The 2030 Sustainable Development Agenda (SDA) adopted by the United Nations General Assembly in September 2015 asserts that it shall leave no-one behind in its global push for social and economic development[1]. As part of the agenda for ending poverty and inequality, ‘decent work for all’ has been promoted in Sustainable Development Goal (SDG) Eight as a key tool for inclusive economic development[1]. This rhetoric is of crucial importance in relation to the estimated one billion people living with disabilities globally, 80% of whom live in low and middle income countries (LMICs) [2, 3].

Persons with disabilities are described in the United Nations Convention on the Rights of Persons with Disabilities (UNCRPD) as those who have ‘long-term physical, mental, intellectual or sensory impairments which in interaction with various barriers may hinder their full and effective participation in society on an equal basis with others’ [4]. Disability was largely absent from the international development agenda set by the precedent 2000–2015 Millennium Development Goals, which arguably led to persons with disabilities being excluded from development efforts and widening the poverty gap between persons with and without disabilities [5–7]. In response to criticism on this issue, the 2015–2030 SDGs have placed a greater emphasis on inclusive development, with calls for tracking equity in progress towards each goal through disaggregation of data by disability.

Poverty and disability are interlinked; a recent systematic review showed a positive association between disability and economic poverty in 80% of the 78 included studies[8]. The relationship between poverty and disability is theorized as cyclical [9, 10]. Poverty is posited to increase the risk of disability via exclusion from health care or health information, and through heightened exposure to risk factors for poor health and impairment (including trauma, infectious disease, unsafe environments, poor sanitation and malnutrition) [11, 12]. Conversely, disability can lead to or exacerbate poverty through participation restrictions, including exclusion from education, and barriers to engaging in decent work [13]. Exclusion from livelihood opportunities has also been shown to negatively impact on psychosocial wellbeing, identity and social inclusion [14].

Livelihoods can be defined as the means through which individuals or households are able to meet their basic needs. It encompasses not only remunerated labour, but also an individual’s capabilities (e.g. level of education, skills), assets and participation in other productive activities (e.g. farming for direct consumption)[15]. Building upon this definition, the Sustainable Livelihood Approach promotes the idea that for a livelihood to be sustainable, individuals must be able to both maintain a basic standard of living through times of stress and shock (e.g. natural disasters, economic upheaval) and to have opportunities for livelihood improvement (e.g. through education and productive investments). The Sustainable Livelihood Approach has been a fundamental cornerstone of international development and poverty reduction, as it emphasizes a shift beyond the subsistence level, toward long-term poverty alleviation. A key component for sustainable livelihoods is engagement in decent work: work that is stable, respects an individual’s dignity, provides safe conditions and has fair remuneration.

Persons with disabilities are believed to face widespread exclusion from livelihood opportunities in many settings. While there is clear evidence from high-income countries of a gap in the employment rate between persons with and without disabilities–averaging 40% of the rate of persons without disabilities–equivalent analyses from LMICs are more challenging given the complexity of livelihood situations in many of these settings [16–18]. Notably, in many LMICs the vast majority of the labour force participates in the informal economy, in subsistence agriculture, or in economic activities that are not well monitored and difficult to measure [19]. Still, existing evidence points to substantial inequalities: for example, in the 2002–2003 World Health Surveys significant employment gaps between persons with and without disabilities were found across nine of 15 LMICs, with persons with multiple impairments and men experiencing the highest gaps[20]. Lower rates of employment among persons with disabilities have been found consistently in other studies, although many focus on formal sector employment only[21, 22].

Considering the over-representation of persons with disabilities amongst the poor[23] and the SDA’s focus on decent work for all as a tool for inclusive economic development, there is an urgent need for data on access to livelihood opportunities amongst persons with disabilities. Understanding this relationship is key if the focus on inclusive development and elimination of poverty within the SDGs is to be achieved.

Cameroon and India are two countries classified by the World Bank as lower middle income, on account of gross national income per capita, with approximately a quarter of the population in both countries living below $1.25 per day [24]. The United Nations Human Development Index—which generates country-level composite scores using indicators of health, life expectancy, education, and standard of living—ranked India 131st and Cameroon 153rd in 2016[24].

The UNCRPD was signed by Cameroon in 2008, but is yet to be ratified[25]. Limited data on disability in Cameroon are available. However, a recent review by Ray et al. (2017) identified five studies that addressed access to work amongst persons with specific impairments (for example visual impairments or related to HIV) in Cameroon, all of which documented unequal employment opportunities related to a combination of stigma, participation restrictions and restricted access to education[26].

The Persons with Disabilities Act in India, which legislates the right to equal opportunities and full participation amongst persons with disabilities, was enforced by the National Government of India in 1995[27]. This was followed by a National Policy for Persons with Disabilities in 2006 and ratification of the UNCRPD in 2007[27] [25]. However, ineffective programmes, insufficient funding and complexities in resource mobilisation are all stated barriers to the realisation of rights as set out in Indian inclusive legislature, leading to lower employment participation rates amongst persons with disabilities in the country [28–30].

The aim of this study was to build evidence on access to livelihoods amongst adults with and without disabilities in one district each of two LMICs: Cameroon and India.

In this two-setting study, persons with disabilities in both settings were five times less likely to be working compared to age, sex and community matched controls without disabilities, and this relationship held across age groups, sex, marital status, and education level. Among persons with disabilities, key predictors of working were lower current age, being married and not having either a physical impairment in both settings, or multiple impairments in Cameroon. Even in the oldest age group of 65 and above, persons with disabilities were substantially less likely to be working than persons without disabilities.

The evidence of substantially lower participation in work among persons with disabilities compared to their non-disabled peers in this study corroborates the limited literature on the negative relationship between disability and access to livelihoods in LMICs [13, 44, 45]. The very limited available evidence from Cameroon and India specifically also corroborates our findings[46]. This reinforces the theorized pathway to poverty via barriers to decent work for persons with disabilities and their families and is contrary to Article 27 of the UNCRPD on the right of persons with disabilities to work on an equal basis as others[4].

Overcoming the gap in access to livelihoods between persons with and without disabilities is essential in view of the international mandates put forward by the SDA and UNCRPD. Labour market analyses in high-income countries have highlighted numerous components underpinning the employment gap between persons with and without disabilities. These include employer misconceptions about the productive capacity of persons with disabilities, insufficient environmental or physical accommodations to the individual’s needs, and increased reservation wages (the lowest wage at which a person will work) affected by unbalanced benefit policies that may dis-incentivise persons with disabilities to join or remain in the labour market[16, 17]. Such dimensions are less well explored or understood in LMICs and in the context of more complex livelihood mechanisms. One qualitative study by Palmer et al. (2015) in Vietnam cited low educational attainment and discrimination as the biggest barriers to both formal and informal work for persons with disabilities in that setting [47].

Possible mechanisms for promoting greater participation of persons with disabilities in livelihoods include improved access to social protection systems, healthcare, rehabilitation and assistive devices, education and vocational training [48, 49]. Furthermore, Article 2 of the UNCRPD outlines key obligations of governments to ensure equal opportunities for persons with disabilities. These commitments include establishing anti-discrimination laws, ensuring the accessibility of workplaces and, together with employers, providing reasonable accommodations to workers with disabilities[50]. However, evidence on both the availability and impact of these different interventions on improving access to livelihoods for persons with disabilities in LMICs is extremely minimal, and in urgent need of prioritisation [51].

Persons with disabilities were slightly more likely to be in the poorest socio-economic quartile in India, but in Cameroon, no differences in socio-economic status between groups was observed. This contrasts to prevailing literature that has shown an association between livelihoods and poverty, but is similar to findings shown in Afghanistan, Zambia and Rwanda[52, 53]. Reasons are unclear, but may reflect very high levels of poverty across the population making it more difficult to detect differences between groups. Similarly, it exposes the need to explore more nuanced measures of multidimensional poverty incorporating additional dimensions such as living standards and empowerment, in addition to measures of economic poverty.

The similarity in education and literacy levels between cases and controls in both settings, while seemingly contrasting to the growing evidence on the widespread exclusion of children with disabilities from school [54] should be interpreted in light of age of onset of disability. The majority of cases (51% in India and 77% in Cameroon) reported disability onset beyond school age. This serves as a reminder of the need to be cognisant of the potentially varying implications of disability as acquired at different time-points in the life-cycle.

Exploring predictors of access to livelihoods amongst persons with disabilities highlighted additional trends and the heterogeneity of the lived experience of disability. The finding that in India, women with disabilities were twice as likely not to be working as men with disabilities, supports the theorized ‘double discrimination’ experienced by women with disabilities [55, 56]. In contrast, in Cameroon no difference was observed by sex. This may be related to the high proportion of both cases and controls working in agrarian livelihoods in Cameroon (77%, compared with 50% in India), which may be less vulnerable to external barriers (e.g. accessibility, stigma) than those seeking livelihood opportunities through an employer or customer-facing business.

Our study found that marital status was strongly associated with disability and access to livelihoods. First, persons with disabilities were less likely to be married than persons without disabilities in both settings, which supports previous literature on disability stigma and societal misconceptions of asexuality of persons with disabilities [3, 57]. Second, among persons with disabilities, those who were married were more likely to work even after adjustment for confounders. This may be related to the psychological benefits of cohabiting with a partner, as opposed to being single, widowed or divorced, which has long been established to build human and social capital, improve psychological wellbeing and provide resilience [58]. Conversely, the reverse causality may hold, in that there may be increased likelihood of marriage amongst persons with disabilities who work. Further, ideally longitudinal, research is needed in this area.

Age of onset and type of functional limitation affected likelihood of access to livelihoods in different ways in the two settings. In terms of onset, amongst persons with disabilities, those who had acquired their disability aged fifty and above in India, and below five years of age in Cameroon were the least likely to be working. Physical limitations were associated with lower likelihood of working in both settings, alongside depression in India, and the presence of multiple limitations in Cameroon. As over half of persons with disabilities that worked in India and three quarters in Cameroon were small-scale farmers, the inherently physically-demanding nature of the predominant livelihood may explain why those with physical limitations in each setting were less likely to work. The finding that persons with depression were least likely to be working in India is in line with findings from a systematic review highlighting a strong association between common mental disorders and poverty–including but not restricted to economic poverty and employment–in LMICs [59].

Taken in aggregate, these findings substantiate arguments related to the heterogeneity of disability and the lived experiences of persons with disabilities, and highlights the importance of disability data disaggregation in research findings[16, 60, 61]. Moreover, in relation to the Sustainable Development Agenda, it necessitates responsiveness and reactivity even within the context of inclusive programme design to meet diverse needs, capacities and environmental contexts, and break down the barriers to engaging in sustainable livelihoods experienced by persons with disabilities in different contexts and settings.

The relatively high proportion of persons with disabilities receiving state-sponsored livelihood support in India, and non-state support in Cameroon, is encouraging, particularly in light of the evolving discourse and evidence related to social protection as a mechanism for mitigating and preventing poverty. Social protection will be most transformative when it addresses drivers of poverty and barriers to decent work, such as poor access to timely, affordable healthcare and quality education; however, the nature and impact of state and non-state supports in these contexts was not a focus of this research. The role of social protection in reducing poverty and improving livelihoods is a complex and nuanced research area, which deserves further attention in future studies[62].

This study provides empirical evidence of the exclusion of persons with disabilities from livelihood opportunities in one district each of Cameroon and India. Moreover, the findings highlight the heterogeneity of that exclusion amongst persons with disabilities. This necessitates both adequately disaggregated quantitative data that fully reflects the spectrum of experiences by persons with disabilities in accessing livelihood opportunities, and appropriately reactive inclusive programmes that can meet diverse needs. The coverage of livelihood protective programmes and benefits in both settings was encouraging, and should be promoted within the context of sustainability and access to work.