Comparison of approaches to control for intracranial volume in research on the association of brain volumes with cognitive outcomes

Background Consistent methods for evaluating the link between brain structure and cognition are essential for understanding determinants of neurologic outcomes. Studies examining associations between brain volumetric measures and cognition use various statistical approaches to account for variation in intracranial volume (ICV). It is unclear if commonly used approaches yield consistent results. Methods Using a brain-wide association approach in the MRI substudy of UK Biobank (N=41,964; mean age=64.5 years), we used regression models to estimate the associations of 58 regional brain volumetric measures with eight cognitive outcomes, comparing no correction and five ICV correction approaches. Approaches evaluated included: no correction; dividing regional volumes by ICV, with and without further adjustment for ICV (proportional approach); including ICV as a covariate in the regression (adjustment approach); and regressing the regional volumes against ICV in different normative samples and using calculated residuals to determine associations (residual approach). We used Spearman-rank correlations and two consistency measures to quantify the extent to which associations were inconsistent across ICV correction approaches for each possible brain region and cognitive outcome pair across 2,784 regression models. Findings The adjustment and residual approaches typically produced similar estimates, which were inconsistent with results from the crude and proportional approaches. Inconsistencies across approaches were largest when estimates from the adjustment and residual approaches were further from the null. That is, the approach used was least important when the association between brain volume and cognitive performance was close to null; in this case, all approaches tend to estimate a null association. Interpretation Commonly used methods to correct for ICV yield inconsistent results and the proportional method diverges from other methods. Adjustment methods are the simplest to implement while producing biologically plausible associations.