S1 Data -

Aim
To investigate the effectiveness of a semantic-based memory-encoding strategy intervention and cognitive stimulation that enhances function for older adults with mild cognitive impairment.
Methods
A two-armed single-blind non-randomised cluster controlled trial was conducted. Participants in two centres were allocated to the semantic-based memory-encoding experimental group and those in the other two centres received cognitive stimulation. In both groups, one centre- or community-based session and one home-based session were provided weekly for 10 weeks. Outcome measures included attention, memory and general cognitive function (Word List Memory and Word List Recall of the Consortium to Establish a Registry for the Alzheimer’s disease, Digit Span Forwards and Backwards and the Cognistat), and daily task performance (Disability Assessment for Dementia and Lawton Instrumental Activities of Daily Living Scale). They were administered pre- and post-intervention.
Results
Thirty-nine participants completed the study. No significant differences were revealed in the demographic or baseline data. The experimental group showed significant improvements in daily task performance (Disability Assessment for Dementia; p = 0.003), memory outcomes (Word List Recall; p < 0.001), general cognitive function (Cognistat subtests of Memory and Similarity; ps = 0.002 and < 0.001). The cognitive stimulation control group did not show any significant improvement in the measures. Between-group analysis showed significant differences in favour of the experimental group for the outcome measures of the Word List Recall and Cognistat Similarity subtest (ps < 0.001).
Conclusions
This study shows that the semantic-based memory-encoding strategy is more superior than cognitive stimulation with improvements in attention, memory, general cognitive function and daily task performance for people with a mild cognitive impairment.
Trial registration
ClinicalTrials.gov Protocol Registration and Results System (NCT02953964).