Detection of neurodevelopmental diversity in memory clinics—Validation of a self-report measure

Highlights

• Childhood cognitive difficulty is common in adults with memory disorders.

• Undiagnosed LD/ADHD may increase susceptibility to age-related neurodegeneration.

• We propose a self-report measure to elicit past childhood cognitive difficulties.

• Discriminant analysis supported our model’s predictive validity.

Abstract

Background

Neurodevelopmental learning and attentional disorders (NLAD) such as dyslexia, dyscalculia and attention deficit hyperactivity disorder (ADHD) affect at least 6% of the adult population or more. They are associated with atypical cognitive patterns in early and adult life. The cognitive patterns of affected individuals in late life have never been described. One main challenge is detecting individuals in clinical settings during which mild cognitive changes could be confounding the clinical presentation. This is a critical research gap because these conditions interact, across the life course, with an individual’s risk for dementia. Also, learning disabilities which present in childhood pose persistent cognitive differences in areas involving executive function, reading and math.

Clinicians lack tools to detect undiagnosed neurodevelopmental in adults with memory disorders. The majority of patients presenting at memory clinics today come from a generation during which NLAD were not yet clinically recognized. In this study, we hypothesized that a self-report scale can detect NLAD in a memory clinic population.

Methods

We developed a self-report, retrospective childhood cognitive questionnaire including key attributes adapted from prior validated measures. 233 participants were included in the primary analysis.

Results

Confirmatory Factor Analysis resulted in a best-fit model with six labelled factors (Math, Language, Attention, Working Memory, Sequential Processing, and Executive Function) and 15 total question items. The model demonstrated unidimensionality, reliability, convergent validity, discriminant validity, and predictive validity. Using 1.5 standard deviations as the cut-off, subjects were categorized into: Normal (n = 169), Language (n = 10), Math (n = 12), Attention (n = 10) or Other/Mixed (n = 32).

Conclusion

A self-report measure can be a useful tool to elicit childhood cognitive susceptibilities in various domains that could represent NLAD among patients in a memory clinic setting, even in the presence of mild cognitive impairment.