15
Review The effectiveness of psychosocial interventions for stroke family caregivers and stroke survivors: A systematic review and meta-analysis § Ho Yu Cheng *, Sek Ying Chair, Janita Pak-Chun Chau The Nethersole School of Nursing, The Chinese University of Hong Kong, Hong Kong Special Administrative Region Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.1. Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2. Study selection and appraisal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2.1. Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.2. Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.3. Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.4. Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.5. Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.2.6. Exclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.3. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4. Data synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.1. Search results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.2. Methodological quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Patient Education and Counseling 95 (2014) 30–44 A R T I C L E I N F O Article history: Received 23 August 2013 Received in revised form 17 December 2013 Accepted 7 January 2014 Keywords: Stroke Family caregivers Psychosocial interventions Systematic review Meta-analysis Physical health Psychosocial wellbeing Quality of life A B S T R A C T Objective: To evaluate the effectiveness of psychosocial interventions for family caregivers on their psychosocial and physical wellbeing, quality of life, and the use of healthcare resources by stroke survivors. Methods: Electronic English and Chinese bibliographic databases were searched (inception to January 2012) for clinical trials. Two reviewers independently selected and appraised study quality. When possible, data from randomized controlled trials (RCTs) were statistically pooled. Otherwise, a narrative summary was used. Results: Eighteen studies (psychoeducation and social support group) were included. Pooled analysis of two individual psychoeducation programs showed a small effect on improving family functioning (SMD: 0.12; 95% CI: 0.23 to 0.01; p = 0.03). Caregivers receiving psychoeducation that aimed at equipping caregivers with the skills of problem-solving, caregiving, and stress-coping appeared to have a more positive influence on the caregivers’ psychosocial wellbeing and a reduced use of healthcare resources by stroke survivors. Conclusion: Evidence on the effects of psychosocial interventions was limited. More RCTs of multifaceted psychoeducation programs are needed to further examine the optimal dose and format. Practical implications: To support caregivers across the stroke trajectory, the core skills of problem- solving and stress-coping should be included in the psychosocial interventions. ß 2014 Elsevier Ireland Ltd. All rights reserved. § ‘‘I confirm all the patient/person(s) have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story.’’. * Corresponding author at: The Nethersole School of Nursing, 8/F, Esther Lee Building, The Chinese University of Hong Kong, Shatin, N. T., Hong Kong Special Administrative Region. Tel.: +852 3943 433; fax: +852 2603 5935. E-mail addresses: [email protected], [email protected] (H.Y. Cheng). Contents lists available at ScienceDirect Patient Education and Counseling jo ur n al h o mep ag e: w ww .elsevier .co m /loc ate/p ated u co u 0738-3991/$ see front matter ß 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.pec.2014.01.005

Cheng 2014 Stroke Family Caregivers and Stroke Survivors

  • Upload
    dorka9

  • View
    9

  • Download
    0

Embed Size (px)

DESCRIPTION

stroke family caregivers and stroke survivors.pdf

Citation preview

Page 1: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Patient Education and Counseling 95 (2014) 30–44

Review

The effectiveness of psychosocial interventions for strokefamily caregivers and stroke survivors: A systematic reviewand meta-analysis§

Ho Yu Cheng *, Sek Ying Chair, Janita Pak-Chun Chau

The Nethersole School of Nursing, The Chinese University of Hong Kong, Hong Kong Special Administrative Region

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.1. Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.2. Study selection and appraisal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

2.2.1. Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.2.2. Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.2.3. Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.2.4. Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.2.5. Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.2.6. Exclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.3. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2.4. Data synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.1. Search results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3.2. Methodological quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

A R T I C L E I N F O

Article history:

Received 23 August 2013

Received in revised form 17 December 2013

Accepted 7 January 2014

Keywords:

Stroke

Family caregivers

Psychosocial interventions

Systematic review

Meta-analysis

Physical health

Psychosocial wellbeing

Quality of life

A B S T R A C T

Objective: To evaluate the effectiveness of psychosocial interventions for family caregivers on their

psychosocial and physical wellbeing, quality of life, and the use of healthcare resources by stroke

survivors.

Methods: Electronic English and Chinese bibliographic databases were searched (inception to January

2012) for clinical trials. Two reviewers independently selected and appraised study quality. When

possible, data from randomized controlled trials (RCTs) were statistically pooled. Otherwise, a narrative

summary was used.

Results: Eighteen studies (psychoeducation and social support group) were included. Pooled analysis of

two individual psychoeducation programs showed a small effect on improving family functioning (SMD:

�0.12; 95% CI: �0.23 to �0.01; p = 0.03). Caregivers receiving psychoeducation that aimed at equipping

caregivers with the skills of problem-solving, caregiving, and stress-coping appeared to have a more

positive influence on the caregivers’ psychosocial wellbeing and a reduced use of healthcare resources by

stroke survivors.

Conclusion: Evidence on the effects of psychosocial interventions was limited. More RCTs of multifaceted

psychoeducation programs are needed to further examine the optimal dose and format.

Practical implications: To support caregivers across the stroke trajectory, the core skills of problem-

solving and stress-coping should be included in the psychosocial interventions.

� 2014 Elsevier Ireland Ltd. All rights reserved.

§

Contents lists available at ScienceDirect

Patient Education and Counseling

jo ur n al h o mep ag e: w ww .e lsev ier . co m / loc ate /p ated u co u

‘‘I confirm all the patient/person(s) have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of

the story.’’.

* Corresponding author at: The Nethersole School of Nursing, 8/F, Esther Lee Building, The Chinese University of Hong Kong, Shatin, N. T., Hong Kong Special Administrative

Region. Tel.: +852 3943 433; fax: +852 2603 5935.

E-mail addresses: [email protected], [email protected] (H.Y. Cheng).

0738-3991/$ – see front matter � 2014 Elsevier Ireland Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.pec.2014.01.005

Page 2: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 31

3.3. Study characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.4. Details of the interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.4.1. Psychoeducation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

3.4.2. Social support group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5. Effects of interventions on family caregivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.1. Psychosocial wellbeing: burden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.2. Psychosocial wellbeing: caregiving competency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.3. Psychosocial wellbeing: depression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.5.4. Psychosocial wellbeing: anxiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

3.5.5. Psychosocial wellbeing: social support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.5.6. Family functioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.5.7. Physical health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.5.8. Quality of life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.5.9. Intervention satisfaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

3.6. Effects of interventions on stroke survivors: use of healthcare resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

4. Discussion and conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

4.1. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

4.2. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4.3. Practical implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

1. Introduction

Stroke is a major cause of disability worldwide. Its abrupt onsetand long-term neurological or cognitive impairments imposevarious challenges on the life situations of stroke survivors andtheir family members [1,2]. Stroke not only calls for readjustmentsin a caregiver’s life, but also requires modifications of theinterfamilial relationships [3].

The new role of being a stroke caregiver comes unexpectedly tothe family members [4]. The amount and duration of care, theresponsibility to stroke survivors, and the lack of time to devote totheir own health behaviors and social activities drain the physical,psychological and social resources of family caregivers [5], and alsoaffect their quality of life (QOL) [6]. The actual and perceivedimbalance between the demands of caregiving and the support forcaregivers is linked to psychological morbidities, including burden,psychological distress, and depressive symptomatology. Approxi-mately 30–68% of stroke family caregivers were found to haveanxiety or depressive symptoms [7,8]. In one study, the caregiversof stroke survivors were found to have higher levels of depressionand anxiety compared to caregivers of other neurological diseases[9]. The strain experienced by caregivers increases their risk ofcardiovascular diseases and their mortality risk by 23% and 63%,respectively, when compared with non-caregivers [10,11]. Thesenegative impacts on the caregivers’ wellbeing could result in thebreakdown of care and the institutionalization of stroke survivors,which imposes a substantial cost on healthcare systems [12,13].Thus, improving the family caregivers’ ability to provide quality ofcare without detriment to their own health, and empowering thefamily caregivers to have a life of their own are indispensable [14].

A meta-analysis of information provision interventions ofstroke knowledge and practical caregiving skills was found to havea small effect on improving the caregivers’ knowledge, but not ontheir psychological health [15]. Different types of psychosocialinterventions, such as counseling, psychoeducation or supportgroups, showed mixed effects on the psychological and socialaspects of caregivers in previous systematic reviews [16–18]. Arecent systematic review of eight studies of non-pharmacologicalinterventions for stroke family caregivers showed that psychoso-cial interventions had no significant effect on reducing thecaregivers’ strain and their level of depression [18]. Visser-Meileyet al. [17] also examined the effectiveness of interventions forstroke family caregivers and reported that counseling interven-tions appeared to have the most positive outcome on thecaregivers’ psychological health. Besides, Brereton et al. [16]

suggested that the teaching of problem-solving techniques and thetraining in caregiving skills could improve the caregivers’ QOL andwellbeing. However, the evidence was based on one to two studiesin these reviews. There was insufficient evidence on which optimalformat, type and essential components of the psychosocialinterventions should be provided to the stroke family caregivers.There was also no synthesized evidence of the effect on thecaregivers’ physical health [17,19], despite the fact that thepresence of psychological distress might contribute to a deterio-ration in the caregivers’ physical health [20].

Today, we are facing the challenge of the increasing prevalenceof stroke-related disability because of the growth of the agingpopulation and the prolonged life span [21]. Therefore, supportingthe needs of stroke family caregivers at the right time, right placeand in the right format throughout the transition of care is timely.The aim of this systematic review is to identify the best availableevidence related to the effectiveness and dose of caregiverpsychosocial interventions on the psychosocial wellbeing, physicalhealth and QOL of family caregivers, and the use of healthcareresources by the stroke survivors.

2. Methods

We followed the guidelines recommended by the Joanna BriggsInstitute when we conducted this review.

2.1. Search strategy

Clinical trials were searched from electronic English and Chinesebibliographic databases from their inception to January 2012, byusing title, abstracts, keywords, MeSH, and index terms (Table 1).Besides, a manual search of relevant journals, gray literatures, onlinedatabases search and websites was also performed. Reference listsand bibliographies of all relevant articles were also reviewed.Appendix A shows a typical searching strategy.

2.2. Study selection and appraisal

Two independent reviewers (HYC and SYC) screened the titlesof all identified papers for their relevance. If the title wasinconclusive, the abstracts or full texts of the study were retrievedfor further analysis. Both reviewers independently assessed allabstracts or full texts of the potential studies by using the studyeligibility verification form developed for this systematic review.Any discrepancies between the reviewers were resolved by

Page 3: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 1Data sources for the review.

Databases searched for English publicationAcademic Search Alumni Edition

Academic Search Premise

British Nursing Index and Archive

CINAHL Plus

ERIC

Controlled Clinical Trials Database

Cochrane Library

EMBASE

Global Health

Stroke Trials Registry (www.strokecenter.org)

Ovid MEDLINE1

PsycINFO

SCOPUS

Stroke Registry in Cochrane Stroke Group

Web of Science-with conference proceedings

Databases searched for Chinese publicationChina Journal Net

Chinese Biomedical Literature Database

Chinese Medical Current Content,

HyRead

Taiwan Electronic Periodical Services

WanFang Data

Databases searched for gray literature or unpublished studiesAgency for Healthcare Research and Quality,

Academic Archive On-line

Chinese Electronic in Thesis and Dissertation Service

Digital Dissertation Consortium

Gray Literature Report (via New York Academy of Medicine)

Index to Thesis

Lancashire Care Library and Information Service

MEDNAR

National Library of Medicine Gateway

Netting the Evidence

ProQuest Dissertation and Thesis

The Networked Digital Library of Theses and Dissertations

WanFang Data: Chinese Dissertation Database

Searching terms used(stroke* OR ((cerebrovascular OR cerebro vascular OR cerebral vascular OR

brain vascular) AND (accident* OR event* OR stroke* OR disorder* OR

disease* OR infarct OR apoplex*)) OR CVA OR apoplexy*)

AND (Caregiver* OR care giver* OR carer* OR ((family OR informal OR

primary OR spous*) AND (caregiver* OR carer* OR care giver*)))

AND (Psychotherap* OR counsel* OR cognitive therapy OR behavio?ral

therapy OR ((psychosocial OR cognit* OR behavio?r* OR cognitive

behavior?ral OR peer OR problem solving) AND (intervention* OR therap* OR

support* OR partnership*)))

AND (health OR quality of life OR depression* OR anxiety* ((physical OR

psychosocial OR psychological OR care* OR social) AND (wellbeing* OR

health* OR stress* OR strain* OR burden OR Burthen OR support OR

condition* OR network* OR competenc*))

Chinese search terms were based on the terminology used in Taiwan and

China, included: ‘‘ ’’ (cerebral stroke), ‘‘ ’’(stroke), ‘‘ / ’’

(cerebrovascular accident/diseases), ‘‘ ’’ (intervention), ‘‘ ’’

(primary caregiver), ‘‘ ’’(support), ‘‘ ’’ (psychological), ‘‘ ’’ (social

activities), ‘‘ / ’’ (quality of life), ‘‘ ’’ (family), ‘‘ ’’

(rehabilitation).

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4432

discussion with the third reviewer (JPC). The eligible studiesshould have the following features:

2.2.1. Population

Adult patients (�18 years old) with a clinical diagnosis ofischemic or hemorrhagic stroke and who were community-dwelling with their family caregivers post-stroke. Adult familycaregivers (�18 years old) were the family members of strokesurvivors and they were the primary caregivers. In this review, the

primary caregiver was defined as the person who spent either themost time in providing care to stroke survivor or who wasidentified as such by the stroke survivor.

2.2.2. Intervention

In this review, ‘psychosocial interventions’ referred to thoseinterventions applying the cognitive, behavioral and/or socialmechanisms of action (e.g., counseling, psychoeducation, behav-ioral or cognitive interventions, social support group) [22] thataimed to improve the psychosocial and physical wellbeing ofcaregivers and the outcomes of stroke survivors. There was norestriction on the formats (individual, group, telephone or web-based), setting (hospital, home or community), duration andfrequency of intervention.

2.2.3. Comparison

The comparator group could either be ‘usual care’ or ‘withoutany active intervention’.

2.2.4. Outcomes

The primary outcomes of interest included (a) psychosocialwellbeing (specific outcomes of interest included but were notlimited to anxiety, depression, caregiving burden and strain,caregiving competency, and social support), (b) physical health,and (c) the QOL of family caregivers. The secondary outcomes ofinterest included family functioning, intervention satisfaction,hospital re-admissions, the placement to a residential or nursingcare home of the stroke survivors, outpatient service contacts bythe family caregivers and the stroke survivors.

2.2.5. Studies

The review included all experimental studies that allocatedparticipants randomly (RCTs) and those that were not trulyrandom in their allocation of participants (quasi-RCTs). Prospec-tive observational studies which observed the entire course ofpsychosocial interventions and the outcomes of participants werealso considered.

2.2.6. Exclusion criteria

(a) Interventions that were targeted exclusively on improvingpractical caregiving skills, for example, caregiver training on liftingand transferring skills [23], as the improvement in psychosocialwellbeing might be indirect, (b) interventions that provided purelystroke-related information to caregivers or focused on liaison withother services [24,25], (c) studies that described only thepsychosocial intervention of stroke family caregivers and withoutan analysis of the intervention outcome, (d) case studies, non-English and non-Chinese studies.

Two independent reviewers (HYC and SYC) assessed themethodological quality of all eligible studies by using The CriticalAppraisal Checklists from the Joanna Briggs Institute-MetaAnalysis of Statistical Assessment and Review Instrument (JBI-MAStARI) [26] for randomized and pseudo-randomized studies,cohort/case control studies, and descriptive/case series studies.Any disagreement was resolved through joint assessment anddiscussion with the third reviewer (JPC).

2.3. Data extraction

Two reviewers (HYC, JPC) independently extracted the precisedetails of the included study (e.g., the study’s setting and method;characteristics of participant and intervention; the outcomemeasures and the results) by using the modified version of theJBI-MAStARI data extraction form. Five potential articles were usedin pilot testing to ensure the consistency of data extraction. Whenthe studies were evaluating two interventions (e.g., education vs.

Page 4: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Citations ident ifi ed thro ugh databases sea rching (n = 8,039)

English citations: 7,332; Chinese citations: 707

Removal of duplica tes and irrelevant citations

(n = 7,884)

English citations: 7,199; Chinese citations: 685

Citation potential ly rele vant and ful l text retrieved (n = 155 )

English citation s: 133; Chinese citation s: 22

Addition al citations identifi ed from

references and bibliographies of retrie ved

full text (n = 12 English citations)Studies excluded due to irrele vance to the

systematic review (n = 125)

English citations: 114; Chinese citation s: 11

Ful l Text ass essed for eligib ility and appraisal of methodologic al

quality (n = 42 )

English citations: 3 1; Chinese citation s: 11

Stud ies included: 24 citations of 18 studies in English

No citations in Chinese

Studies excluded (n = 18)

English citations:

- No sub -group an alysis on stroke caregiver (n = 5)

- Insu fficient information on interven tion (n = 1)

- No publi cation ide ntified (n = 1)

Chinese citations:

- Insuffici ent infor mation on the type s and

componen ts of interven tion, the

socio-de mograph ic da ta of the parti cipants an d/ or

the method s of outcome measu remen ts (n = 1 0)

- No sub group an alysis on stroke c aregivers (n = 1)

Fig. 1. Flow diagram of study retrieval and selection.

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 33

counseling [27], and information provision vs. cognitive behavioralintervention [28]), only the data from counseling/cognitivebehavioral intervention and the no-treatment control groupswere extracted. Besides, if the study consisted of two controlgroups [29] (the attention-control group and the concurrentcontrol group), only the data from the attention-control groupwere extracted.

2.4. Data synthesis

The included studies were categorized in accordance with thetypes and formats of the interventions being conducted. Resultsfrom comparable groups of studies were pooled in a statisticalmeta-analysis using RevMan 5.1 [30]. In this review, an a prioriselection plan of outcome measures was used. The most commonlyused measure (e.g., the Caregiver Strain Index for caregiver burden)was selected for data synthesis when there was more than oneinstrument measuring the same outcome in order to minimize thepotential bias in selecting an outcome measure that achievedbetter results. For the Medical Outcome Short Form-36, if thecalculation of the physical component score and the mentalcomponent score was not possible [29,31], the domain scores ofphysical functioning and mental health were selected; this wasbecause they were the most valid measures of physical health andmental health, respectively [32]. For any missing data from thepublished reports, the principal investigators were contacted in anattempt to obtain the original data. No imputation of missingvalues was attempted in this review.

For the RCTs, the standardized mean difference or meandifference, and a 95% confidence interval (95% CI) were calculatedfor the summary effect of continuous data measured by differentscales and the same scale, respectively. Data from each RCT werepooled as appropriate by using a fixed effect model, except wheresubstantial heterogeneity existed according to the I2 statistic(I2 > 50%) [33]. When I2 > 50%, the studies were combined byusing random-effects meta-analysis models if the studies wereconsidered to be clinically similar. Otherwise, a narrative summarywas used.

3. Results

3.1. Search results

A total of 8039 English and Chinese citations were identifiedfrom the search (Fig. 1). Of these, the full texts of 155 potentiallyrelevant studies were retrieved for more detailed review. Twelvemore articles were identified and retrieved after reviewing thereference lists and bibliographies of the retrieved articles. A total of125 studies were excluded because they were irrelevant to thereview. Among the 42 studies which underwent a full assessmentof eligibility and appraisal of methodology, seven relevant studiespublished in English were excluded because there was noinformation of the intervention effects on the subgroup of strokeparticipants, and attempts made to contact the authors to seekfurther information were unsuccessful [34–40]. In addition, 11citations of eight studies published in Chinese were excluded [41–51] due to insufficient information on the design of interventionsand/or the methods of outcome measurement were provided.Thus, 24 articles of 18 studies that met the inclusion criteria wereincluded in this review.

Of these 18 included studies, we attempted to contact elevenauthors for further details on the outcome data and subgroupanalysis on stroke caregivers, of which four were not contactable[29,52–54]. For the remaining seven successful attempts, theauthors of three studies provided further details of their studies[55–57].

3.2. Methodological quality

The methodological quality of the 13 RCTs or quasi-RCTs issummarized in Table 2. Incomplete details on methods of randomassignment and lack of allocation concealment were the concernson the risk of selection bias among these studies. Besides,participants actually knew about their group allocation becausethey received active intervention when compared with theparticipants of the no-treatment control group [31,52,53,58–60]or the waitlist control group [61,62]. Thus, the blinding ofparticipants was not feasible among eight included randomizedstudies which caused concern about the risk of performance bias.In addition, detection bias may also be present because theevaluations of psychosocial outcomes in these studies were mainlysubjective and reported by the participants. There were only twostudies [29,53] which clearly stated the method of blinding to theassessor. The risk of attrition bias due to insufficient informationabout the missing data (including the number and reasons of drop-out of participants in the intervention group and the control group,the statistical method used in dealing with missing data) was alsonoted. On the other hand, two studies [29,54] employed anattention-control group, in which performance bias might beminimized. Also, two studies [61,62] adopted a waitlist controlgroup, so that the effects of delayed treatment and participantexpectancy could be investigated [63].

Of the remaining studies, two of them were non-RCTs [56,60],two were historical cohort studies [28,64], and one was a pre- andpost-test study [57]. The inclusion criteria were clearly stated ineach study, yet there was insufficient information related to thesample characteristics in relation to the study population andthe sample size calculation. Although these non-randomizedstudies imposed the risk of selection bias, the historical cohortstudies employed a matched cohort group to deal with theconfounding factors, such as age, gender and depression level.

Page 5: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 2Methodological quality of the 13 RCTs/quasi-RCTs according to the criteria of JBI critical appraisal (‘‘H’’ = ‘‘Yes’’; ‘‘�’’ = ‘‘No’’; ‘‘?’’ = ‘‘unclear’’).

Bakas

et al.

[54]

Clark

et al.

[52]

Draper

et al.

[61]

Evans

et al.

[27]

Forster and

Young [65]

Grant

et al.

[29]

Grasel

et al.

[58,75]

Hartke and

King [66]

Johnston

et al. [53]

Marsden

et al. [62]

Perrin

et al.

[67]

Pierce

et al.

[59]

Shyu

et al.

[55,31]

1. Was the assignment to

treatment group

truly random?

H H ? H H ? � ? H H H ? ?

2. Were participants

blinded to

treatment allocation?

H � � � � H � � � � � � ?

3. Was allocation to

treatment groups

concealed from

the allocator?

? H ? H H � � � ? H H � �

4. Were the outcomes of

people who withdrew

described and included

in the analysis?

� � � � � ? � � H H � � �

5. Were those assessing

outcome blind to the

treatment allocation?

? ? � H ? H ? ? H � ? ? �

6. Were the control and

treatment groups

comparable at entry?

H H H H H � H H H H ? H H

7. Were groups treated

identically other

than for the named

interventions?

? ? H ? H H H H ? H ? ? H

8. Were outcomes measured

in the same way for

all groups?

H H H H H H H H H � H ? H

9. Were outcomes measured

in a reliable way?

H ? ? ? ? H ? ? H ? ? ? H

10. Was appropriate

statistical analysis used?

H H H H H H H H ? H ? H H

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4434

However, the reasons for selecting the confounders were notmentioned [28,64]. In addition, there was a risk of attrition biasamong these non-randomized studies, except for the study carriedout by van den Heuvel et al. [60], in which they conducted anintention-to-treat analysis which included participants who didnot complete the intervention. The overall inter-rater agreementbetween the two independent reviewers on the study appraisalwas good (Kappa statistic = 0.80).

3.3. Study characteristics

Eighteen studies involving 1723 family caregivers and 1836stroke survivors were included in this review. The important studycharacteristics are summarized in Table 3. Among these 18 studiesconducted between 1988 and 2010, six involved the participationof stroke survivors and family caregivers [52,53,56,60,62,65],whereas the remaining 12 studies involved the family caregiversonly.

Seven studies originated from North America[27,29,54,59,64,66,67], seven were from Europe [28,53,56–58,60,65], three were from Australia [52,61,62], and one was fromAsia [55]. The study settings varied from acute inpatient settings[55], home settings [27,29,52–54,56,59,65–67], community settings[28,60–62], and from inpatient to community settings [58,64].

The mean age of the caregivers ranged from 47.9 years [27] to71.3 years [52]. The majority of them were female and they werethe spouses of the stroke survivors, expect for the study of Shyuet al. (sons > daughters) [55]. The mean age of the stroke survivorsranged from 61.1 years [27] to 74.4 years [58], and 67% were male.For studies that considered the number of stroke episodes ofsurvivors, most of the stroke survivors were recovering from theirfirst stroke [28,53,55,56,59,60,62,65]. In addition, the time sinceonset of the stroke ranged from 17.57 days [55] to more than threeyears [60].

3.4. Details of the interventions

Of the 18 included studies, they were all multifaceted, and theyincluded information provision relating to stroke and its caregiv-ing. Sixteen studies were grouped as psychoeducation [27–29,52–56,58,60–62,64–67], which applies educational techniques andmethods in combination with psychotherapeutic strategies such ascounseling, to assist caregivers to understand the process of thedisease, as well as to acknowledge and cope with their emotionsand reactions [68]. The remaining two studies [57,59] were ofsocial support groups. No studies on the effects of psychodynamicinterventions for family stroke caregivers were identified. Table 4summarizes the type, format, conceptual framework and dose ofintervention of all the included studies.

3.4.1. Psychoeducation

Of 16 studies of psychoeducation, general information aboutstroke, its care, and its community resources were providedthrough the following means: an information package[29,52,54,60,64,65,67]; individual education (face-to-face interac-tion [55,56,62]; telephone contact [53]); or group education (class[27,28,58,61]; telephone conference [66]). Besides, two studies[55,58] also provided individual hands-on caregiving skill trainingto caregivers.

For the psychotherapeutic techniques, stress-coping andproblem-solving skills were commonly used in the psychoeduca-tion. Eight studies focused on enhancing stress-coping skills[52,53,55,58,60–62,66], five studies focused on improving prob-lem-solving abilities [27,29,56,65,67], whereas the remainingthree studies focused on both techniques [28,54,64]. Stress CopingTheory [69] and Cognitive Behavioral Theory [70] were the mostcommon theoretical underpinnings of psychoeducation [27–29,53,54,60,64,66]. Overall, there was no exact match betweenthe duration and dosage of the intervention.

Page 6: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 3Characteristics of the included studies.

Study

(Year/Country)

Study setting/

Type of study

Participants/

Intervention provider

Interventions Outcome measure

(a) short-term follow-up

(b) long-term follow-up

Treatment group Control group

Psychoeducation (individual format)Bakas et al. [54,73]

(2009/USA)

Home/

RCT

Caregivers (mean age

57.1; female: 72.5%;

spouse: 57.5%)/

Nurse

Telephone psychoeducation (N = 21)- Information provision: TASK notebook: 38

written tip sheets on five areas of skill building, a

stress management workbook and a brochure on

family caregiving from the American Stroke

Association

- Telephone contact: 8 weekly calls by a nurse to:

(1) facilitate caregiver’s weekly assessment of

skill needs using the CNCC (2) provided

individualized interventions that addressed

priority skill needs identified by the caregivers

Attention-controlgroup (N = 19) and

a brochure on

family caregiving

from the American

Stroke Association

(a) Outcomes measured

immediately, 1-month post-

intervention

Caregivers:Burden (Task difficulty

subscale of OCBS, Appraisal of

Caregiving Threat Subscale),

Depression (PHQ-9),

Intervention satisfaction

(self-developed rating form)

Bjorkdahl et al. [56]

(2007/Sweden)

Home/

Sub-study nested

in a larger RCT

Caregivers (spouse: 94%)

and stroke survivors

(median age: 53; male:

83.3%)/

Physiotherapist and

occupational therapist

Home-based psychoeducation and physicalrehabilitation training (N = 17)- Counseling: information provision related to

stroke, its consequences and how to deal with

problems commonly experienced by caregivers

and patients

- Physical Rehabilitation Training: a top-down

approach of individually tailored training based

on patients’ needs and desire, focus the activities

in the natural context

Usual care: (N = 15)Physical

Rehabilitation

Training

(a) Outcomes measured

immediately post-

intervention

Caregivers: Caregiver burden

(CBS)

Clark et al. [52]

(2003/Australia)

Home/

RCT

Caregivers (mean age:

70.3; female: 61.3%;

spouse: 100%) and stroke

survivors (mean age: 72.3

and/

Social worker

Family-based psychoeducation (N = 32)- Information package related general

information on stroke and its consequences,

measures for reducing the risk of further stroke,

practical coping suggestions, community services

and support structures

- Structured but not standardized family

counseling on stroke-related stresses during

home visit: Advising how to obtain appropriate

assistance and reinforcing the information from

the package

No-treatmentcontrol (N = 30)

(a) Outcomes measured

immediately post-

intervention

Caregivers: Family

functioning (FAD), QOL (SF-

36)

Forster and Young [65]

(1996/UK)

Home/

RCT

Caregiver and stroke

survivors (age: median:

73; male: 52.9%)/

Specialist nurse

Home-based psychoeducation (N = 120)- Information pack: detailed lists of locally

available services booklets published by the

Stroke Association and information on benefits

was given to patients and families in the first visit

- Six home-based counseling: included (1) goal

setting, (2) problem-solving, (3) advice on social

support, benefits, driving and holiday, (4) reviews

of patients’ needs, (5) providing information and

support to improve patients’ satisfaction and

relief for caregivers (6) review of progress.

- In addition to usual treatment

Usual care(N = 120)

(a) Outcome measured

immediately post-

intervention

Caregivers: Distress (GHQ-

28)

Grant et al. [29]

(2002/USA)

Home/

RCT

Caregivers (female: 90.5%;

spouse: 39.5%)/

Nurse

Social problem-solving telephone partnership- One face-to-face session at home and taught to

use positive problem orientation and 4

systematic steps in solving care-giving problems.

Application of the social problem-solving skills to

problem of safety and 2 additional problems

identified by the caregivers

- Eight telephone contact for developing and

maintaining the skill

Attention-controlgroup

(a) Outcome measured

immediately post-

intervention

Caregivers:Competency (Preparedness

for Caregiving Scale), Burden

(difficultly subscale of the

CBS), Depression (CES-D),

QOL (SF-36), Intervention

satisfaction (Client

satisfaction questionnaire)

Johnston et al. [53]

(2007/UK)

Home/

RCT

Caregivers (mean age:

61.7; female: 65.6%) and

stroke survivors (mean

age: 69.0; male: 61.1%)/

Nurse

Workbook-based behavioral intervention(N = 75)- Workbook contained information related to

stroke and recovery, guidance on coping skills

and self-management instructions which was

drew on cognitive behavioral techniques; diary

sheets; an audio relaxation cassette tape that

describes simple body relaxation and breathing

exercises

- Five contacts by workbook implementer to

present the workbook and instruction in its use,

answer questions, provide encouragement, offer

more information on stroke risk factors, monitor

goals and achievement, provide encouragement,

and record the number of completed quizzes and

tasks, diary days and set goals

No-treatmentcontrol (N = 85)

(a) Outcomes measured

immediately and 4-months

post-intervention

Caregivers: Competency

(RLOC), Intervention

satisfaction (10-point rating

scale)

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 35

Page 7: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 3 (Continued )

Study

(Year/Country)

Study setting/

Type of study

Participants/

Intervention provider

Interventions Outcome measure

(a) short-term follow-up

(b) long-term follow-up

Treatment group Control group

King et al. [64]

(2007/USA)

Inpatient to community/

Historical cohort design

Caregivers (mean age

62.5; female: 66.7%;

spouse: 3.3%)/

Nurse

Caregiver Problem-solving intervention(N = 15)- Combination of problem-solving (problem

orientation and problem-solving skill) and

cognitive behavioral techniques (relaxation

training, reframing negative thoughts)

- A manual containing the contents of each

session and exercises to tailor the content to each

caregiver and a stress management booklet was

given to caregivers

- In each session: participants identified their

most difficult problem with the counselor and

developed plans to solve them using the problem-

solving steps

- Home assignments e.g., apply problem-solving

to problems, rating their mood, and practicing

relaxation

Historical cohortgroup (N = 15)matched on

baseline CES-D

score, gender, age,

race and caregiving

relationship)

(a) Outcome measured

immediately and at 2-months

post-intervention

Caregivers:Anxiety (tension anxiety

subscale of Profile Mood Scale

short form), Burden

(difficulty subscale of CBS),

Intervention satisfaction

(Self-developed measure),

Competency (Caregiver

Competency and Confidence

Scale, Preparation for

Caregiving Scale)

Perrin et al. [67]

(2010/USA)

Home/

RCT

Caregivers (mean age:

58.5; female: 91.8%;

spouse:

70.5%)/

A master level

physiotherapist or a

doctoral-level nurse

practitioner

Telephone psychoeducation- Providing information about stroke, stroke

caregiving, and an overview of the recovery

experiences of individuals post-stroke and

common experience of stroke

- Review of current caregivers’ issues related to

the stroke survivors’ recovery and caregiver tasks

and activities

- Refer caregivers to the relevant guidebook

section and engage in supportive problem-

solving

Usual care (a) Outcomes measured

immediately post-

intervention

Caregivers:Burden (CSI), Depression

(CES-D), Intervention

satisfaction

Shyu et al. [31,55]

(2010, 2008/Taiwan)

Acute inpatient setting/

Cluster RCT

Caregivers (mean age:

47.8; female: 60.8%;

children: 41.1%)/

Nurses

Caregiver-oriented discharge planning andcounseling: (N = 72)- Provide information and skill training focused

on health education and caregiving task at home

in the individualized intervention and counseling

aimed at improving ability of caregiver to balance

competing needs before patient was discharged

- In addition with usual care

Usual care (N = 86) (a) Outcomes measured at

immediately, 2-months and

5-months post-intervention

Caregivers:Competency (Preparedness

for caregiving scale-

caregivers perspective),

Burden (Competing Needs

Scale), Intervention

satisfaction (caregiver

discharge needs satisfaction

scale)

(b) Outcomes measured at

11-months post-intervention

Caregivers: QOL (SF-36)

Psychoeducation (group format)Draper et al. [61]

(2007/Australia)

Outpatient or

day hospital/

RCT

Caregivers (mean age:

62.5)/

Speech therapist, social

worker and clinical

psychologist

Group psychoeducation (N = 17)- 6–11 caregivers/group

- Each session involved an element of education

and support, skill training, stress management

and group dynamics

- Cover impact of stroke, managing the resulting

life changes, communication strategies,

relaxation and stress management, managing

emotions, accessing community service and

relapse prevention strategies

- Caregivers were encouraged to join a self-help

group after the intervention

Waitlist control(N = 11)

(a) Outcomes measured

immediately, 3-months post-

intervention of experimental

and waitlist control group

Caregivers: Distress (GHQ-

28), Burden (Relative Stress

Scale)

Hartke and King [66]

(2003/USA)

Home/

RCT

Caregivers (mean age:

69.7; female: 76.1%;

spouse: 100%)/

Nurse, social worker or

psychologists

Telephone group psychoeducation (N = 43)- Topics of discussion: ‘Facts about stroke and

caregiving’, ‘Communicating with your spouse’,

‘Dealing with your spouse’s problematic feelings

and behavior’, ‘Your stress as a caregiver’, ‘Taking

care of yourself as a caregiver’, ‘Resources for you

in the community as a caregiver’

- An audiotape of a relaxation procedure and a

publication on stress management were given to

participants. A list of the stressful problems in

caregiving that participant had identified was

given during the initial assessment

- Peer support was emphasized

No-treatmentcontrol (N = 45)

(a) Outcomes measured

immediately and 4-months

post-intervention

Caregivers: Burden (Burden

Interview, Pressing Problem

Index), Depression (CES-D)

Competency (Caregiver

Competency Scale)

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4436

Page 8: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 3 (Continued )

Study

(Year/Country)

Study setting/

Type of study

Participants/

Intervention provider

Interventions Outcome measure

(a) short-term follow-up

(b) long-term follow-up

Treatment group Control group

Marsden et al. [62]

(2010/Australia)

Community/

RCT

Caregivers (mean age:

68.0; female: 88.2%;

spouse: 82.4%) and stroke

survivors (mean age:

71.6; male: 76.0%) and/

Multi-disciplinary team

CLASSiC program (N = 9) consisted of 3 parts:

1 h physical activity training on lower limb

functional task, strength training and balance

training

1/2 h ‘healthy option’ morning tea for teaching on

healthy eating, to share recipes, and engage in a

social activity

1 h psychoeducation about health-related goal

setting, secondary prevention of stroke,

management of fatigue and stress, relaxation

activity, memory and talking to a health

professional. A separate session was organized by

a social worker for caregivers to vent their

feelings and discuss issues related to being carers

Waitlist controlgroup (N = 8)

(a) Outcomes measured

immediately and at 3-months

post-intervention

Caregivers: Physical

functioning (6-min walk test,

Timed Up and Go), Burden

(CSI), QOL (Health Impact

Scale)

Wilz and Barskova [28]

(2007/Germany)

Community/

Historical cohort design

Caregivers (mean age:

62.1; female: 73.0%;

spouse/partner: 100%)/

Clinical

psychologist � public

nurse

Cognitive behavioral intervention group(N = 38)- Each group with 4–8 spouses

- Four main goals: expressing emotions and social

sharing; comprehension and acceptance of a

changed post-stroke life situation; improving

competence in active problem-solving;

encouraging healthy behavior: learning of

relaxation techniques, awareness of own burden,

acceptance of relief through professional support

Historical cohortgroup (N = 51)- matching on

selection criteria,

time points, and

measurement

instruments

Outcomes measured (a)

immediately and (b) at 6-

months post-intervention

Caregivers: Anxiety (BAI),

Depression (BDI), QOL (World

Health Organization Quality

of Life Questionnaire-BREF)

Psychoeducation (mixed format)Evan et al. [27]

(1988/USA)

Inpatient to home/

RCT

Caregivers (mean age:

49.3; female: 90.3%;)/

Occupational therapist

and social worker

Inpatient group education and individualoutpatient counseling (N = 61)- Two sessions of group education: (1)

Lecture + video tape ‘‘living with stroke’’ (basic

information about the consequence of stroke); (2)

Explained treatment unique to family’s situation

and respond to questions

- Seven sessions of individual counseling: To

apply the knowledge learnt in the class, to

develop coping and problem-solving strategies

post-discharge

Usual care (N = 63) Outcomes measured at (a) 1–

2 months and (b) 7–8 months

post-intervention

Caregivers: Family

functioning (FAD), Use of

social resources (ESCROW

profile)

Grasel et al. [58,75]

(2005, 2006/Germany)

In-patient setting

to community/

Quasi RCT

Caregivers (mean age:

59.6; female: 74.2%;

spouse: 71%)/

Not mentioned

Intensified transition support: (N = 33)- Psychoeducation seminar: information giving

related to caregiving at home, burden and

outpatient service care, self- help group

- Individual bedside caregiving skill training

- Therapeutic weekend care: home leave before

discharge to allow the family caregiver to show

their competence in caregiving and to receive

assistance in checking to what extent adaptation

of the home environment was necessary

- Telephone follow-up: providing assistance if

problem had arisen

- In addition with usual care

Usual care (N = 29) (a) Outcomes measured

immediately and 3-months

post-intervention

Caregivers: Burden (Burden

Scale for Family Caregivers),

Depression (Zerssen

Depression Scale), Somatic

complaint (Short version of

Giessen Symptom List)

(b) Outcome measured at 28-

months post-intervention

Stroke survivors: Use of social

resources (readmission or

deterioration of general

condition, use of outpatient

care service, institutionalized

to nursing home)

van den Heuvel et al.

(2000, 2002) [60,72]

Schure et al. (2006) [74]

Netherland/

Community/

non-RCT

Group program:Caregiver only (mean age:

66.4; female: 74.5%);

Individual home visits:Caregiver (mean age:

63.2; female: 71.7%) and

stroke survivors (male:

71.7%)/

Nurses

Information pack + individual (N = 60) or group(N = 110) psychoeducation- Information pack: information material

compiled by the stroke patient association

- Individual or group psychoeducation: same

content (use of education and counseling

strategies for helping family caregivers to deal

with emotional and practical problems of

caregiving, active coping and problem-solving

skills)

For group program: telephone network between

caregivers were developed

For home visit: patient was not present at the

session that dealt with problems experienced by

caregivers in caring for the patient and patient’s

behavior

No-treatmentcontrol (N = 42)

(a, b) Outcomes measured at

1-month and 7-months post-

intervention

Caregivers: Burden (CSI),

Competency (instrument for

confidence of knowledge and

perception of self-efficacy),

Social support (Adapted

version of Social Support List-

interaction, discrepancy),

Intervention satisfaction

(mechanisms of change

questionnaire)

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 37

Page 9: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 3 (Continued )

Study

(Year/Country)

Study setting/

Type of study

Participants/

Intervention provider

Interventions Outcome measure

(a) short-term follow-up

(b) long-term follow-up

Treatment group Control group

Social support groupsPierce et al. (2009) [59]

Steiner et al. [71] (2008)

USA/

Home/

RCT

Caregivers (mean age

54.5; female: 75.3%;

spouse: 68.5%)/

Nurse and assistance from

rehabilitation team

(therapists, pharmacist,

dietitian, social worker

and physician)

Caring�Web: (N = 36)-Web-based education and support in home

settings

- Four interrelated components:

1) linked web-sites about stroke and caring

2) customized educational information or tips

specific to caregivers’ needs

3) an email forum to ask a nurse specialist and a

rehabilitation team (therapists, pharmacist,

dietitian, social worker and physician) any

questions in private

4) a non-structured email discussion amongst all

participants facilitated by the nurse

No-treatmentcontrol (N = 37)

(a) Outcome measured

immediately post-

intervention

Caregivers:Perceived health status (5-

point Likert scale),

Depression (CES-D), Social

support (emotional support

and physical help from family

members or friends).

Stroke survivors:

Use of social resources

(number of hospital

readmissions, visits to

emergency department,

institutionalized to nursing

home)

Torp et al. [57]

(2008/Norway)

Home/

Pre- and post-test design

Caregivers (mean age:

74.4; female: 42.9%;

spouse/partner: 100%)/

Nurse

Web-based social support group- Computer-based information provision on

topics of stroke, care for patients, nutrition and

social services

- 3-h training on the use of internet several

months after the commencement of intervention

- Online discussion forum: direct visual and

verbal contact with each other using videophone

- Telephone contacts with nurses via videophone

or telephone in answering question related to

caring situation or the use of ICT

- Social group gathering: formal meeting on every

alternate month with an agenda directing

discussion about the progress of the project and

suggestions for future improvement; lectures by

professionals on the topics of interest to the

caregivers

Not applicable (a) Outcomes measured

immediately post-

intervention

Caregiver: Burden (Relative

Stress Scale), Distress (GHQ-

20), Social functioning

(Family and Friendship

contacts scale)

BAI: Beck Anxiety Inventory; BDI: Beck Depression Inventory; CBS: Caregiver Burden Scale; CES-D: Center for Epidemiological studies depression; CSI: Caregiver Strain Index;

FAD: McMaster Family Assessment Device; GHQ-20: 20-item General Health Questionnaire; GHQ-28: 28-item General Health Questionnaire; NHP: Nottingham Health

Profile; PHQ-9: Patient Health Questionnaire-9; OCBS: Oberst Caregiving Burden Scale; QOL: Quality of Life; RCT: Randomized control trial; RLOC: Recovery Locus of Control;

RSS: The Relative’s Stress Scale; SF-36: Short Form 36 Health survey Questionnaire.

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4438

3.4.2. Social support group

The two studies of social support group [57,59] were conductedthrough the use of videophone and Internet with differentconceptual underpinnings. In addition, with the stroke informationprovision via the Internet or computer, both studies promoted thesharing of experiences and feelings between participants throughonline discussion forums facilitated by nurses. Besides, Torp et al.[57] also included a face-to-face caregiver group gathering.

3.5. Effects of interventions on family caregivers

Data of seven RCTs was sufficiently reported and was pooled inmeta-analyses [27,52,53,58,61,62,66]. For the remaining studies,narrative summaries of the findings are presented.

3.5.1. Psychosocial wellbeing: burden

Twelve studies (individual psychoeducation: seven [29,54–56,60,64,67]; group psychoeducation: four [58,61,62,66]; socialsupport group: one [57]) investigated the effects on the caregivers’burden. The pooled analysis of three studies of group psychoe-ducation [58,62,66], where sufficient data were available, demon-strated a small but not significant effect on reducing the caregivers’burden (SMD: 0.18, 95% CI: �0.13 to 0.48, p = 0.25) (Fig. 2). Inaddition, five studies of individual psychoeducation and one studyof group psychoeducation [61] showed that the level of burden ofthe caregivers in the intervention group was similar to that in thecontrol group [29,54–56,64].

In the study of the social support group, caregivers had asignificant reduction of their burden level when compared withtheir baseline data [57]. Also, the caregivers who receivedproblem-solving skill training via telephone contacts experienceda significantly lower level of burden when compared with thecaregivers in the control group [67].

3.5.2. Psychosocial wellbeing: caregiving competency

Caregiving competency refers to the self-perceived adequacy ofown performance as a caregiver in this review. Six studies ofpsychoeducation (individual format: four [29,53,55,64]; groupformat: two [60,66]) investigated the immediate post-interventioneffect upon it. The two studies included in the pooled analysisshowed no beneficial effect on improving caregiving competency(SMD: �0.09, 95% CI: �0.49 to 0.31, p = 0.12) (Fig. 3).

Of the remaining studies, three out of the four studiesdemonstrated significant effects on improving the caregivingcompetency of family caregivers [29,60,64]. The duration of thesesinterventions lasted for two to three months, and aimed to equipcaregivers with stress-coping strategies and problem-solvingskills.

3.5.3. Psychosocial wellbeing: depression

Eleven studies (individual psychoeducation: six[29,53,54,64,65,67]; group psychoeducation: four [28,58,61,66];social support group: one [59]) investigated the effects ondepression. Sufficient information was available in three studies

Page 10: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Table 4Type, dose and conceptual framework of the psychosocial intervention.

Authors Approaches Format Conceptual

framework

Intervention time Intervention dose

(minutes/sessions/interval)

PsychoeducationBjorkdahl et al. [56] Face-to-face Individual Not mentioned 180 min/session, 3 sessions/week for 3 weeks 180 min/session/week

Clark et al. [52] Face-to-face Individual

family

Family System

Theory

60 min/session, 3 sessions in 5 months 9 min/session/week

Draper et al. [61] Face-to-face Group Not mentioned 120 min/session, 4 sessions in 4 weeks 120 min/session/week

Evan et al. [27] Face-to-face Mixed Cognitive

Behavioral Theory

2 group education sessions and 7 individual

psychoeducational sessions/from 3 weeks of

hospitalization to 3-months post discharge

Insufficient information

Forster and Young [65] Face-to-face Individual Not mentioned 7 home visits within 12 months Insufficient information

Marsden et al. [62] Face-to-face Group Not mentioned 150 min/session, 7 sessions in 7 weeks 150 min/session/week

Bakas et al. [54,73] Telephone Individual Lazarus’s

Transaction

Theory of stress

1 call/week, 8 calls in 8 weeks (average:

236.8 min in 8 calls)

29.6 min/session/week

Grant et al. [29] Telephone Individual Social Problem-

Solving Theory

1� 3 h home visit before hospital discharge

8 telephone calls post discharge: 1st month:

weekly

2nd and 3rd month: bi-weekly

Insufficient information

Hartke and King [66] Telephone Group Stress Coping

Theory

60 min/session, 8 sessions in 8 weeks 60 min/session/week

King et al. [64] Telephone Individual Social Problem-

Solving Theory and

Cognitive

Behavioral

Theory

Each session: 45–60 min

During inpatient period: 2–3 in-person

sessions

Post-discharge: 7–8 weekly telephone

sessions

Telephone sessions:

45–60 min/session/week

Grasel et al. [58,75] Face-to-face and

telephone contact

Mixed Not mentioned 1� 1-h psychoeducation seminar

3 � 45–60 min individual training course

1� therapeutic weekend care

1� telephone follow-up at 3rd month

post discharge

From hospitalization to 3rd month post

discharge

Insufficient information

Johnston et al. [53] Face-to-face and

telephone contact

Individual Behavioral Theory 3 home visits and 2 telephone contacts

over 5 weeks

Insufficient information

Perrin et al. [67] Face-to-face and

telephone

Individual Not mentioned Prior to discharge: 25 min individual

interview

Post-discharge:1st and 2nd week: 30 mins/videophone call,

one phone call/week

4th to 6th week: 30 min/videophone call, one

call/2 weeks

Telephone sessions:

20 min/session/week

Shyu et al. [31,55] Face-to-face and

telephone contact

Individual Transitions

Theory

Face-to-face education(30 min/4–5 sessions/inpatient period)

Counseling(telephone: 30–45 mins/1 session/1st week of

discharge)

(home visit: 30–60 mins/2 sessions/every

2 weeks)

Insufficient information

van den Heuvel

et al. [60,72]

Face-to-face Individual

or group

Stress Coping

Theory

Home visit: 120 min/session, 4 sessions in

8 weeks

Group support: 120 min/session, 8 sessions in

8 weeks

Home visit: 60 min/session/

week

Group support: 120 min/

session/week

Wilz and Barskova [28] Face-to-face Group Cognitive

Behavioral

Theory

90 min/session, 2 session/month, 15 sessions

in 7 months

45 min/session/week

Social Support GroupPierce et al. [59] Web-based Group Friedman’s

Framework

of Systematic

Organization

Duration: 12 months, the use of internet was

in accordance with the participants’ needs,

average 1–2 h/week

Insufficient information

Torp et al. [57] Web-based,

videophone and

face-to-face

Group Temporal Model of

Family Caring

Duration: 12 months

Social support group: monthly face-to-face

meeting

Frequency of the use of internet is not

mentioned

Insufficient information

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 39

of group psychoeducation [58,61,66] for a pooled analysis, and itdemonstrated a small but not significant effect on reducing thelevel of depression (SMD: 0.19, 95% CI: �0.11 to 0.48, p = 0.22)(Fig. 4).

Of the remaining studies, psychoeducation that aimed atequipping the caregivers with problem-solving skills and stress-coping techniques reduced the caregivers’ depression level

significantly [28,29,64]. Caregivers participating in the socialsupport group have a similar level of depression with those in thecontrol group [59].

3.5.4. Psychosocial wellbeing: anxiety

Three studies of psychoeducation (individual format: two[53,64]; group: one [28]) reported the anxiety level of caregivers.

Page 11: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

Fig. 2. Forest plot of studies of group psychoeducation reporting caregiver burden immediately post-intervention.

Fig. 3. Forest plot of studies of psychoeducation reporting caregiving competence immediately post-intervention.

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4440

Although the anxiety level of the caregivers in the interventiongroup was reduced, the difference did not reach a statisticalsignificance when compared to the caregivers in the control group[28,53].

3.5.5. Psychosocial wellbeing: social support

Three studies (group psychoeducation: one [60]; social supportgroup: two [57,71]) included social support as their outcomes. Theimmediate post-intervention effects on improving the socialsupport of caregivers were not significant. However, the face-to-face psychoeducation conducted in a group format demonstrated abeneficial effect on the seeking of and the satisfaction with socialsupport at 7-months post-intervention [72].

3.5.6. Family functioning

The pooled analysis of individual face-to-face psychoeducation[27,52] showed that there was a very small effect on improvingfamily functioning (SMD: �0.12, 95% CI: �0.23 to �0.01, p = 0.03)(Fig. 5). The intervention dose could not be determined becausethere was insufficient information about the duration of each

Fig. 4. Forest plot of studies of psychoeducation repo

Fig. 5. Forest plot of studies of psychoedu

session in Evan et al.’s study [27]. The whole intervention durationranged from seven sessions in three months [27] to three sessionsin five months [52].

3.5.7. Physical health

The psychosocial interventions did not show any effect onimproving the caregivers’ physical health (perceived health statusand somatic complaints) [29,31,54,58,60,71]. Psychoeducationwhich included three sessions of bedside caregiving skill training[31,58] also showed no significant effect on reducing thecaregivers’ somatic complaints.

3.5.8. Quality of life

The short-term effects both of individual psychoeducation[29,31,52] and group psychoeducation [28,62] on improving thephysical, psychological and social domains of QOL were notsignificant. However, a significant improvement in the caregivers’psychological and social domains of QOL was found in the grouppsychoeducation conducted via face-to-face interaction at6-months post-intervention [28], whereas no significant beneficial

rting depression immediately post-intervention.

cation reporting family functioning.

Page 12: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 41

effects were found in the individual psychoeducation at 11-months post-intervention [31].

3.5.9. Intervention satisfaction

Caregivers in the psychoeducation group were more satisfiedthan those in the control group [29,73,74]. Caregivers rated theintervention as helpful irrespective of whether it was conductedindividually or in a group. Besides, group psychoeducation wasperceived as more helpful in the emotional support andinformational support than individual psychoeducation(p < 0.01) [74].

3.6. Effects of interventions on stroke survivors: use of healthcare

resources

Individual studies of psychoeducation and the social supportgroup demonstrated a significant effect on the survivors’ use ofhealthcare resources. For the short-term effects (within 6-monthspost-intervention), the stroke survivors in the social support group[59] had significantly fewer hospital readmissions (66%) than thosein the control group (p < 0.01). Also, significantly fewer strokesurvivors in the group psychoeducation visited the physicians at 3-months post-intervention (p < 0.01) [58], and fewer strokesurvivors in the social support group visited the emergencydepartment within the 12-months of intervention (p < 0.01) [59].However, neither the psychoeducation in individual format [31]and group format [75] nor the social support group [59] couldsignificantly reduce the number of placements in a residential ornursing care home of stroke survivors.

Two studies of psychoeducation examined the long-termeffects on the use of healthcare resources [31,75]. Stroke survivorsin the individual psychoeducation group had significantly fewerplacements in a residential or nursing care home (p = 0.03) at 11-months post-intervention [31], and also more stroke survivors inthe group psychoeducation remained at home, when comparedwith the control group (p = 0.04) at 2-years post-intervention[75]. Although there were fewer hospital readmissions amongstroke survivors of the individual psychoeducation group, thedifference was not statistically significant at 1-year post-intervention [31].

4. Discussion and conclusion

4.1. Discussion

Stroke is a family disease, and the breakdown of functioningof the individual family member can extend to the whole family[76]. This systematic review aims to evaluate the effects ofdifferent types and doses of psychosocial interventions inenhancing the psychosocial and physical wellbeing, as well asin evaluating the QOL of family caregivers and the use ofhealthcare resources by stroke survivors. This is the firstsystematic review to investigate the effect on the caregivers’physical health, including somatic complaints (e.g., pain, head-aches, exhaustion, heart complaints, and stomach complaints),and the effects on stroke survivors, as reported in studiespublished in English and Chinese.

In this review, synthesized evidence shows that equippingcaregivers with problem-solving skills and stress-coping strate-gies via face-to-face psychoeducation had a small beneficialeffect on the caregivers’ family functioning, as measured by theself-reported McMaster Family Assessment Device [77]. Besides,the significant reduction of the caregivers’ depression level byproblem-solving training that was underpinned by SocialProblem Solving Theory [78] also concurred with the previousreview [79]. One of the objectives of supporting family

caregivers was to reduce the use of healthcare resources bystroke survivors, but only limited studies examined this outcome[31,58,59,74]. The effect of psychosocial interventions withongoing support via telephone [31] or Internet [59] on reducingthe number of hospital readmission was encouraging. It isbelieved that the ongoing supports helped the caregivers indecision-making so that minor health issues could be dealt withbefore becoming major healthcare crises. Moreover, psychoe-ducation that was composed of stress-coping skill training andbedside caregiving skill training during the hospital stay ofstroke survivors appeared to have a partial benefit on remainingthe stroke survivors at home [31,74].

Similar to the results of the previous reviews, which looked atthe effects on the stroke family caregivers’ psychosocialwellbeing and QOL [16–18], the social and emotional cost onfamily caregivers exacted by caregiving did not appear to beminimized either by psychoeducation or by a social supportgroup alone. In addition, this review found no significant effectof psychosocial interventions on improving the caregivers’physical health. Concurring with the systematic reviews ofinterventions for family caregivers of dementia [80] andterminal illnesses [81], it was found that it was difficult toalleviate the burden imposed on caregivers by means ofpsychosocial interventions. The multi-dimensionality of care-giver burden, including physical, emotional, social and financialaspects, presents a great challenge for healthcare professions insupporting family caregivers. The highly individualized psycho-social intervention for stroke family caregivers increases theclinical heterogeneity among the included studies [82]. In orderto achieve the power of the synthesized evidence, the meta-analyses were performed in accordance with the format ofdelivery and it specific components. This might explain why nosignificant improvement in the caregivers’ mental health wasfound in this review, whereas a significant beneficial effect wasshown in another meta-analysis that pooled the results fromeducation groups and support groups together [83].

The total duration of psychosocial interventions ranged fromthree weeks [56] to seven months [28], with the average contacttime with participants ranging from 2.42 h [67] to 104 h [59].Although the effect of the intervention dose could not be analyzedowing to insufficient information from the included studies, thisreview suggests a 3-months psychoeducation with eight sessionsmight improve the caregivers’ family functioning, psychologicalwellbeing and sense of competence, and also enhance the chanceof the stroke survivors remaining at home. For the delivery ofpsychosocial interventions, in addition to traditional face-to-faceinteraction sessions, the use of the telephone or the Internetshould be considered, so that more caregivers can receivecontinuing support which is less time-consuming and lessexpensive [84].

Some limitations in this review need to be addressed. In thisreview, only the studies published in Chinese or English wereincluded and 18 relevant studies were excluded owing either to theinsufficient information on the effects of the intervention on thesubgroup of stroke caregivers, or to the insufficient information onthe specific component and outcome measurement of theinterventions. In addition, only a small number of studies (usually2 or 3) were included in the meta-analysis owing to the diversity ofthe intervention and its outcome measurement. Thus, thesubgroup analyses of the effect on the dosage, providers, anduse of the conceptual framework of intervention were not feasiblein this review. Beside, these 18 included studies were not of a highquality and a number of methodological weaknesses were foundand have been mentioned before. The summary of this reviewshould be interpreted with caution owing to the methodologicalheterogeneity among the included studies. Moreover, the lack of a

Page 13: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4442

precise definition of family caregivers and the caregiving conditionmight impose the risk of providing the psychosocial intervention toinappropriate participants and thus reduce the chance of detectingthe benefit of the intervention [18]. To improve the generalizabilityof the research findings, these methodological pitfalls should beconsidered when designing the RCTs. Inadequate reporting ofstudies was also another concern when conducting this systematicreview. Attempts were made to contact the authors for additionalinformation, yet only a few authors responded. The reporting ofclinical trials should follow the guidelines of the CONSORTstatement [85] to enable readers to determine their validity andreliability.

4.2. Conclusion

To conclude, the effectiveness of different types of psychosocialinterventions could not be drawn from this systematic reviewbecause there were only a limited amount of studies of each type ofinterventions. Synthesized evidence demonstrated that the effectof psychoeducation on improving the caregivers’ family function-ing was very small. There is a trend that psychoeducation,consisting of training in problem solving and stress coping, offeredby telephone, reduced the depression level and improved the senseof competency of caregivers, although the summary of the effectwas not statistically significant. In addition, equipping caregiverswith caregiving skills had a more favorable effect on reducing theuse of healthcare resources of stroke survivors. Future research onpsychosocial intervention in supporting stroke family caregivers isexigent.

4.3. Practical implications

The readjustment of a caregiver’s life occurs throughout thetrajectory of the stroke. To equip family caregivers to face thechallenges of caregiving, the core skills of problem-solving andstress-coping, stroke-related information, as well as caregivingskill training, are suggested to be included in the psychoeducationprogram. The duration of the psychoeducation program rangesfrom two to three months would have a better effect in improvingthe caregivers’ competency and reducing the use of healthcareresources by stroke survivors in the long term. For the delivery ofthe psychosocial interventions, in addition to traditional face-to-face interaction sessions, use of telephone or Internet should beconsidered as an alternative, so that more caregivers could receivecontinuing support; this is less time-consuming and lessexpensive [84]. More RCTs of multifaceted psychoeducationprograms, focused on problem-solving training, during thetransitions from hospital to home care, are needed beforerecommendations of the optimal time, format and dosage ofthe psychosocial interventions can be made to the healthcareprofessionals.

Role of funding

This systematic review was funded by the Joanna BriggsInstitute. The funding source has no involvement in the collection,analysis and interpretation of data, in the writing of this report, andin the decision to submit the paper for publication.

Conflict of interest

All authors state that there are no conflicts of interestconcerning financial, personal or other relationships with otherpeople or organizations within three years of beginning thesubmitted work that could inappropriately influence, or beperceived to inappropriately influence, their work.

Appendix A. An example of typical searching strategy

Ovid MEDLINE(R)

1. exp Stroke/2. (Brain Ischemia or cerebrovascular disorder* or cerebrovascu-

lar disease* or cerebral hemorrhage or stroke* or cva or CVAs orapoplex*).mp.

3. ((cerebrovascular* or cerebro vascular* or cerebral vascular* orbrain vascular) adj3 (accident* or event* or incident* or stroke*or disorder* or disease* or infarct* or apoplex*)).mp.

4. or/1-35. exp Caregivers/6. (caregiver* or care giver* or carer*).mp.7. ((informal or family or primary or spouse) adj3 (carer* or

caregiver* or care giver*)).mp.8. (5 or 6 or 7) and 49. exp Psychotherapy/

10. exp Cognitive Therapy/11. exp Behavior Therapy/12. exp Counseling/13. (rehabilitation* or intervention* or management* or treat-

ment* or psychotherap* or counseling* or cbt).mp.14. ((psycho* or psychosocial or social or cognitive or cognition or

behavioral or cognitive behavioral or peer or problem solving)adj2 (education or modification* or psychotherap* or inter-vention* or support* or therap*or partnership)).mp.

15. ((telephone* or phone* or computer or internet or web) adj3(intervention* or support or counseling*)).mp.

16. or/9-1517. 8 and 1618. exp Health/19. (health* or hlth* or wellness or physical health* or physical

condition* or fatique* or headache* or pain* or accident* orinjur*).mp.

20. ((psychosocial or psychological or caregiving) adj3 (wellbeing*or health* or stress* or strain* or burden or burthen)).mp.

21. exp Social Support/22. ((psychosocial or social) adj3 (support or network* or

system*)).mp.23. exp Depression/24. (depression* or sad*).mp.25. ((emotional or depressive or low or morose or feeling) adj3

(symptom* or illness* or mood or depression or sad*)).mp.26. exp Anxiety/27. (anxiet* or anxiety reaction or angst or anxiousness or

nervousness).mp.28. exp Quality of Life/29. (qualit* of life or life qualit* or quality of life satisfaction).mp.30. or/18-2931. 8 and 3032. 17 or 3133. clinical trial.mp. or exp Clinical Trial/34. placebo.mp.35. (rct or rct$).mp.36. random$.mp.37. (control$ adj3 (stud$ or trial$)).mp.38. ((singl$ or double$ or trebl$ or tripl$) adj (blind$ or

mask$)).mp.39. (comparison stud$ or comparison group$).mp.40. research design.mp. or exp Research Design/41. exp Follow-up studies/42. exp Prospective studies/43. or/33-4244. limit 43 to human45. 32 and 44

Page 14: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–44 43

References

[1] Sit JWH, Wong TKS, Clinton M, Li LSW, Fong Y. Stroke care in the home: theimpact of social support on the general health of family caregivers. J Clin Nurs2004;13:816–24.

[2] Ski C, O’Connell B. Stroke: the increasing complexity of carer needs. J NeurosciNurs 2007;39:172–9.

[3] D’Zurilla TJ, Nezu AM. Problem-solving therapy: a positive approach to clinicalintervention. New York: Springer; 2007.

[4] O’Connell B, Baker L. Managing as carers of stroke survivors: strategies fromthe field. Int J Nurs Pract 2004;10:121–6.

[5] Pierce LL, Thompson TL, Govoni AL, Steiner V. Caregivers’ incongruence:emotional strain in caring for persons with stroke. Rehabil Nurs2012;37:258–66.

[6] Greenwood N, Mackenzie A, Cloud GC, Wilson N. Informal carers of strokesurvivors-factors influencing carers: a systematic review of quantitative stud-ies. Disabil Rehabil 2008;30:1329–49.

[7] Wilz G, Kalytta T. Anxiety symptoms in spouses of stroke patients. CerebrovasDis 2008;25:311–5.

[8] Qiu Y, Li S. Stroke: coping strategies and depression among Chinese caregiversof survivors during hospitalisation. J Clin Nurs 2008;17:1563–73.

[9] Chow SK, Wong FK, Poon CY. Coping and caring: support for family caregiversof stroke survivors. J Clin Nurs 2007;16:133–43.

[10] Schulz R, Beach SR. Caregiving as a risk factor for mortality. J Amer Med Assoc1999;282:2215–9.

[11] Haley WE, Roth DL, Howard G, Safford MM. Caregiving strain and estimatedrisk for stroke and coronary heart disease among spouse caregivers differentialeffects by race and sex. Stroke 2010;41:331–6.

[12] Meijer R, van Limbeek J, Kriek B, Ihnenfeldt D, Vermeulen M, de Haan R.Prognostic social factors in the subacute phase after a stroke for the dischargedestination from the hospital stroke-unit. A systematic review of the litera-ture. Disabil Rehabil 2004;26:191-7.

[13] McLennon SM, Habermann B, Davis LL. Deciding to institutionalize: why dofamily members cease caregiving at home? J Neurosci Nurs 2010;42:95–103.

[14] Department of Health (DH): Carers at the heart of 21st century families andcommunities:acaring systemonyourside,a lifeofyourown2008. [updated2008Jun 10; cited 2011 Dec 10]. Available from: http://www.dh.gov.uk/en/Publica-tionsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_085345.

[15] Smith J, Forster A, House A, Knapp P, Wright J, Young J. Information provisionfor stroke patients and their caregivers. Cochrane Database Syst Rev 2008;2.CD001919. DOI: 10.1002/14651858.CD001919.pub2.

[16] Brereton L, Carroll C, Barnston S. Interventions for adult family carers of peoplewho have had a stroke: a systematic review. Clin Rehabil 2007;21:867–84.

[17] Visser-Meily A, van Heugten C, Post M, Schepers V, Lindeman E. Interventionstudies for caregivers of stroke survivors: a critical review. Patient Educ Couns2005;257–67.

[18] Legg LA, Quinn TJ, Mahmood F, Weir CJ, Tierney J, Stott DJ, Smith LN, LanghorneP. Non-pharmacological interventions for caregivers of stroke survivors.Cochrane Database Syst Rev 2011;5. CD008179. DOI: 10.1002/14651858.CD008179.pub2.

[19] Eldred C, Sykes C. Psychosocial interventions for carers of survivors of stroke: asystematic review of interventions based on psychological principles andtheoretical frameworks. Br J Health Psychol 2008;13:563–81.

[20] Salter K, Zettler L, Foley N, Teasell R. Impact of caring for individuals withstroke on perceived physical health of informal caregivers. Disabil Rehabil2010;32:273–81.

[21] Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G,Ferguson TB, Ford E, Furie K, Gillespie C. Heart disease and stroke statis-tics—2010 update. Circulation 2010;121:e46–215.

[22] Cooke D, McNally L, Mulligan K, Harrison M, Newman SP. Psychosocialinterventions for caregivers of people with dementia: a systematic review.Aging Ment Health 2001;5:120–35.

[23] Kalra L, Evans A, Perez I, Melbourn A, Patel A, Knapp M, Donaldson N. Trainingcarers of stroke patients: randomised controlled trial. Brit Med J2004;328:1099. DOI 10.1136/bmj.328.7448.1099.

[24] Dennis M, O’Rourke S, Slattery J, Staniforth T, Warlow C. Evaluation of a strokefamily care worker: results of a randomised controlled trial. Brit Med J1997;314:1071–7.

[25] Mant J, Carter J, Wade DT, Winner S. Family support for stroke: a randomisedcontrolled trial. The Lancet 2000;356:808–13.

[26] MAStARI version 4.0. Available at: http://mastari.joannabriggs.edu.au.[27] Evans RL, Matlock AL, Bishop DS, Stranahan S, Pederson C. Family intervention

after stroke: does counseling or education help. Stroke 1988;19:1243–9.[28] Wilz G, Barskova T. Evaluation of a cognitive behavioral group intervention

program for spouses of stroke patients. Behav Res Ther 2007;45:2508–17.[29] Grant JS, Elliott TR, Weaver M, Bartolucci AA, Giger JN. Telephone intervention

with family caregivers of stroke survivors after rehabilitation. Stroke2002;33:2060–5.

[30] RevMan 5.1. 2011; Available at: http://ims.cochrane.org/revman.[31] Shyu YIL, Kuo LM, Chen MC, Chen ST. A clinical trial of an individualised

intervention programme for family caregivers of older stroke victims inTaiwan. J Clin Nurs 2010;19:1675–85.

[32] Ware JE, Gandek B. Overview of the SF-36 health survey and the internationalquality of life assessment (IQOLA) project. J Clin Epidemiol 1998;51:903–12.

[33] Higgins J, Green S. Handbook for Systematic Reviews of Interventions Version5.1. 0 [updated March 2011]. The Cochrane Collaboration 2011.

[34] Brown R, Pain K, Berwald C, Hirschi P, Delehanty R, Miller H. Distanceeducation and caregiver support groups: comparison of traditional and tele-phone groups. J Head Trauma Rehabil 1999;14:257–68.

[35] Downes B, Rooney V, Roper-Hall A, Oyebode J, Main A, Mayer P. The effec-tiveness of counselling stroke survivors and their carers in the community. AgeAgeing 1993;22:28–9.

[36] Elliott TR, Berry JW, Grant JS. Problem-solving training for family caregivers ofwomen with disabilities: a randomized clinical trial. Behav Res Ther2009;47:548–58.

[37] Hoen B, Thelander M, Worsley J. Improvement in psychological well-being ofpeople with aphasia and their families: evaluation of a community-basedprogramme. Aphasiology 1997;11:681–91.

[38] Magnusson L, Hanson E, Brito L, Berthold H, Chambers M, Daly T. Supportingfamily carers through the use of information and communication technology—the EU project ACTION. Int J Nurs Stud 2002;39:369–81.

[39] Printz-Feddersen V. Group process effect on caregiver burden. J Neurosci Nurs1990;22:164–8.

[40] Won CW, Fitts SS, Favaro S, Olsen P, Phelan EA. Community-based powerfultools intervention enhances health of caregivers. Arch Gerontol Geriatr2008;46:89-100.

[41] Ling CJ. The effect of a psychological intervention for stroke survivors and theircaregivers on the rehabilitation of post-stroke depression. Nurs Pract Res2008;5:113–4 (published in Chinese only).

[42] Zhang H, Feng ZY, Hu YS, Wang BL, Qian XL. Effects of home-based rehabilita-tion nursing intervention on self-care ability of patients after stroke. Chin J ClinRehabil 2003;7:3844–55 (published in Chinese only).

[43] Chia SL, Li IC. The effectiveness of interventions in ameliorating the quality oflife of home care patient caregivers. J Nurs 1999;46:31–43 (published inChinese only).

[44] Chiu TS, Lin CW, Kan PG, Lin CM. The effectiveness of family support groupintervention in the quality of life of primary caregivers of stroke patients. VGHNurs 2006;23:109–17 (published in Chinese only).

[45] Shao JH, Yeh HF. Evaluation of carer readiness and care outcome following acaregiver needs directed discharge program for elderly stroke patients. J Long-term Care 2006;10:238–56 (published in Chinese only).

[46] Zheng XW, Jiang XY, Zheng CH, Zhang XY, Xu Y. Influence of carrying out healtheducation for family members on quality of life of cerebral infarction patients.Chin Nurs Res 2009;23:326–8 (published in Chinese only).

[47] Feng ZY, Zhang H, Hu YS, Wang BL, Qian XL, Lu HJ. Efficacy study on familialrehabilitation nursing of cerebral apoplexy patient in community. Chin NursRes 2002;16:627–9 (published in Chinese only).

[48] Liu ZX, Lin L, Yin X. Influence of nursing intervention on quality of life of familycaregivers of cerebral apoplexy patients. Fam Nurse 2008;6:661–2 (publishedin Chinese only).

[49] Zhou XY, Zhang YL. Effect of family support training on quality of life forpatients with post-stroke depression. Chin J Rehabil Theory Pract2007;13:1092–3 (published in Chinese only).

[50] Tang YM. Effect of nursing intervention on mental health of community strokepatient’s relatives. Pract Prev Med 2010;17:1448–9 (published in Chineseonly).

[51] Feng ZY, Zhang H, Hu YS, Wang BL, Qian XL, Lu HJ. The affection of home-basedrehabilitation nursing intervention on motor function of daily living inpatients with stroke. Chin J Clin Rehabil 2003;7:730–1 (published in Chineseonly).

[52] ClarkMS,RubenachS,WinsorA.Arandomizedcontrolledtrialofaneducationandcounselling intervention for families after stroke. Clin Rehabil 2003;17:703–12.

[53] Johnston M, Bonetti D, Joice S, Pollard B, Morrison V, Francis JJ, Macwalter R.Recovery from disability after stroke as a target for a behavioural intervention:results of a randomized controlled trial. Disabil Rehabil 2007;29:1117–27.

[54] Bakas T, Farran CJ, Austin JK, Given BA, Johnson EA, Williams LS. Strokecaregiver outcomes from the telephone assessment and skill-building kit(TASK). Top Stroke Rehabil 2009;16:105–21.

[55] Shyu YIL, Chen MC, Chen ST, Wang HP, Shao JH. A family caregiver-orienteddischarge planning program for older stroke patients and their family care-givers. J Clin Nurs 2008;17:2497–508.

[56] Bjorkdahl A, Lundgren Nilsson A, Stibrant Sunnerhagen K. Can rehabilitation inthe home setting reduce the burden of care for the next-of-kin of strokevictims? J Rehabil Med 2007;39:27–32.

[57] Torp S, Hanson E, Hauge S, Ulstein I, Magnusson L. A pilot study of howinformation and communication technology may contribute to health pro-motion among elderly spousal carers in Norway. Health Socl Care Community2008;16:75–85.

[58] Grasel E, Biehler J, Schmidt R, Schupp W. Intensification of the transitionbetween inpatient neurological rehabilitation and home care of strokepatients. Controlled clinical trial with follow-up assessment six months afterdischarge. Clin Rehabil 2005;19:725–36.

[59] Pierce LL, Steiner VL, Khuder SA, Govoni AL, Horn LJ. The effect of a Web-basedstroke intervention on carers’ well-being and survivors’ use of healthcareservices. Disabil Rehabil 2009;31:1676–84.

[60] van den Heuvel ETP, de Witte LP, Nooyen-Haazen I, Sanderman R, Meyboom-de Jong B. Short-term effects of a group support program and an individualsupport program for caregivers of stroke patients. Patient Educ Couns2000;40:109–20.

[61] Draper B, Bowring G, Thompson C, Van Heyst J, Conroy P, Thompson J. Stress incaregivers of aphasic stroke patients: a randomized controlled trial. ClinRehabil 2007;21:122–30.

Page 15: Cheng 2014 Stroke Family Caregivers and Stroke Survivors

H.Y. Cheng et al. / Patient Education and Counseling 95 (2014) 30–4444

[62] Marsden D, Quinn R, Pond N, Golledge R, Neilson C, White J, McElduff P, PollackM. A multidisciplinary group programme in rural settings for community-dwelling chronic stroke survivors and their carers: a pilot randomized con-trolled trial. Clin Rehabil 2010;24:328–41.

[63] Barkauskas VH, Lusk SL, Eakin BL. Selecting control interventions for clinicaloutcome studies. West J Nurs Res 2005;27:346–63.

[64] King RB, Hartke RJ, Denby F. Problem-solving early intervention: a pilot studyof stroke caregivers. Rehabil Nurs 2007;32:68–76.

[65] Forster A, Young J. Specialist nurse support for patients with stroke in thecommunity: a randomised controlled trial. Brit Med J 1996;312:1642–6.

[66] Hartke RJ, King RB. Telephone group intervention for older stroke caregivers.Top Stroke Rehabil 2003;9:65–81.

[67] Perrin PB, Johnston A, Vogel B, Heesacker M, Vega-Trujillo M, Anderson J,Rittman M. A culturally sensitive Transition Assistance Program for strokecaregivers: examining caregiver mental health and stroke rehabilitation. JRehabil Res Dev 2010;47:605–16.

[68] Gleason-Wynn P. Psychoeducation with caregivers of older adults. In: Cor-coran J, editor. Clinical applications of evidence-based family interventions.United Kingdom: Oxford University Press; 2003. p. 297–320.

[69] Lazarus RS, Folkman S. Stress, appraisal, and coping. New York: SpringerPublishing Company; 1984.

[70] Meichenbaum D. Cognitive-behavior modification: an integrative approach.New York: Plenum Press; 1977.

[71] Steiner V, Pierce L, Drahuschak S, Nofziger E, Buchman D, Szirony T. Emotionalsupport, physical help, and health of caregivers of stroke survivors. J NeurosciNurs 2008;40:48–54.

[72] van den Heuvel ETP, Witte LP, Stewart RE, Schure LM, Sanderman R, Mey-boom-de Jong B. Long-term effects of a group support program and anindividual support program for informal caregivers of stroke patients: whichcaregivers benefit the most. Patient Educ Couns 2002;47:291–9.

[73] Bakas T, Farran CJ, Austin JK, Given BA, Johnson EA, Williams LS. Contentvalidity and satisfaction with a stroke caregiver intervention program. J NursScholarsh 2009;41:368–75.

[74] Schure LM, van den Heuvel ET, Stewart RE, Sanderman R, de Witte LP,Meyboom-de Jong B. Beyond stroke: description and evaluation of an effectiveintervention to support family caregivers of stroke patients. Patient EducCouns 2006;62:46–55.

[75] Grasel E, Schmidt R, Biehler J, Schupp W. Long-term effects of the intensifica-tion of the transition between inpatient neurological rehabilitation and homecare of stroke patients. Clin Rehabil 2006;20:577–83.

[76] Walsh F. Conceptualization of normal family functioning. In: Walsh F, editor.Normal family process. New York: Guilford Press; 1982.

[77] Epstein NB, Baldwin LM, Bishop DS. The McMaster Family Assessment Device.J Marital Fam Ther 1983;9:171–80.

[78] D’Zurilla TJ. Problem-solving therapy: a social competence approach to clinicalintervention. New York: Springer Publishing Company; 1986.

[79] Lui MHL, Ross FM, Thompson DR. Supporting family caregivers in stroke care:a review of the evidence for problem solving. Stroke 2005;36:2514–22.

[80] Selwood A, Johnston K, Katona C, Lyketsos C, Livingston G. Systematic reviewof the effect of psychological interventions on family caregivers of people withdementia. J Affect Disord 2007;101:75–89.

[81] Candy B, Jones L, Drake R, Leurent B, King M. Interventions for supportinginformal caregivers of patients in the terminal phase of a disease. CochraneDatabase Syst Rev 2011;6. CD007617. DOI: 10.1002/14651858.CD007617.pub2.

[82] Santacroce SJ, Maccarelli LM, Grey M. Intervention fidelity. Nurs Res2004;53:63–6.

[83] Lee JH, Soeken K, Picot SJ. A meta-analysis of interventions for informal strokecaregivers. West J Nurs Res 2007;29:344–56.

[84] Buckley KM, Tran BQ, Prandoni CM. Receptiveness, use and acceptance oftelehealth by caregivers of stroke patients in the home. Online J Issues Nurs2004;9:9.

[85] Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC, Devereaux P,Elbourne D, Egger M, Altman DG. CONSORT 2010 Explanation and Elaboration:updated guidelines for reporting parallel group randomised trials. J ClinEpidemiol 2010;63:e1–37.