PROTOCOL: Community-Based Rehabilitation for People with Disabilities in Low- and Middle-Income Countries

Abstract

1 Background

1.1 DESCRIPTION OF THE PROBLEM

Disability is an umbrella term for impairments, activity limitations, and participation restrictions, denoting the negative aspects of the interaction between an individual (with a health condition) and that individual's contextual factors (environmental and personal factors) (WHO 2011; WHO 2001). People with disabilities (PWD) therefore include those who have long-term physical, mental, intellectual or sensory impairments resulting from any physical or mental health conditions which, in interaction with various barriers, may hinder their full and effective participation in society on an equal basis with others (UN 2008). This view of disability is therefore an expansion beyond the traditional view, which focused on impairments only.

The World Disability Report estimates that there are over one billion people with disabilities in the world, of who between 110–190 million experience very significant difficulties (WHO 2011). This corresponds to about 15% of the world's population, and is higher than previous World Health Organization's (WHO) estimates. These figures therefore suggest an increase in the prevalence of disability, potentially due to population ageing and the rise in chronic conditions. However, the data underlying these estimates is sparse making it difficult to gauge trends over time or their causes.

It is widely reported that PWD are excluded from education, health, and employment and other aspects of society and that this can potentially lead to or exacerbate poverty (WHO 2011). This exclusion is contrary to the essence of the United Nations (UN) Convention on the Rights of Persons with Disabilities, which is an international human rights instrument of the UN intended to protect the rights and dignities of PWD (UN 2008). This Convention calls upon all countries to respect and ensure the equal rights and participation of all PWD to education, health care, employment and inclusion in all aspects of society. The text was adopted by the UN General Assembly in 2006, and came into force in 2008. By April 2012, it had 153 signatories and 112 parties. Effective interventions therefore need to be identified that will enhance participation in society by PWD and thereby enforce the convention.

1.2 DESCRIPTION OF THE INTERVENTION

The UN Convention states that comprehensive rehabilitation services including health, employment, education and social services are needed ‘to enable PWD to attain and maintain maximum independence, full physical, mental, social and vocational ability, and full inclusion and participation in all aspects of life’ (UN 2008). A range of interventions can be made available to PWD, extending from purely medical (e.g. hospital treatments) to exclusively social (e.g. inclusion in family events). Comprehensive rehabilitation services may be preferred to isolated interventions, given the recommendation of the UN convention and the wide range of needs of PWD to enable participation.

Community-based rehabilitation (CBR) is the strategy endorsed by WHO (WHO 2010a) for general community development for the rehabilitation, poverty reduction, equalization of opportunities, and social inclusion of all PWD. The concept was firstly introduced in an unpublished WHO report in 1976 (WHO 1976; Finkenflugel 2004) as a promising strategy to provide rehabilitation for people with disability in developing countries and part of the broader goal of reach ‘Health for All by the year 2000‘ (WHO 1978). Since the first training manual published in 1980 (Helander 1980) and updated in 1989 (Helander 1989), the concept has evolved to become a multi-sectoral strategy. CBR is implemented through the combined efforts of PWD themselves, their families and communities, and the relevant governmental and non-governmental health, educational, vocational, social and other services. CBR is delivered within the community using predominantly local resources.

The CBR matrix (WHO 2010a) provides a basic framework for CBR programmes. It highlights the need to target rehabilitation at different aspects of life including the five key components: health, education, livelihood, social, and empowerment. Each component consists in five elements where the different activities are classified. A CBR programme is formed by one or more activities in one or more of the five components. Thus, a CBR programme is not expected to implement every component of the CBR matrix, and not all PWD require assistance in each component of the matrix. However, a CBR programme should be developed in partnership with PWD to best meet local needs, priorities and resources.

The CBR guidelines were launched in October 2010 to provide further direction on how CBR programmes should be developed and implemented (WHO 2010a). Although CBR is currently implemented in over 90 countries, in reality only 2% of PWD are estimated to have access even to basic health and rehabilitation services (Meikle 2002). The scaling up of CBR is therefore urgently needed, but there is also a need for a stronger evidence base on the efficacy and effectiveness of CBR programs (Finkenflugel 2005; Hartley 2009; WHO 2011) to support the expansion in coverage of CBR.

1.3 HOW THE INTERVENTION MIGHT WORK

A health condition may lead to an impairment, which could restrict full participation in aspects of society, thus resulting in disability. Providing CBR may reduce some of the consequences of the impairment, by facilitating participation by PWD in the domains of health, education, livelihood, social activities, and empowerment. CBR could therefore range from providing assistive devices in the community to increase mobility, to coordinating with local schools to ensure inclusion of children with disability, offering vocational rehabilitation to increase wage employment, family counselling to improve relationships, and the establishment of self-help groups to improve political participation. The outcomes of CBR will therefore vary depending on the targets of specific programmes, but could include improving social participation, clinical outcome and quality of life among PWD.

1.4 WHY IT IS IMPORTANT TO DO THIS REVIEW

There are estimated to be at least 1 billion PWD in the world. Many of these PWD will require CBR to meet their basic needs, ensure inclusion and participation, and enhance the quality of life of PWD and their family, their caregivers or their community (WHO 2011). Unfortunately the coverage of CBR is currently very low (Meikle 2002), and the evidence has not been comprehensively assessed to identify whether CBR is effective and under which circumstances. Establishing an evidence base for the effectiveness of CBR is inherently difficult (Hartley 2009). Each individual programme is tailored to the specific needs and setting and therefore may include a different focus, different components and different client types. Furthermore, the impact of CBR can be measured in a variety of domains. The only available literature review on CBR in developing countries (Finkenflugel 2005) found that the impact evidence base is ‘fragmented and incoherent’ on almost all aspects of CBR and noted methodological concerns with many studies. However, the authors did not assess the overall impact of CBR in their review. Other literature reviews have reported more positively on the literature, but were more limited in scope, focusing on specific geographical location (Velema 2008) or types of disability (Robertson 2012; Wiley-Exley 2007; Evans 2008). Available systematic reviews are also limited in scope, covering either single community-based rehabilitation interventions or single aspects of disability. For instance a co-registered Cochrane/Campbell systematic review focuses on personal assistance in adults (Mayo-Wilson 2008a; Mayo-Wilson 2008b) and children (Mayo-Wilson 2008c; Mayo-Wilson 2008d; Mayo-Wilson 2008e) in both developed and developing countries.

There is therefore a need to assess the full evidence base, updating previous reviews comprehensively and providing an overview assessment, to address the question ‘What are the impacts of CBR for people with disabilities in low- and middle-income countries?‘ This will be the first systematic review to our knowledge to address this question comprehensively.

2 Objective of the review

Primary objective: To assess the effectiveness and cost-effectiveness of community-based rehabilitation for people with disabilities in low- and middle-income countries.

3 Methods

3.1 CRITERIA FOR INCLUDING STUDIES IN THE REVIEW [PICOS]

3.1.1 Participants

People with disability, and/or their family, their caregivers, their community living in low- and middle-income countries.

Disability is defined as impairments, activity limitations, and participation restrictions denoting the negative aspects of the interaction between an individual (with a health condition) and that individual's contextual factors (environmental and personal factors) (WHO 2011; WHO 2001).

We will include participants from low- and middle –income countries only, as this was the original commitment of CBR (Helander 1989).

3.1.2 Interventions

After the definition provided within the CBR Guidelines (WHO 2010a) and its recent operationalisation (Lukersmith in press), we defined community-based rehabilitation as:

•

program for people with disabilities;

•

delivered at the community level;

•

implemented through the combined efforts of at least two of the following: PWD, their family, their caregivers or their community, and the relevant governmental and nongovernmental health, education, vocational, social, and other services; and

•

focusing on rehabilitation, equalization of opportunities, poverty reduction, and social inclusion of PWD within general community development.

Due to the lack of a recognised list of long-term physical or mental health conditions associated with disability, the advisory group and experts were consulted and such a list was created (Appendix 1).

A CBR programme is formed by one or more activities in one or more of the five components (health, education, livelihood, social, and empowerment). List of activities for each element of the five components are presented within the CBR Guidelines under the section ‘Suggested activities’ (WHO 2010a). The following activities are here given as examples:

•

Health: training PWD in the use of assistive devices; providing information to PWD and their family or their caregivers about time and location of activities for screening health conditions and impairments associated with disabilities.

•

Education: providing education and training for families or caregivers of PWD; installing ramps in schools to make them accessible to PWD using wheelchairs.

•

Livelihood: linking the jobseeker with disability to existing support services; advocating before relevant public and private agencies to ensure accessible housing for PWD.

•

Social: converting institutions for PWD in rehabilitation centres; providing information to PWD about the sports opportunities available within the community.

•

Empowerment: helping PWD running meetings of new self-help group; involving disabled's people organizations in CBR planning, implementation, and monitoring.

CBR interventions will be compared with:

•

facility-based interventions;

•

other types of CBR interventions;

•

other interventions;

•

any mix of the above;

•

no intervention.

Trials will be excluded if:

•

the CBR intervention takes place only in health facilities or schools.

3.1.3 Outcomes

3.1.3.1 Primary Outcomes

•

Functional outcomes, including education, employment status, social participation, empowerment.

•

Disability outcomes, such as extent of disability, measured using validated instruments (e.g. Disability Rating Scale (DRS); Expanded Disability Status Scale (EDSS); Global Mental Health Assessment Tool (GMHAT); Clinical Global Impressions Scale (CGIS)).

3.1.3.2 Secondary Outcomes

•

Quality of life, measured using validated instruments (e.g. WHO Quality of Life-BREF (WHOQOL-BREF); Health-Related Quality of Life (HRQoL); Global Assessment of Functioning (GAF); Medical Outcome Study Short Form 36 (SF36)).

•

Economic impact, including cost-effectiveness, cost-utility, cost-benefit.

•

Adverse effects.

3.1.4 Study Types

Randomised controlled trials, non-randomised controlled trials, controlled before-after studies, controlled interrupted time series studies, economic studies (cost-effectiveness analysis, cost-utility analysis, cost-benefit analysis, economic modelling) of CBR for people with disabilities in low- and middle-income countries in which the outcome is measured before and after the intervention or an intervention is studied against another intervention with baseline.

3.2 SEARCH METHODS FOR IDENTIFICATION OF STUDIES

We will not restrict the search for studies by language or publication status. We aim to include studies regardless of whether they are published or unpublished. Searches will be limited to studies published after 1976 as this is the year when the concept of community-based rehabilitation was first introduced (WHO 1976; Finkenflugel 2004). Low- and middle-income countries were identified using the World Bank Atlas method (World Bank 2012) (Appendix 2).

3.2.1 Electronic searches

We will search the following electronic databases:

Biomedical databases

•

AIM (African Index Medicus) (Global Health Library)

•

CENTRAL (Cochrane Register of Controlled Trials) (The Cochrane Library)

•

CINHAL Plus (Cumulative Index to Nursing and Allied Health Literature) (EBSCO)

•

Cochrane Database of Systematic Reviews (The Cochrane Library)

•

EMBASE (OvidSP)

•

Global Health (OvidSP)

•

IMEMR (Index Medicus for the Eastern Mediterranean Region) (Global Health Library)

•

IMSEAR (Index Medicus for South East Asia Region) (Global Health Library)

•

LILACS (Latin American and Caribbean Health Sciences Literature) (Global Health Library)

•

MEDLINE (OvidSP)

•

PsycINFO (OvidSP)

•

WHOLIS (World Health Organisation Library Information System) (Global Health Library)

•

WPRIM (Western Pacific Region Index Medicus) (Global Health Library)

Social sciences databases

•

CAB Abstract (OvidSP)

•

DARE (Database of Abstracts of Reviews of Effectiveness) (The Cochrane Library)

•

EconLit (OvidSP)

•

ERIC (ProQuest)

•

HTA Database (The Cochrane Library)

•

IBSS (International Bibliography of the Social Sciences) (ProQuest)

•

NHSEED (NHS Economic Evaluation Database) (The Cochrane Library)

•

PAIS International (Public Affairs Information Services) (ProQuest)

•

The Campbell Collaboration Library of Systematic Reviews (The Campbell Library)

•

Web of Science (Web of Knowledge)

We will base the electronic databases searches on the MEDLINE strategy in Appendix 3 which will be adapted as necessary, for use in each of the other databases.

3.2.2 Other Searches

We will search relevant websites from governmental and non-governmental organisations, academics, and users using Advanced Google Search (Appendix 4). Relevant embedded databases and libraries within the websites will be searched manually.

We will contact key authors and institutions to request details on recently published, in press, unpublished or ongoing studies on the topic.

We will search reference lists of included studies and literature reviews.

We will track citations of included studies using Google Scholar.

3.3 DATA COLLECTION AND ANALYSIS

3.3.1 Selection of studies

Title and abstract of studies stemmed from the electronic searches will be independently screened by couples of reviewers against the inclusion criteria. Whether unclear from the title and abstract, whether the study will meet the inclusion criteria, we will include this in the next screening. Disagreements will be solved through consultation with a third reviewer.

Full-text of studies meeting the inclusion criteria will be retrieved and then screened by couples of reviewers against the inclusion criteria. Disagreements will be solved through consultation with a third reviewer. Missing information necessary for screening will be obtained contacting the authors of the study. If the information cannot be obtained, the study will be listed under ‘Studies awaiting classification’.

In order to avoid language bias, studies with full-text in a language different from the ones available (English, French, Spanish, German, Italian) will not be excluded but they will be also listed under ‘Studies awaiting classification’. Excluded studies will be listed under ‘Excluded studies’ and the reason for their exclusion (methods, participants, interventions, publication date, language) will be recorded within the table ‘Characteristics of excluded studies’. Reviewers will not be blind to any information of studies screened, as for the name of the authors and their affiliations. In order to avoid the outcomes reporting bias, studies will not be excluded on the basis of outcomes only. If the study meets all inclusion criteria but the outcome searched is not reported, the authors of the study will be contacted to obtain missing information.

Full-text of studies in languages other than English (French, Spanish, German, and Italian) will be screened by one reviewer only.

Relevant literature reviews will not be included but they will be identified and recorded in a separate library. Their full-text will be retrieved and reference lists searched.

3.3.2 Data extraction and management

Data extraction will be jointly performed by two reviewers: a first reviewer will extract data into a data extraction form and a second reviewer will verify the correctness of data extracted by the first reviewer. Disagreements will be solved through consultation with a third reviewer. Missing information will be obtained by contacting the authors of the study. Review Manager 5 will be used to organise extracted data that will be reported as tables under ‘Characteristics of included studies’ including the ‘Risk of bias table’ and ‘Data and analyses’.

The data extraction form will be developed a priori and it will include the following information:

•

Methods: including study design and duration of the study.

•

Participants: including type of disability, age, sex, country.

•

Interventions: details on both intervention and comparison; including type(s) of CBR, intervention (or comparison) details (i.e. intensity, frequency), agent(s), setting(s).

•

Outcomes: including type of outcome(s), measurement instrument(s) (i.e. scale, questionnaire), and time-points measured.

•

Funding: including types of funder of the study.

•

Publication: including publication type (i.e. article, report), publication language.

•

Notes: including comments on the study not covered by the previous categories.

Data extraction from studies in languages other than English (French, Spanish, German, and Italian) will be done by one reviewer only.

3.3.3 Assessment of risk of bias in included studies

As for the data extraction, the assessment of the methodological quality of selected studies will be jointly performed by two reviewers: a first reviewer will assess risk of bias using the data extraction form and a second reviewer will verify the correctness of data extracted by the first reviewer. Disagreements will be solved through consultation with a third reviewer. Assessment the methodological quality of studies in other languages than English (French, Spanish, German, and Italian) will be done by one reviewer only.

We will use the ‘Risk of Bias’ tool from section 8.5 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) to assess the risk that a study over or under-estimates the true intervention effect. We will assess specific potential sources of bias, including: sequence generation, allocation sequence concealment, blinding of outcome assessment; incomplete outcome data; selective outcome reporting; intention-to-treat analyses and ‘other’ identified concerns about sources of bias such as baseline imbalance and protection against contamination (Lundh 2008). Review authors' judgments regarding risk of bias will be graded for each criterion as low, high, or unclear risk of bias. Risk of bias graph will be developed using Review Manager 5 to summarise methodological quality of the studies included. We will assess missing data and attrition rates for each of the included studies, and report the number of participants who were included in the final analysis as a proportion of all participants in the study. Reasons given for missing data will be provided in the narrative summary and we will ascertain the extent to which the results are altered by missing data in order to offer possible explanation for differences between studies when interpreting the results of the review (Schulz 1995).

We will assess the risk of bias in economic studies using the Drummond checklist (Drummond 1996) and the Evers checklist (Evers 2005) for economic evaluations, and the Philips checklist (Philips 2004) for economic modelling.

3.3.4 Measures of treatment effect

We will analyse dichotomous (binary) outcomes by calculating incidence rate ratios (IRR), cumulative incidence ratios (CIR) or odds ratios (OR) for each study with the uncertainty in each result being expressed using 95% confidence intervals (CI). Where multiple measures of effect are available for the study we will choose the CIR or IRR over the OR, because it is more accessible to understanding and interpretation by non-research/statistically trained stakeholders. When overall results are significant, we will determine the number needed to treat (NNT) or number needed to harm (NNH).

For continuous data, including measurements on scales, we will report the mean score and standard deviation for each outcome as determined by a standardised tool. These outcome measures will be compared between the two groups to give a mean difference (MD), with a 95% confidence interval. Continuous data that are skewed will be reported separately. Skew will be identified when, for a scale or measure with positive values and a minimum value of zero, the mean is less than twice the standard deviation (Altman 1996).

Where scales measuring the same outcome have different directions of benefit, a minus sign will be added to that measuring a negative direction to ensure that all measurements can be read in the same direction.

3.3.5 Unit of analysis issues

Where a study presents results for several periods of follow-up for the same outcome we will only include endpoint data, to avoid double counting of the participants in studies. We will focus on endpoint data because it is more clinically relevant and if change data were to be presented along with endpoint data it would be given undeserved, equal prominence.

Where multiple treatment/control group types are presented in study reports, we will aim to present the data from each study as consistently as possible with the primary comparison of treatment compared with control group. We will conduct a separate sub-group analysis of studies comparing different types of interventions for different types of disabilities.

3.3.6 Dealing with missing data and incomplete data

We will contact the original investigators to request any missing data as well as information on whether or not it can be assumed to be missing at random. In addition, as mentioned above (see Assessment of risk of bias in included studies), proportions of missing participants will be reported in the risk of bias assessment, reasons given for missing data will be provided in the narrative summary and the extent to which the results are altered by missing data will be ascertain.

We will report separately all data from studies where more than 50% of participants in any group were lost to follow-up, and explore the impact of this on the review findings by means of sensitivity analysis.

3.4 DATA SYNTHESIS

3.4.1 Data Synthesis

Data analysis will be performed using Review Manager 5. If visual examination of results and test statistics (e.g. Chi2 test and I2 statistic) suggest homogeneity, we will quantitatively combine results for each primary outcome for meta-analysis using a random effects model. The weight given to each study will be the inverse of the variance so that the more precise estimates (from larger studies with more events) are given more weight.

If results are too heterogeneous for meta-analysis or if insufficient data are available to meta-analyse, then the authors will write a narrative synthesis for the results and forest plots will be used to show each study's point estimates and error measurements for each primary outcome. Unless the reason for leaving the study early is clearly reported, we will assume that participants who dropped out had no change in level of baseline physical and psychosocial function. When information provided is insufficient to define the original group size prior to leaving the study, we will contact the authors of the study. We will test the sensitivity of results using the number of patients who completed each study and comparing trials using intention-to-treat analysis with those that did not.

3.4.1.1 Assessment of heterogeneity

We will assess heterogeneity in the results of the studies by visual inspection of the graphical presentations, by performing the Chi2 test of heterogeneity (where a significance level less than 0.10 will be interpreted as evidence of heterogeneity), and by examining the I2 statistic (Deeks 2008). We will consider I2 values less than 30% as indicating low levels of heterogeneity, values in the range of 31% to 69% as indicating moderate heterogeneity, and values greater than 70% as indicating high levels of heterogeneity. We will evaluate four possible reasons for heterogeneity for each study through comparing separate subgroups of studies: (i) different quality of the study; (ii) different types of community-based rehabilitation used; (iii) different types of disability; (iv) different baseline levels of symptoms and functioning of participants.

3.4.1.2 Investigation of Heterogeneity

If sufficient studies (more than five) are found, we will undertake subgroup analysis to examine the effect on primary outcomes of: (i) type of CBR; (ii) disability type (physical/mental); (iii) severity of disability; (iv) age (children/adults); (v) geographical location (low-/middle-income countries).

3.4.1.3 Sensitivity Analysis

If there are sufficient data, we will undertake sensitivity analyses to investigate the robustness of the overall findings in relation to aspects of methodological quality.

3.4.1.4 Assessment of Reporting Biases

If sufficient studies are identified (more than five) we will enter data from all selected studies into a funnel graph (study effect versus study size) in an attempt to investigate the likelihood of overt publication and related biases.

4 Acknowledgements

The authors of the protocol will like to thank all members of the advisory group and contacted experts for their valuable support in designing the protocol.

Footnotes

8 Apendices

9 Contribution of Authors

Valentina Iemmi: coordinating the review; designing the review; writing the protocol.

Suresh Kumar: designing the review; writing the protocol.

Karl Blanchet: conceiving the review; designing the review; securing funding for the review.

Sally Hartley: providing advice on the review.

Gudlavaletti Murthy: providing advice on the review.

Vikram Patel: providing advice on the review.

Joerg Weber: providing advice on the review.

Richard Wormald: providing advice on the review.

Hannah Kuper: conceiving the review; designing the review; writing the protocol; securing funding for the review.

10 Declarations of Interest

Professor Patel has a Wellcome Trust grant for a randomised controlled trial for a CBR intervention for schizophrenia in India.

Several members of the group have previously undertaken systematic reviews on related subjects but not on this particular topic.

There are no further conflicts of interest.

11 Published Notes

This review is part of a joint systematic review between Cochrane (Injuries Group) and Campbell Collaboration (International Development Coordinating Group), funded by the International Initiative for Impact Evaluation (3ie). A version of this review will be published in the Cochrane Collaboration Library of Systematic Reviews (http://www.thecochranelibrary.com/view/0/index.html). Another version will be published in the 3ie database of systematic reviews ().