Objective: Premorbid estimates of intellectual functioning are a key to assessment. This study aimed to compare 3 common measures and assess their accuracy: the Test of Premorbid Functioning (TOPF), Oklahoma Premorbid Intelligence Estimate (OPIE-3), and what is commonly referred to as the Barona equation. We also sought to provide appropriate adjustment considering the Flynn effect.
Method: The sample consisted of a cross-section of 189 outpatient veterans receiving neuropsychological assessment including the TOPF and Wechsler Adult Intelligence Scale, 4th ed. (WAIS-IV). Paired sample t tests assessed differences between IQ models. Correlations for all models and actual WAIS-IV Full Scale IQ (FSIQ) to establish which model best predicted variance in current IQ. Mean differences were evaluated to establish how closely the models approximated WAIS-IV FSIQ.
Results: The Barona equation estimated higher premorbid IQ than TOPF Simple Demographics Model; however, differences between the models were nonsignificant after a Flynn effect correction for the Barona equation (.23 IQ points per year). The OPIE-3 correlated with FSIQ but overestimated the FSIQ, demonstrating the Flynn effect. TOPF performance models (include word reading) characterized the variance of IQ scores best, but the Flynn-adjusted Barona equation had the smallest mean difference from the actual WAIS-IV FSIQ of any prediction model.
Conclusion: Demographic models for premorbid IQ accurately estimate IQ in adult populations when normed on the test used to measure IQ, or when adjusted for the Flynn effect. A Flynn-corrected Barona score provided a more accurate estimation of WAIS-IV FSIQ than the TOPF or the OPIE-3. (PsycINFO Database Record (c) 2020 APA, all rights reserved).