Resilience Factors in Youth Transitioning Out of Foster Care: A Systematic Review

Abstract

Objective:

Youth who emancipate from the foster care system often experience poor outcomes during their transition into independent living (e.g., criminal justice involvement and homelessness). Yet, some youth are resilient and achieve positive outcomes. The purpose of this study is to review the resilience factors identified in the literature for youth transitioning out of care.

Method:

A systematic review of the literature was conducted using PsycINFO, PubMed, and EMBASE databases. A total of 12 studies met our inclusion criteria.

Results:

A total of 38 different resilience factors were identified across the studies reviewed. From these resilience factors, 18 were statistically significant, including four assets and 14 resources.

Conclusion:

Understanding the assets and resources that promote resilience can aid clinicians in assessing and capitalizing on youths’ strengths and can help researchers develop effective interventions and target gaps in the literature. Further, it can assist policymakers in generating legislation to improve outcomes for foster youth transitioning out of care. To assist in this process, we systematically reviewed studies to provide a summary of resilience factors currently identified in the literature.

Roughly 40,000 youth, ages 16 to 20, transition out of foster care every year across the country (U.S. Department of Health and Human Services [DHHS], 2017). Foster youth experience extraordinary adversity because of the factors that led them into foster care (e.g., physical abuse, abandonment), as well as processes within the child protection system (e.g., multiple placements). Unfortunately, as older youth transition out of foster care and embark on independent living, they encounter additional hardships (e.g., homelessness; Pecora et al., 2006) and are often unable to achieve similarly to their same-age peers (e.g., educational attainment; Courtney & Hook, 2017).¹ Yet, some foster alumni are able to overcome these challenges and achieve outcomes comparable to the general population (Yates & Grey, 2012). Despite their adversities, these resilient youth capitalize on assets and resources to attain positive outcomes. The aim of this systematic review is to compile the resilience factors that promote positive adaptation in youth transitioning out of foster care.

Experiences and Outcomes of Foster Youth Transitioning Out of Care

Over 250,000 children enter the foster care system every year due to adverse experiences, ranging from neglect and abandonment to physical and sexual abuse (DHHS, 2017). The state recognizes the importance of protecting youth from egregious circumstances. As such, the foster care system is meant to provide a stable and secure environment. Despite this, about a third of youth report mal-treatment during care (Pecora et al., 2006) and a majority experience significant placement instability as wards of the state (Cusick et al., 2010). The impact of their experiences is captured in their outcomes. For example, foster youth have higher rates of mental illness and substance use disorders compared to the general population (Braciszewski & Stout, 2012; Pecora et al., 2006). On top of that, the foster care system often fails to meet their mental health needs (Landsverk et al., 2006). Although all the aforementioned would be difficult for any child or adolescent to experience, foster youth are further expected to achieve complete independence between the ages of 18 and 21. As such, transitioning out of care marks a critical developmental period of time in which youth attempt to become self-sufficient. Compared to their foster youth peers still in care, those transitioning into independent living often have either limited or a lack of services provided by the state. In other words, not only is this period critical for their long-term success but it is also one of the most challenging. Not surprisingly, they feel unprepared for their transition into independent living (Courtney et al., 2001), and the prevalence of negative outcomes during their transition underscores their claim.

Foster alumni demonstrate poor outcomes across a variety of life domains. For example, despite foster youth having high educational aspirations (Barth, 1990; McMillen et al., 2003), the estimates for college graduation are as low as 2% (Casey Family Programs, 2010). Foster alumni also struggle with financial insecurity and employment opportunities during their transition to independent living (Pecora et al., 2006), tending to have lower earnings than their peers in the general population (Courtney & Dworsky, 2006; Okpych & Courtney, 2014). Likely, this impacts their housing situation as nearly a third of foster youth experience homelessness within 12 months of exiting care (Shah et al., 2017). By the time foster alumni reach the age of 26, between 31% and 46% of them will have experienced homelessness (Dworsky et al., 2013). Criminal behaviors are also higher among foster youth than the general population. Although foster youths’ criminal behavior appears to peak during late adolescence and decline into adulthood (similar to the general population), their rates of offending remain higher than their nonfoster youth peers (Cusick & Courtney, 2007).

Resilience

Despite the numerous negative outcomes among foster alumni, there are some youth that perform on par, if not better, than their nonfoster youth peers. Resilience refers broadly to the dynamic process by which individuals achieve successful adaption (or maintain healthy functioning) following exposure to significant adversity. While there is a lack of consistency in the definition and measurement of resilience (Luthar et al., 2000), a key requirement for the resilience process to take place is the presence of factors that increase risk and factors that promote adaptation, with the latter being the focus of the present review.

Resilience Factors

The literature has used different terms to describe intra- and interpersonal characteristics of an individual that promote adaptation. These terms are referred to as protective factors, promotive factors, positive factors, and strengths (Fergus & Zimmerman, 2005; Fletcher & Sarkar, 2013; Grych et al., 2015; Luthar et al., 2000). While the terms tend to have a specific meaning and origin in their niche field (Fletcher & Sarkar, 2013), researchers have often used the terms interchangeably. In order to avoid the complexity of this terminology across studies reviewed and potential misclassification, we refer to these characteristics under the umbrella of resilience factors. Consistent with previous literature (e.g., Fergus & Zimmerman, 2005; Grych et al., 2015) resilience factors are conceptualized across two categories: assets and resources. Assets refer to characteristics of an individual that promote adaptation (e.g., emotion regulation), whereas resources refer to external sources of support that promote adaptation (e.g., social support).

Delineation of Resilience Factors

Consistent with recommendations from the literature, we conceptualized resilience factors to exclude static variables (e.g., ethnicity; Hamby et al., 2018; Luthar et al., 2000). This is because such factors are not amenable to interventions and thus have limited clinical implications (Grych et al., 2015). Further, previous studies on foster youth have tested variables as both predictors and outcomes (e.g., educational attainment; Courtney & Hook, 2017; Okpych & Courtney, 2017). Because resilience factors are meant to be mechanisms in the process of attaining positive outcomes, we excluded variables widely used as outcomes from our conceptualization of resilience factors. For example, variables like educational attainment are not considered resilience factors but variables like educational aspirations or academic performance, that have traditionally been used as predictors in resilience research in foster youth, meet criteria. Another example is reading ability, which has been used as a predictor (e.g., Courtney & Hook, 2017) and an outcome (e.g., Shin, 2003). In this case, reading ability is conceptualized as an asset. Studies that use reading ability as an outcome are not discussed in the present review.

Another issue arises with factors considered protective but that actually represent the opposite effect of an established risk factor (Grych et al., 2015; Hamby et al., 2018). Despite placement stability representing the opposite of a risk factor (i.e., placement instability), we conceptualized it as a resilience factor. This is because the foster care system presents unique challenges for youth to develop close relationships with their parents, foster caregivers, as well as other adults (Storer et al., 2014). Further, the majority of foster youth experience placement instability (Courtney et al., 2004; DHHS, 2017), making it the norm in the population. In scenarios where studies measured the total number of placements youth experienced but tested placement instability (e.g., Ahrens et al., 2013; Crawford et al., 2018), we made inferences about the protective effects of placement stability based on their results. Similarly, school stability was conceptualized as a resilience factor. The present review also conceptualized kinship care, but no other types of placements, as a resilience factor given that policies tend to favor relative care for youth.

Purpose of Review

The purpose of this review is to be the first to compile the assets and resources among foster youth transitioning out of care. To our knowledge, there is only one previous systematic review that discusses resilience factors in foster youth, including foster alumni (Braciszewski & Stout, 2012). However, this study focused exclusively on substance use. Our review expands the literature by focusing exclusively on resilience factors in foster youth transitioning out of care, including the areas of mental health, physical health, substance use, education, employment, and criminal justice involvement. The review will provide a better understanding of the resilience factors that promote successful adaptation despite the experience of significant adversity in order to inform clinical care and legislation aimed at helping this population achieve similar outcomes as their nonfoster youth peers. Given our conceptualization of resilience, we opted to focus on studies that conducted multivariate analyses. Subsequently, a systematic review was more appropriate than a meta-analysis as our ability to compare analyses across studies was limited.

Method

Inclusion and Exclusion Criteria

Inclusion criteria for this systematic review were as follows: (a) empirical studies (not including pilot studies) published in a peer-reviewed journal (b) originally written in English that (c) included foster youth within the ages of 16 to 26 (e.g., a study including ages 19 to 22 met the inclusion criterion but studies that included ages 14 to 20 did not), (d) measured at least one resilience factor quantitatively, and (e) included an outcome measure. Articles were excluded if (a) the study did not present original results (e.g., review articles), (b) the data was not collected in the United States, (c) the sample included a mixture of foster youth and nonfoster youth that only reported results in aggregate form, (d) the study reported results separately based on a factor (e.g., gender) but did not report the aggregate results for the whole sample, (e) the study sample consisted of a specific subset of the foster youth population (e.g., kinship-care only, a specific race/ethnicity, foster youth with a disability), (f) the study only conducted bivariate analyses, (g) the study measured resilience through a resilience composite score, and (h) the study utilized person-centered approaches to examine resilience. Notably, research using person-centered approaches and resilience composite scores provide a unique and important contribution to the field of resilience; however, they are aimed at distinct types of questions that fall outside the purview of this systematic review (Luthar et al., 2000).

Search Strategy

PsycINFO (k = 431), PubMed (k = 2,527), and EMBASE (k = 319) databases were systematically searched in March 2020 using the search terms (foster youth, foster child*, foster teen, foster adolescent, foster care, foster system, foster home, foster family), and (resilience, resiliency, resilient, protective factor, promotive factor, strength). Filters were applied to the searches (if available) including (a) peer-reviewed, (b) journal article, and (c) English language. Given the limited amount of research focusing on resilience in foster youth transitioning out of care, we did not set time limits on our searches. After removing duplicates, articles were screened by title and abstract for relevance by the first author. In cases when there was doubt about whether a study met criteria for inclusion, the matter was discussed with the second and third author to ensure agreement. If relevant, then their full text was reviewed for compliance with the inclusion/exclusion criteria. The reference list of studies that met eligibility were manually searched for additional relevant studies. The first author was responsible for extracting all relevant variables, including sample sizes, study design, measures, outcome variables, multivariate analyses, and results that included a resilience factor.

Study Selection

The database search yielded 3,277 studies with an additional 14 studies identified through the manual search of reference lists. Following the removal of duplicates, 2,659 articles were screened for relevance yielding a total of 188 articles that were assessed through full text. From these, a total of 12 articles met the inclusion criteria (see Figure 1; Moher et al., 2009).

Figure 1.

PRISMA Flow Diagram

Results

Sample characteristics for the included studies can be found in Table 1. For brevity, only statistically significant results are discussed. However, in situations where mixed findings are present, both statistically significant and nonsignificant findings are discussed. A list of the resilience factors and outcomes from each study, including all statistically significant and nonsignificant results, can be found in Table 2. Results reported by authors as trending toward significant (e.g., “marginally,” “borderline”) were considered nonsignificant. The present review does not discuss risk factors; however, instances are reported were a factor typically considered protective was found to increase the likelihood of a negative outcome. Results are reported by domain, including (a) mental health, (b) physical health, (c) substance use, (d) education, (e) employment, and (f) criminal justice involvement.

Table 1.

Study Sample Descriptions

Study	N	Age range	Location
Ahrens et al. (2013)	732	17–26	Illinois, Iowa, and Wisconsin
Courtney & Hook (2017)	732	17–26	Illinois, Iowa, and Wisconsin
Crawford et al. (2018) a	1,420	18	Oklahoma
Dworsky & Courtney (2009)	603	17–19	Illinois, Iowa, and Wisconsin
Hook & Courtney (2011)	512–732b	17–24	Illinois, Iowa, and Wisconsin
Katz et al. (2017)	579	17–25	Illinois, Iowa, and Wisconsin
Kennedy et al. (2019)	33	18–26	Florida
Munson & McMillen (2009)	406	17–19	Missouri
Okpych & Courtney (2017)	712	19–17	California
Prince et al. (2019) c	7,449	17–19	Nationwide
Shah et al. (2017)	1,202	17–21	Washington
Tyrell & Yates (2017)	172	18–21	California

^aAge range and location of study were not reported in the article. Based on the authors’ institutions at the time of publication, it is assumed that the study took place in Oklahoma. Furthermore, because Oklahoma did not offer extended care to foster youth at the time of publication, it is assumed that youth were 18 years of age.

^bTwo different sample sizes were used in multivariate analyses based on the outcome.

^cThe study included data from all U.S. states and the District of Columbia, except New York, Connecticut, and Puerto Rico.

Table 2.

Resilience Factors (Assets, Resources) and Outcomes Included in Review

Study	Resilience factors		Outcome
	Asset	Resource
Ahrens et al. (2013)		Kinship care	Depressiona
		Close relationship with caregiver1	Posttraumatic stressa
		Placement stability	Substance usea
			Delinquent behavior1a
			Anxious attachmenta
			Avoidant attachment1a
		Extended care3b	Five or more sexual partners
			Inconsistent condom use3
			Sexual acts for money
Courtney & Hook (2017)	Educational aspirations1	Extended care3	Educational attainment1,2,3
	Reading ability2	Placement stability
		Kinship care
		Educational support services
		Services receiveda
		Employment
		Financial Literacy
		Health
		Housing
Crawford et al. (2018)		Kinship care	Adjudicated felony1
		Services received
		Public aid
		ETV
		Transition plan
		Systems of care case approach
		Mental health/substance use
		Placement stability1
Dworsky & Courtney (2009)	Perceived social support	Placement stability	Homelessness1
		Kinship care
		Close relationship with caregiver
		Close relationship with family member1
		Housing services (prior to baseline)
Hook & Courtney (2011)	Reading ability1	Kinship care	Employment1
		Placement stability2	Hourly wages2
		Extended care
Katz et al. (2017)		Placement stability1	No relationship
		Close relationship with family member	Nonviolent relationship
			Violent relationship
			Victim
			Perpetrator
			Bidirectional1
Kennedy et al. (2019)	Emotional intelligence1		Quality of life1
			Psychological distress1
Munson & McMillen (2009)		Presence of natural mentor1	Perceived stress1
			Life satisfaction1
		Mentoringb	Depressive symptoms2
		Duration2	Perceived stress
		Frequency	Life satisfaction
		Relationship quality
		Mentoring (over time)b	Depressive symptoms
		Short-term	Perceived stress3
		Long-term3	Employment
			Alcohol use
			Marijuana use
			Arrest history3
Okpych & Courtney (2017)	Reading ability	Institutional agents	College enrollment1,2,3,4,5,6,7
	Academic performance1	Tangible support/advice3
	Educational aspirations2	School encouragement4
	Educational preparedness	Network size
		Educational services received
		School stability
		Kinship care5
		Help with college preparation6a
		Extended care7a
Prince et al. (2019)		Close relationship with adult1	Homelessness1,3,4
		Placement stability2	Incarceration3
		Kinship care	Substance abuse referral
		Extended care3	Childbirth2
		Number of financial services received
		Number of wellbeing services received
		State CFCIP spending4
Shah et al. (2017)	Academic performance1	Placement stability	Homelessness1,2,3
		Kinship care2
		School stability3
Tyrell & Yates (2017)		Extended care1	Housing quality1
		Kinship care

Note. Significant relations between resilience factors and outcomes are marked by matching number superscripts for each of the studies.

^aThe resilience factor or outcome was analyzed in the respective study as a mediator.

^bResilience factors had different outcomes from other resilience factors in the study (outcomes are repeated for readability).

Mental Health

Assets

There was a paucity of research on assets that protect mental health, with only one study meeting criteria. Kennedy et al. (2019) showed that emotional intelligence was predictive of both quality of life and psychological distress while controlling for general intelligence (Kennedy et al., 2019).

Resources

Two studies found social relationships were a resource for foster youth, specifically mentoring relationships, and a close relationship with a foster caregiver. The presence of a mentor was associated with lower levels of stress and higher life satisfaction over and above demographic characteristics including race/ethnic group and gender (Munson & McMillen, 2009). While also controlling for the same demographic characteristics, the duration of the relationship had a negative association with depression symptoms and long-term mentoring was associated with lower levels of perceived stress (Munson & McMillen, 2009). Other research has also shown that having a close relationship with a foster caregiver can protect against an avoidant attachment style (Ahrens et al., 2013).

Physical Health

Assets

The systematic literature search yielded no studies that identified assets protective of physical health among youth transitioning out of foster care.

Resources

Foster care characteristics, specifically placement stability and extended care, protected foster youth against poor physical health outcomes. Research has shown that placement stability can protect against bidirectional violent relationships (Katz et al., 2017) and becoming a parent at age 19 (Prince et al., 2019).² Participation in extended care also protected against inconsistent condom use (Ahrens et al., 2013). In other words, participation in extended care seemed to protect individuals from some risky sexual behaviors. One study found that being very close to an adult relative did not protect against one-directional violent relationships; but rather, increased the likelihood of a bidirectionally violent relationship (Katz et al., 2017).

Substance Use

Assets

There was a complete lack of research on assets that protect foster youth transitioning out of care against substance use.

Resources

Despite existent research studies on this topic, researchers have yet to identify a resource that is protective against substance use. On the other hand, there were several statistically nonsignificant findings (see Table 2). Notably, one study found that extended care increased the likelihood of a substance abuse referral (Prince et al., 2019). However, this finding may actually represent the support available from extended care to address substance use.

Education

Assets

Youths’ educational aspirations increased the likelihood of college enrollment (Courtney & Hook, 2017; Okpych & Courtney, 2017). Notably, Okpych and Courtney (2017) found that youth must desire more than a college degree (e.g., not just some college) for the likelihood of college enrollment to increase. They also found that academic performance predicted college enrollment. However, academic performance did not have a straightforward effect. They compared students with mostly As to those with mostly Cs and found those with higher grades were more likely to enroll in college. However, the relation was no longer present when they compared students with mostly As to students with Ds and lower. Reading ability was also a statistically significant predictor of educational attainment (Courtney & Hook, 2017).

Resources

Two studies found extended care to promote positive educational outcomes. However, one should interpret these findings with caution as one of the requirements for participating in extended care is school attendance. Notwithstanding, youth who participated in extended care had a higher chance of college enrollment (Okpych & Courtney, 2017), as well as reaching the next level of education (Courtney & Hook, 2017). Kinship care increased the likelihood of youth enrolling in college (Okpych & Courtney, 2017), but ultimately had no effect on their educational attainment (Courtney & Hook, 2017). The number of institutional agents available for tangible support and advice, along with the amount of school encouragement, increased the likelihood of college enrollment (Okpych & Courtney, 2017). However, these associations were mediated by participation in extended care and the amount of help that youth received with college preparation. Notably, school encouragement was no longer significant after help with college preparation was added as a mediator (Okpych & Courtney, 2017).

Employment

Assets

The limited research regarding assets and employment demonstrated that reading ability can promote employment at age 24; however, it does not affect hourly wage (Hook & Courtney, 2011).

Resources

Research has shown that placement stability can predict hourly wage but not employment (Hook & Courtney, 2011). Contrary to the conceptualization of relative care as promotive to positive outcomes, one study found that traditional foster homes promoted employment compared to kinship care (Hook & Courtney, 2011).

Housing

Assets

We found one study assessing an asset in relation to housing. Using a stepwise procedure for analysis, one study found that high academic performance protects against homelessness (Shah et al., 2017).

Resources

There were mixed findings related to resources and housing outcomes. A consistent finding was extended care, which protected against homelessness (Prince et al., 2019) and a decline in housing quality (Tyrell & Yates, 2017). Notably, prioritizing funds for housing is part of the policy in extended care, so these results should be interpreted with caution. Youth who lived in a state that spent higher than average of their John H. Chafee Foster Care Independence Act (CFCIP) allotments on housing support were less likely to become homeless (Prince et al., 2019).³ Regarding less consistent findings, having a close connection to a caring adult or family member protected against homelessness (Dworsky & Courtney, 2009; Prince et al., 2019). However, perceived social support and a close relationship with a foster caregiver had no protective effect on homelessness (Dworsky & Courtney, 2009). Shah et al. (2017) found that kinship care was protective against homelessness. However, other studies found not only a lack of effect on homelessness (Dworsky & Courtney, 2009; Prince et al., 2019), but also an increase in the likelihood of experiencing housing quality decline over time (Tyrell & Yates, 2017). Whereas school stability was found to protect against homelessness (Shah et al., 2017), placement stability did not have a statistically significant effect on housing (Dworsky & Courtney, 2009; Prince et al., 2019; Shah et al., 2017).

Criminal Justice Involvement

Assets

Similar to other areas, there was a paucity of research on assets that protect youth transitioning out of care against criminal justice involvement.

Resources

State action and support appear to be key resources against criminal justice involvement for youth transitioning out of care. Extended care and state CFCIP spending on housing support were negatively associated with foster youth experiencing incarceration (Prince et al., 2019). However, these findings should be interpreted with caution as felony convictions and current incarcerations often make youth ineligible for federal extended care programs. The role of social support was mixed. Some researchers found that being close to a foster caregiver (Ahrens et al., 2013) and having long-term mentoring relationships (Munson & McMillen, 2009) reduced the likelihood of delinquent behaviors and arrest. However, other studies found that neither a connection to a caring adult (Prince et al., 2019) nor a short-term mentorship relationship (Munson & McMillen, 2009) protected against arrest. Similarly, some research showed that placement stability decreases the likelihood of an adjudicated felony (Crawford et al., 2018), but other research found that it does not protect against incarceration (Prince et al., 2019) or delinquent behavior (Ahrens et al., 2013). Notably, emancipating with a history of mental health or substance abuse services seemed to increase the likelihood of a future adjudicated felony (Crawford et al., 2018).

Discussion

Foster youth experience numerous adversities throughout life. They are admitted as wards of the state due to experiencing egregious childhood adversity. During their time in care, they are faced with additional challenges such as placement instability and additional abuse (Pecora et al., 2006). When they reach adulthood, they are expected to transition into independent living without supports typically present for nonfoster youth. Unsurprisingly, their experiences place them at increased risk for multiple negative outcomes during their transition into independent living (Courtney et al., 2001). However, some foster youth capitalize on resilience factors to attain positive outcomes. The study of resilience in youth transitioning out of care is essential to understanding the mechanisms that help them overcome these challenges and helping clinicians and researchers promote positive outcomes in this population. No previous study has reviewed the factors which promote resilience in youth transitioning out of care. As such, the purpose of the current review was to gain an understanding of the assets and resources that promote successful adaptation subsequent to the experience of adversity.

Based on the results from this systematic search, the current literature describes a total of 65 resilience factors, including 10 assets and 55 resources (see Table 2). Accounting for resilience factors analyzed across multiple studies, we identified 38 unique resilience factors in the current body of research. A total of 18 out of the 38 resilience factors were statistically significant, including four assets and 14 resources. The assets include educational aspirations, reading ability, emotional intelligence, and academic performance. The resources include kinship care, extended care, placement stability, a close relationship with a caregiver, family member or adult, presence of a natural mentor (especially a long-term mentorship and a prior history of relationship with the mentor), institutional agents available for tangible support and/or advice, receiving school encouragement, school stability, receiving help with college preparation, and residing in a state with above average CFCIP spending on housing support.

Limitations of the Systematic Review

One of the key limitations for the present study involves terminology. Whereas efforts were made to capture all available research on the topic, both the areas of resilience and research on the foster youth population use varying terminology. For example, foster youth, looked after children, wards of the state, and children in care, are some of the terms used in reference to children in out-of-home care. Capturing all possible terminology was simply unfeasible for the present review. Relatedly, research studies that included a variable as a predictor (e.g., placement stability, extended care) of positive outcomes, without describing its role as protective or related to resilience, are unlikely to have been identified in the systematic search. Further, we acknowledge the possibility that the studies included may have a risk of bias that we did not account for given our lack of risk bias analysis. The present review is also limited by its exclusion of qualitative studies. Therefore, it is possible that some of the research gaps (e.g., lack of significant resilience factors against substance use) may have been filled through the inclusion of the qualitative literature.

Clinical and Policy Implications

While some researchers have recommended integrating measures of resilience into clinical practice (Traub & Boynton-Jarrett, 2017), there is currently no agreed upon method of assessing resilience. The difficulty arises from resilience being context specific (Luthar et al., 2000), which means that a one-size-fits-all measure of resilience is often not applicable across different groups. Instead, the assessment of resilience should examine assets and resources empirically found to promote resilience in foster youth. In the case of youth transitioning out of care, that would mean assessing the presence of the 18 resilience factors identified in the literature.

Clinicians should explore the intersectionality of foster youth to identify additional resilience factors. For example, research has identified resilience factors across different subsets of the foster youth population, including ethnic minority youth, (e.g., Ashley & Brown, 2015), youth with a physical disability (e.g., Lee et al., 2018), and youth who are parents (e.g., Hook & Courtney, 2013). By combining resilience factors in both broad and more narrow foster youth populations, clinicians will gain greater understanding of the assets and resources available to promote resilience. Clinicians should also address risks, as this differentiates resilience-based interventions from purely strengths-based approaches (Thomas et al., 2005). Examples of risks in foster youth include type of abuse experienced (Ahrens et al., 2013; Dworsky et al., 2013), placement in special education (Okpych & Courtney, 2017), behavioral problems in childhood (Prince et al., 2019), and age of entry into foster care (Tyrell & Yates, 2017).

Moving from the assessment of resilience to interventions designed to enhance it, therapists could help capitalize on strengths already present and help develop new strengths shown to lessen the negative impact of adversity. In the case of youth transitioning out of care, that would mean promoting the 18 resilience factors identified in the literature. The promotion of the resilience factors should include an innovative, trauma-informed, and eclectic therapeutic approach, to maintain engagement with foster youth transitioning out of care. Innovative approaches to interventions that appeal to younger generations should be explored. For example, relationships with natural mentors and institutional agents can be promoted through the use of technology. Some researchers have already started exploring how the use of smartphones can help with relationship building (Denby et al., 2016).

While some of the identified resilience factors can be enhanced with clinical interventions, others require policy-level changes. Extended care is the most prominent resilience factor requiring policy-level change. In order for youth to even have the option of remaining in care past the age of 18, legislators must first make extended care possible in every state. Foster youth are children of the state and the system often fails them by abruptly discontinuing support services as soon as they reach the age of majority. By providing youth with support through extended care, they may be more likely to attain higher educational outcomes, as well as decrease risky sexual behaviors, criminal justice system involvement and homelessness (Ahrens et al., 2013; Okpych & Courtney, 2017; Prince et al., 2019). Other studies found kinship care and placement stability to be protective. Capitalizing on these findings means continuing efforts to minimize disruptions in care, as well as promoting kinship care for children while supporting the family members providing the care. Last, one study found that state spending on housing support from their CFCIP allotment is related to housing outcomes. States should review their current CFCIP allocation to support youth transitioning out of care.

Enhancing other resilience factors requires a combined effort from multiple parties. For example, five resilience factors were found to occur within the academic context. Some of these require the help of school staff, clinicians, and legislators to be optimally developed. For example, clinicians can help youth explore educational aspirations and school staff can provide academic support to improve school performance and reading ability. These resilience factors are not possible without policy that supports these assets and resources through funding. Through the aforementioned clinical and policy implications, youth transitioning out of care will be better equipped to cope with adversity and successfully transition into independent living.

Future Research Directions

Three primary gaps in the literature were identified based on the findings. The first is a lack of statistically significant resilience factors identified against substance use. The present review is not the first to find a lack of research in this area. Eight years ago, a systematic review on substance use was also unable to identify resilience factors for foster care alumni (i.e., Braciszewski & Stout, 2012). Future research should attempt to identify resilience factors that can help decrease the high substance use among foster youth transitioning out of care.

The second gap in the literature is the lack of geographic diversity among included studies. Five of the twelve studies reviewed derive from the Midwest Evaluation of the Adult Functioning of Former Foster Youth which used data from Illinois, Iowa, and Wisconsin (i.e., Ahrens et al., 2013; Courtney & Hook, 2017; Dworsky & Courtney, 2009; Hook & Courtney, 2011; Katz et al., 2017). An additional study was also conducted with a sample from the Midwest (i.e., Munson & McMillen, 2009). In other words, half of the research that met our criteria is based upon findings from foster youth in the Midwest (see Table 1). From the remaining six studies, three were conducted in the West (i.e., Okpych & Courtney, 2017; Shah et al., 2017; Tyrell & Yates, 2017), one in the Southwest (i.e., Crawford et al., 2018), and one in the Southeast (i.e., Kennedy et al., 2019). There were no studies representing the Northeast except from the one study that used national data (i.e., Prince et al., 2019). Future research samples should be drawn from a wider variety of geographic locations across the United States to provide a better understanding of the policy-level changes that can promote resilience.

The third gap in the literature is the emphasis on resources over assets in the study of resilience factors in foster alumni. A total of 10 assets were analyzed across studies reviewed compared to 55 resources. This imbalance may be due to the number of studies which used the same dataset, subsequently limiting the possible number of assets that could be studied. Future quantitative research should consider both assets and resources in the conceptualization of the resilience process.

Conclusion

Understanding the resilience factors that help youth successfully transition out of care can help the assessment of resilience, the development of interventions and the implementation of policy changes to prepare other foster youth in their pursuit of independent living. This systematic review is the first to provide a synthesis on resilience factors in youth transitioning out of care. The results show a total of 18 different resilience factors, four assets and 14 resources, identified by the literature that can help youth successfully overcome adversity. Although there are areas for future research on the topic, the studies reviewed demonstrate a most impressive pattern: the ability of foster youth to be resilient.

Clinical Impact Statement.

Resilience is a necessary ability for foster youth to achieve independent living after they transition out of the foster care system. This review provides a summary of the assets and resources known to promote resilience in youth transitioning out of foster care. This information can help assess for resilience, aid in the development of interventions and legislation that aim to increase it, and guide future research in this area.