Poster Session E, Monday, March 25, 2:30 – 4:30 pm, Pacific Concourse
Predictive validity of word reading tests for estimating premorbid IQ
Peter Bright1,2, Izobel Clegg1, Farah Hina1, Ian van der Linde2,3; 1Department of Psychology, Anglia Ruskin University, Cambridge, UK, 2Vison & Eye Research Unit (VERU), School of Medicine, Anglia Ruskin University, 3Department of Computing & Technology, Anglia Ruskin University
The 50-item National Adult Reading Test (NART; Nelson, 1982; NART–R; Nelson & Willison, 1991), first published in 1982, remains a widely used method for estimating premorbid intelligence in neurological patients. However, using 92 neurologically healthy British participants, it was only recently restandardised against the most recent revision of the Wechsler Adult Intelligence Scale (WAIS-IV; Bright et al., 2018). In the current study we analysed NART and WAIS-IV data in a larger sample (N = 145) to assess reliability of our published estimates, and address whether inclusion of demographic data might improve precision of estimates. We also compared NART estimates in a smaller sample (N=45) with those derived from the 70-item UK version of the Test of Premorbid Functioning (TOPF, 2011). In the larger sample, NART-FSIQ linear correlations remained strong (although marginally lower) and the regression equations produced estimates within one FSIQ point of our published figures. Inclusion of age yielded a modest but significant improvement (from R2=.43 to R2=.45). In our smaller sample the NART outperformed the TOPF, indicating that inclusion of more test items in this case does not lead to better precision of FSIQ estimates. Application of a recent genetic algorithm approach for optimising the relationship between neuropsychological test data (van der Linde & Bright, 2018) reinforces this view, showing not only that more than half the stimuli included in the reading tests provide minimal or no diagnostic contribution, but that FSIQ estimates can be substantially improved while reducing arduousness and test duration for patients.
Topic Area: METHODS: Other