1A large majority of the estimated one billion persons with disabilities around the world live in resource-limited countries, where they experience challenging living conditions and difficulties in achieving full social participation (World Health Organization & World Bank Group, 2011). In recent years, there has been an emergent literature on their situation in these contexts, with an important focus on the link between poverty and disability (Braithwaite & Mont, 2009; Groce et al., 2011; Mitra, 2018). Despite this important development, few studies have adopted a life-course perspective to examine the social experience of persons with disabilities in resource-limited countries. The transition to adulthood is an important stage of human development in resource-limited countries as in high-resource countries, with a potential impact on social participation in adult life (Arnett, 2014; Galland, 1984; Gaudet, 2007; Gillis, 1974; National Research Council, 2005). It is therefore particularly interesting to examine how disability affects this period of transition from social and economic dependence to independence and the acquisition of social roles. The transition to adulthood is characterized by different social markers, five of which have been identified as particularly important in both high- and low-income countries: leaving the parental home, which indicates movement towards independent living; the end of one’s formal education and securing a first job, which mark the establishment of financial independence; and living in a partnership and having children, which mark the beginning of one’s own family (Billari, 2001; Calvès et al., 2006; Coles, 1995; Galland, 2016; National Research Council, 2005; Van de Velde, 2014). There are also significant gender-related differences in the transitional pathways to adulthood, although some recent trends towards convergence have been noticed (Ibid). Events related to family life among women (e.g., marriage, birth) often impact their professional life, while greater disconnect between those two life domains is observed in men (Antoine, 2002; Hamel, Moguérou & Santelli, 2011). Furthermore, women experience an earlier transition than men from their parental family to their own family (Singh et al., 2000). In contrast, men usually start working before initiating a common life with a partner, while the opposite is more frequently observed among women (Antoine, 2002).
2Although there is a vast literature on the transition to adulthood, the specific case of young people with disabilities has received limited attention. Moreover, most of the available data is limited to high-income countries where the environment is more supportive, notably because of the important role played by social services and welfare programmes and because of the difference in the overall economic situation (Stewart, 2006). This study aims to fill this gap and to examine the effect of disability on the social markers of the transition to adulthood in an African urban context.
3The data used for this analysis were collected during the HandiVIH ANRS 12302 study, conducted in Yaoundé, Cameroon, between November 2014 and December 2015 (DeBeaudrap et al., 2016). Located in Central Africa, Cameroon ranks 151st in terms of human development and has a population of 24 million inhabitants, 66% of whom live below the international poverty line of $1.90 per day (United Nations Development Programme, 2018). Figures on disability prevalence vary widely according to the method and definition used, ranging from 1.5% to 10.4% (Simo Fotso, Duthé & Odimegwu, 2019). Cameroon is a signatory to the United Nations Convention on the Rights of Persons with Disabilities and its associated protocol (United Nations, 2006). However, it has not ratified either text. In 2010, a law for the protection and promotion of persons with disabilities was promulgated (République du Cameroun, 2010), and the decree on its implementation was published in 2018 (République du Cameroun, 2018). Like in many other countries in sub-Saharan Africa, programmes to improve education, health, and livelihoods for persons with disabilities are mostly small-scale programmes entrusted to faith-based or charitable organizations (Jolley et al., 2018).
4In this survey, two-phase cluster random sampling was used to recruit 807 persons with disabilities aged 15 to 49 years and 807 persons without disabilities of the same age, sex and residential location among households of the general population. Participants with disabilities were identified based on functional limitations measured with the Washington Group Short Set (WGSS) questionnaire (Altman, 2006). This tool includes a small number of questions asking the respondent to rate on a four-point scale how much difficulty he/she has experienced in six functional domains or basic actions: seeing, hearing, mobility, cognition, self-care, and communication. For each question, respondents are asked to choose one of the following: 1 – no difficulty, 2 – some difficulty, 3 – a lot of difficulty, or 4 – unable to do. Participants who reported a difficulty level greater than “a lot of difficulty” were considered to have severe limitations, while those who reported only “some difficulty” were considered to have mild disability.
5Face-to-face structured interviews were conducted to retrospectively collect data on participants’ life-course including family story and environment, education and professional activities, resources, intimate partners relationships and fertility using the life-grid method (Freedman et al., 1988). The life-grid used in this study was made of two A3 sheets divided into several columns (Figure 4 in Appendix). The vertical axis represents the time; the first column represents the time in years from birth to the current year; the second column represents age (from 0 to current age); and the third column represents the time elapsed. The other columns were used to report events according to their nature (people with whom the subject lived, main occupations/activities, resources, quality of life, intimate relationships, periods of transactional sex or sexual violence, pregnancies, children and disability onset). With this method, information on the timing and nature of the events forming the different dimensions studied were reported on a unique calendar (Lelièvre & Vivier, 2001). Data related to family environment were collected from birth, while data on the other dimensions were collected from 10 years of age. Details of the survey methods and procedures have been described elsewhere (DeBeaudrap et al., 2016).
6The final protocol of the study was approved by the Comité d’éthique pour la recherche en santé humaine in Cameroon and the Comité consultatif de déontologie et d’éthique from the Institut de recherche pour le développement in France. Informed consent was obtained from each participant (or from the legal representative when needed) prior to research. Because some questions explore sensitive topics, efforts were made to ensure confidentiality and privacy, and interviewers were trained to provide psychological assistance and to refer people in need of psychosocial support.
7This analysis includes cohorts born between 1966 and 1990 aged between 15 and 49 years old when data were collected. To preserve temporal ordering between the onset of disability and the transitions to adulthood, participants with disabilities were excluded from the analysis when the onset of disability occurred after 10 years of age. Five social markers of the transition to adulthood were examined:
-
establishment of independent living away from the parental home, as measured by the first time the individual lived without any of the parents (mother and father). Participants who were not raised by any of them were excluded from this analysis;
-
initiation of an independent family, as measured by the first occurrence of living in an intimate partner relationship (cohabitation or marriage);
-
end of education, as measured by the last transition from school or higher education to any other activity among participants who ever went to school;
-
work initiation, as measured by the first occurrence of a work activity in the formal or informal labour market. Only work activities occurring after 10 years of age were considered in this analysis;
-
parenthood initiation, as measured by the birth of the first child.
8Survival analysis was used to assess the time to each transition to adulthood, accounting for the unequal duration of observation (i.e., right censoring) (Therneau, 2015). The Cox model was used to estimate the association between the probability of experiencing the transitions and disability, overall and for sub-groups defined by the nature of the main activity limitation (mobility, visual, hearing, cognitive) and by its severity. This analysis was adjusted for the birth cohort (1966-80, 1981-90 or 1991-2000) to account for the important changes in the transitions to adulthood observed over the last decades (Juarez & Gayet, 2014; National Research Council, 2005). In addition, the transitions to first work and to living in an intimate partner relationship were jointly analysed to assess the probabilities of achieving:
-
both transitions;
-
none of them, according to the disability characteristics.
9In a second step, the associations among the characteristics of a new residential environment after leaving the parental home (cohabitation with family/friends/partner/alone), the educational level achieved, or the nature of the first job (formal versus informal work) and disability were assessed, overall and by sub-groups as defined above, using logistic regression. The analyses were initially conducted separately for women and men because of the gender-related differences observed in life-course trajectories (Barber & Axinn, 1998; Furstenberg, 1999). Differences between men and women in the timing and nature of the transitions were formally assessed in a second step using the data for men and women and including an interaction term for sex in the models. Data analysis was performed using R version 3.5.1 (R Core Team, 2018).
10The study population consisted of 420 participants with disabilities (51% were women) and 420 without disability of same sex and similar age and residential location (Fig. 1). The median age at data collection was 26 years without a significant difference according to sex (p = 0.4). As shown in Table 1, most women and men with disabilities reported severe limitations (84% and 85%, respectively). Mobility difficulties were the most common limitation reported for both sexes (40% and 43%, respectively), followed by visual difficulties in women (24%) and cognitive difficulties in men (31%). Overall, both men and women with disabilities were less likely than their counterparts to have left the parental home (65% versus 73%, p = 0.05 and 69% versus 77%, p = 0.05, for men and women, respectively), to have left school earlier (56% versus 62%, p = 0.14 and 59% versus 66%, p = 0.10), to have found a job (52% versus 60%, p = 0.07 and 46% versus 55%, p = 0.04), to have begun living in a partnership (17% versus 36%, p < 0.001 and 26% versus 48%, p < 0.001) and to have had a child (27% versus 48%, p < 0.001 and 49% versus 66%, p < 0.001). Fifty women (96% of women with disabilities) and sixty men (93% of whom were disabled) who were not in school at the beginning of the observation were excluded from the analysis of the “end of education” transition but not from the other analyses.
Figure 1: Study flowchart.
Table 1: Study population characteristics.
|
Women n (%)
|
|
|
Men n (%)
|
|
|
Without disability
|
With disability
|
|
Without disability
|
With disability
|
|
(n = 212)
|
(n = 212)
|
(n = 208)
|
(n = 208)
|
|
Birth cohorta
|
|
|
|
|
|
1966–1980
|
42 (20%)
|
38 (18%)
|
49 (24%)
|
46 (22%)
|
|
1981–1990
|
74 (35%)
|
79 (37%)
|
73 (35%)
|
68 (33%)
|
|
1991–2000
|
96 (45%)
|
95 (45%)
|
86 (41%)
|
94 (45%)
|
|
Limitation severity
|
|
|
|
|
|
Mild
|
na
|
35 (16%)
|
na
|
36 (17%)
|
|
Severe
|
na
|
177 (84%)
|
na
|
172 (83%)
|
|
Type of limitationb
|
|
|
|
|
|
Visual
|
na
|
50 (24%)
|
na
|
25 (12%)
|
|
Hearing
|
na
|
39 (18%)
|
na
|
29 (14%)
|
|
Mobility
|
na
|
84 (40%)
|
na
|
89 (43%)
|
|
Cognitive
|
na
|
43 (20%)
|
na
|
64 (31%)
|
|
Markers of transition to adulthood
|
|
|
|
|
|
Leaving parental
|
163 (77%)
|
147 (69%)
|
152 (73%)
|
136 (65%)
|
|
home
|
|
|
|
|
|
End of education
|
139 (66%)
|
125 (59%)
|
128 (62%)
|
116 (56%)
|
|
First job
|
117 (55%)
|
98 (46%)
|
125 (60%)
|
109 (52%)
|
|
Living in partnership
|
101 (48%)
|
56 (26%)
|
75 (36%)
|
35 (17%)
|
|
First child
|
139 (66%)
|
103 (49%)
|
100 (48%)
|
57 (27%)
|
na: not applicable.
a When data was collected: the 1991–2000 cohort was 15–24 years old, the 1981–1990 cohort was 25–34 years old, and the 1966–1980 cohort was 35–49 years old.
b Types of limitations are non-exclusive. There were 14 cases of double limitation and two of triple limitation.
11Overall, as shown in Fig. 2, severe functional limitations were associated with significant delays in the transitional age of work initiation, living in an intimate partner relationship and with parenthood. The magnitudes of these delays were similar in men and women (p-values for interaction: 0.6, 0.6 and 0.8 for work initiation, living in an intimate partner relationship, and parenthood, respectively). In addition, women with severe limitations (and, to a lesser extent, men) were at higher risk of leaving school at an earlier age (p = 0.01 and 0.09 for women and men, respectively). Men [but not women (p = 0.19)] with severe limitations (p = 0.002) left the parental home later than their counterparts without disabilities. A more complex pattern was observed among participants with moderate limitations. In women, moderate limitations had the opposite effect on the age at first job than did severe limitations (i.e., reducing age at first job) but were not associated with a significant change in the age of living in an intimate partner relationship (p = 0.43) or in the age of parenthood (p = 0.69). In contrast, men with moderate limitations were less likely to start living in an intimate partner relationship (p = 0.01) or to have a child (p = 0.01) but started working at a similar age as men without disabilities (p = 0.53).
Figure 2: Changes in age of transitions to adulthood with respect to sex and disability severity.
12As shown in Fig. 3, there were also significant variations in the association between disability and the transitions to adulthood across the different types of limitations. Mobility limitations were associated with delayed transitions to independent living in an intimate partner relationship and to parenthood for both men and women (p < 0.001 in both cases). There was also a trend towards older age at first work in women (p = 0.04) but not in men (p = 0.3). Women and men with visual and/or hearing limitations were more likely to leave school earlier than their counterparts without disability. However, only participants with visual limitations started a job earlier, meaning that participants with hearing limitations experienced a longer gap between the end of their education and their first job (1.1 versus 0.3 years, p = 0.05). Participants with hearing limitations were also more likely to live in an intimate partner relationship later, with a trend towards longer delays in men than in women (p-value for interaction: 0.1). All the transitions to adulthood were strongly delayed for men and women with cognitive limitations.
Figure 3: Changes in age of transitions to adulthood with respect to sex and disability characteristics.
N1: number of people with a disability at the beginning of the event observation; N2: number of people without a disability at the beginning of the event observation; n1: number of people with a disability who experienced the event; n2: number of people without a disability who experienced the event; HR: hazard ratio; 95% confidence interval.
13When the transitions to work and living in partnership were analysed jointly, women [but not men (p = 0.3)] with severe limitations were significantly more likely to not have completed any transition [hazard ratio (HR): 1.60, 95% CI: 1.17-2.18], and both women and men with severe limitations were less likely to have made both transitions (HR: 0.31, 95% CI: 0.20-0.48 and 0.44, 95% CI: 0.29-0.66, respectively). In contrast, there was no significant difference regarding the risk of not achieving any transition between women and men with mild limitations and those without (p = 0.3 and 0.7, respectively); furthermore, men [but not women (p = 0.9)] with mild limitation were at higher risk of not achieving both (HR: 0.18, 95% CI: 0.04-0.73).
14After leaving the parental home, women with disabilities were less likely than those without disabilities to live with a partner (11% versus 22%, p = 0.006, Table 2). This difference was observed among participants with severe limitations and among those with mobility limitations (respectively, 9% versus 22%, p = 0.003 and 8% versus 22%, p = 0.02) but not in the other sub-groups. In contrast, there was no significant difference between men with and without disabilities with respect to the nature of the residential environment after leaving the parental home (p = 0.13).
Table 2: Residential environment after leaving parental home with respect to disability characteristics and sex.
|
Family (%)
|
Husband/wife (%)
|
Friends (%)
|
Alone (%)
|
|
Women
|
|
|
|
|
|
|
No disability
|
163
|
68
|
22
|
1
|
9
|
|
Any disability
|
147
|
82
|
11
|
3
|
4
|
|
Mild limitations
|
25
|
72
|
20
|
0
|
8
|
|
Severe limitations
|
122
|
84
|
9
|
4
|
3
|
|
Visual limitations
|
39
|
80
|
15
|
0
|
5
|
|
Hearing limitations
|
26
|
81
|
11
|
4
|
4
|
|
Mobility limitations
|
63
|
82
|
8
|
5
|
5
|
|
Cognitive limitations
|
20
|
85
|
5
|
10
|
0
|
|
Men
|
|
|
|
|
|
|
No disability
|
152
|
73
|
2
|
5
|
20
|
|
Any disability
|
135
|
77
|
5
|
6
|
12
|
|
Mild limitations
|
23
|
83
|
0
|
13
|
4
|
|
Severe limitations
|
112
|
77
|
6
|
4
|
13
|
|
Visual limitations
|
16
|
63
|
6
|
12
|
19
|
|
Hearing limitations
|
17
|
88
|
0
|
0
|
12
|
|
Mobility limitations
|
73
|
69
|
8
|
7
|
16
|
|
Cognitive limitations
|
23
|
96
|
0
|
4
|
0
|
[hazard ratio (HR): 1.60, 95% CI: 1.17-2.18], and both women and men with severe limitations were less likely to have made both transitions (HR: 0.31, 95% CI: 0.20-0.48 and 0.44, 95% CI: 0.29-0.66, respectively). In contrast, there was no significant difference regarding the risk of not achieving any transition between women and men with mild limitations and those without (p = 0.3 and 0.7, respec- tively); furthermore, men [but not women (p = 0.9)] with mild limitation were at higher risk of not achieving both (HR: 0.18, 95% CI: 0.04-0.73).
15Educational attainment was lower in women and men with disabilities than in their counterparts without disabilities (Table 3): 19% of the women with disabilities and 17% of men with disabilities had an educational level of less than primary compared to 6% and 1% of women and men, respectively, without disabilities (p < 0.001). However, there was substantial heterogeneity across the sub-groups: for instance, women with mobility limitations and women with moderate limitations had an educational attainment similar to that of those without disabilities (p = 0.4 in both cases).
Table 3: Educational level at the end of the education period with respect to disability characteristics and sex.
|
n
|
Educational level
|
|
|
< Primary (%)
|
Primary (%)
|
Secondary (%)
|
Higher (%)
|
|
Women
|
|
|
|
|
|
|
No disability
|
210
|
6
|
26
|
48
|
20
|
|
Any disability
|
164
|
19
|
26
|
43
|
12
|
|
Mild limitations
|
32
|
9
|
22
|
60
|
9
|
|
Severe limitations
|
132
|
22
|
27
|
39
|
12
|
|
Visual limitations
|
49
|
14
|
16
|
58
|
12
|
|
Hearing limitations
|
32
|
16
|
41
|
34
|
9
|
|
Mobility limitations
|
62
|
13
|
24
|
44
|
19
|
|
Cognitive limitations
|
19
|
53
|
31
|
16
|
0
|
|
Men
|
|
|
|
|
|
|
No disability
|
204
|
1
|
19
|
48
|
32
|
|
Any disability
|
152
|
17
|
30
|
34
|
19
|
|
Mild limitations
|
29
|
13
|
21
|
45
|
21
|
|
Severe limitations
|
123
|
18
|
32
|
31
|
19
|
|
Visual limitations
|
23
|
17
|
36
|
17
|
30
|
|
Hearing limitations
|
22
|
18
|
50
|
23
|
9
|
|
Mobility limitations
|
74
|
8
|
22
|
46
|
24
|
|
Cognitive limitations
|
27
|
41
|
33
|
22
|
4
|
16The characteristics of the first job were significantly different between men and women with disabilities but not between men and women without disabilities (Table 4): a majority of participants (men or women) with disabilities were working in the informal labour market, while a majority of those without disabilities were working in the formal labour market. These differences were also found among men and women with severe limitations (p = 0.02), men with moderate limitations (p = 0.01), and men and women with cognitive limitations (p < 0.001).
Table 4: Characteristics of first work with respect to disability characteristics and sex.
|
n
|
Formal labour market (%)
|
Informal labour market (%)
|
|
Women
|
|
|
|
|
No disability
|
117
|
53
|
47
|
|
Any disability
|
98
|
36
|
64
|
|
Mild limitations
|
21
|
43
|
57
|
|
Severe limitations
|
77
|
34
|
66
|
|
Visual limitations
|
34
|
38
|
62
|
|
Hearing limitations
|
15
|
40
|
60
|
|
Mobility limitations
|
45
|
36
|
64
|
|
Cognitive limitations
|
4
|
25
|
75
|
|
Men
|
|
|
|
|
No disability
|
125
|
55
|
45
|
|
Any disability
|
109
|
44
|
56
|
|
Mild limitations
|
19
|
21
|
79
|
|
Severe limitations
|
90
|
49
|
51
|
|
Visual limitations
|
14
|
43
|
57
|
|
Hearing limitations
|
14
|
29
|
71
|
|
Mobility limitations
|
63
|
54
|
46
|
|
Cognitive limitations
|
15
|
7
|
93
|
17Our study is one of the first to examine how growing with a disability in the context of an African city impacts the transition to adulthood, an important period of human development that shapes the future life of young people (Arnett, 2014). Our results show that men and women with disabilities experience significant delays in completing the different transitions to adulthood. These results are consistent with the growing evidence already available from high-resource context showing significant differences in educational achievements and access to employment, as well as in marriage or independent living, between youth with and without disabilities (Filmer, 2008; Janus, 2009; Mactaggart et al., 2018; Simo Fotso et al., 2018; Wells, Sandefur & Hogan, 2003). However, it is noteworthy that structural factors, which are key to understanding these transitions, differ dramatically between high- and low-resource settings. In the former, institutions play an important role in the lives of young people with disabilities and their families (Stewart, 2006), while restricted resources and economical adjustments have limited the place of institutional support in many low-resource countries (Jolley et al., 2018).
18In this study, young people with disabilities were less likely to access secondary or higher educational levels and to be employed, which corroborates results from various surveys indicating that people with disabilities are at increased risk of “multidimensional” poverty, a term introduced for deprivations in different domains (e.g., employment, education, health) (Mitra, 2018). Our results also reveal complex variations in the effect of disability on these transitions with respect to disability characteristics and gender, which deserves full attention.
19First, we found that the severity of the limitations modulates gender differences in the transitions to adulthood. In the population without disabilities in this study, men were more likely than women to achieve higher levels of education and to access paid work, while women lived in an intimate partner relationship and had children earlier, which is consistent with data from other sources (Juarez & Gayet, 2014; National Research Council, 2005). The presence of severe difficulties had a strong effect on all transitions, and there was no evidence for heterogeneity associated with sex, which suggests that severe disabilities compound gender inequalities. Some caution is nevertheless required, as the power for these modification analyses was limited. In contrast, women with moderate difficulties showed a faster transition to work, mostly in the informal sector, without a significant delay in the initiation of their own family compared to their non-disabled counterparts. Although they stopped attending school earlier than non-disabled young women, most of them completed the secondary level, but few reached higher education. Men with moderate difficulties started their own families later than men without disabilities but worked at a similar age. These results illustrate the complexity of the intersectionality between disability and gender and outline the need for further research, including qualitative studies, to better understand the diversity of the situation and the determinants thereof (Mohamed & Shefer, 2015). Most studies available have focused on people with severe disabilities; it would also be interesting to collect information on the group with milder functional limitations, which appears to experience barriers to full social participation.
20Our study also highlights that the effect of disability on the transition to adulthood is highly variable across the different types of limitations. Cognitive difficulties had the strongest impact on the transitions, with all domains of transition delayed in both men and women. Indeed, the transition to adulthood is a challenging stage in the development of people with cognitive disabilities (Redgrove, Jewell & Ellison, 2016; Riddell, Wilson & Baron, 1999), in the first place because of the strong belief that they are childlike and dependent (Groce, 2004; Redgrove et al., 2016). In resource-rich settings, external services can provide support to families (Baldwin & Carlisle, 1994). By contrast, in resource-limited countries such as Cameroon, which are characterized by a lack of services for people with intellectual disabilities, the transition must operate within the family (Jolley et al., 2018).
21In contrast to the cognitive limitations, we found that the other types of limitations were associated with more selective effects on the transitions to adulthood. For instance, mobility difficulties had a strong impact on transitions in the family domain (first marriage, first birth) but a limited effect on those in the professional domain (end of education and first job). These difficulties experienced by young people with mobility difficulties starting their own family could be related to the persistent belief that people with disabilities, especially those with more visible physical functional limitations, are asexual and not valuable partners (Milligan & Neufeldt, 2001). Sensory difficulties, however, mainly impacted the educational dimension, which reflects the barriers related to the ability of the educational system to accommodate individuals’ difficulties. The lack of adapted teaching for hearing-impaired children and not having eyeglasses are strong barriers to educational success in many resource-limited countries; these absences of these resources may, in turn, reduce work opportunities (World Health Organization & World Bank Group, 2011).
22Through its random population-based selection of participants with and without disabilities, this study provides important empirical evidence on the overarching inequalities in the transition to adulthood related to disability. However, it also has limitations that should be taken into account in the interpretation of the results. First, the limited sample size results in low power for the sub-group analyses. Consequently, it was not possible to investigate the effect of the severity and type of functional limitation jointly. Information on the transitions was collected retrospectively using the life-grid method. Although this method has been shown to be reliable for important life events, such as those examined in this analysis (Caspi et al., 1996), it could be prone to memory bias. In addition, the observation period of this study was relatively short because of the relatively young age of its population (median 26 years), resulting in an incomplete picture of the transition to adulthood.
23To provide a detailed description of the association between disability and the different transitions to adulthood with respect to functional limitation type, severity and gender, the transitions were examined independently. However, there are inter-relations between the different transitions (Arnett, 2014), and future research should complement these findings by adopting a sequential approach and using qualitative methods to incorporate the perspectives of people with disabilities.
24Our results showed significant delays in the participation of young African people with disabilities in different activities of adulthood, but our findings also highlighted the diversity of the situations. To achieve the goal of equal opportunity as envisioned in the sustainable development goals, gender- and disability-sensitive transitional programmes that take into account the different types of impairment are needed.