests) and good test-retest reliability (scores on the same test one month apart were highly correlated). For other tests, the reliability and validity was moderate.
‘These results provide previously lacking information for researchers on the reliability and validity of the UK Biobank cognitive tests.’
The research was conducted as part of DPUK’s cognitive assessment work package.
Read the full study in the journal PLOS ONE.
Access DPUK's cognitive test guides and resources for researchers.