Cognitive tests on which performance is unrelated to brain pathology are considered “hold” tests and are often used to estimate cognitive abilities prior to injury or disease. Amongst the most commonly used “hold” tests are measures of irregular word reading, such as the Test of Premorbid Functioning (TOPF). Measures of irregular word reading assess ability to accurately pronounce phonetic irregularities based on prior experience and word knowledge, and tend to be insensitive to most forms of brain pathology (Lezak, 2012). However, research examining whether a relationship exists between neurodegenerative diseases and decline in irregular word reading is limited. The few studies completed have demonstrated a decline in irregular word reading in neurodegenerative disease in general (Berg, Durant, Banks, & Miller, 2016) and Alzheimer’s dementia specifically (McFarlane, Welch, & Rodgers, 2006). However, no known research has been published examining whether irregular word reading and TOPF scores differ depending on cognitive classifications commensurate with DSM-V diagnoses (i.e., mild or major neurocognitive disorder, etc.), or presumed neurological etiology.

Patients were enrolled from the University of Colorado Hospital Neuropsychology Clinic. This study was a retrospective review of consecutive referrals over the age of 65 to the University of Colorado Hospital Neuropsychology Clinic from 2019 to present. The TOPF was administered along with a full neuropsychological battery and patients were clinically classified by severity of cognitive impairment (e.g., Normal, Mild Neurocognitive Disorder, Major Neurocognitive Disorder) and presumed neurologic etiology (e.g., Alzheimer’s disease (AD), Parkinson’s disease (PD), vascular cognitive impairment (VCI), and mixed dementia (AD and VCI). TOPF Raw scores were used for all analyses. Correlation analysis was conducted to determine significant relationships between various demographic variables and TOPF performance. ANCOVA analyses were conducted to examine differences on TOPF performance by diagnostic classification and differences on TOPF performance by presumed neurologic etiology.

Correlation determined a significant relationship between TOPF performance and education (r = .51, p < .001), but not age (p = .092) or gender (p = .680). ANCOVA revealed a significant effect of TOPF performance on diagnostic group classification after controlling for education, F(2, 504)= 26.45, p < .001. Post hoc analysis revealed that those diagnosed with Major Neurocognitive Disorder performed the worst on the TOPF (M=39.801 ± .958), followed by those diagnosed with Mild Neurocognitive Disorder (M= 45.371 ± .767), while those diagnosed as cognitively normal performed the best (M= 49.826 ± .993). Additional ANCOVA analysis revealed a significant effect of TOPF performance on presumed neurologic etiology after controlling for education, F(3,148)=6.07, p = .001. Post hoc analyses revealed that participants with suspected AD (M= 40.728 ± 1.613) and those with suspected VCI (M= 32.804 ± 3.480) performed worse on the TOPF compared to those with suspected PD (M=46.964 ± 1.506), (p=.042 and p = .004, respectively).

Results suggest that TOPF performance in older individuals is sensitive to cognitive impairment. Furthermore, these results suggest that this sensitivity may be further influenced by presumed neurologic etiology. These findings are consistent with prior studies which demonstrated a decline in irregular word reading in individuals with neurodegenerative diseases.