Little is known about the details of the journey out of care towards independence. Longitudinal research in South Africa finds that residential care-leaving outcomes do not improve significantly between one and two years out of care. This suggests that the outcomes achieved during the first 12 months set a pathway that extends to at least two years, and implies that child welfare services need to provide intensive support to care-leavers during the transition out of the care system.
Keywords
Leaving care; out of home care; transitional outcomes; longitudinal research; youth transitions
Introduction
The transition out of adolescence and towards independent living is particularly challenging for young people aging out of care. While many youth continue to enjoy the extended support of family, care-leavers typically have little family support, leading some to refer to this transition as “compressed and accelerated” (Stein, 2008, p. 39). Evidence from the UK (National Audit Office, 2015), USA (Courtney & Dworsky, 2006) and Australia (Mendes, Johnson, & Moslehuddin, 2011) is consistent in showing that care-leavers have poor transitional outcomes – they may be accelerated, but they are not effective.
In South Africa, independent living outcomes of young people transitioning out of residential care appear to be similarly disappointing. One year out of care, approximately a third of care-leavers are NEET (not in employment, education or training) and more than three quarters earn less than the minimum wage (Dickens, 2016). While these findings are not dissimilar from those of the population of South African youth, they do not correspond to the ‘better life’ that child welfare staff aspire to for the children in their care. The question thus arises whether the outcomes seen at one year out of care improve over time.
My colleagues and I had anticipated, based on our practice experience and research, that care- leavers might experience a slow start in achieving independent living outcomes, with initially poor accommodation and substance abuse (for example), but that these would probably pick up from the second year out of care. We expected that the shock of the transition out of care and the lack of support networks and aftercare services would disrupt the transition out of care initially, but that care-leavers would find their feet after a short while and begin to establish themselves in independent living.
Care-leaving research in South Africa is still in its infancy (Van Breda & Dickens, 2016). Prior to 2012, there were no more than two research outputs per year, dating back to 2003, prior to which there appears to have been no research on care-leaving. Since 2012, however, research on care- leavers in South Africa has rapidly increased, reaching 19 outputs over 2015/6. With one exception, all of the research has focused on young people leaving residential care. However, other than the study reported in this article, there has been no longitudinal research that tracks the journey of young people out of care and towards independence. There have been a handful of studies of care-leavers a few years after leaving care, but these are cross-sectional studies, unable to show changes in independent living outcomes over time.
This article thus examines the shifts in transitional or independent living outcomes of a sample of residential care-leavers over their first and second years out of care. The article draws on data from a larger longitudinal study on care-leaving called Growth Beyond the Town, which is being conducted among youth exiting the residential care of Girls and Boys Town South Africa (GBT). The article aims to describe the changes, or lack of changes, in a range of independent living outcomes from one to two years out of care, with the expectation that we will see an inclining pattern of independence over time.
Leaving care in the South African context
In South Africa, children come into the care system (i.e. foster care, which includes kinship care, or residential care) through the Children’s Court (RSA, 2005). This requires a social work investigation and report, recommending to the Magistrate of the Children’s Court that a child be found in need of care and protection. The Children’s Act (RSA, 2005) champions children remaining in the family context, with foster care as the preferred form of alternative care, and residential care as a last resort and only when in the best interests of the child. Over 21,000 children lived in residential care in 2011/2012 (Jamieson, 2017). For the most part, children leave care at the end of their 18th year.
There is, in South Africa, very little in the way of aftercare services, and the Children’s Act is almost silent on services to care-leavers (Van Breda & Dickens, 2016). Consequently, young people leaving care typically exit with little or no aftercare support, frequently into contexts of severe deprivation: youth unemployment in South Africa exceeds 50% (ibid.). Young people therefore have to make the transition to independence very rapidly and be particularly resourceful to avoid the unemployment and poverty trap.
While the term ‘independent living’ continues to be used as a shorthand for ‘young adult life after leaving care’, the lack of formal social welfare services and networks and the preponderance of indigenous informal family-, kin- and community-based resources mean young people actually transition into ‘interdependent living’ (Tanur, 2012). It is the young person’s ability to network with a variety of people and social systems, which range from enduring relationships to complete strangers, that appears to enable them to make the transition towards young adulthood more successfully, and which forms a foundation of resilience for South African care-leavers (Van Breda, 2015).
Longitudinal research on care-leaving
Very few longitudinal studies have been conducted on care-leavers globally, where the same cohort is measured on the same transitional outcomes on more than one occasion after leaving care (Jones, 2011). Most longitudinal studies on care-leaving report two points of data collection – just before transitioning out of care and at some point after leaving care (e.g. Dinisman, 2016). Only four published studies with two or more data points after leaving care could be located, viz. Cashmore and Paxman, Jones, McMillen, and Courtney. Three of these studies are located in the USA and one in Australia; no longitudinal studies on care-leaving have been conducted in South Africa, or elsewhere in Africa.
Cashmore and Paxman (2007) studied a group of 45 young people transitioning out of residential and foster care in New South Wales, Australia, in 1992/3. They interviewed the care-leavers just before leaving care, three months after leaving care and again at 12 months and 4-5 years out of care. Their report showed changes in several independent living outcomes over time. For example, three months out of care, 26% of participants were living independently or sharing accommodation, compared with 42% at 12 months and 76% at 4-5 years – a steady increase over time. During their first year out of care, participants changed accommodation on average 3.3 times. About a quarter of participants left care without completing their education and did not complete any qualification over the following 4-5 years. The focus of their report, however, was not so much on tracking changes in independent living outcomes over time. Rather, they focused on comparing care-leavers with age- equivalent cohorts of non-cared-for young people, on describing their outcomes at 4-5 years out of care (Cashmore, Paxman, & Townsend, 2007) and on identifying factors that predict better transitional outcomes (Cashmore et al., 2007). In addition, their data is now almost 20 years old.
Jones (2011) conducted a longitudinal study of care-leavers transitioning out of a ‘residential education’ programme in the USA between 2001 and 2004. Of the initial 42 participants, 16 participated in follow-up interviews at one, two and three years out of care. Most of the other 26 participants were lost to follow up due to moving. Data were analysed on a case-by-case basis, rather than as an aggregate group, enabling close observation of each young person’s care-leaving journey over time.
McMillen and colleagues conducted a longitudinal study with 325 young people transitioning out of foster and residential care in Missouri, USA, between 2001 and 2003, interviewing them three- monthly from the month they turned 17 until age 19. They found that a little over half the participants had disengaged from care by the age of 19 (McCoy, McMillen, & Spitznagel, 2008).
Youth who left care younger (soon after turning 17) were more likely to return to live with family, while those who left after 18 tended to move directly to independent living. The percentage of youth utilising mental health services dropped precipitously at the point of leaving care and care status was the strongest predictor of continued service use (McMillen & Raghavan, 2009). The researchers found that care-leavers who had a stable mentor over the course of the interviews had lower stress and were less likely to be in conflict with the law by age 19 (Munson & McMillen, 2009). Further analysis of the data reveals that the first year out of care, regardless of the age of disengagement, is a vulnerable year for increased substance use (Narendorf & McMillen, 2010), and that remaining in care longer may protect against substance use.
Courtney’s Midwest Study in the USA is the largest and longest longitudinal study of care-leaving to date, with 736 youth in foster and residential care first interviewed at age 17 between 2002 and 2003, and followed up at 19, 21, 23/24 and 26 years (Courtney et al., 2011). While all participants were in care at age 17, almost half (53%) had disengaged from care by age 19, and by age 21 all had been disengaged. Looking at the final dataset, the research finds a sharp increase in employment from 32-41% at ages 17 and 19 to 49-54% at ages 21 to 26. Among men, arrests peaked slightly at 43% at age 21 and decreased thereafter to 38% at age 26, while incarcerations peaked at 43-44% at ages 21 and 23/24 and decreased to 40% by age 26. NEET rates were lowest at age 17 while still in care (4-5%), increased sharply by age 19 (32-34%), and continued to increase over time to 38% among women and 51% among men by age 26. This study is now generating nuanced research into care-leaving pathways, for example, Lee, Courtney, Harachi, and Tajima (2015) show how criminal involvement in adolescence decreases the likelihood of attaining a high school diploma by age 19, which subsequently reduces employment and increases criminal activity by age 21.
The Growth Beyond the Town study is distinct from these four studies in a number of ways. First, it specifically focuses on young people transitioning out of residential care (like Jones’ study), rather than foster care more inclusively defined (the other three studies). This distinction is important, because foster care is frequently intended as a relationship that lasts a lifetime, thus with a less clearly demarcated disengagement than residential care. The GBT participants therefore had to leave care earlier than those in some of the longitudinal studies reported here. Second, as with the studies by Cashmore and Paxman and Jones, this study enrolled participants at the point of disengaging from care, rather than at a particular age, as with McMillen’s and Courtney’s studies, making it less about age and more deliberately about leaving the care system. Third, the focus of this study is solely on examining changes in care-leaving outcomes between one and two years out of care, rather than on comparisons with age-equivalent samples or even making observations about the state of care-leavers at one or two years post care. In this way, the current study may make a significant contribution not only to understanding care-leaving in South Africa, but also to understanding care-leaving pathways internationally.
Methodology
The data for this article was drawn from the dataset of a larger ongoing study on care-leaving in South Africa, called Growth Beyond the Town. This is a mixed methods, longitudinal, rolling cohort study of young people transitioning out of Girls and Boys Town South Africa’s (GBT) residential children’s homes. The larger study aims to describe care-leavers’ journey towards independent living over time and to identify resilience processes that facilitate more ‘successful’ transitioning towards young adulthood. All young people disengaging from GBT at age 16 and above are invited to participate in the study. Participants are recruited in the year of disengagement and participate in a baseline interview shortly before leaving care. Annually thereafter they are invited to participate in a follow-up interview. At disengagement, they complete a resilience assessment and a semi- structured interview that focuses on their preparation and readiness for leaving care. Their social worker also completes a questionnaire about their care history. At the annual follow-up interviews, they participate in an unstructured interview focused on their care-leaving narrative since the last interview and complete a structured interview and self-administered questionnaire on independent living outcomes. This article uses just the quantitative assessment of independent living outcomes, collected at one and two years out of care.
Fifty-eight young people enrolled between the study’s inception in 2012 and the end of 2014. Of these, 35 participated in both one- and two-year interviews between 2013 and 2016, representing a 60% retention rate. The 23 care-leavers who dropped out of the study did so because they could not be traced (n=15), withdrew from the study (n=4), were readmitted into care (n=3) or died (n=1).
Eleven of these dropped out during the first year out of care and a further 12 during the second year out of care. No demographic differences (race, gender, disability, care facility, home province, age at first care placement, number of care placements, age at disengagement or educational attainment) were found between the 35 care-leavers who completed the one- and two-year interviews and the 23 who did not, suggesting that this sample is representative of the population from which it was drawn.
Participants were assessed using a structured interview schedule and self-administered scale, designed for this study, which assess eight outcomes measured dichotomously (i.e. each participant is scored as either meeting or not meeting the outcome criteria) and 11 on a continuous scale (i.e. each participant obtains a score that can range from zero to 100). These cover the independent living outcomes widely used in care-leaving studies, viz. accommodation, employment, education, finances, substance use, crime, health, well-being, relationships with family, friends and lovers, and NEET. Some of these constructs are measured as both dichotomous and continuous variables (e.g. finances), while some are measured as only dichotomous (e.g. education for employment) or continuous (e.g. health) variables.
The domains were selected based on a review of the outcomes addressed in international care- leaving studies (Dickens, 2016). The items for each domain were constructed through benchmarking against these international studies, together with South African national surveys, including the Census, conducted by Statistics South Africa. In addition, the team that designed the items drew on their prior research on care-leavers in South Africa as well as their experience working in the Child and Youth Care sector. Table 1 provides the type and definitions of the variables.
Interviews were conducted face to face, except in rare instances when a participant was in a remote rural area, when the interview was conducted telephonically. Fieldworkers completed the structured interview schedule, while the participants completed the self-administered scale. All data was captured in a Microsoft Access database that is used to manage the data for the entire study. For the purposes of this article, one- and two-year outcome data for the identified sample was merged and exported to SPSS v24 for analysis.
Table 1. Definitions of Outcome Variables
| Title | Type | Definition |
| Accommodation | Continuous | The extent to which care-leavers live independently (or with a partner) in self-funded, formal housing, with no moves or periods of homelessness since their last interview. |
| Paid Employment | Continuous | The extent to which working care-leavers have stable employment, with reasonable working hours and perform well in their jobs. |
| Studying | Continuous | The extent to which studying care-leavers persist in, commit to and perform well in their studies. |
| Financial Security | Continuous | The extent to which care-leavers are financially independent, with a well-paying job, their own bank account, sufficient savings and no ‘bad’ debt. |
| Drugs & Alcohol * | Continuous | The extent to which care-leavers used cigarettes, alcohol, cannabis and hard drugs over the past 2-4 weeks. |
| Crime * | Continuous | The extent to which care-leavers engaged in vandalism, theft and violence and have had trouble with the law since their last interview. |
| Physical health | Continuous | The extent to which care-leavers feel healthy (e.g. good energy, mobility, sleep and absence of pain), so that they can function in daily life. |
| Psychological health | Continuous | The extent to which care-leavers experience well-being (e.g. good body image, self-esteem, concentration, meaning in life and absence of negative emotions), so that they can function in daily life. |
| Family relationships | Continuous | Relationships with family members are experienced as caring and supportive. |
| Friend relationships | Continuous | Relationships with friends are experienced as pro-social, caring and supportive. |
| Love relationships | Continuous | A romantic relationship that is experienced as intimate and characterised by mutual understanding. |
| Self-supporting Accommodation | Dichotomous | The percentage of care-leavers who are paying for, or own, their own accommodation, or receive accommodation in exchange for work |
| Education for Employment | Dichotomous | The percentage of care-leavers who have completed, or are busy with, secondary education or a trade qualification. |
| NEET * | Dichotomous | The percentage of care-leavers who are not working, studying, or in training |
| Reliable Employment | Dichotomous | The percentage of employed care-leavers who have maintained a reliable work record |
| Diligent Education | Dichotomous | The percentage of studying care-leavers who attend class and have not failed their modules during the past year |
| Liveable income | Dichotomous | The percentage of care-leavers earning above R1600 per month through employment and with no short term loans (other than from the bank, friends or family) Note: minimum wage for domestic workers for 2015 = R2000/month |
| Drug & Alcohol ‘Free’ | Dichotomous | The percentage of care-leavers who, during the past 2-4 weeks, avoided binge drinking more than once a week, who used dagga no more than twice a week, and who did not use hard drugs |
| Crime ‘Free’ | Dichotomous | The percentage of care-leavers who avoided any serious crime or trouble with the law during the past year |
* These outcomes are negatively scored – a high score indicates negative outcomes.
Because of the small sample size and the non-normal distribution of data, non-parametric statistics were used for the analysis (Pett, 2016). While non-parametric statistics have less power to detect differences or relationships in the data, a benefit of this is to strengthen the confidence we can have in significant results. Because of the small sample size and the exploratory nature of the study, two- tailed significance was set at the more permissive level of p < .05.
For continuous variables, the Wilcoxon signed-rank test was used to compare paired differences between one- and two-year outcomes (Corder & Foreman, 2009). The effect size of the resultant z score was calculated for each test. For the dichotomous variables (2x2 contingency tables), McNemar’s chi-square test was used because the same participants were measured twice on a dichotomous variable (Elliott & Woodward, 2016). McNemar is concerned to determine whether a significant number of individuals change their outcome status from time one to time two, e.g. a decrease in NEET rates. Because the numbers of participants who changed status were low (i.e. less than 25 in all tests), McNemar’s binomial distribution was used to generate an exact p statistic.
Ethical approval for the study was provided by the Faculty of Humanities Academic Research Ethics Committee of the University of Johannesburg on 20 September 2012. GBT, as legal guardian, provided consent for the study to take place and for the participation of each young person. Participants were provided with information about the study and signed a consent form at each interview. In addition, parental consent was obtained for those under 18, since in South Africa the parenting role of parents is encouraged even when children have been removed from their care. The consent form clarified the limits of confidentiality (e.g. in the case of disclosed abuse or intent to harm someone else). Each interview incorporated an assessment of the participant’s need for a counselling referral and GBT professionals were on standby. Participants were provided with a snack and drink during the interview (which takes about two hours) and a R100 ($7.50) compensation for their time. Every interview incorporated an unstructured narrative interview, which provided participants with an opportunity to reflect on their past year of life experience – the vast majority of participants rated this highly, as a unique opportunity to take stock of their journey to date. All data was anonymized before analysis.
Results
The sample was predominantly male (33 of 35 participants), due to GBT taking girls into their programme only recently. The majority of participants were African (n=19), seven mixed-race, six white, and three Indian. All the participants were South African citizens, and all but one came from the three provinces where GBT sites are located (Gauteng, Western Cape and KwaZulu-Natal).
Participants ranged in age from 16 to 21 at the time of leaving care, with the majority (n=33) aged 17-19. None of the participants was disabled. Almost half the participants (n=15) had had no previous placements before GBT and a third (n=12) had had one; the remaining eight participants had had two to four previous placements. The majority of the participants had come into GBT’s care at ages 14-16 (n=27) and spent more than two years at GBT (n=27). Most of the young people had been in care due to behavioural problems (e.g. substance abuse, aggression, truancy, and theft); smaller numbers were in care due to neglect or abuse (n=6), previous placement breakdown (n=5), or being orphaned or abandoned (n=3).
Table 2 presents the results of the Wilcoxon signed-rank test, to indicate which of the continuous outcome variables differed significantly over the two interviews. In addition to the z score, the significance (p), sample size (number of matched participants) and the effect size (r) are provided. Lastly, the median outcome scores at one and two years are provided to enable a visual inspection of the results, although the Wilcoxon signed-rank test makes use of the mean ranks for its analysis.
Table 2. Wilcoxon signed-rank test for changes in continuous outcomes
| Continuous Outcome Variables | z | p | n | r | Mdn 1-year | Mdn 2-year |
| Accommodation | -.359 | .720 | 35 | -0.06 | 61.9 | 61.9 |
| Paid Employment | -.360 | .719 | 9 | -0.12 | 70.6 | 70.6 |
| Studying | -.368 | .713 | 6 | -0.15 | 76.7 | 76.7 |
| Financial Security | -1.002 | .316 | 35 | -0.17 | 60.0 | 53.3 |
| Drugs & Alcohol * | -.228 | .820 | 35 | -0.04 | 8.8 | 5.9 |
| Crime * | -2.002 | .045 | 35 | -0.34 | .0 | .0 |
| Health | -.640 | .522 | 35 | -0.11 | 82.1 | 78.6 |
| Well-being | -1.182 | .237 | 35 | -0.20 | 75.0 | 75.0 |
| Family Relationships | -.343 | .731 | 35 | -0.06 | 80.0 | 75.0 |
| Friends Relationships | -1.104 | .269 | 35 | -0.19 | 75.0 | 70.8 |
| Lover Relationships | -.843 | .399 | 10 | -0.27 | 87.5 | 92.5 |
* These outcomes are negatively scored – a high score indicates negative outcomes.
Only Crime changed statistically significantly (z = -2.0, n = 35, p < .05), with ranked scores increasing for 11 participants, decreasing for six and tying for 18. This indicates an increase over time in the extent to which care-leavers engaged in vandalism, theft and violence and had had trouble with the law since their last interview. However, it possible that this result emerged by chance. With significance set at .05, there is a 5% chance of a spurious statistically significant result. Nineteen statistical tests were conducted, suggesting it is likely that one significant result could occur by chance. Considering this, and given that the p statistic is barely less than .05, it is likely that this deterioration in the crime score is not significant.
A visual examination of the median scores at one and two years reveals only very small changes between the two times. Four scores deteriorated slightly, two improved slightly and five stayed the same. This confirms the results of the Wilcoxon tests, that there are no meaningful changes in the continuous independent living outcome variables between one and two years.
Table 3 presents the results of McNemar’s chi-square test, to indicate which of the dichotomous outcome variables differed significantly over the two interviews. The number (and percentage) of participants meeting the criteria at each time are also presented.
Table 3. McNemar’s chi-square test for changes in dichotomous outcomes
| Dichotomous Outcome Variables | p | n | One Year | Two Years |
| Self-supporting Accommodation | .388 | 35 | 18 (51%) | |
| Education for Employment | 1.000 | 35 | 19 (54%) | |
| NEET * | .344 | 35 | 9 (26%) | |
| Reliable Employment | ** | 9 | 6 (67%) | |
| Diligent Education | ** | 5 | 1 (20%) | |
| Liveable income | 1.000 | 35 | 7 (20%) | 7 (20%) |
| Drug & Alcohol ‘Free’ | 1.000 | 35 | One year | 30 (86%) |
| Crime ‘Free’ | .508 | 35 | 14 (40%) | 24 (69%) |
| 19 (54%) | ||||
| 13 (37%) | ||||
| 5 (56%) | ||||
| 2 (40%) |
* This outcome is negatively scored – a high score indicates negative outcomes.
** Too few participants were employed or studying at both one and two years to run the chi-square tests on these outcomes.
While there are some observed changes in the dichotomous variables over time, none of them reached statistical significance. In the case of NEET, for example, there is an improvement in the NEET rate from 37% at one year to 26% at two years. This represents a change of four individuals moving from NEET to not-NEET, which was insufficient to reach statistical significance. A visual inspection of the frequencies shows improvements of no more than four participants for any outcome, accounting for a change in about one tenth of the participants. Improvements were noted in three of the measures, deterioration in three and no change in two.
Limitations
A primary limitation of this study is the small sample size, which restricts the power of the statistical analysis. For this reason, a visual inspection of the data was also undertaken, which appears to confirm the statistical findings. The small sample size also limits the kinds of statistical analyses that can be performed on the data, which in turn limits the kinds of research questions that can be answered.
Due to studying the whole population and given the reasonable retention rate of 60% and the absence of demographic differences between those who continued in the study and those who dropped out, the data can be regarded as representative of GBT care-leavers. Nevertheless, the findings cannot be generalised with confidence beyond GBT, because the sample represents only GBT care-leavers. Given that care-leavers from different organizations in South Africa transition into the same socioeconomic contexts, however, it can be argued that these results might be similar for care-leavers from other organisations. While some organisations provide more expansive after- care services than GBT, it appears that very few of these are substantial programmes, which may result in more similarities than differences across programmes in South Africa.
The study is further limited by the predominance of male participants. The Midwest study in the USA (Courtney et al., 2011) shows quite marked gender differences, with males evidencing generally poorer outcomes than females. If this holds true in South Africa, and there is as yet no data to investigate gender differences in care-leaving outcomes in South Africa, this suggests that these South African results may be more negative than they would be if the sample had a better gender balance.
Discussion
The results did not support our hunch, viz. that there would be a steady, even if not statistically significant, improvement in outcomes between one- and two-years out of care. There were no statistically significant improvements. A visual analysis of the data suggested a fairly even spread of very slight improvements and deteriorations, equally balanced by no changes at all. There is, in short, no evidence to support the view that the outcomes of this sample of care-leavers improve, even marginally, over the first two years out of care.
This absence of improvements has, however, important implications for understanding the care- leaving journey of these participants. It suggests that the degree of independent living achieved by youth one year out of care (whether positive or negative) appears likely to remain unchanged at two years out of care. Only a small proportion of participants evidenced changes (both improvements and deteriorations) in independent living outcomes between one and two years out of care, and when aggregated, the changes were largely cancelled out. The majority of participants showed stable outcomes over the first two years out of care.
This implies a critical window of opportunity to set a care-leaver on a positive path towards young adulthood around the time of transitioning out of care. If, by the end of the first year out of care, a care-leaver has, for example, enrolled in education or started work, the chances are s/he will continue in education or work a year later. Conversely, if s/he finds him/herself NEET or in conflict with the law by the end of their first year out of care, the chances are s/he will continue to be NEET or in conflict with the law a year later. The first year out of care appears to set a course for the first two years out of care, and potentially beyond.
Some similar evidence is generated by the other longitudinal studies. For example, Cashmore et al. (2007) found that almost all Australian care-leavers who left care with a completed secondary education continued their education later, while only half of those who left care without completing secondary education continued their education later. McMillen and Raghavan’s (2009) research shows a dramatic drop in mental health service use over the period of one month before to one month after disengagement, but fairly stable service use both before and especially after disengagement. The Midwest study data (Courtney et al., 2011) suggests that age 21 is the year at which the path is fairly set, as outcomes vary considerably from 17 to 19 to 21 years, but after that the changes are comparatively small. The interpretation of the Midwest data is complicated, however, because the data points are linked to age, not years out of care. Nevertheless, there are hints in these international longitudinal studies that confirm that future pathways are, to some extent, established by initial pathways.
Recommendations and conclusion
Further longitudinal research on care-leaving is required, not only in South Africa but elsewhere in the world too. Many researchers have stated this, and for good reason. We need greater insight into the detail of the pathway out of care. In particular, we need to see where the critical moments are for intervention. This data, small and limited as it is, suggests that the critical moment is in the transition out of care, from shortly before leaving care to shortly thereafter. It is at this point that there is the greatest opportunity to structure the young person’s care-leaving arrangements. If the path over the first two years is established in the first months after leaving care, it is vital that child welfare service providers ensure that young people leave care on the best foot possible, so that they are on the path towards better transitional outcomes.
This has important implications for child welfare practice. Preparation of young people for leaving care is a subject that has been well documented (Freundlich & Avery, 2006) and is widely recognised as a critical requirement. A care-leaving plan, formulated collaboratively with the young person, needs to specify education, employment and accommodation for at least the first year out of care. Sulimani-Aidan (2015) has shown that the future expectations of care-leavers at the time of leaving care predict better post-care outcomes in terms of accommodation, economics and education. Thus, preparation for care-leaving is a crucial task, and needs to be detailed and specific. For example, a plan for a young person to take up employment should include visits to the place of employment and meetings with the employer prior to the young person’s disengagement from care.
Moreover, care-leavers need aftercare support during the first several months of being out of care to actualise their plan. Merely having a care-leaving plan is insufficient – the participants in the GBT study all had a care-leaving plan at the time of disengagement. But the absence of policy and funding for aftercare services means these plans may fall into disuse soon after leaving care. If left to do it alone, many young people may be unable to translate the plan into reality, setting them on a pathway to negative transitional outcomes. Services during the critical period of the first several months out of care thus appear to be particularly important.
This study suggests that the transitional phase is crucial for care-leavers and that practitioners who provide intensified services to care-leavers to establish positive transitional outcomes soon after leaving care may be setting them up for better independent living outcomes in the years to come.
Acknowledgements
The author thanks Dr Lisa Dickens, Ms Lee Loynes and Mr Peter Marx at Girls and Boys Town for their support on this research, as well as the fieldworkers who assisted with data collection, and the youth who participated in the research. In addition, the author thanks the reviewers for the valuable comments on an earlier draft of this manuscript. The author thanks the Anglo American Chairman’s Fund Trust for financial support for the Growth Beyond the Town study. This work is based on research supported in part by the National Research Foundation of South Africa for the Grant No. 93634. Any opinion, finding and conclusion or recommendation expressed in this material is that of the author and the NRF does not accept any liability in this regard.
References
Cashmore, J., & Paxman, M. (2007). Longitudinal study of wards leaving care: Four to five years on. Sydney: Social Policy Research Centre.
Cashmore, J., Paxman, M., & Townsend, M. (2007). The educational outcomes of young people 4-5 years after leaving care: An Australian perspective. Adoption & Fostering, 31(1), 50-61.
Corder, G. W., & Foreman, D. I. (2009). Nonparametric statistics: A step-by-step approach. Hoboken, NJ: John Wiley & Sons.
Courtney, M., Dworsky, A., Brown, A., Cary, C., Love, K., & Vorhies, V. (2011). Midwest evaluation of the adult functioning of former foster youth: Outcomes at age 26. Chicago, USA: Chapin Hall at the University of Chicago.
Courtney, M. E., & Dworsky, A. (2006). Early outcomes for young adults transitioning from out-of- home care in the USA. Child & Family Social Work, 11(3), 209-219. doi: 10.1111/j.1365- 2206.2006.00433.x
Dickens, L. F. (2016). The contribution of resilience to the 12-month transitional outcomes of care- leavers in South Africa. Doctoral Thesis, University of Johannesburg.
Dinisman, T. (2016). Life satisfaction in the transition from care to adulthood: The contribution of readiness to leave care and social support. Child & Family Social Work, 21(4), 401-411. doi: 10.1111/cfs.12156
Elliott, A. C., & Woodward, W. A. (2016). IBM SPSS by example: A practical guide to statistical data analysis (2nd ed.). Thousand Oaks, CA: Sage.
Freundlich, M., & Avery, R. J. (2006). Transitioning from congregate care: Preparation and outcomes. Journal of Child & Family Studies, 15(4), 503-514. doi: 10.1007/s10826-006- 9023-3
Jamieson, L. (2017). Children and young people’s right to participate in residential care in South Africa. The International Journal of Human Rights, 21(1), 89-102. doi: 10.1080/13642987.2016.1248126
Jones, L. P. (2011). The first three years after foster care: A longitudinal look at the adaptation of 16 youth to emerging adulthood. Children and Youth Services Review, 33(10), 1919-1929. doi: 10.1016/j.childyouth.2011.05.018
Lee, J. S., Courtney, M. E., Harachi, T. W., & Tajima, E. A. (2015). Labeling and the effect of adolescent legal system involvement on adult outcomes for foster youth aging out of care. American Journal of Orthopsychiatry, 85(5), 441-451. doi: 10.1037/ort0000090
McCoy, H., McMillen, J. C., & Spitznagel, E. L. (2008). Older youth leaving the foster care system: Who, what, when, where, and why? Children and Youth Services Review, 30(7), 735-745. doi: 10.1016/j.childyouth.2007.12.003
McMillen, J. C., & Raghavan, R. (2009). Pediatric to adult mental health service use of young people leaving the foster care system. Journal of Adolescent Health, 44(1), 7-13. doi: 10.1016/j.jadohealth.2008.04.015
Mendes, P., Johnson, G., & Moslehuddin, B. (2011). Young people leaving state out-of-home care. Melbourne, Australia: Australian Scholarly Publishing.
Munson, M. R., & McMillen, J. C. (2009). Natural mentoring and psychosocial outcomes among older youth transitioning from foster care. Children and Youth Services Review, 31(1), 104- 111.
Narendorf, S. C., & McMillen, J. C. (2010). Substance use and substance use disorders as foster youth transition to adulthood. Children and Youth Services Review, 32(1), 113-119. doi: 10.1016/j.childyouth.2009.07.021
National Audit Office. (2015). Care leavers’ transition to adulthood. London, UK: National Audit Office.
Pett, M. A. (2016). Nonparametric statistics for health care research: Statistics for small samples and unusual distributions (2nd ed.). Thousand Oaks, CA: Sage.
RSA. (2005). Children's Act, No 38 of 2005, amended 2007. Pretoria, South Africa: Government Printers.
Stein, M. (2008). Young people leaving care. Child & Family Social Work, 11(3), 273-279. Sulimani-Aidan, Y. (2015). Do they get what they expect? The connection between young adults' future expectations before leaving care and outcomes after leaving care. Children and Youth Services Review, 55, 193-200.
Tanur, C. (2012). Project Lungisela: Supporting young people leaving state care in South Africa. Child Care in Practice, 18(4), 325-340.
Van Breda, A. D. (2015). Journey towards independent living: A grounded theory investigation of leaving the care of Girls & Boys Town South Africa. Journal of Youth Studies, 18(3), 322- 337.
Van Breda, A. D., & Dickens, L. F. (2016). Young people transitioning from residential care in South Africa: Welfare contexts, resilience, research and practice. In P. Mendes & P. Snow (Eds.), Young people transitioning from care: International research, policy and practice (pp. 349-366). London: Palgrave.
From: Child Welfare, 2017