Effects of different modes of computerised cognitive training in older adults with reversible cognitive frailty: a randomised controlled trial in nursing homes

We aimed to evaluate the effects of computerised cognitive training (CCT) on cognitive functions in older adults with reversible cognitive frailty (RCF) and compare memory-focused training (MT) with multi-domain cognitive training (MDCT). In this double-blind, CONSORT-compliant randomised controlled trial (RCT), 52 older adults with RCF living in a nursing home were stratified by frailty status (pre-frail/frail) and randomly allocated to MT (active control) or MDCT for 12 weeks (two sessions per week). Outcomes were assessed at baseline and 3 months (post-intervention), and 4 months (1-month follow-up), using generalised estimating equations (GEE) with intention-to-treat analysis. Both groups demonstrated substantial and statistically significant improvements across all cognitive measures over time (Time effect: P < 0.001). Montreal Cognitive Assessment (MoCA) scores increased significantly (Wald χ² = 20.211, P < 0.001), with a mean gain of 1.02 points from baseline to 4 months (95% CI: 0.46–1.58, P < 0.001, effect size r = 0.369), with no significant between-group differences (P = 0.568). Subjective Cognitive Decline (SCD) scores also improved significantly (Wald χ² = 73.374, P < 0.001), with a mean reduction of 2.71 points (95% CI: 2.03–3.39, P < 0.001). STT-A and B completion times decreased markedly by averages of 27.57 and 92.99 s, respectively. Notably, for STT-A, performance was significantly faster in the MT group compared to the MDCT group (64.79 vs. 80.46 s, P = 0.038). regarding STT-B, although the interaction was not significant, the MDCT group showed a substantial magnitude of recovery. Clock Drawing Test (CDT) scores improved (Wald χ² = 11.229, P = 0.004), with a mean increase of 0.21 points (95% CI: 0.06–0.36, P = 0.002). The MDCT group showed marginally higher CDT scores, though not statistically significant (P = 0.068). Both MT and MDCT were feasible and yielded comparable short-term global cognitive gains. However, they exhibited distinct therapeutic profiles where MT facilitated transfer effects in processing speed while MDCT effectively restored executive function. Concurrent improvements in sleep and mood were also observed. These findings support a flexible implementation strategy based on available resources, allowing clinicians to select the modality that best targets an individual’s specific cognitive profile. This study was registered with the Chinese Clinical Trial Registry on 8 February 2023 (ChiCTR2300068132).