O4.3. SHARED AND DIFFERENTIAL NEUROCOGNITIVE EFFECTS COMPARING HIGH-INTENSITY INTERVAL TRAINING AND ACTIVE VIDEO-GAMING IN SCHIZOPHRENIA – RESULTS FROM AN RCT

BACKGROUND: Studies on the efficiency and efficacy of aerobic exercise (AE) as a neurocognitive intervention in schizophrenia are promising, but there is a need for larger and well-designed randomized controlled trials (RCT) to draw more certain conclusions. The aim of this RCT was to study the effect of high-intensity interval training (HIIT), on neurocognitive functions in schizophrenia. A comparison group performed active video-gaming (AVG) using Nintendo WII sport. We hypothesized some effect in both groups, but anticipated that the effect of HIIT would reach beyond AVG. However, in recent studies AVG interventions with perceptual and motor-coordination demands were found to have neurocognitive effects equivalent to AE, challenging our hypothesis and adding new relevance to our study. To our knowledge, this is the first study to compare the effects of HIIT and AVG on neurocognition in schizophrenia. METHODS: The current study was a single-blinded RCT. Eighty-two participants (61% males, median age 32 (24.3)) diagnosed with schizophrenia were recruited from outpatient clinics at Vestfold Hospital Trust and randomly allocated to HIIT (n=43) or AVG (n=39). Both interventions consisted of 45 minutes supervised sessions two times per week for 12 weeks. The primary outcome neurocognition was measured with the MATRICS consensus cognitive battery by blinded assessors at baseline, post-intervention and follow-up. In sum 88% were protocol-compliant and attrition rate was 30.5%, resulting in 65% protocol-compliant study-completers (HIIT n=25/AVG n=28). Intervention effects were estimated using repeated linear mixed model analysis (LMM): Intention-to-treat (ITT) analyses of all randomized participants and per-protocol (PP) analyses of protocol-compliant study-completing participants. Per-protocol between-group effect sizes (Cohen’s d) and within-group pretest-posttest effect sizes were calculated. RESULTS: We present post-intervention results. In the between-group analysis (ITT and PP) there was a main effect of time (EST 2.71, 95% CI: 1.45–3.98, p <.001), but no significant interaction effect of group and time (EST 0.36, 95% CI: -1.45–2.17, p=.696), for the neurocognitive composite T-score. The same pattern of results was found in the subdomains, indicating an increase in functioning over time for both groups on several subdomains. Per-protocol effect-sizes indicated a small to medium effect on verbal learning favoring HIIT (Cohen’s d =.46). Within-group analysis (PP) and pre-post effect sizes (PP) indicated small effects from both interventions on the neurocognitive composite score and the subdomains attention and working memory. There was a small effect on speed from HIIT. In the learning domain, HIIT had significant medium effect on verbal learning while AVG had a significant small effect on visual learning. DISCUSSION: Contrary to our hypothesis, the main analyses did not indicate superiority of HIIT compared to AVG on neurocognitive functions. Despite this, effect sizes indicated a small/moderate effect on verbal learning favoring HIIT. Within-group analysis showed increase in the neurocognitive composite score and the subdomains attention and working memory following both interventions. There was a small increase in speed following HIIT. Differential effects within the learning domain as indicated by increased verbal learning following HIIT and increased visual learning following AVG was unexpected, but replicate studies showing differential neurocognitive effects from different training modes. Our results show shared and differential positive effects from the two exercise modes HIIT and AVG on neurocognition. Both show potential as clinical interventions targeting neurocognition in schizophrenia.