Leung 2015.

Methods	Design: 2‐arm RCT with parallel‐group design Recruitment period: not reported No. of centres involved: 1 Unit of randomisation: individuals No. randomised: not reported Number of arms considered in this review: 2 Maximum trial duration: 3 months Funding by non‐profit organisation: this work was supported by Health and Health Services Research Fund (No. 09100911) Funding by commercial organisation: none reported Publication status: full‐text report
Participants	Patient flow: unclear number randomised; 109 described at baseline in experimental group; 100 described at baseline in control group Number of females: 87 of 109 (80%) in experimental group 1; 77 of 100 (77%) in control group 1 Average age (SD): 70 (6.2) years in experimental group 1; 70 (6.6) years in control group 1 Average (SD) education: 8.71 (3.84) years in experimental group; 9.49 (4.44) years in control group Baseline cognitive function: selection criteria on cognition in experimental group: right‐handed community‐dwelling Chinese older adults at risk of cognitive decline, as indicated by their Montreal Cognitive Assessment (MoCA) scores falling into the range of 19 to 26. MoCA, Hong Kong version, mean total score 23.6 (SD 1.88). Selection criteria on cognition in control group: right‐handed community‐dwelling Chinese older adults at risk of cognitive decline, as indicated by their MoCA scores falling into the range of 19 to 26. MoCA, Hong Kong version, mean total score 23.8 (SD 1.97) Ethnicity: experimental group: white, Indian 109; Asian, Black, other unclear. Control group: white, Indian 100; Asian, Black, other unclear APOE: number of participants positive for APOE not reported
Interventions	Type of experimental intervention: computerised CT group; treatment duration 13 weeks; intervention provided in group format, under supervision Details of experimental intervention: for both CT and AC groups, each participant was assigned a laptop, a headset, and a mouse, all of which were used for performing cognitive exercises. They used the same laptop for their entire training Type of concomitant treatment provided: none reported Session duration: 60 minutes in experimental group Number of treatment sessions: 39 in experimental group Treatment frequency: 3/week in experimental group Maximum treatment duration in weeks: 13 in experimental group Type of control intervention: other; treatment duration 13 weeks; intervention provided in group format, under supervision Details of control intervention: participants in active control group "..were shown educational programs covering diverse topics (e.g., history, science, health information, and local social issues) on a group basis. Immediately after watching the video, they were instructed to answer several questions that were related to the video content" Type of concomitant treatment provided: none reported Session duration: 60 minutes in control group Number of treatment sessions: 39 in control group Treatment frequency: 3/week in control group Maximum treatment duration in weeks: 13 in control group
Outcomes	Cognitive functioning outcomes considered Episodic memory measured with WMS‐III Logical Memory Delayed recall at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit* Working memory measured with Digit Span, total at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit Physical functioning outcome considered: none reported Quality of life outcome considered: none reported Safety outcome considered: none reported Depression outcome considered: none reported Other outcome data on cognitive functioning, not considered in our meta‐analyses Episodic memory measured with WMS‐III Logical Memory Immediate recall at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit* Episodic memory measured with WMS‐III Family Pictures Delayed recall at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit Episodic memory measured with WMS‐III Family Pictures Immediate recall at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit Working memory measured with Digit Vigilance Test at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit Working memory measured with Visual Spatial Span, total at 3.25 months, on a scale from not reported to not reported with higher values indicating benefit
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Judgement: adequate random sequence generation Quote(s): "the 209 participants were randomly assigned to the CT and AC groups by an experimenter blind to the cognitive status of the participants using computer‐generated random sequences of numbers"
Allocation concealment (selection bias)	Unclear risk	Judgement: the method of concealment is unclear as it is not understandable why participants had to be "rearranged", and we suspect that participants were not allocated consecutively, while it remains unclear if allocation of a participant could be foreseen by the researcher Quote(s): "specifically, each participant ID was paired with a random number, and the order of the participants was rearranged based on the value of the assigned number (from smallest to largest)"
Blinding of participants (performance bias)	High risk	Judgement: patients were not blinded to treatment
Blinding of personnel (performance bias)	High risk	Judgement: research assistants could not be blinded, and they both supervised training and performed post‐training assessments Quote(s): "a research assistant was present in each training session to keep track of their attendance and address any questions pertaining to the task instruction raised by the participants. These research assistants were also responsible for conducting the post‐training assessments"
Blinding of outcome assessment (detection bias) All outcomes	High risk	Judgement: therapists, who could not be blinded, supervised training and perform post‐training assessments Quote(s): "a research assistant was present in each training session to keep track of their attendance and address any questions pertaining to the task instruction raised by the participants. These research assistants were also responsible for conducting the post‐training assessments"
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Judgement: reporting is inconclusive. Although according to the flow diagram it seems that 109 were randomised to experimental and 100 to control, we wonder if this merely reflects the number randomised who completed the follow‐up assessment. There is no mention of intent‐to‐treat analyses, missing data, dropouts, or withdrawals, so that we judged risk as unclear Quote(s): "our final sample consisted of 209 older adults (…) who successfully completed the pre‐ and post‐training assessment, of which 109 older adults were randomly assigned to the CT group (…) and 100 older adults were in the AC group".
Selective reporting (reporting bias)	Low risk	Judgement: all outcomes described in the methods section are adequately addressed in the results section
Other bias	Unclear risk	Judgment: the selection process is not clear because study authors do not indicate the number of participants actually screened, those excluded, and reasons for exclusion. It is not clear whether inclusion was consecutive, and study authors mention that baseline characteristics were "matched". With the latter, we assume they meant "comparable" Quote(s): "participants from the CT and AC groups were matched for their demographic characteristics"