In secondary analyses, we examined the association of the AD PRS (for only the single most significant p-value threshold as selected on the basis of the EEG coherence association findings) with neuropsychological performance, including planning and problem-solving skills on the Tower of London task and visuospatial memory and working memory on the visual span test (forwards span and backwards span), and DSM-5 symptom counts (lifetime) for alcohol, cannabis, cocaine, and opioid use disorder. We also conducted sensitivity analyses in which all models were re-examined with the inclusion of alcohol consumption (i.e., maximum number of drinks consumed in a typical week, past 12 months) and DSM-5 AUD diagnosis in the model to determine if the effects of alcohol consumption and/or AUD were driving associations. Finally, we re-ran the model stratified by those with and without a DSM-5 AUD diagnosis (lifetime) to determine if developmental patterns of PRS-EEG coherence associations differed among these groups.
