28: Implementation of Nonlinear Z-score Analysis for Cognitive Abnormality Detection

Music City Center 

Traditional linear methods for generating age, sex, and education level corrected Z-scores in neuropsychological assessments can be problematic because of nonlinearity and bounded test scores. We propose a nonlinear censored regression model for generating Z-scores that adjusts for age, sex, education, and race, while incorporating age-varying residual standard deviations. This approach addresses non-normal score distributions and boundary censoring, enhancing the detection of abnormal cognitive performance. Application to diverse normative datasets demonstrates improved accuracy and sensitivity over traditional methods, as corroborated by clinician feedback. Our results advocate for adopting this model to refine neuropsychological evaluations across varied populations.

Neuropsychological Testing


Z-Score Adjustment

Censored Regression

Nonlinear Modeling

Cognitive Assessment 

Mental Health Statistics Section