Interventions to Improve the Labour Market Situation of Adults with Physical and/or Sensory Disabilities in Low‐ and Middle‐Income Countries: A Systematic Review

The objectives of this Campbell systematic review were: 1) To describe the range and diversity of interventions available for addressing the low labour market participation of adults with physical and/or sensory disabilities in developing country contexts. 2) To systematically identify, assess, and synthesise the evidence on the effects of interventions on labour market outcomes for disabled adults in low‐ and middleincome countries. As part of this, to critically analyse the evidence along the causal chain framework, linking interventions with intermediate outcomes and final impacts, and document the level/strength of evidence on potential pathways of impact using the framework. 3) To assess if effects are moderated by characteristics of the participants, interventions, and/or settings. 4) To provide an explanation for the intervention effects by examining what participants in the included studies reported about why the interventions did, or did not, work for them. A total of 14 studies met the eligibility criteria. Publication dates of included studies ranged between 1992 and 2012, with six studies published in the four‐year period 2010‐2013. A key finding of this review is the overall scarcity of robust evidence, as indicated by the relatively few studies that met the inclusion criteria. Although the evidence in general showed positive results, we need to be wary of drawing strong inferences from the findings of this body of literature. Not only is the number of impact evaluations limited, but most used designs in which conclusively attributing causality is not possible. Our assessment of the evidence does not allow us to develop practical suggestions on what interventions are likely to work, for whom, and when. Clearly, there is an urgent need for investment in high quality impact evaluations of interventions to support people with disabilities in accessing the labour market in low‐ and middle income settings. To build the evidence base further, it is therefore important that many more of the interventions currently in existence in low‐ and middle‐income countries are rigorously evaluated, and the results are reported and disseminated widely. The methodological inconsistencies and weaknesses of the current evidence base, and specific knowledge gaps, suggest a number of future research priorities. Executive Summary BACKGROUND Disability is a development issue, with widespread poverty, inequality and violation of human rights. Recent estimates suggest that more than one billion people are living with some form of disability. Persons with disabilities are over‐represented among the world's poor, and significant labour market disadvantage helps maintain the link between poverty and disability in many country contexts. The costs of disability are particularly acute in low‐ and middle‐income countries (those with gross national income per capita of less than $12,616), where up to 80% of people with disabilities of working age can be unemployed, around twice that for their counterparts in high‐income countries. When people with disabilities do work, they generally do so for longer hours and lower incomes, have fewer chances of promotion, are more likely to work in the informal labour market, and are at greater risk of becoming unemployed for longer periods. The barriers faced by people with disabilities globally in accessing and sustaining paid work is a profound social challenge. There is now growing recognition of employment as a key factor in the process of empowerment and inclusion into society, and the role of interventions to improve labour market outcomes for disabled people is receiving increased international attention. It is therefore both vital and timely to increase understanding of the impacts of available programmes, in order to ensure that they are effective in delivering positive outcomes for people with disabilities and provide value for money. Although several reviews have attempted to summarise the existing research in this area, there are a number of substantive and methodological limitations to these reviews. Thus, there is a need to systematically examine the evidence base to provide an overview of the types of interventions being used to improve employment outcomes, to identify those that are effective and ineffective, and to identify areas in which more research needs to be conducted. OBJECTIVES To describe the range and diversity of interventions available for addressing the low labour market participation of adults with physical and/or sensory disabilities in developing country contexts. To systematically identify, assess, and synthesise the evidence on the effects of interventions on labour market outcomes for disabled adults in low‐ and middle‐income countries. As part of this, to critically analyse the evidence along the causal chain framework, linking interventions with intermediate outcomes and final impacts, and document the level/strength of evidence on potential pathways of impact using the framework. To assess if effects are moderated by characteristics of the participants, interventions, and/or settings. To provide an explanation for the intervention effects by examining what participants in the included studies reported about why the interventions did, or did not, work for them. SELECTION CRITERIA To be included in the review, studies were required to meet several eligibility criteria. First, studies must have evaluated an intervention with the means to improve the labour market situation of adults with disabilities. Such interventions could take the form of a device, policy, programme, strategy, or other type of action. Second, studies must have investigated outcomes for adults aged 16‐65 years with physical and/or sensory impairments associated with disability. Third, the study setting must have been a low‐or middle‐income country (LMIC). Fourth, studies must have utilised one of the following: (a) randomised experimental design, (b) rigorous quasi‐experimental design that used robust methods for removing biases due to non‐random assignment of treatment, or (c) quasi‐experimental design that used less rigorous methods for constructing the counterfactual, including uncontrolled studies. Fifth, studies must have reported at least one quantitative employment‐related outcome variable. Sixth, the date of publication or reporting of the study must have been within the period 1 January 1990 to 31 December 2013. Finally, no language or form of publication restrictions was applied. SEARCH STRATEGY A systematic and comprehensive search was used to locate both published and unpublished studies. Ten major bibliographic databases were electronically searched, along with 32 specialist databases and library catalogues, and 59 websites of relevant organisations. Six journals were manually searched and search engines used. The reference lists of previous reviews and included studies were examined, and forward citation checking exercises were conducted. Finally, information was requested from authors of included studies and other relevant stakeholders. DATA COLLECTION AND ANALYSIS Each study was subject to a rigorous process of data extraction and quality assessment, conducted independently by pairs of reviewers using a coding tool specifically designed for this review. Descriptive analysis was undertaken to examine and describe data related to the characteristics of the included studies and interventions. The findings from the included studies were combined descriptively using a narrative approach to synthesis. RESULTS The literature search yielded a total of 20,417 potentially relevant reports, 479 of which were retrieved for full‐text screening. A total of 14 studies met the eligibility criteria. Publication dates of included studies ranged between 1992 and 2012, with six studies published in the four‐year period 2010‐2013. Studies were conducted in nine different LMICs in Asia, Africa and Latin America: Bangladesh (three studies); Brazil (two studies); China (one study); India (four studies); Kenya (one study); Nigeria (one study); Philippines (one study); Vietnam (one study) and Zimbabwe (one study). The majority of studies examined outcomes for adults with physical impairments. There was variation in sample sizes. One study had a sample size greater than 500 participants, the sample size was between 251 and 500 in three studies, and the remaining ten studies had a sample size of less than 250. Different methodologies were employed to construct the counterfactual and evaluate the impacts of the interventions. The majority were uncontrolled before‐and‐after studies. One quasi‐experiment (ex‐post) utilised propensity score matching techniques and one study applied logistic regression to pre‐test/post‐test data. The remaining studies used a non‐equivalent groups design. All 14 studies were assessed as high risk of bias. The 14 studies examined 15 different interventions grouped as follows: treatment & therapy (four interventions); assistive devices and accommodations (two interventions); occupational rehabilitation services (four interventions); financial services (one intervention); and community‐based rehabilitation (four interventions). Thirteen were multi‐component programmes. All 15 interventions were targeted at people with disabilities, with some designed for people with a specific impairment or diagnosis. Six interventions targeted persons with specific types of physical impairment, and a further two interventions were available to adults with any type of physical impairment. Three interventions were targeted at persons with visual impairments. Finally, four interventions were available to persons with any/multiple impairments. Non‐governmental organisations (NGOs) were the most common source of funding. The main aim of eight interventions was to improve employment prospects for persons with disabilities. The other interventions sought improvements in a wider range of outcomes. The interventions were designed and implemented on different scales, with the majority available over a large geographical area, such as one or more districts, provinces or regions. Information about duration of the interventions was often not reported, but typically they were available for periods of less than six months. All 14 studies measured relevant labour market outcomes and assessed the following impacts: motivation to work (one study); professional social skills (one study); employment participation (12 studies); self‐employment (two studies); income (four studies) and hours worked (one study). Five studies measured additional outcomes: health‐related outcome (four studies); social outcomes (five studies); and empowerment‐related outcome (one study). Several studies relied solely on self‐reported outcome data. The majority of study reports did not provide clear information about the timing of outcome measurement. Only one study examined longer‐term outcomes, evaluating impacts after two and four years of participants entering the programme. In all 14 studies, the direction of effect was positive for the outcome variables measured. Five studies reported results of tests for statistical significance and indicated study findings were significant. Effects on motivation to work. One study measured this outcome. It investigated two interventions for visually impaired students. Effects on professional social skills. One study measured this outcome. It assessed a programme for persons with any type of physical impairment. Effects on paid employment. Twelve studies measured this outcome. Of these, seven studies evaluated different types of support for persons with physical disabilities, with five designed for people with a specific diagnosis or impairment. One study investigated an intervention for the visually impaired. The remaining four studies in this category evaluated interventions that were open to individuals with any/multiple types of impairments. Effects on self‐employment. Two studies measured this outcome. Both studies evaluated interventions available to persons with any type of physical impairment. Effects on income. Four studies measured this outcome. Of these, three studies evaluated interventions designed for persons with physical disabilities. The remaining study focused on an intervention for the visually impaired. Effects on hours worked. One study measured this outcome. It evaluated a programme for persons with any type of physical impairment. Seven of the 14 included studies explored variation in treatment effects. The variables considered were gender (three studies), participants' size of business (one study), impairment severity (one study), type of intervention (one study), and duration of follow‐up (two studies). Four of the seven studies tested whether results were statistically significant. Overall, these seven studies were not sufficiently similar to detect meaningful differences in outcomes. Two studies investigating occupational rehabilitation services reported participants' observations, experiences and views about why the intervention they received had worked for them. The following factors were cited: general health & well‐being; cooperation in the family/community; motivation; attitudes in the workplace; attitudes in the community; and appropriateness of the training. Three studies reported participants' observations, experiences and views about why the intervention they received had not worked for them. Two examined occupational rehabilitation services and the other evaluated the provision of free wheelchairs. The following barriers to the success of the interventions were cited: discriminatory attitudes of prospective employers; attitudes of family members and/or wider community; health and well‐being; physical inaccessibility (workplace and/or broader environment); lack of ‘start‐up’ funds for self‐employment; shortcomings of the training (i.e., mismatch between it and participant's skills, abilities and financial resources); lack of education and skills; and motivation. AUTHORS' CONCLUSIONS A key finding of this review is the overall scarcity of robust evidence, as indicated by the relatively few studies that met the inclusion criteria. Although the evidence in general showed positive results, we need to be wary of drawing strong inferences from the findings of this body of literature. Not only is the number of impact evaluations limited, but most used designs in which conclusively attributing causality is not possible. Our assessment of the evidence does not allow us to develop practical suggestions on what interventions are likely to work, for whom, and when. Clearly, there is an urgent need for investment in high quality impact evaluations of interventions to support people with disabilities in accessing the labour market in low‐ and middle‐income settings. To build the evidence base further, it is therefore important that many more of the interventions currently in existence in low‐ and middle‐income countries are rigorously evaluated, and the results are reported and disseminated widely. The methodological inconsistencies and weaknesses of the current evidence base, and specific knowledge gaps, suggest a number of future research priorities.