Cognitive intra-individual variability as an outcome or moderator of speed of processing training in aging adults with HIV-associated neurocognitive disorder: A secondary data analysis of a 2-year longitudinal randomized clinical trial

Objective: Cognitive intra-individual variability (IIV), dispersion of cognitive performance across cognitive domains, indicates diminished cognitive control resulting from brain pathology or aging. For people with HIV-Associated Neurocognitive Disorder (HAND), cognitive IIV may interfere with and/or be improved by cognitive training. In this current secondary data analysis, we explored whether cognitive IIV functions as an outcome or moderator variable in speed of processing (SOP) training in middle-aged and older people living with HIV (PLWH). Method: In this randomized clinical trial, 216 PLWH with HAND or borderline HAND were randomized to either: 1) 10 h of SOP training; 2) 20 h of SOP training, or 3) 10 h of Internet Navigation Control Training. Participants completed a seven-domain cognitive battery at baseline, posttest after training, and years 1 and 2 follow up; from this, the coefficient of variation (CoV) served as the cognitive IIV dispersion score. Participants also completed a gold-standard measure of SOP (i.e., Useful Field of View Test (UFOV®), a targeted cognitive outcomes of the SOP training paradigm) at each time point. Results: Using linear mixed-effect models, no strong pattern of SOP training effects were detected on cognitive IIV. However, a three-way interaction tests between time, group, and baseline cognitive IIV CoV suggested potential moderation effects. Stratified analyses showed differential treatment effects between the lower cognitive IIV and higher cognitive IIV group. Conclusions: Although SOP training did not improve cognitive IIV, cognitive IIV may be a modifier of SOP training. Implications for future research and other types of training are provided.