| Methods |
Design: 3‐arm randomised controlled trial with parallel‐group design Recruitment period: 2009 to 2009 No. of centres involved: not reported Unit of randomisation: individuals No. randomised: 6742 Number of arms considered in this review: 3 Maximum trial duration: 6 months Funding by non‐profit organisation: This research was funded by Alzheimer’s Society UK and Medical Research Council Funding by commercial organisation: none reported Publication status: full text report |
| Participants |
Patients: 2557 randomised, 2557 described at baseline in the experimental group; 2432 randomised, 2432 described at baseline in the experimental group; 1753 randomised, 1753 described at baseline in the control group Number of females: 1752 of 2557 (69%) in experimental group 1; 1676 of 2432 (69%) in experimental group 2; 1093 of 1753 (62%) in control group Average age (SD): 58.5 (6.5) years in the experimental group 1; 59.1 (6.4) years in the experimental group 2; 59.1 (6.6) years in the control group Average (SD) education: Experimental group 1: none: 44 (1.7%); primary school: 14 (0.6%); secondary school: 400 (15.6%); further education (A level): 777 (30.4%); university graduate/PG: 1322 (51.7%). Experimental group 2: none: 55 (2.3%); primary school: 10 (0.4%); secondary school: 418 (17.2%); further education (A level): 717 (29.5%); university graduate/PG: 1230 (50.6%). Control group: none: 37 (2.1%); primary school: 9 (0.5%); secondary school: 320 (18.3%); further education (A level): 556 (31.7%); university graduate/PG: 831 (47.4%) Baseline cognitive function: Baddeley Grammatical Reasoning Test 14.4 (5.3); Paired Associate Learning 3.5 (0.6); digit span 4.8 ladder (1.1). Overall, up to 2873/6742 (43%) of participants had age‐associated impairment in reasoning Ethnicity: Experimental group 1: 2478 white; 0 Indian; 25 Asian; 7 black; 47 other; 0 unclear. Experimental group 2: 2359 white; 0 Indian; 31 Asian; 4 black; 36 other; 2 unclear. Control group: 1707 white; 0 Indian; 10 Asian; 4 black; 32 other; 0 unclear APOE: number of participants positive for APOE not reported |
| Interventions |
Type of experimental intervention: computerised CT, individualised; treatment duration 6 months. Intervention provided as individual training, without supervision Details of experimental intervention: ReaCT focussed on 3 reasoning tasks and 3 problem‐solving tasks Session duration: 10 minutes daily in the experimental group Number of treatment sessions: on average, 112 training sessions over 6 months Type of experimental intervention 2: computerised CT, individualised; treatment duration not reported. Intervention provided as individual training, without supervision Details of experimental intervention 2: GCT involved 6 cognitive tasks covering mathematics, attention, memory, and visuospatial ability Session duration: 10 minutes daily in experimental group 2 Number of treatment sessions: on average, 112 training sessions over 6 months Details of control intervention: the control group performed equivalent Internet‐based tasks involving a game in which people were asked to put a series of statements in correct numerical order Session duration: 10 minutes daily in the control group Number of treatment sessions: on average, 112 training sessions over 6 months |
| Outcomes |
-
Cognitive functioning outcomes:
Episodic memory measured with verbal short‐term memory at 6 months, on a scale from not reported to not reported with higher values indicating benefit Executive functioning measured with Baddeley Grammatical Reasoning Test at 6 months, on a scale from not reported to not reported with higher values indicating benefit Working memory measured with digit span at 6 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 extracted: none reported Safety outcome extracted: none reported Available cognitive outcome, not considered in this review: spatial working memory (SWM) with higher values indicating benefit Available physical functioning outcome, not considered for this review: daily function measured with instrumental activities of daily living (IADLs) at 6 months with higher values indicating benefit. Outcome data on IADLs were reported for only a subgroup of participants 60 years of age or older |
| Notes |
|
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
Judgment: random sequence adequately generated
Quote(s): "Participants were randomly assigned in equal proportions via simple randomization to receive ReaCT, GCT, or control. This was achieved by using a computer‐generated randomization sequence to eliminate allocation bias" |
| Allocation concealment (selection bias) |
Low risk |
Judgment: adequate method of allocation concealment
Quote(s): "The online format enabled complete allocation concealment from investigators" |
| Blinding of participants (performance bias) |
Low risk |
Judgment: Study authors report that participants were blinded to treatment assigned
Quote(s): "Participants were blind to which group they were allocated"; "This was a double‐blind 6‐month online randomised 3‐arm controlled trial" |
| Blinding of personnel (performance bias) |
Low risk |
Judgment: home‐based; no involvement of therapists
Quote(s): "The online format enabled complete allocation concealment from investigators" |
| Blinding of outcome assessment (detection bias) All outcomes |
Low risk |
Judgment: outcomes based on computer tests
Quote(s): "The online format enabled complete allocation concealment from investigators" |
| Incomplete outcome data (attrition bias) All outcomes |
High risk |
Judgment: we judged high risk of bias for all outcomes, as the imputation technique (last observation carried forward) is likely to yield biased estimates in the presence of observed fractions of participants with missing outcome data at 6 months
Comparison ReaCT reasoning and planning vs computerised tasks
Outcome episodic memory: 1369 out of 2557 (54%) randomised were analysed in the experimental group, and 591 out of 1753 (34%) randomised were analysed in the control group Outcome executive functioning: 1501 out of 2557 (59%) randomised were analysed in the experimental group, and 1059 out of 1753 (60%) randomised were analysed in the control group Outcome working memory: 2236 out of 2557 (87%) randomised were analysed in the experimental group, and 1499 out of 1753 (86%) randomised were analysed in the control group
Comparison GCT multi‐domain vs computerised tasks
Outcome episodic memory: 1130 out of 2432 (46%) randomised were analysed in the experimental group, and 591 out of 1753 (34%) randomised were analysed in the control group Outcome executive functioning: 1434 out of 2432 (59%) randomised were analysed in the experimental group, and 1059 out of 1753 (60%) randomised were analysed in the control group Outcome working memory: 2096 out of 2432 (86%) randomised were analysed in the experimental group, and 1499 out of 1753 (86%) randomised were analysed in the control group
Comment: no data at 12 weeks were available for extraction
Quote(s): "The primary analysis was intention‐to‐treat and involved all participants who were randomized"; "Missing values were imputed by last observation carried forward for the 6‐month outcome for individuals who completed the 3‐month outcome assessment" |
| Selective reporting (reporting bias) |
Low risk |
Judgment: all outcomes indicated in the methods are reported in the results |
| Other bias |
Low risk |
Judgment: no other sources of bias are important |
