Walker 2012.

Study characteristics
Methods	Design: parallel RCT Duration of trial: 2 years (1 March 2008 to 28 February 2010) Unit of randomisation: people with stroke with persistent dressing problems and accompanying cognitive difficulties at 2 weeks after their stroke Recruitment and allocation: 965 consecutive inpatients were monitored by the ward occupational therapists over 2 weeks of conventional rehabilitation to identify those people with persistent dressing difficulties. Of 110 screened, 70 people met inclusion criteria and were randomised by Internet randomisation to the IG (n = 36) or CG (n = 34). Data from 33 participants from the IG and 31 participants from the CG were analysed.
Participants	Setting: stroke rehabilitation wards Country: UK Sample: 70 adults, 41% men; allocated: IG: 36; CG: 34; analysed: IG: 33; CG: 31 Inclusion: scoring < 100% on the Nottingham Stroke Dressing Assessment and having confirmed cognitive impairment from a brief cognitive screening test Exclusion: unable to tolerate sitting in a chair for 15 minutes, premorbid disability (Rankin > 3), known diagnosis of depression or dementia, unable to understand English if it was not their first language Age: median: IG: 77 (range 47–93) years; CG: 81 (range 41–96) years Time since stroke onset: median: IG: 26 (range 12–139) days; CG: 22 (range 13–99) days
Interventions	Intervention group Brief name: neuropsychological approach to dressing (DRESS) Recipients: inpatient adults with cognitive impairments after acute stroke Why: to select tailored, evidence‐based techniques for dressing impairment from a preprepared neuropsychological treatment manual based on detailed cognitive testing that assessed the impact of deficits on performance using error analysis. What (materials): neuropsychological treatment manual of evidence‐based techniques culled from the wider neuropsychological literature and based on comprehensive literature searches, survey results (Walker 2003), and occupational therapy textbooks (Edmans 2001); standard T‐shirt available in different sizes (Sunderland 2006); error analysis rating form (see Fletcher‐Smith 2010) What (procedures): participants received further detailed cognitive testing and an assessment of the impact of cognitive deficits on dressing by observation of a standard task of putting on a t‐shirt (Sunderland 2006), with performance scored using an error analysis rating form (Fletcher‐Smith 2010). Based on the test results and observed errors, the occupational therapists selected interventions from a menu of evidence‐based techniques described in the neuropsychological treatment manual. Commonly used techniques included cueing and alerting procedures to combat neglect or attentional difficulties, systematic laying out of clothing to reduce spatial confusion and graded errorless learning strategies to enhance acquisition of dressing skills. Who provided: 2 research occupational therapists experienced in the treatment of people with stroke How: face‐to‐face and individually Where: in a stroke rehabilitation ward and in participants' homes if they were discharged from hospital before the end of the treatment period. When and how much: the aim was for 3 times per week for 6 weeks based on previous single case experiments (Sunderland 2006). They received a median of 13 sessions (minimum 0, maximum 18) of 18 possible sessions. Tailoring: interventions were selected on the basis of participants' test results and observed errors in the dressing assessment and provided at home if participants were discharged early. Modifications: none reported How well (planned): a random sample of treatment sessions were observed by an independent researcher to ensure the manuals were adhered to and that they included the actual treatment prescribed in the manual. How well (actual): the authors reported that they found a high level of fidelity of treatment but details were not specifically reported. Comparator group Brief name: functional group Recipients: inpatient adults with cognitive impairments after acute stroke Why: to provide usual care in addressing dressing difficulties by providing dressing practice using a problem‐solving approach, with assistance when required. What (materials): functional treatment manual What (procedures): dressing interventions included components such as putting the affected arm into the sleeve first, crossing affected leg over other leg to reach feet, energy conservation techniques, etc. There was no attempt to formally assess the participant's cognitive difficulties or relate them to evidence on which approach to training might be the most successful Who provided: 2 research occupational therapists experienced in the treatment of people with stroke How: face‐to‐face and individually Where: in a stroke rehabilitation ward and in participants' homes if they were discharged from hospital before the end of the treatment period When and how much: the aim was for 3 times per week for 6 weeks; received a median number of 12 sessions (minimum 0, maximum 18). Modifications: none reported Tailoring: none reported How well (planned): a random sample of treatment sessions was observed by an independent researcher to ensure the manuals were adhered to and that they included the actual treatment prescribed in the manual. How well (actual): the authors reported that they found a high level of fidelity of treatment but details were not specifically reported.
Outcomes	Primary Nottingham Stroke Dressing Assessment Secondary: none Other Line Cancellation test Object Decision test Pegs per second Gesture Imitation test Methods of data collection: baseline assessments were conducted prior to randomisation and participants were assessed 6 weeks after randomisation by an independent assessor. Data collection time points: at baseline and 6 weeks
Notes	Funding: yes Conflict of interest: none Published trial protocol: no Trial registration: yes; Dressing Rehabilitation Evaluation Stroke Study: ISRCTN14430342 Ethics approval: not stated
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "… using … the University of Nottingham Clinical Trials Unit Internet randomization service, patients were randomized to one of two treatment groups … Patients were stratified by side of stroke and severity of their dressing problem".
Allocation concealment (selection bias)	Low risk	Quote: "… concealed allocation via the University of Nottingham Clinical Trials Unit internet randomization service".
Blinding of participants and personnel (performance bias) All outcomes	High risk	The 2 groups continued with their usual rehabilitation therapy and nursing care and only differed in the type of dressing practice provided by the trial occupational therapists. Both interventions were delivered by 2 research occupational therapists experienced in the treatment of people with stroke.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Patients were assessed … by an independent assessor who was masked to the patient's treatment group allocation. Masking of the independent assessor was monitored by completion of a best guess form. Masking of the outcome assessor was tested and found to be compromised for only six patients".
Incomplete outcome data (attrition bias) All outcomes	Low risk	Lost to follow‐up was reported and managed in analysis.
Selective reporting (reporting bias)	Low risk	All outcome measures identified in registered protocol reported on in the paper  (ISRCTN14430342).
Other bias	Low risk	No other identifiable bias.