Trebbastoni 2018.

Methods	A single‐blind RCT comparing effects on cognition of a group CT programme vs a no intervention control condition for persons with mild to moderate AD
Participants	140 community‐dwelling participants (78 women) between 50 and 85 years of age with a diagnosis of Alzheimer‐type dementia, according to NINCDS‐ADRDA, were recruited in Rome, Italy
Interventions	Participants in the intervention condition (n = 54) participated in group CT sessions that involved paper‐and‐pencil tasks, as well as verbal‐learning exercises. Participants were trained twice a week over 24 weeks Participants assigned to the control condition (n = 86) received usual care at the hospital
Outcomes	Oucomes included performance on several cognitive domains, including memory, attention, language, visuospatial functions, frontal functions, and praxis Participants were assessed at baseline, post intervention, and 6 months later
Country	Italy
Registration status	No information provided; presumed to be unregistered
Conflict of Interests	No
Notes	Six participants from the experimental group (n = 54) were included only in the first part of the study (from T0 to T1), but not from T1 to T2. They were excluded because they had attended less than 80% of the sessions. Between T1 and T2, 3 participants were excluded from this group (1 loss to follow‐up; 1 decision to withdraw; 1 death). Therefore, 54 participants in the experimental group were assessed at T1, but only 45 were assessed at T2 (6 months' follow‐up) Study authors provided the original paper along with clarification on a discrepancy between the same score in 2 different tables
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The ratio of intended numbers of participants in each of the comparison groups depended on the resources available at our site. Hence, we used a random numbering with an unequal allocation ratio of 1:2 (1 treated:2 untreated). Ten days prior to the commencement of the training, we computer‐generated a randomisation list that assigned the patients belonging to the TG to nine treatment sub‐groups"
Allocation concealment (selection bias)	Unclear risk	Study authors stated that they used a computerised randomisation system. It is likely that allocation concealment was done, but this is not specified
Blinding of participants and personnel (performance bias) All outcomes	High risk	Study authors did not mention blinding of participants; the study included a passive control condition, so blinding was not possible
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"Independent expert evaluators, who were blinded to the treated or untreated status of the patients, recorded the outcome measures at T0, T1 and T2"
Incomplete outcome data (attrition bias) All outcomes	High risk	Researchers stated, "We analysed the data of all the participants who concluded the study and were tested at T2" Nine participants in the experimental condition and one in the control condition dropped out during the study; their data were not reported, which is likely to introduce attrition bias
Selective reporting (reporting bias)	Low risk	All tests mentioned in the "Methods" section were reported in the "Results" section
Other bias	Low risk	Study appears to be free of other sources of bias