Occupational therapy for adults with problems in activities of daily living after stroke

doi:10.1002/14651858.CD003585.pub3

Cochrane Database Syst Rev. 2017 Jul; 2017(7): CD003585.

Published online 2017 Jul 19. doi: 10.1002/14651858.CD003585.pub3

PMCID: PMC6483548

Occupational therapy for adults with problems in activities of daily living after stroke

Monitoring Editor: Lynn A Legg, Sharon R Lewis, Oliver J Schofield‐Robinson, Avril Drummond, Peter Langhorne, and Cochrane Stroke Group

NHS Greater Glasgow and Clyde Health Board, Royal Alexandra Hospital, PaisleyUK, PA2 9PN

Royal Lancaster Infirmary, Patient Safety Research Department, Pointer Court 1, Ashton Road, LancasterUK, LA1 4RP

Royal Lancaster Infirmary, University Hospitals of Morecambe Bay, NHS, Research and Development, LancasterUK, LA1 4RP

University of Nottingham, Faculty of Medicine and Health Sciences, QMC Campus, NottinghamNG7 2HAUK

ICAMS, University of Glasgow, Academic Section of Geriatric Medicine, Level 2, New Lister Building, Glasgow Royal Infirmary, GlasgowUK, G31 2ER

Lynn A Legg, Email: moc.liamg@ggelannyl.

Author information Copyright and License information Disclaimer

Abstract

Background

A stroke occurs when the blood supply to part of the brain is cut off. Activities of daily living (ADL) are daily home‐based activities that people carry out to maintain health and well‐being. ADLs include the ability to: eat and drink unassisted, move, go to the toilet, carry out personal hygiene tasks, dress unassisted, and groom. Stroke causes impairment‐related functional limitations that may result in difficulties participating in ADLs independent of supervision, direction, or physical assistance.

For adults with stroke, the goal of occupational therapy is to improve their ability to carry out activities of daily living. Strategies used by occupational therapists include assessment, treatment, adaptive techniques, assistive technology, and environmental adaptations. This is an update of the Cochrane review first published in 2006.

Objectives

To assess the effects of occupational therapy interventions on the functional ability of adults with stroke in the domain of activities of daily living, compared with no intervention or standard care/practice.

Search methods

For this update, we searched the Cochrane Stroke Group Trials Register (last searched 30 January 2017), the Cochrane Controlled Trials Register (The Cochrane Library, January 2017), MEDLINE (1946 to 5 January 2017), Embase (1974 to 5 January 2017), CINAHL (1937 to January 2017), PsycINFO (1806 to 2 November 2016), AMED (1985 to 1 November 2016), and Web of Science (1900 to 6 January 2017). We also searched grey literature and clinical trials registers.

Selection criteria

We identified randomised controlled trials of an occupational therapy intervention (compared with no intervention or standard care/practice) where people with stroke practiced activities of daily living, or where performance in activities of daily living was the focus of the occupational therapy intervention.

Data collection and analysis

Two review authors independently selected trials, assessed risk of bias, and extracted data for prespecified outcomes. The primary outcomes were the proportion of participants who had deteriorated or were dependent in personal activities of daily living and performance in activities of daily living at the end of follow‐up.

Main results

We included nine studies with 994 participants in this update. Occupational therapy targeted towards activities of daily living after stroke increased performance scores (standardised mean difference (SMD) 0.17, 95% confidence interval (CI) 0.03 to 0.31, P = 0.02; 7 studies; 749 participants; low‐quality evidence) and reduced the risk of poor outcome (death, deterioration or dependency in personal activities of daily living) (odds ratio (OR) 0.71, 95% CI 0.52 to 0.96; P = 0.03; 5 studies; 771 participants; low‐quality evidence). We also found that those who received occupational therapy were more independent in extended activities of daily living (OR 0.22 (95% CI 0.07 to 0.37); P = 0.005; 5 studies; 665 participants; low‐quality evidence). Occupational therapy did not influence mortality (OR: 1.02 (95% CI 0.65 to 1.61); P = 0.93; 8 studies; 950 participants), or reduce the combined odds of death and institutionalisation (OR 0.89 (95% CI 0.60 to 1.32); P = 0.55; 4 studies; 671 participants), or death and dependency (OR 0.89 (95% CI 0.64 to 1.23); P = 0.47; 4 trials; 659 participants). Occupational therapy did not improve mood or distress scores (OR 0.08 (95% CI ‐0.09 to 0.26); P = 0.35; 4 studies; 519 participants; low‐quality evidence). There were insufficient data to determine the effects of occupational therapy on health‐related quality of life. We found no studies of consenting carers prior to study participation and therefore there were no carer‐related outcomes in our review. There were insufficient data to determine participants' and carers' satisfaction with services.

Using GRADE, the quality of evidence was low. The major limitation was the number of studies at unclear risk of selection bias and an inevitable high risk of performance and detection bias, as both participants and occupational therapists could not be blinded to the intervention. In addition, there was a sparseness of data for our outcomes of interest and we downgraded the quality of our evidence for these reasons.

Authors' conclusions

We found low‐quality evidence that occupational therapy targeted towards activities of daily living after stroke can improve performance in activities of daily living and reduce the risk of deterioration in these abilities. Because the included studies had methodological flaws, this research does not provide a reliable indication of the likely effect of occupational therapy for adults with stroke.

Plain language summary

Occupational therapy for adults with problems in activities of daily living after stroke

Review question
What are the effects of occupational therapy for adults with stroke on activities of daily living?

Background
Different parts of the brain carry out different functions: seeing, sensation, balance, movement, understanding language, behaviour, problem solving, and emotion. A stroke occurs when the blood supply to part of the brain is cut off. If the blood supply is cut off to a part of the brain that carries out a particular function (such as seeing, moving arms and legs, or speaking), then these body parts or body functions will not work as they should.

Activities of daily living (ADLs) are daily household‐based activities that people carry out to maintain health and well‐being. ADLs include eating and drinking, moving about, going to the toilet, personal hygiene, dressing and undressing, and grooming. When stroke changes how body parts or functions work, then the ability to carry out ADLs can become affected.

For adults with stroke, the goal of occupational therapy is to improve ability to carry out ADLs. Strategies used by occupational therapists include activity‐based interventions, adaptive techniques, assistive technology, and environmental adaptations.

Study characteristics
We found nine studies up to January 2017, involving 994 participants, that looked at the benefits of occupational therapy interventions for adults with stroke who had problems with activities of daily living. This is an update of the Cochrane review first published in 2006.

Key result
We found that occupational therapy for people with stroke can improve their ability to carry out these daily activities and stop them deteriorating in those abilities. We found no evidence that occupational therapy reduced rates of death or the need to be cared for in an institution, or affected mood or distress of the participant. We did not collect data on carer‐related outcomes or participant satisfaction with the service.

Quality of the evidence
There were few studies measuring our outcomes of interest and we judged the quality of the evidence to be of low‐quality. Many of the studies did not report methods sufficiently clearly and it was not possible to mask the occupational therapy from the person giving or receiving the treatment; this could also have influenced the results in our studies. We did not have sufficient good‐quality evidence to be certain of our results and we cannot be certain that future studies will not change these conclusions.

Summary of findings

Background

Description of the condition

Stroke is the death of brain tissue as a result of an interruption of the blood flow leading to or within the brain, or by an aneurysm or ruptured blood vessel in the brain. Activities of daily living (ADLs) are normal day‐to‐day household‐based tasks and activities that people carry out to maintain health and well‐being. ADLs include eating and drinking, moving about (such as getting into and out of bed, turning over in bed, getting on and off a chair, moving between rooms, moving between levels (using stairs)), going to the toilet (including carrying out toilet hygiene), carrying out personal hygiene tasks (including washing hair and body, and drying oneself), getting dressed and undressed, and grooming activities (including performing oral hygiene activities, combing hair, shaving). Stroke causes impairment‐related functional limitations that may result in difficulties carrying out ADLs independent of supervision, direction, or physical assistance.

Description of the intervention

For adults with stroke, the goal of occupational therapy is to improve the ability to carry out activities of daily living. Strategies used by occupational therapists include assessment, treatment, adaptive techniques, assistive technology, and environmental adaptations.

How the intervention might work

The central aspects (or core components) of occupational therapy for adults with stroke include the following.

Assessment

Assessment allows occupational therapists to gather important information about the adult with stroke and the difficulties they are facing. The purpose of assessment is to:

determine the impact of stroke‐related changes in body structures and body functions on the adult's skills, capacities, and abilities to independently initiate, sustain, and complete actions and tasks necessary to carry out activities of daily living (i.e. stroke‐related impairment‐related functional limitations) (Gillen 2016a; ICF 2001; OTfPD 2008a).
identify the adult with stroke's perceived problems and priorities and to prioritise main concerns.
learn about stroke‐related impairment‐related functional limitations by observing the adult with stroke carrying out meaningful tasks, such as getting dressed or eating (Gillen 2016e; OTfPD 2008e).

All assessments are used to analyse how the adult with stroke completes activities of daily living, the results of which are used to create a therapy plan.

Partnership

Occupational therapists work in partnership with adults with stroke and make joint decisions about what therapeutic interventions and individual goals are appropriate and relevant (OTfPD 2008c).

Programme plan

There is a programme plan for each adult with stroke and this plan includes all activities necessary to achieve clinical recovery and performance improvement for that given individual (OTfPD 2008c).

The programme plan includes detailed specifications of individual short‐term and long‐term goals, and treatment approaches and methods to be employed to achieve the goals. The nature and definition of individual goals is a shared decision between the adult with stroke and occupational therapist and is based on the self‐determined needs and preferences of the adult with stroke combined with the best available information from assessments, current best scientific information, and clinical judgement (OTfPD 2008c).

Occupational therapy treatment approaches and methods are selected based on intentionality, appropriateness, and resources available (OTfPD 2008b). Short‐term and long‐term goals are the stages into which the programme is divided for monitoring, and evaluation and measurement of the impact of therapy (OTfPD 2008c).

Activities of daily living as a therapeutic medium and goal

Occupational therapists use activities of daily living as a means to achieve therapeutic change (improvement in capacities and abilities) and as the goal of therapy (i.e. to achieve improvements in activities and participation) (ICF 2001; OTfPD 2008d).

Intentionality

Interventions are directed towards activities that adults with stroke believe are worthwhile investing effort in and are keen to improve (Gillen 2016b; OTfPD 2008d).

Appropriate

Interventions are designed to be appropriate to the level of skill, capacities, abilities, intentions, interests and intra‐psychological processes of adults with stroke (OTfPD 2008c; OTfPD 2008d).

Incremental and progressive

Interventions targeted towards clinical recovery are designed to be progressive with incremental increases in activity and difficulty levels (OTfPD 2008e).

Healthy perceptions, attitudes, and behaviour‐focused

Therapeutic contacts emphasise and encourage healthy perception, attitudes, and behaviour through concrete experience, positive feedback, and reinforcement (Gillen 2016c; OTfPD 2008d).

Active engagement in authentic practice

Interventions are designed to engage the adult with stroke in authentic practice of actions, activities, and interactions (OTfPD 2008d; Gillen 2016b; Gillen 2016c).

Enduring and applicable

Interventions are designed to promote therapy generalisation and long‐term maintenance of therapeutic change by empowering stroke patients (Gillen 2016c).

Evaluation and accountability

Intervention effectiveness is evaluated continuously by both the occupational therapist and stroke patient. The occupational therapist assumes accountability for achieving outcomes (Gillen 2016a; OTfPD 2008b; OTfPD 2008c).

Why it is important to do this review

As mortality rates from stroke continue to improve, so do the odds that several generations of stroke survivors are alive at the same time. Many stroke survivors have marked limitations in ability to perform routine ADLs as a result of stroke‐related physical or mental impairment. As a consequence, many stroke survivors require some form of long‐term care, which includes unpaid care (provided by family and friends), government or funded personal care services, assisted living, and residential or nursing care. The significant costs associated with providing this support is likely to be borne by families and society.

Occupational therapy aims to help people reach and maintain their maximum level of function and self‐reliance in ADL.

The original Cochrane review (Legg 2006), including eight studies involving 1258 participants, found that occupational therapy interventions reduced the odds of a poor outcome (dependency or deterioration) and increased personal ADL scores. This is an update of the original review. A GRADE (Grades of Recommendation, Assessment, Development and Evaluation Working Group) approach was not used to assess the quality of the evidence in the original review.

Objectives

To assess the effects of occupational therapy interventions on the functional ability of adults with stroke in the domain of activities of daily living, compared with no intervention or standard care/practice.

Methods

Criteria for considering studies for this review

Types of studies

We sought all randomised controlled trials (RCTs) of adults with stroke receiving an intervention provided by an occupational therapist, or under the supervision of an occupational therapist, with the specific aim of improving functional capacity or performance in activities of daily living (ADL) compared with no intervention or standard care/practice. We excluded studies which compared different types of occupational therapy intervention. We included randomised cross‐over studies.

Types of participants

We included trials that recruited adults (aged 18 years and above) who met a clinical definition of stroke (focal neurological deficit caused by cerebrovascular disease).

We excluded trials of mixed aetiology where the percentage of adults with stroke was less than 50%. We also excluded trials that only included adults with specific stroke‐related impairments (i.e. neuromusculoskeletal and movement‐related impairments, or changes in muscle power, or mental impairment, or sensory impairment, or voice and speech function impairment).

Types of interventions

We included trials of occupational therapy interventions, which had the following features.

Described as an occupational therapy intervention.
Delivered by an occupational therapist or under the supervision of an occupational therapist.
Focused on activities of daily living (ADL), which are eating and drinking, functional mobility, going to the toilet, dressing, carrying out personal hygiene (including washing body, bathing/showering), grooming (including oral hygiene) (Gillen 2016d).
Occupational therapy interventions fell under one or more of the following categories:
- treatments focused on remediating impaired capacities or abilities in the context of ADLs (OTfPD 2008c) such as task‐orientated approaches (Gillen 2016b) or activity‐based interventions (Gillen 2016d);
- the use of adaptive (compensatory) techniques as an alternative way to effectively complete ADLs (OTfPD 2008c);
- the use of assistive technology (AT) in which the adult with stroke has the use of any item, piece of equipment, or product system to increase, maintain, or improve their functional capacity (ATiA); or
- environmental adaptations in which the participant's physical environment is modified with ramps, electric hoists, stair lifts, handrails, level access, or wet‐room wheelchair accessible shower (Age UK 2017) in order to restore or enable self‐reliance, privacy, confidence, or dignity for the adult with stroke.

We included trials in which the control group received no intervention or standard care/practice.

We excluded trials:

of occupational therapy interventions delivered to adults with stroke in care or nursing homes, as this is covered in a separate review (Fletcher‐Smith 2013);
comparing occupational therapy interventions being delivered in different settings, e.g. occupational therapy in clinic versus occupational therapy at home;
that included occupational therapists as part of a multidisciplinary team as they are or will be covered in other reviews and therefore are beyond the remit of this review;
that included occupational therapy interventions in combination with other interventions, for example, occupational therapy and brain stimulation;
testing specific treatment approaches, e.g. task‐specific training or cognitive training.

If multi‐arm studies included more than one intervention performed by an occupational therapist, we selected the intervention arm that was most closely related to our review objectives (Higgins 2011c).

Types of outcome measures

We were interested in outcomes that assessed whether occupational therapy improves activities in daily living (ADL), as reflected in our primary outcomes. We collected data which were reported and measured by study authors using appropriate measurement tools, at the end of scheduled study follow‐up.

We required that study authors reported that unpaid carers had provided consent prior to study participation for inclusion of carer‐related outcomes in our review.

Primary outcomes

Performance in personal ADL at the end of scheduled follow‐up. Measures of ADL included Barthel Index or Functional Independence Measure (FIM).
Death or a poor outcome at the end of follow‐up*. We defined this as the combined outcome of being dead or:
- having deteriorated, characterised by experiencing a deterioration in ability to perform personal activities of daily living (i.e. experiencing a drop in ADL score); or
- being dependent, characterised by lying above or below a predefined cut‐off point on a given ADL scale; or
- requiring institutional care at the end of scheduled follow‐up.

* The primary outcome 'Death or a poor outcome' is a composite outcome with some components appearing as secondary outcomes.

Secondary outcomes

Death at the end of scheduled follow‐up.
Number of participants dead or physically dependent at the end of scheduled follow‐up.
Number of participants dead or requiring institutional care at the end of scheduled follow‐up.
Performance in extended activities of daily living (EADL) (community and domestic activities) at the end of scheduled follow‐up. Measure of EADL included Frenchay Activities Index or Nottingham Extended ADL.
Participant mood at the end of scheduled follow‐up. Measures of mood included Beck Depression Inventory, or Hospital Anxiety and Depression Scale.
Participant health‐related quality of life (HRQOL) at the end of scheduled follow‐up. Measures included EuroQOL, or Nottingham Health Profile.
Carer mood at the end of scheduled follow‐up. Measures included Beck Depression Inventory, or Hospital Anxiety and Depression Scale.
Carer health‐related quality of life at the end of scheduled follow‐up. Measures included EuroQOL, or Nottingham Health Profile.
Participant and carer satisfaction with services.

Search methods for identification of studies

See the 'Specialized register' section in the Cochrane Stroke Group module. We attempted to identify all relevant trials, regardless of language or publication status, and arranged translation of relevant papers, where necessary.

Electronic searches

For this update we searched the following:

Cochrane Stroke Group Trials Register (searched 31 January 2017);
Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 1) in the Cochrane Library (searched 11 January 2017) (Appendix 1);
MEDLINE Ovid (1946 to 5 January 2017) (Appendix 2);
Embase Ovid (1974 to 5 January 2017) (Appendix 3);
CINAHL EBSCO (Cumulative Index to Nursing and Allied Health Literature; 1937 to 5 January 2017) (Appendix 4);
PsycINFO Ovid (1806 to 2 November 2016) (Appendix 5);
AMED Ovid (The Allied and Complementary Medicine Database; 1985 to 1 November 2016) (Appendix 6);
Web of Science (limited to Science Citation Index Expanded, Social Sciences Citation Index, Conference Proceedings Citation Index ‐ Science, and Conference Proceedings Citation Index ‐ Social Science & Humanities; 1900 to 6 January 2017) (Appendix 7);
OpenGrey (System for Information on Grey Literature in Europe; http://www.opengrey.eu; 1980 to January 2017) (Appendix 8).

We developed the MEDLINE search strategy (Appendix 2), which included controlled vocabulary and free‐text terms, with the help of the Stroke Group’s Cochrane Information Specialist (Joshua Cheyne) and adapted this for the other databases. Where appropriate, they were combined with subject strategy adaptations and modifications of the highly sensitive search strategy, designed by the Cochrane Collaboration for identifying randomised controlled trials and controlled clinical trials (as described in Higgins 2011a Chapter 6). Where necessary, we updated our search terms by adding relevant new database controlled vocabulary terms.

We also searched the following trials registries.

US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov; searched 31 January 2017) (Appendix 9);
World Health Organization Clinical Trials Registry Platform (app.who.int/trialsearch; searched 31 January 2017) (Appendix 9).

Evidence for this review included search results from the previous version of this review (Legg 2006) combined with results from the above searches. In Legg 2006, there were no date limits and searches were applied from inception of databases.

Databases searched in addition to the above, and other differences in searches since the previous review, are reported in Differences between protocol and review.

Searching other resources

We searched the reference lists of papers reporting studies selected for inclusion in this review in order to identify additional relevant trials.

Data collection and analysis

Selection of studies

We downloaded all titles and abstracts retrieved from the electronic searches into reference management software (Endnote). One review author (SL) identified and removed duplicate records. Once duplicate records were removed, one reviewer (SL) uploaded all remaining records into Covidence (Covidence). Two of three review authors (SL, LL, OSR) independently examined titles and abstracts and removed any obviously irrelevant reports. We obtained full‐text reports of potentially relevant reports. At this stage, we attempted to link multiple reports of the same study. Two review authors (SL, LL) examined the full reports for compliance of studies with review eligibility criteria. We resolved differences in opinion regarding trial eligibility through discussion with review coauthors (PL, AD).

Data extraction and management

Two of three review authors (LL, PL, SL) independently extracted data from published reports using a standard data recording form. Where differences occurred between investigators, we resolved these through discussion. We contacted the study authors of three studies to request additional information. We did not contact all trialists, as had been the intention in the original review (Legg 2006).

In studies where a cross‐over design was used, we only extracted data from the first treatment phase after randomisation.

Assessment of risk of bias in included studies

We assessed risk of bias using the Cochrane 'Risk of bias' tool (Higgins 2011b). We assessed the methods used in each study to control for the following potential sources of bias: sequence generation (selection bias); allocation concealment (selection bias); blinding of participants, personnel and outcome assessors (performance and detection bias); incomplete outcome data (attrition bias); selective outcome reporting (reporting bias); and other sources of bias.

It was not feasible to blind participants and personnel to the intervention and therefore we agreed to judge all outcomes as having a high risk of performance bias. For detection bias, we considered this by outcome. We judged that all outcomes relating to death were unlikely to be influenced by lack of blinding of outcome assessors and, even if outcome assessors for death were described as not blinded, we judged this bias to be low risk. For detection bias relating to all other outcomes, we considered three aspects from the information presented in the paper: whether the tool that was used to assess the outcome used observation, interview or self‐completion (e.g. questionnaire); whether the assessment was made by the participant or an assessor; and whether the assessor was blinded. If the assessment tool involved self‐assessment by the participant, we judged detection bias to be unclear because the participant could not be blinded to the intervention and we did not know if lack of blinding would influence participant responses to self‐assessment (Higgins 2011b). If the assessment tool was used by an independent assessor and only included observation, we judged detection bias to be low. Where assessment was mixed either by type of assessor, type of tool, or type of outcome, we judged detection bias to be unclear.

For the purpose of this review, a true intention‐to‐treat analysis was defined as following three principles: 1) participants were kept in the intervention groups to which they were randomised, regardless of the intervention they actually received; 2) there was measured outcome data on all participants; and 3) all randomised participants were included in the analysis (Higgins 2011b). We made judgements on attrition bias from these three criteria.

For reporting bias, we attempted to source prospective registration reports through information presented in study reports or from sifting of clinical trial register searches. We intended to judge bias by comparing prospective registration reports or protocols with outcomes reported in the full report.

Measures of treatment effect

We collected binary data for our outcomes related to death and analysed these using the Peto odds ratio (OR) and 95% confidence interval (CI). We collected continuous data reported as means (with standard deviations) for outcomes relating to activities of daily living, extended activities of daily living, health‐related quality of life (participant and carers), mood and distress scores (participant and carers). We calculated the standardised mean difference from continuous data using inverse variance, to account for differences in the assessment tools used to measure the same outcome.

Unit of analysis issues

In the event we had included a trial using a cluster design (in which participants were randomised at group level) we would have used the intra‐cluster correlation coefficient (ICC) to estimate the effective sample size (Higgins 2011c).

Dealing with missing data

The primary aim of this review was to obtain standardised data through collaboration with the original investigators. Incomplete data are relatively common in trials of rehabilitation. It is difficult to impute missing values for continuous outcomes. Where data were missing from a published report, we contacted the primary investigators in an attempt to get this information.

Assessment of heterogeneity

We assessed whether there was evidence of inconsistency in our results by considering possible methodological, clinical, and statistical heterogeneity. We assessed clinical and methodological heterogeneity by comparing similarities in our included studies between study designs, participants, interventions, and outcomes.

We assessed statistical heterogeneity by visually examining forest plots, and by performing the Chi² test using a P value of less than 0.1 to indicate statistically significant heterogeneity. We used a P value of less than 0.1 rather than the conventional cut‐off point of 0.05 because of the low power of this test. We quantified the effect of heterogeneity using the I² statistic including its 95% CI. We used the following cut‐offs as a guide to interpretation: I² at 0% to 40% was not considered important, 30% to 60% suggested moderate heterogeneity, 50% to 90% suggested substantial heterogeneity, and 75% to 100% was considerable heterogeneity (Higgins 2011c).

Assessment of reporting biases

We searched clinical trials registers to identify published protocols for each of our included studies. We intended to compare published protocols with published study results to assess the risk of selective reporting bias. If we had identified sufficient included studies (i.e. more than 10 studies (Higgins 2011a)), we would have generated a funnel plot to assess risk of publication bias in the review; an asymmetrical funnel plot may suggest publication of only positive results (Egger 1997).

We employed a comprehensive search strategy in an effort to avoid reporting biases in our review methodology. See Search methods for identification of studies.

Data synthesis

We completed meta‐analysis of outcomes for which we had comparable effect measures from more than one study, and when measures of heterogeneity indicated that pooling of results was appropriate. We used the statistical calculator provided in RevMan (RevMan 5.3) to perform meta‐analysis.

We used a fixed‐effect model (Mantel 1959). We assessed the robustness of the results to choice of model using sensitivity analysis.

We limited the primary analysis to studies with an overall low or unclear risk of bias. We reached decisions on overall risk of bias per study by consideration of four risk of bias domains: sequence generation, allocation concealment, outcome assessment, and incomplete outcome data. We required a study to have a judgment of low risk of bias in all four domains in order to categorise it as having an overall low risk of bias. We categorised a study that had both low and unclear judgements to have an overall unclear risk of bias. We did not include studies with high risk of bias judgements in these four domains in any of our primary analyses; we reconsidered this decision in sensitivity analysis.

Subgroup analysis and investigation of heterogeneity

If there had been sufficient studies, we would have explored variability in the participants, interventions, and outcomes among studies using the following subgroups:

hospital‐based occupational therapy interventions versus community‐based occupational therapy interventions within three months of stroke onset;
single versus multiple staff providing the experimental intervention (Higgins 2011a); and
randomisation to treatment within one year of stroke onset versus randomisation to treatment more than one year after stroke onset.

Sensitivity analysis

We explored the potential effects of decisions made as part of the review process as follows.

We included all studies regardless of risk of bias judgement.
We conducted meta‐analysis using the alternate meta‐analytical effects model (fixed‐effect or random‐effects).
We conducted a sensitivity to missing data analysis by first adding missing participants to the denominator in the intervention and control groups (missing participants better) and second, adding missing participants to the denominator and the numerator in the intervention and control groups (missing participants worse).

We compared effect estimates from the above results with effect estimates from the main analysis. We aimed to report differences that altered interpretation of effects.

We performed sensitivity analyses, where appropriate, on the following outcomes:

performance in activities of daily living;
odds of poor outcome (death and deterioration or death and dependency).

GRADE and 'Summary of findings'

We assessed the overall quality of the evidence for our outcome using the GRADE (Grades of Recommendation, Assessment, Development and Evaluation Working Group) approach, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011d). The quality of the evidence for a specific outcome is based on the performance of the studies against five criteria: study design and limitations, directness of evidence, consistency of results, precision (sufficient data), and reporting of results across all studies that measure a specific outcome. This approach yields an overall measure of how confident we can be that our estimate of effect is correct (Guyatt 2008). We used the principles of the GRADE system to prepare an overall assessment of evidence related to each of the following outcomes.

Performance in activities of daily living.
Death or 'poor outcome' (deterioration or dependency).
Extended activities of daily living.
Mood or distress

We (LL and SL) independently used GRADEpro software (http://www.guidelinedevelopment.org/) to create a 'Summary of findings’ table for each comparison. We reached consensus and resolved disagreements by consulting a third review author, if required.

Study design and limitations refers to the results of the 'risk of bias' assessment (Assessment of risk of bias in included studies). Directness of evidence refers to the comparability of population, interventions, and outcomes of interest in the included studies. We described any concerns regarding comparability of the included studies. Consistency of results refers to similarity in treatment effect in results across the included studies and includes the direction and size of effect and statistical significance. We considered overlap of confidence intervals to indicate statistical homogeneity. We also used the Chi² test of significance for heterogeneity using a P value of less than 0.10 to indicate statistical significance and used I² as an index of heterogeneity. Precision refers to the effect of study size on the precision of the estimate of treatment effect. For each outcome, we noted the sample size and magnitude and direction of the experimental intervention's effect compared with the control intervention ‐ specifically the width of the confidence interval surrounding the estimate i.e. whether the width was narrow indicating more precision) or wider (indicating more uncertainty) (Higgins 2011d). Publication bias refers to the likelihood of selective publication of studies or outcomes. Had there been a sufficient number of studies we would have assessed probability of selective outcome reporting as part of the assessment of risk of bias using funnel plots as visual tools for examining publication and other bias (i.e. small study effects).

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies

Results of the search

For this update, we screened 38,671 references from database searches and accessed the available full‐text reports for 32 studies. We screened 191 studies in the Cochrane Stroke Group Trials Register and accessed the available full‐text reports for three studies. We screened 1069 reports from clinical trial registers and 116 reports available from grey literature searching.

We considered nine studies identified as eligible in the previous version of this review and considered two studies that were awaiting classification in the previous version (Legg 2006).

The flow of search results for the previous version of the review are reported in Legg 2006. We reported details of the search for this update in a PRISMA flow chart. See Figure 1.

Occupational therapy compared to usual or no care for stroke
Patient or population: adults with stroke Setting: any (with the exception of care‐ or nursing‐home settings). Included studies conducted in: Hong Kong, UK, and USA Intervention: occupational therapy Comparison: no intervention or standard care/practice
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Usual or no care	Occupational therapy
Activities of daily living at end of scheduled follow‐up.	The mean activities of daily living score was 80.4³	The mean activities of daily living score in the intervention groups was 0.17 standard deviations higher (0.03 to 0.31 higher)		749 (7 studies)	⊕⊕⊝⊝ low^1,2	A standard deviation of 0.17 represents a small difference between groups
Odds of death or a poor outcome at end of scheduled follow‐up. Combined odds of death and deterioration, or death and dependence, or death and institutional care	Study population		Peto OR 0.71 (0.52 to 0.96)	771 (5 studies)	⊕⊕⊝⊝ low^1,2
	440 per 1000	313 per 1000 (229 to 423)
	Moderate

Extended Activities of Daily Living at end of scheduled follow‐up. Measures of Extended Activities of Daily Living	The mean Extended Activities of Daily Living score was 33.3³	The mean Extended Activities of Daily Living score in the intervention groups was 0.22 standard deviations higher (0.07 to 0.37 higher)		665 (5 studies)	⊕⊕⊝⊝ low^1,2	A standard deviation of 0.22 represents a small difference between groups
Mood or distress scores Measures of mood or distress	The mean depression score was 19.8³	The mean mood or distress scores in the intervention groups was 0.08 standard deviations higher (‐0.09 lower to 0.26 higher)		519 (4 studies)	⊕⊕⊝⊝ low^1,2	A standard deviation of 0.08 represents a small difference between groups
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval;
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

Study	N (I)	n (I)	Dead (I)	Missing (I)	N (C)	n (C)	Dead (C)	Missing (C)
Chui 2004	30	30	0	0	23	23	0	0
Corr 1995	55	46	9	0	55	39	11	5
Gilbertson 2000	67	60	6	1	71	62	5	4
Logan 1997	53	45	5	3	58	38	7	13
Parker 2001	156	106	15	35	157	110	11	36
Radomski 2007	5	5	0	0	5	0	0	0
Walker 1996	15	12	0	3	15	15	0	0
Walker 1999	94	84	6	4	91	79	7	5

Date	Event	Description
31 March 2017	New citation required and conclusions have changed	Occupational therapy increases independence in activities of daily living and reduces the odds of a poor outcome (deterioration or dependency). However, the evidence was rated low‐quality. This is a change to the conclusions of the original review. The changes to the conclusions are a result of incorporating 'risk of bias' assessments.
2 January 2017	New search has been performed	We changed the Background section, to add further explanation of how the intervention might work. We included one new trial with 10 participants. We need further information on five trials to assess eligibility. Two trials are ongoing. The review now has 9 included studies, involving 994 participants. We excluded any trials that tested interventions delivered to people with stroke living in long‐term care. Therefore, we excluded one previously included study (Sackley 2004). We improved the search strategies and history. We did not handsearch journals. We extended our search of trial registers. Previously, we had included each intervention arm as a combined dataset compared to the control group. In this 2017 update, we selected the intervention that most closely related to our intended intervention. Therefore, we excluded data from leisure promotion intervention arms from two studies: Drummond 1996 and Parker 2001. We incorporated Cochrane's tool for assessing risk of bias into this update. We revisited the critical appraisal of all studies included in previous versions of both reviews updating all assessments of risk of bias. We incorporated 'risk of bias' assessments in analyses by restricting the primary analysis to studies at low and unclear risk of bias. We adhered to the MECIR standards for conduct and reporting.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Activities of daily living	7	749	Std. Mean Difference (IV, Fixed, 95% CI)	0.17 [0.03, 0.31]
2 Death or 'poor outcome' (deterioration or dependency)	5	771	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.71 [0.52, 0.96]
3 Death by the end of scheduled follow‐up	8	950	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.02 [0.65, 1.61]
4 Death or requiring institutional care by the end of scheduled follow up	4	671	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.89 [0.60, 1.32]
5 Death or dependency by the end of scheduled follow‐up	4	659	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.89 [0.64, 1.23]
6 Extended activities of daily living scores	5	665	Std. Mean Difference (IV, Fixed, 95% CI)	0.22 [0.07, 0.37]
7 Mood or distress scores	4	519	Std. Mean Difference (IV, Fixed, 95% CI)	0.08 [‐0.09, 0.26]
8 Sensitivity to missing data (odds of poor outcome: better)	5	857	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.75 [0.56, 1.00]
9 Sensitivity to missing data (odds of poor outcome: worse)	5	857	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.75 [0.56, 0.99]

Methods	RCT
Participants	Hong Kong Total number of randomised participants: 53 Mean age: 72.1 years 66% men BI score at baseline: not available Definition of stroke: unclear Recruitment: inpatients and outpatients who had been discharged from hospital for less than 2 weeks Inclusion criteria: aged over 55, diagnosis of stroke; able to follow instructions; able to communicate using speech; family support at home; required bathing device
Interventions	Additional home‐based intervention in the use of bathing devices (n = 30) versus no intervention (n = 23)
Outcomes	Outcomes were recorded at 3 months after discharge Relevant review outcomes: ADL (measured using FIM) (study authors did not report who carried out assessment); and death at end of follow‐up Study authors also report: users evaluation of satisfaction with AT
Notes	Funding sources/Declarations of interest: not reported Study dates: not reported
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"Subjects in need of bathing devices were randomly assigned to the two groups. The occupational therapists, who were involved in the random assignment procedures, were blind to the study's purpose" Insufficient information about the sequence generation process to permit judgement of yes or no
Allocation concealment (selection bias)	Unclear risk	"The occupational therapists, who were involved in the random assignment procedures, were blind to the study’s purpose" Method of concealment was not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this would introduce bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcome reported using FIM. Study authors did not state who carried out FIM assessments and whether they were blinded. Insufficient information to permit judgement of 'low risk' or 'high risk'
Incomplete outcome data (attrition bias) All outcomes	Low risk	No apparent loss of data. No reports of attrition due to death
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'
Other bias	Low risk	No other sources of bias were identified

Methods	RCT
Participants	UK Total number of participants randomised: 110 Mean age: 75.5 years 37% men BI score at baseline: intervention group median 15 (IQR 2 to 20); control group median 14 (IQR 0 to 20) Definition of stroke: clinical definition of stroke Recruitment: participants recruited prior to discharge from inpatient facility Inclusion criteria: discharged alive from 1 of 2 units regardless of discharge destination
Interventions	Rehabilitation at home by occupational therapists versus usual care. Input at 2, 8, 16 and 24 weeks. Intervention based on the model of human occupation. Interventions included: teaching new skills; facilitating more independence in activities of daily living; facilitating return of function; enabling participants to use equipment supplied by other agencies; information provision to participant and carer; referring to or liaison with other agencies. Service provided by a qualified occupational therapist
Outcomes	Outcomes were recorded at 12 months Relevant review outcomes: ADL (assessed using BI, completion of postal questionnaire by participant), death, extended ADL (assessed using Nottingham Extended ADL Index, completion of postal questionnaire by participant), mood or distress scores (assessed using Geriatric Depression Scale, completion of postal questionnaire by participant), HRQOL (Pearlman's 6‐point Quality of Life Scale) carers' quality of life (assessed using Pearlman's 6‐point Quality of Life Scale) Follow‐up period used in analysis: 12 months
Notes	Funding sources/declarations of interest: 1 author received funding from the Stroke Association Study dates: April 1991 to January 1992 Note: Not explicit that unpaid carers were consented and recruited at baseline. Therefore, we did not perform analysis on carer‐reported outcomes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Participants were randomly allocated to either a control or intervention group. Allocation was carried out by an administrative assistant not otherwise involved with participant care or management, using a predetermined code in individual sealed envelopes; unclear how predetermined code was generated
Allocation concealment (selection bias)	Unclear risk	Participants were randomly allocated to either a control or intervention group. Allocation was carried out by an administrative assistant not otherwise involved with participant care or management, using a predetermined code in individual sealed envelopes. The method of concealment was not described in sufficient detail to allow a definite judgement – the use of sealed assignment envelopes was described but it remained unclear whether envelopes were opaque and sequentially numbered
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind participants and personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	"The status (dead or alive) ... was determined at one year post stroke by a research assistant, not otherwise involved with the patients, by contact with the patients' general practitioner."
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (BI): participant‐reported, completion of postal questionnaire EADL (measured using Nottingham Extended ADL Index): participant‐reported, completion of postal questionnaire; mood or distress scores (measured using Geriatric Depression Scale) participant‐reported, completion of postal questionnaire HRQOL (assessed by Pearlman's 6‐point Quality of Life Scale): participant‐reported, completion of postal questionnaire; carers' quality of life (assessed using Pearlman's 6‐point Quality of Life Scale): carer‐reported, completion of postal questionnaire Follow‐up period used in analysis: 12 months Knowledge of the assigned intervention (see performance bias above, blinding of participants and personnel) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b)
Incomplete outcome data (attrition bias) All outcomes	Low risk	At 1 year, 9/55 of the intervention group were dead and 11/55 of the control group were dead. Of those who were alive, 46/55 of the intervention group and 43/55 of the control group returned their questionnaires. 1 person in the control group moved away and was lost to follow‐up. Missing outcome data balanced in numbers across groups with same reason for missing data
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'.
Other bias	Unclear risk	Significant gender imbalance at baseline. Unclear if gender might influence response to intervention and outcome data

Methods	RCT
Participants	UK Total number of randomised participants: 44 (note: Drummond 1996 included an additional intervention arm with 21 participants who were supported with leisure activities. Participants in the leisure intervention arm were not included in this review update. See Differences between protocol and review) Mean age: 66 years 57% men Definition of stroke: not reported. Participants recruited at discharge from inpatient facility Inclusion criteria: admitted to City Hospital Nottingham Stroke Unit Exclusion criteria: severe comprehension difficulties i.e. score < 3 on Speech Therapy Boston Diagnostic Aphasic Examination; a documented history of dementia; no English language
Interventions	Conventional occupational therapy (n = 21): participants were seen by OT for a minimum of 30 minutes per week for 12 weeks after discharge from hospital and 30 minutes every 14 days for the next 12 weeks. The focus was practice of activities of daily living (washing, dressing, transfers) and where appropriate, treatment of perceptual problems No additional occupational therapy over usual care from health or social services (n = 23)
Outcomes	Outcomes were recorded at 3/6 months: ADL (assessed by Rivermead ADL self‐care section), EADL (assessed by Nottingham EADL), mood/distress scores (assessed by Wakefield Depression) and HRQOL (assessed by Nottingham HealthProfile) Study authors also reported the number and amount of time spent in leisure activities (assessed by Nottingham Leisure Questionnaire) Follow‐up period used in analysis: 6 months Knowledge of the assigned intervention (see performance bias above, blinding of participants and personnel) may have impacted on participant‐reported outcomes and we were unable to judge whether this would have influenced outcome data (Higgins 2011b)
Notes	Funding sources/declarations of interest: funding from the Stroke Association and the Nottingham Fights Stroke Association Study dates: 30 October 1990 to 31 July 1992 Note: data included in the previous review (Legg 2006) were unpublished data from the study authors. We attempted to source the data for the single‐pair wise comparison for this update; the data were no longer available and we were not able to include outcome data for the single intervention arm for all outcomes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"... subjects were randomly allocated to one of the three study groups using restricted randomisation." However, the study authors did not provide sufficient information on the methods used to control the random allocation procedure to permit judgement of 'low risk' or 'high risk'
Allocation concealment (selection bias)	Unclear risk	"Sealed envelopes were opened by administrative staff who then indicated to which group a subject was to be assigned." Unclear if assignment envelopes were opaque and sequentially numbered
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this would introduce bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (measured by Rivermead ADL self‐care section, unclear if completed by independent assessor using objective observation or participant‐reported (interview or completion of questionnaire) EADL (measured by Nottingham EADL): participant‐reported ‐ unclear if interview or self‐completed questionnaire Mood or distress scores (measured by Wakefield Depression Inventory): participant‐reported: unclear if interview or self‐completed questionnaire HRQOL (measured by Nottingham Health Profile): participant‐reported: unclear if interview or self‐completed questionnaire "At three and six months from the date of discharge, all subjects were visited by an independent assessor who did not know to which group they had been allocated." HRQOL and mood were participant‐reported outcomes. Unclear how ADL and EADL data were obtained e.g. objective observation or participant‐reported (interview or completion of questionnaire) Follow‐up period used in analysis: 6 months Knowledge of the assigned intervention (see blinding of participants and personnel above) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Difficult to work out who was missing from what group and why. However, numbers missing at 6 months, were 1 (group 1) and 3 (group 2)
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'. Only data for the Nottingham Leisure Questionnaire were presented for each of the groups. There was selective reporting of comparisons of intervention arms for Nottingham EADL, Nottingham Health Profile, and Wakefield Depression Inventory
Other bias	Low risk	No other sources of risk were identified

Methods	RCT
Participants	UK 111 participants: 53 intervention, 58 control Mean age: 55 years 43% men Clinical definition of stroke Inclusion criteria: first stroke and discharged from hospital and referred to the Social Services occupational therapy department
Interventions	Experimental intervention: enhanced occupational therapy service provided by social services, included provision of equipment; single therapist Comparator intervention: usual care
Outcomes	Outcomes were recorded at 3 and 6 months Relevant review outcomes: performance in ADL (measured by BI), performance in EADL (measured by Nottingham Extended Activities of Daily Living); mood or distress (measured by General Health Questionnaire); carer mood (measured by General Health Questionnaire)
Notes	Funding sources/declarations of interest: financial support from the Stroke Association Study dates: not reported Follow‐up period used in analysis: 6 months Note: not explicit that unpaid carers were consented and recruited at baseline.Therefore, we did not perform analysis on carer‐reported outcomes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Participants randomly allocated by administration clerk; no further details.Insufficient information about the sequence generation process to permit judgement of 'low risk' or 'high risk'
Allocation concealment (selection bias)	Unclear risk	"They were randomly allocated by the clerk using prepared sealed envelopes". Method of concealment not described in sufficient detail to permit judgement – unclear if envelopes were opaque or sequentially numbered
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this introduced bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (measured by BI): unclear how outcome data were obtained i.e. by observation, interview or participant‐completed questionnaire EADL (measured by Nottingham Extended Activities of Daily Living): participant‐reported ‐ unclear if interview or self‐completed questionnaire Mood or distress (measured by General Health Questionnaire): participant‐reported ‐ unclear if interview or self‐completed questionnaire Carer mood (measured by General Health Questionnaire completed by carer) "Six months after entry to the study the patients were assessed at home by an independent assessor, who had not been informed which treatment the patients had received. The EADL scale was sent prior to the visit, to be completed by the patient and picked up at interview. The independent assessor administered the Barthel Index and the GHQ" Follow‐up period used in analysis: 6 months Knowledge of the assigned intervention (see blinding of participants and personnel above) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b)
Incomplete outcome data (attrition bias) All outcomes	Low risk	At 6 months, 45/53 of the experimental group returned the questionnaires. Reported reasons for attrition were death (5), in nursing home (1) withdrew consent (2) At 6 months, 38/58 of the comparator group returned the questionnaires. Reported reasons for attrition were death (7), withdrew consent (6), in nursing home (6), in hospital (1) At 6 months, a greater (non‐significant) number of participants in the usual service group were dead or dependent (nursing home) (n = 13) compared with the enhanced group (n = 6) (Chi² = 2.4, P = 0.1; this result is not significant at P < 0.05). 2/45 in the enhanced group and 6/38 in the usual care withdrew consent. The difference in number of participants who withdrew was not significant (Chi² = 3.04, P = 0.08; this result is not significant at P < 0.05). Missing outcome data balanced across groups with similar reasons for missing data across groups
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'
Other bias	Low risk	No other sources of bias identified

Methods	RCT
Participants	USA 15 participants: experimental 10 (5 participants in 'habit training'; 5 participants in comparison group 1). 5 in comparison group 2 Mean age: 59 years (SD 14) 80% men All participants admitted to Sister Kenny Rehabilitation Institute (SKRI) Abbott‐Northwestern and United Hospitals between 6 November 2006 and 15 March 2007 who met the inclusion/exclusion criteria were given the opportunity to participate in the study Inclusion criteria: participants had the following characteristics: 1. admitted to SKRI for inpatient rehabilitation after onset of first ever stroke, 2. pre‐discharge FIM scores of ≥ 4, 3. discharged to his or her home, 4. had a family carer who was willing to participate Exclusion criteria: 1. a history of stroke, 2. supervision required with cognition and communication (FIM score < 5) at or before discharge, 3. assistance needed to carry out hygiene and dressing tasks ((FIM) scores of < 4)
Interventions	Experimental intervention 1: 'habit training': use of a checklist outlining an individualised, contextual morning self‐care routine + daily adherence reinforcement via a wireless device during a 4 to 5‐week practice period Comparator group 1: use of a checklist outlining an individualised, contextual morning self‐care routine. Participants were given a wireless pocket computer and had to respond to questions about energy levels 3 times per week Both the experimental group and the comparator 1 group worked with the researcher (an occupational therapist) to develop their individualised morning self‐care routine checklist. In addition, the occupational therapist researcher discussed the value of re‐establishing and practicing a regular pattern of day‐to‐day activities as a means of maintaining and achieving independence Comparator group 2: no therapeutic intervention
Outcomes	Outcomes were recorded at the end of the intervention (5 weeks) Relevant review outcomes: performance in ADL (measured FIM, performance in EADL (measured by Frenchay Activities Index)) The study authors also reported: Caregiver Burden Scale
Notes	Funding sources/declarations of interest: none reported Study dates: 6 November 2006 and 15 March 2007
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Prior to the beginning to the study, the principal investigator used a computer‐based random numbers generator to create two tables of three sets of random numbers (corresponding to the three conditions). One table was used to randomise persons with a Medicare case mix index (CMI) of 101‐105 (lower levels of disability at admission to SKRI) and the other table for persons with a CMI of 106–110 (greater degree of disability at admission to SKRI)."
Allocation concealment (selection bias)	High risk	"Once notified that a given patient‐caregiver dyad had consented to participate in the study, the researcher drew a numbered card from an envelope and looked up the number on the appropriate table to determine the condition to which the participant was assigned". Researcher allocating participants could possibly see the next assignment ‐ assignment envelopes used without proper safeguards and use of an open allocation schedule
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this would introduce bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (measured by Functional Independence measure): participant‐reported outcomes by interview. EADL (measured by Frenchay Activities Index): participant‐reported outcomes by interview. "Within one‐week of discharge from SKRI, former patient‐caregiver dyads participated in a telephone interview with a member of the research team in which the Functional Independence Measure, a modified version of the Frenchay Activities Index, and the Caregivers Burden Scale were administered. The caller was an SKRI rehabilitation therapist and member of the research team who was blinded to the participants’ group assignment.... These instruments were re‐administered via telephone four‐five weeks later by the same caller to the same person answering the questions at pretest." Follow‐up period used in analysis: 5 weeks Knowledge of the assigned intervention (see blinding of participants and personnel above) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b)
Incomplete outcome data (attrition bias) All outcomes	Low risk	17 dyads consented to participate. 2 participants did not receive the intervention after randomisation (1 = return to hospital; 1 = difficulties in scheduling visits) – not included in any data collection. Did not state which groups they were randomised to, therefore not clear if this was balanced between groups. Large percentage loss but only very small number of participants, judged to be low risk of attrition bias
Selective reporting (reporting bias)	Unclear risk	Study author did not report prospective registration with clinical trials register. Not feasible to judge reporting bias
Other bias	Low risk	No other sources of bias identified

Methods	Cross‐over RCT
Participants	UK Total number of participants recruited: 30 Mean age: 68 years 53% men Definition of stroke: not reported Recruitment: participants recruited at discharge from inpatient facility Exclusion criteria: blind; deaf; unable to understand or speak English prior to stroke onset
Interventions	Domiciliary occupational therapy over a 3‐month period provided by a senior occupational therapist (n = 15). Amount of therapy provided at therapist's discretion. Components of intervention: dressing practice on a regular basis; teaching participants and carers specific dressing techniques, energy conservation techniques, advice on clothing adaptation. Relative/carer involvement in therapy programme and between therapy sessions homework. Single therapist. No contact with occupational therapist. Usual care (n = 15)
Outcomes	Outcomes were recorded at 3/6 months Relevant review outcomes: performance in activities of daily living (measured by Rivermead ADL self‐care section); HRQOL (assessed using Nottingham Health Profile). Other reported outcomes included: Nottingham stroke dressing assessment
Notes	Outcome data recorded at 3 months used in analyses i.e. before cross‐over period Funding sources/declarations of interest: financial support from the Stroke Association Study dates: not reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information about the sequence generation process to permit judgement of 'low risk' or 'high risk'
Allocation concealment (selection bias)	Unclear risk	The method of concealment was not described to permit judgement of 'low risk' or 'high risk'
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind study participants and treating therapists delivering the intervention to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this would introduce bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (measured by Rivermead ADL self‐care section): unclear how outcome data were obtained i.e. by observation, interview, or participant‐completed questionnaire. HRQOL (assessed using Nottingham Health Profile): participant‐reported outcome, unclear if interview or self‐completed questionnaire "At three months and six months both groups of patients were assessed by an independent assessor who was unaware of the group to which the patient had been allocated." Follow‐up period used in analysis: 6 months Knowledge of the assigned intervention (see blinding of participants and personnel above) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b).
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Number of participants allocated to the AB study arm received treatment A (dressing) first, followed by treatment B (no intervention) – number of participants at end of treatment A and treatment B not explicitly stated. Insufficient information to make a decision of 'high' or 'low' risk of bias
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'
Other bias	Low risk	No other sources of bias identified

Methods	RCT
Participants	UK 185 participants: 94 intervention, 91 control Mean age: 74 years 51% men Baseline functional status: Median BI score at baseline: intervention group 18 (IQR 15 to 20); control group 18 (IQR 15 to 20) Clinical definition of stroke Participants were recruited < 1 month after stroke onset from home Exclusion criteria: > 1 month after stroke onset; history of dementia; living in a nursing or residential home; unable to speak or understand English prior to stroke onset
Interventions	Experimental intervention: occupational therapy intervention for a period of 5 months. Frequency of visits arranged between therapist, participant, and carer (if appropriate). Mean of 5.8 visits per participant. Aim of therapy was to achieve independence in personal (bathing, dressing, feeding, stair mobility) and instrumental activities of daily living (outdoor mobility, driving a car, using public transport, household chores). Homework tasks were set in between therapy sessions Comparator intervention: no occupational therapy
Outcomes	Outcomes were recorded at 6 months Performance in ADL (measured by BI); performance in ADL (measured by Nottingham Extended Activities of Daily Living Index); mood or distress (measured by General Health Questionnaire‐28 item) Other reported outcomes included: London Handicap Scale, Rivermead motor assessment (gross function), carer mood (General Health Questionnaire‐28) and carer Strain Index
Notes	Funding sources/declarations of interest: financial support from the Stroke Association Study dates: February 1994 to March 1998 Follow‐up period used in analysis: 6 months Unpaid carers: not explicit that unpaid carers were consented and recruited at baseline.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Our single‐blind randomised controlled trial. Randomisation was by numbered, sealed, opaque envelopes prepared from random‐number tables."
Allocation concealment (selection bias)	Low risk	"Patients were then allocated randomly to occupational therapy or to no intervention (control group). Randomisation was by numbered, sealed, opaque envelopes prepared from random‐number tables."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind personnel to group allocation due to nature of intervention
Blinding (mortality)	Low risk	Method of obtaining data on death not reported; review authors did not believe this would introduce bias
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Outcomes assessed: ADL (measured by BI): unclear how outcome data were obtained (i.e. by observation, interview, or participant‐completed questionnaire) NEADL (measured by Nottingham Extended Activities of Daily Living Index): participant‐reported outcome: unclear if interview or self‐completed by participant Mood (measured by General Health Questionnaire‐28 item): participant‐reported outcome; unclear if interview or self‐completed by participant An independent assessor who was unaware of treatment allocation assessed the participants in their homes Follow‐up period used in analysis: 6 months Knowledge of the assigned intervention (see blinding of participants and personnel above) may have impacted on participant‐reported outcomes and we were unable to judge whether this would influence outcome data (Higgins 2011b)
Incomplete outcome data (attrition bias) All outcomes	Low risk	At 6 months, 84/94 of the experimental group completed the outcomes assessment. The reasons for attrition (withdrew consent, died) reported. At 6 months, 79/91 of the comparator group completed the outcomes assessment. The reasons for attrition (withdrew consent, died) reported. Reasons for missing outcome data unlikely to be related to true outcome
Selective reporting (reporting bias)	Unclear risk	Clinical trials registration or prospective publication of a protocol not reported in the full text. Insufficient information to permit judgement of 'low risk' or 'high risk'
Other bias	Low risk	No other sources of bias identified

Study	Reason for exclusion
Abizanda 2011	RCT, occupational therapy intervention, population of stroke participants < 50%
Andrea 2003	Identified from Cochrane Stroke Group Trials Register, RCT, intervention not performed by occupational therapist
Bai 2012	RCT, intervention for people with stroke, rehabilitation programme included physiotherapy and occupational therapy
Chaiyawat 2012	RCT, intervention for people with stroke, rehabilitation programme performed by a physiotherapist
Cross 2014	Occupational therapy intervention for ADL in people with stroke, not RCT
Desrosiers 2007	RCT, people with stroke, included an occupational therapist as a supervisor but control group also received an active comparison intervention
Egan 2007	RCT, people with stroke, occupational therapy intervention but primary outcome was not ADL, used COPM
Guidetti 2011	RCT, people with stroke, client‐centred self‐care intervention versus ordinary training, therefore comparison of two active occupational therapy interventions.
Jing 2006	RCT. people with stroke, occupational therapy + exercise versus exercise, therefore occupational therapy in combination with other treatment versus active comparator
Kessler 2014	RCT, people with stroke, occupational performance coaching
Landi 2006	RCT, intervention for people with stroke: intervention group received occupational therapy, both groups received physiotherapy. Excluded as control group had an active comparator
Li 2008	RCT, people with stroke, rehabilitative training + occupational therapy twice a day versus rehabilitative training. Control group included an active comparator
Park 2011	RCT, people with stroke but focused on walking only, not occupational therapy intervention
Rasmussen 2016	RCT, people with stroke, rehabilitation with multidisciplinary team, and also active comparison group
Rodgers 2015	Also called EXTRAS study, RCT, intervention for people with stroke. Extended rehabilitation service which might involve occupational therapist as part of a multidisciplinary team
Sackley 2004	Previously listed in 'Studies awaiting classification' (Legg 2006). Occupational therapy in nursing homes, therefore not eligible. Study currently included in Fletcher‐Smith 2013
Sahebalzamani 2009	RCT, intervention for people with stroke, performed by nursing team not occupational therapists
Skidmore 2012	RCT, people with stroke, intervention focused solely on individuals with cognitive impairments after acute stroke
Skidmore 2016	RCT, occupational performance coaching to improve cognitive functioning for people with stroke. Recruited participants with impairment in cognitive functions
Stalhandske 1997	Previously listed in 'Studies awaiting classification' (Legg 2006). Unable to access any full publication of this study and abstract had insufficient information to clarify eligibility (abstract from 1997), therefore excluded
Sun 2001	Previously listed in 'Studies awaiting classification' (Legg 2006). Unable to access any full publication of this study and abstract had insufficient information to clarify eligibility (abstract from 1997), therefore excluded
Tuncay 2006	RCT, people with stroke, self‐care educational intervention with physiotherapists, not occupational therapist
Walker 2012	RCT, compared 2 types of occupational therapy for stroke patients with cognitive impairments. Excluded because it tackled a specific impairment; also comparison of different occupational therapy techniques
Whitehead 2016	RCT, people with stroke. Occupational therapy + home care reablement vs home care reablement, therefore active comparator
Yu 2009	RCT, rehabilitation programme for people with stroke. Study authors did not state involvement of occupational therapist
Zhang 2008a	RCT, people with stroke, intervention not performed by an occupational therapist
Zhu 2007	RCT, people with stroke, intervention included physiotherapy and occupational therapy

Methods	RCT
Participants	164 people with stroke
Interventions	Rehabilitation versus control groups
Outcomes	Neurological deficit scores Barthel Index
Notes	Abstract only. Difficult to ascertain the components of rehabilitation

Methods	RCT
Participants	25 people with stroke
Interventions	Control and intervention wards. In the intervention wards, self‐care activities were handed over from nursing staff to therapy assistants to practice self‐care activities
Outcomes	Motor Activity Log Amount Scale Motor Activity Log How Well Action Research Arm Test (ARAT) Physiotherapy clinical outcome variables Berg Balance Scale Barthel Index
Notes	Abstract only. Unclear whether there was any occupational therapy involvement

Methods	RCT
Participants	80 people with stroke
Interventions	3 grades regular rehabilitation versus no rehabilitation
Outcomes	Fugl‐Myer
Notes	Abstract only. Difficult to ascertain the components of 3 grades regular rehabilitation

Trial name or title	Efficacy of participation in daily life promotion program for patients with chronic stroke
Methods	RCT
Participants	Adults with stroke aged between 20 and 90 years Inclusion criteria: participants must be 6 months post stroke onset participants must be able to follow instructions and have no other cognition impairment, Mini Mental State Examination scale score > 20 participants must be able to communicate with others Exclusion criteria: a clinical diagnosis of: dementia, psychosis musculoskeletal disorders or nervous system diseases. congestive heart failure, hypertension, atrial fibrillation An adult with stroke who was not from Taiwan
Interventions	Experimental intervention: daily life promotion programme Comparator intervention: general rehabilitation treatment
Outcomes	Primary outcome measures at 1 year: Stroke Impact Scale (SIS) Secondary outcome measures at 1 year: Postural Assessment Scale for Stroke patients (PASS) Action Research Arm Test (ARAT) Fugl‐Meyer Assessment (FMA) Self‐rated abilities for health practice scale WHO Quality of Life‐BREF (Taiwan version) Taiwan Instrumental Activities of Daily Living (TIADL) Barthel Index (BI)
Starting date	June 2016
Contact information	I Ming Hsiao (PI) skipbeat227@gmail.com Yi‐Jiun Yang skipbeat227@gmail.com
Notes

Trial name or title	Effectiveness of FACoT for individuals post stroke
Methods	RCT
Participants	Inclusion criteria: age > 18 years minimum of six month post stroke mild to moderate stroke (NIHSS ≤ 10) independent prior to the stroke could understand and speak Hebrew Exclusion criteria: with no other neurological or psychiatric conditions without dementia or depression (GDS < 10)
Interventions	Experimental group: FACoT: 10 treatment sessions are provided on a 1‐to‐1 basis. Each treatment session includes: functional activities, cognitive activities and strategies (pencil‐pen treatment), and behavioural strategies Comparator intervention: control group receiving standard care ‐ cognitive and functional assessment
Outcomes	Primary outcome measures: Change between baseline (week 0) to time 1 (postintervention, 10 to 13 weeks later) and between baseline to time 2 (follow‐up 1 month later): Canadian Occupational Performance Measure (COPM). Secondary outcome measures: Changes in scores between baseline (week 0) to time 1 (postintervention, 10 to 13 weeks later) and between baseline to time 2 (follow‐up 1 month later): Instrumental Activities of Daily Living (IADL) scale Reintegration to Normal Living Index (RNL) Short Form‐12v2 Health Survey (SF‐12v2) The Daily Living Self Efficacy scale (DLSES) Patient competency rating scale University of Rhode Island Change Assessment (URICA) Montreal Cognitive Assessment (MoCA) Trail making test (TMT) Zoo‐map Dysexecutive Questionnaire (DEX)
Starting date	April 2016
Contact information	Tal Adamit, PHD student: adamit33@bezeqint.net Jeffrey Shames: j.shames@gmail.com
Notes

Occupational therapy for adults with problems in activities of daily living after stroke

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Assessment

Partnership

Programme plan

Activities of daily living as a therapeutic medium and goal

Intentionality

Appropriate

Incremental and progressive

Healthy perceptions, attitudes, and behaviour‐focused

Active engagement in authentic practice

Enduring and applicable

Evaluation and accountability

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

GRADE and 'Summary of findings'

Results

Description of studies

Results of the search

Included studies

Excluded studies

Ongoing studies

Studies awaiting classification

Risk of bias in included studies

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Overall risk of bias

Effects of interventions

Summary of findings for the main comparison

Primary outcomes

Personal activities of daily living

1

Death or a poor outcome

2

Secondary outcomes

Death

3

Death or institutional care

4

Death or dependency

5