Background: Attention is essential to daily life and cognitive functioning, and attention deficits can affect daily functional and social behaviour, such as falls, risky driving, and accidental injuries. However, attention function is important yet easily overlooked in older adults with mild cognitive impairment, and evidence is limited. We aimed to explore the pooled effect of cognitive training on domains of attention in older adults with mild cognitive impairment and mild dementia using a meta-analysis of randomised controlled trials.

Methods: We searched PubMed, Embase, Scopus, Web of Science, CINAHL, PsycINFO, and Cochrane Library for randomised controlled trials (RCTs) up to 3 November 2022. We included participants aged ≥50 years diagnosed with cognitive impairment, with various cognitive training interventions as the intervention measures. The primary outcome was overall attention and the secondary outcomes were attention in different domains and global cognitive function. We calculated the Hedges' g and confidence intervals (CIs) using a random-effects model to evaluate the effect size of the outcome measures and evaluated heterogeneity using the χ(2) test and I(2) value.

Results: We included 17 RCTs and found that cognitive training interventions improve overall attention (Hedges' g = 0.41; 95% CI = 0.13, 0.70), selective attention (Hedges' g = 0.37; 95% CI = 0.19, 0.55), divided attention (Hedges' g = 0.38; 95% CI = 0.03, 0.72), and global cognitive function (Hedges' g = 0.30; 95% CI = 0.02, 0.58) in older adults with mild cognitive impairment, but with relatively low effectiveness.

Conclusions: Cognitive training intervention can improve some attention functions in older adults with mild cognitive impairment. Attention function training should also be incorporated into routine activities and long-term sustainability planning to delay the deterioration of attention function in older adults. Besides reducing their risk of abnormal events in daily life (such as falls), it can also improve their quality of life and help reduce the progression of cognitive impairment, achieving early detection of secondary prevention. REGISTRATION: PROSPERO (CRD42022385211).