P3-039: exploring the relationship between iq and other cognitive scores

The relationship between measures of intelligence (IQ) and neuropsychological test performance has been well studied. However there is limited data in adults on whether IQ is predictive of changes in cognition over 15 years. All WHAP participants who attended baseline cognitive assessment and had at least two cognitive assessments within 8 years follow-up window were included in this analysis. Their baseline and follow-up demographic and clinical data as well as cognitive and IQ test results, presence of APOE4 and menopause status were collected. IQ was estimated using 2002 Wechsler Adult Intelligence Scale (WAIS-III); cognition was also assessed using CERAD delay, CVLT immediate and long delay, verbal episodic memory and executive function composite scores. Changes in cognitive scores were calculated as the difference between baseline and follow-up scores. Regression models were created to evaluate the impact of IQ on cognition both cross-sectionally and longitudinally with the adjustment for age, education and presence of APOE4. After applying participants’ selection algorithm outlined in the methodology section, the data from 385 WHAP participants (age 50–60 at baseline) were included in this analysis. Higher IQ scores were associated with higher scores on other cognitive tests at baseline, with the exception of the executive function (EF) composite score. 2002 IQ was predictive of performance on all cognitive measures, including EF, 10–12 years later, while controlling for age, education and APOE4 (p<0.05). In addition, a strong positive association was observed between IQ and EF changes over time (β =0.10, 95%CI 0.03−0.17, p=0.007) after the adjustment for baseline age, education and APOE4. Over and above the education level and other potential confounders, women with high IQ have better cognitive function and show continuous improvement in executive function in particular. Further analysis will investigate the association between IQ and modifiable risk factors and whether the association between IQ and cognition is mediated by these risk factors as recently highlighted by the Lancet commission.