Foster Families: A Systematic Review of Intention and Retention Factors

Background: Compared to residential care, family foster care is the preferable type of alternative care for neglected or abused children as it provides a familiar context that supports children’s developmental needs. New foster families are needed to care for these children. Objective: This systematic review aims to provide a critical analysis of the literature, identifying factors that explain the intention to become and to continue as a foster family. This review was performed following the PRISMA checklist and guidelines, through a search conducted in the following databases (no restrictions were made): PsycArticles, PsycInfo, Psychology and Behavioral Sciences Collection, Academic Search Complete, ERIC, Scopus, and Web of Science. Study eligibility: The review includes empirical quantitative and/or qualitative studies in English, Portuguese, and Spanish, with community and/or foster parents’ samples and explores the factors for becoming and/or retention of foster parents. Results: Forty-nine studies were included. The results revealed that the intention to become a foster parent is largely influenced by motivational factors, personal and family characteristics, individual values and beliefs, social context influences, and perceived familiarity with the child protection system. The retention of foster families is closely related to factors within the child protection system, personal or family characteristics, foster child characteristics, and placement challenges. The relationship with agencies and professional support stands out as the most important factors. Limitations and Implications: This review did not include studies focused on children with specific needs and characteristics, and future research should consider the particular challenges of fostering this group. Practice implications of these findings for the recruitment, selection, and retention of foster families will be discussed. Motivations, personal and family characteristics, and perceived familiarity or support from the child protection system were dimensions identified as associated with intention and retention. The child protection system must be prepared to support families that are highly motivated and meet the necessary conditions to foster. Formal support might reduce foster families’ willingness to discontinue fostering. Motivations, personal and family characteristics, and perceived familiarity or support from the child protection system were dimensions identified as associated with intention and retention. The child protection system must be prepared to support families that are highly motivated and meet the necessary conditions to foster. Formal support might reduce foster families’ willingness to discontinue fostering.

• The child protection system must be prepared to support families that are highly motivated and meet the necessary conditions to foster.
• Formal support might reduce foster families' willingness to discontinue fostering.
Family Foster Care (FFC) aims to provide an alternative family environment for children who were neglected and/or abused and were removed from their families (Thoburn, 2010). Every child has the right to effective and individualized care within a family context that addresses their specific needs, given their development and permanence. Thus, for these same reasons, when this is not possible within their birth family, FFC placement has been considered as preferable in comparison to residential care (Bick et al., 2017;Del Valle & Bravo, 2013;Dozier et al., 2014). Several studies have stressed that FFC contributes positively to children's development (Ghera et al., 2009;Humphreys et al., 2018;Kang et al., 2014;Smyke et al., 2010). Furthermore, in a recent meta-analysis, Li and colleagues (2019) found that children placed in FFC presented better behavioural and psychosocial outcomes comparatively to children placed in residential care. Bick and colleagues (2017) considered that residential care units were not the preferable type of placement for children in care, due to factors such as high child-caregiver ratio, high staff turnover, and/or lack of resources to promote children's socio-emotional and cognitive stimulation.
Across several countries there is a consensus that children should be placed in FFC, rather than in residential care (Del Valle and Bravo 2013). Three specific countries can be highlighted as good examples of having only around 5% of children in out-of-home care who are placed in residential care. These are Australia, Ireland, and the USA (Australian Institute of Health and Welfare [AIHW], 2018a; Child Welfare Information Gateway, 2020; Department of Children and Youth Affairs, 2016). In sharp contrast, there are still some countries, such as in Southern Europe, that are known for having greater percentages of residential care placements (e.g., Italy with 47.4% in 2016;Palareti, 2019). Specifically, Portuguese national data reveals that a mere 2.7% of children removed from their homes was placed in FFC, meaning that the vast majority of children were placed in residential care (Instituto de Segurança Social [ISS] 2020).
Although people might have a positive opinion about foster parents and FFC, the FFC system seems to be less valued compared to other social services (Leber & LeCroy, 2012). One example of this is the difficulty in having enough foster parents available for the children who are in need of an out-of-home care placement. This has been mentioned in different countries, including Australia, Ireland, and the USA (Colton et al., 2008). In the Australian context, available foster parents have been decreasing, with a reduction of around 13% since 2012 (AIHW, 2018b). In Ireland, data from 2017 shows increased difficulty in recruiting general foster parents, as well as foster parents specifically skilled in caring for children with complex needs. The data further shows a decrease in the total number of applicants to become foster parents and in the total number of foster parents (Gilligan, 2019). In the USA, Hebert and Kulkin (2018) stated that there is a shortage of qualified foster parents, who have the skillset to be able to foster children with particular needs.
Children's social care services have the challenging task to recruit and retain suitable foster parents who are able to support children with a history of adverse experiences (Bass et al., 2004;Vasileva & Petermann 2016), and children who consequently might have mental health problems (e.g., mood and anxiety disorders) and/or cognitive developmental deficits (Heim et al., 2010;Sinclair et al., 2005;Vasileva & Petermann, 2016). In order to recruit and retain more foster parents, it is crucial to outline evidence about their motivations and to understand which factors explain the intention to become and remain a foster parent (MacGregor et al., 2006;. The literature suggests that different motivations can explain the intention to become and remain a foster parent--namely, intrinsic and extrinsic motivations (Sebba, 2012). Intrinsic motivation is described as the most enduring type of motivation and relates to individual strengths (Ambrose & Kulik, 1999). Examples of intrinsic motivation for fostering children include helping children in need of care (e.g., Keys et al., 2017) and protecting children from future harm (e.g., . This intrinsic motivation can be viewed as an indicator of altruism, which is associated with greater pro-social behaviors (Bockler et al., 2016). As such, altruistic motives and prosocial behaviors not only benefit others (Bockler et al., 2016;Keltner et al., 2014) but also those who practice them. Altruistic motives to foster are positively associated with higher levels of job satisfaction, which in turn is positively associated with foster parents' retention Sebba, 2012;Cleary et al., 2018). On the contrary, extrinsic motivation refers to rewards or expectations that yield to the subject by performing a certain task (Ambrose & Kulik, 1999). This is viewed as less long-lasting and is related to a lower retention rate. Examples of extrinsic motivation are a family wanting to fill the empty nest or wanting to give a brother to their biological children (Andersson 2001). The desire to counterbalance failed family expansion experiences (e.g., infertility) may also be a motivation to FFC.
However, foster parents can be motivated by a combination of both intrinsic and extrinsic reasons (MacGregor et al., 2006), and FFC calls for highly committed foster parents who are warm and affectionate with children and can effectively deal with different challenges simultaneously (Herczog et al., 2001;Marcellus, 2010;Solomon et al., 2016). Particularly, such challenges may include regular contacts between children and their birth families (Hudson and Levasseur 2002), children's developmental problems (Dubois-Comtois et al., 2015;Hambrick et al., 2016;Sawyer et al., 2007;Turney & Wildeman, 2016;Vanschoonlandt et al., 2013), or feelings of grief and loss due to reunification (Wolf et al., 2013). Dealing with these challenges nonadaptively can lead to placement disruptions (Chamberlain et al., 2006). Furthermore, becoming and remaining a foster family is influenced by contextual factors beyond intrinsic or extrinsic motivations. Support from relatives (Doyle and Melville 2013) together with help, support, and partnership from professionals (Barter & Lutman, 2015;Hudson & Levasseur, 2002;MacGregor et al., 2006;Wolf et al., 2013) are determinant and will increase the satisfaction of foster parents (Denlinger & Dorius, 2018).
It has become evident that it is crucial to recruit and retain foster parents with the needed skillset to care for children who have been removed from their families. Several studies have been exploring factors that contribute to the recruitment and retainment of foster parents; however, there are not systematic reviews in this field. Evidencebased recruitment processes require an in-depth knowledge about the existing evidence, focused on the reasons associated with becoming and retaining foster families. This review may provide an integrated picture of the influencing factors for fostering that could guide policy and practice and enhance FFC system. Also, future research may be informed by the gaps identified through this systematic review. As such, this study aims to systematically review existing literature to address our main research questions: a) What factors are associated with the intention to become a foster parent? b) What factors are associated with the retention of foster parents? c) What are the methodological characteristics (e.g., sample, instruments, and design) of the empirical studies developed to address those questions? Finally, based on those finding we will identify the studies' limitations and future directions as well as the implications for practice in the FFC system.

Research Question and Search Strategies
Our research question was formulated using the SPIDER method (Sample, Phenomenon of Interest, Design, Evaluation and Research Design; Cooke et al., 2012): a) Sample: Foster parents and general population, older than 18 years old; b) Phenomenon of Interest: Intention and retention factors related to being a foster family (i.e., explanatory factors of the decision to become a foster family or to continue fostering, including individual, social, institutional, and macrosystemic factors); c) Design: All designs (except case studies and literature reviews) and methods were considered as long as they were empirical; d) Evaluation: Several outcomes were considered, in particular the decision to become a foster parent, the intention to become a foster parent, or the intention to continue being a foster parent. These can be measured in a dichotomous way (yes/no) or in a continuous measure of intention; e) Research Design: All types of studies, quantitative, qualitative, and mixed methods.
The search was conducted in September 2018, on the following online databases: PsycArticles, PsycInfo, Psychology and Behavioral Sciences Collection, Academic Search Complete, ERIC, Scopus, and Web of Science. The combination of the following terms was used in the search: (a) foster care OR foster families OR foster parent* AND (b) motivation* OR retention OR willingness to foster* OR motivation* factors OR motivation* foster OR reasons for fostering OR predict* foster* care. Specific restrictions were applied in all databases: articles must be (a) published in peer-reviewed academic journals, (b) written in the English, Portuguese, or Spanish language. English, Spanish, or Portuguese papers were included as the authors are proficient in these languages. No restrictions were applied regarding the publication date. A hand search of the reference lists of previous literature reviews, and of all the articles included in these, was performed. As such, some articles that were not identified by our electronic search, but which met the inclusion criteria, were included. Duplicate studies were verified and removed. No registration of protocol was performed.

Inclusion and Exclusion Criteria
Inclusion criteria were as follows: (1) empirical quantitative and/or qualitative studies; (2) studies including a community sample where the factors for becoming foster parents were explored (i.e., what would lead people to become a foster family); (3) studies with foster parents where the factors for becoming foster parents were explored; and (4) studies with foster parents where the factors of retention were explored. Case studies and literature reviews were excluded, as well as studies that explored motivations to become foster parents of children with special needs (e.g., children with different mental and physical abilities) or specific characteristics (e.g., aboriginal children). These studies about children with specific needs or characteristics were not included, as the profile of these foster parents is expected to be different, and these specificities would require a particular review of those articles only.

Study Selection and Data Extraction
This review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist and guidelines (Liberati et al., 2009). The retention or rejection of articles was based on the sequential screening of the title, abstract, and full text. Inter-judge agreement was made by two independent coders. Initially, the search yielded 3378 articles, plus, an additional 42 from a hand search of other articles. After removing duplicates, a total of 2883 articles were screened on the basis of the title and abstract alone. At this phase, an inter-judge agreement of 30% of the articles was achieved, resulting in 96.7% in agreement. All disagreement decisions were reviewed and discussed according to the inclusion and exclusion criteria. After the initial screening, 87 articles were selected for full text analysis (eligibility). The next step included another inter-judge agreement of 30% of the articles, resulting in 73% in agreement and 8% in disagreement. In 19% of articles, one of the coders was undecided whether or not to include the article in the review. Disagreements and uncertainties were subsequently resolved by an in-depth discussion about the specificities of those articles, bearing in mind the inclusion/exclusion criteria. A total of 49 articles were included in this systematic review. The flow diagram of the study selection process is displayed in Fig. 1.
Based on our objectives, the following data extraction procedures were implemented. First, data was taken from the primary studies based on an extraction sheet that included the following sections: the country in which the study was conducted, research design, sample, data collection methods and instruments, and analytic strategies (objective 3; Table S1). Second, all significant results (i.e., indicators on Tables 1 and 2) were extracted to two spreadsheets (one for intentionobjective 1; and another for retentionobjective 2) and coded with (+) or (−) depending on whether it is positively or negatively related to intention and retention. In order to organize the amount of information that was extracted into broad factors that would allow the results' interpretability (Tables 1 and 2), a qualitative analysis of these indicators was conducted. Two researchers have analyzed this material, organizing those indicators into factors and subfactors, and then discussing with each other and with a third researcher. All disagreements were discussed until consensus was reached.

Factors Associated with Becoming a Foster Parent
Results revealed that five main factors may influence the decision to become a foster parent (Table 1): (1) motivational factors; i.e., motives that guide individuals' behaviors, ranging from self-centered motives to those centered on others; (2) foster parents' personal and family characteristics; i.e., personal attributes of foster parents/prospective parents and characteristics of foster families; (3) values and beliefs; i.e., representations and attitudes underlying the decision to become a foster parent; (4) social context influences; i.e., a set of contextual and environmental circumstances that influence the decision to become a foster family; and 5) familiarity with the FFC system; i.e., the extent of people' knowledge about the FFC system. Below, each of these factors is described in detail.

Foster parent's personal and family characteristics
Within this factor (n = 24) the following subfactors were identified: (a) foster family functioning (n = 11; e.g., Doyle & Melville, 2013), (b) failed family expansion (n = 11; e.g., Rhodes et al., 2006), (c) personal experiences and attributes (n = 5; e.g., Goodman et al., 2017), and (d) sociodemographic characteristics (n = 5; e.g., Contreras & Muñoz, 2016). Regarding the sociodemographic characteristics subfactor, studies revealed inconsistent results on age, with some studies suggesting that older people were more likely to become foster parents (e.g., Contreras & Muñoz, 2016), while others identified younger participants as more likely to become foster parents (e.g., . Personal experiences and attributes, as well as family functioning, had both a negative and positive influence on becoming a foster parent. All personal experiences and attributes had a positive impact, such as parental experience and having also been a foster child, with the exception of experiencing abuse, neglect, and violence during childhood, which seemed to prevent this decision. In respect to family functioning, having adequate financial

Values/beliefs
The values and beliefs identified in this review were (n = 20): (a) moral or social responsibility (n = 15; e.g., Howell-Moroney, 2014), (b) religious motives (n = 6; e.g., Tyebjee, 2003), and (c) family-based values (n = 5; e.g., Diogo & Branco, 2017). All these subfactors positively impacted the decision to become a foster parent. This indicates that those who believed in moral and social responsibility and in the positive influence of fostering on children were more likely to become foster parents. Also, those who ascribed to 'family-based values' were more likely to become foster parents.

Social context
The subfactors identified within the social context were (n = 7): (a) social influence (n = 5; e.g., Ramsay, 1996), (b) formal support (n = 2; e.g., Metcalfe & Sanders, 2012), and (c) social commitments (n = 1; . The first and second subfactors positively affected the decision to foster, which meant that positive social influence and supportive formal relationships were associated with becoming a foster parent. On the contrary, having other social commitments seemed to prevent people becoming foster parents. Familiarity with the system Finally, this factor was the least identified in the studies under analysis (n = 6; e.g., Wilson et al., 2007). This factor referred to the knowledge of the child protection system and the context of FFC, which had a positive impact on becoming a foster parent.

Factors Associated with the Retention of Foster Parents
This literature review identified four distinct factors effecting the decision to continue fostering (Table 2): (1) child protection system, (2) foster parent's personal and family characteristics, (3) foster child's characteristics, and (4) placement challenges. Below, each of these factors are described in more detail.

Foster parent´s personal and family characteristics
Within this factor (n = 15), five subfactors were identified: a) personal attributes/characteristics of foster parents (n = 7; e.g., Broady et al., 2010), (b) foster family functioning (n = 7; e.g., , (c) sociodemographic characteristics (n = 6; e.g., Maeyer et al., 2014), (d) personal or family changes (n = 5; e.g., Geiger et al., 2013), and (e) experience as foster family (n = 2; Hendrix & Ford, 2003). Studies revealed that the personal attributes/characteristics subfactor has both a negative and positive impact on retention. For instance, greater feelings of insecurity (Broady et al., 2010) were associated with lower retention; whereas greater empathy, flexibility (Keys et al., 2017), and internal locus of control (Geiger et al., 2013) positively impacted retention. All the sociodemographic characteristics had a negative impact on retention (e.g., employment, marriage status, single parenthood), except for age (Maeyer et al., 2014). Also, both white (Rindfleisch et al., 1998) and nonwhite ethnicities (Ahn et al., 2017; had been associated with lower retention, yielding a contradictory result in the literature. Personal or family changes were negatively associated with foster parent retention. The foster family functioning subfactor impacted both positively and negatively the intention to continue to be a foster parent. If, for instance, having resources was positively associated with continuing to be a foster parent , stressful experiences in the family (Geiger et al., 2013) or receiving inadequate financial reimbursement (Rhodes et al., 2001) were associated with a lower retention rate. The foster parent experience also impacted both positively and negatively the intention to continue to be a foster parent, with being a foster family for more than two years being negatively associated with continuing to be a foster parent (Hendrix & Ford, 2003). While, feelings of commitment to the foster child were associated with a higher retention rate (Eaton & Caltabiano, 2009).

Foster child's characteristics
The foster child's characteristics factor (n = 7) was identified as being strongly related to retention, with the following subfactors: (a) psychological problems (n = 5; e.g., , (b) problems with the child (n = 1; Ahn et al., 2017), and (c) few child's improvements (n = 1; e.g., Broady et al., 2010). Greater psychological problems and problems with the foster child in general were associated with a lower retention rate, as well as was few child's improvements.

Placement Challenges
This factor refers to aspects of the FFC process (n = 3). Only one subfactor was identified: reunification with birth family (n = 3; e.g., Denby & Rindfleisch, 1996). The reunification with the birth family appeared as an inconsistent result, impacting both positively and negatively the decision to continue fostering, due to the perceived difficulties associated with the child leaving (Rhodes et al., 2001).

Discussion
This current systematic review aimed to critically analyze the existing literature to identify factors that contributed to the intention and retention of foster parents. A total of 49 articles were identified and summarized according to the study's sample, context, and main findings. Results will be discussed by integrating and organizing findings that focus on becoming a foster family and foster family retention according to common and interrelated domains: motivations, personal and family characteristics, and child protection system related dimensions (i.e., familiarity with and support from the child protection system).
Evidence indicates that the decision to become a foster parent is primarily related to motivational factors. Both extrinsic and intrinsic factors were identified, and although extrinsic motives like family expansion (e.g., Cole, 2005) and financial reasons (e.g., Howell-Moroney, 2014) were reported, most studies focused on intrinsic motives (e.g., desire to care and love children or desire to help children; Tyebjee, 2003). We know that in the context of fostering children, being guided by intrinsic motivation, in comparison to extrinsic motivation, leads to higher levels of job satisfaction. This is even more critical as higher levels of satisfaction are associated with a greater retention rate (Ambrose & Kulik, 1999;Cleary et al., 2018). Besides that, the desire to care, love, and help children was mostly identified within the motivational factor, which can be framed within the literature on altruism (Metcalfe & Sanders, 2012). Altruism is seen as a disposition to seek and increase the welfare of others (Batson & Powell, 2003). Some findings propose that altruism could be associated with greater prosocial behavior, and that prosocial behavior might significantly benefit others (Böckler et al., 2016;Keltner et al., 2014).
In addition to this perspective, different motivations for being a foster family can be observed in two distinct groups. One group characterized as having child-centered motivations and the other as being more self/family-oriented (Rhodes et al. 2006). When analyzing the data, childcentered motivations narrowly stand out, being mentioned in 23 of the analyzed articles while the self/family-centered motivations were identified in 21 articles. This shows that child-centered motivations are more referred to than the needs of the family itself. Child-centered needs include, for example, the need to protect and prevent children from harm  or the desire to provide children with a positive family experience (e.g., Tyebjee, 2003). In contrast, motivations centered on the family or self refer to family expansion, wanting to be loved by a child (e.g., Maeyer et al., 2014), or even financial reasons (e.g., Howell-Moroney, 2014). Being a foster family, in addition to helping these children, can also enable a sense of personal achievement (Martin et al., 1992).
Furthermore, foster families' personal and family characteristics were also identified as being relevant to the decision of becoming a foster parent. The cases of failed family expansion--particularly childlessness, infertility, and unsuccess in adopting a childcan increase the likelihood of fostering children (e.g., Dando & Minty, 1987). On the contrary, having biological children seemed to prevent individuals from becoming foster parents. However, failed family expansion, despite being related to family characteristics, can also be understood and categorized as a motivational factor, therefore a combination between wanting a larger family and to care for children (e.g., Kozlova, 2013;Maeyer et al., 2014;Rhodes et al., 2006). It can be argued that the intention to foster based on a failed family expansion might be related to the fulfillment of individual needs instead of being focused on the best interest of children (the latter of which should be central to the fostering role). An inconsistent result was identified in relation to age, with some studies reporting that the older the people were, the greater their predisposition to become foster parents (e.g., Contreras & Muñoz, 2016), while other studies reported the opposite finding (i.e., when people are younger their predisposition to become a foster parent increases) (e.g., . These divergent results could be explained in light of the family lifecycle (Carter & McGoldrick, 1988). On one hand, younger people are more focused on establishing a career and obtaining financial independence, which might undermine their ability to become a foster parent. Moreover, family life cycles are changing, e.g., late home-leaving, delay in autonomy processes and the postponement of family formation (Organisation for Economic Co-operation and Development [OECD] 2011). This can weaken the willingness to become a foster parent. On the other hand, people in later life might need to take care of their grandchildren or deal with disabilities and with the decline of abilities (Carter & McGoldrick, 1988), which can negatively influence their time and capacity to care for a foster child.
Another interesting result suggests that former foster children (e.g., Martin et al., 1992) are more likely to be foster parents; however, some studies highlighted that having adverse childhood experiences was associated with lower willingness to foster (e.g., Goodman et al., 2017). Research has highlighted the negative impact of adverse childhood experiences on adulthood (Hughes et al., 2017), but this effect has also been proved to be attenuated by protective factors (Sciaraffa et al., 2018). For instance, having contact with parenting styles that are positive can enhance positive results that in turn last until adulthood (Hamilton & Harris, 2018). This is aligned with research conducted by Vanderfaeillie and colleagues (2013), which associates a positive FFC system experience with children's positive outcomes. In fact, children learn from interacting with others, reproducing behaviors by observing, as described by the social learning theory (Bandura, 1971). Thus, a positive FFC experience could be associated with positive parenting of a foster child in the future. Therefore, this explains that having a challenging childhood (Dando & Minty, 1987) and growing up in a foster family (Peake & Townsend, 2012) increases the motivation to care for children. Moreover, this could also be viewed in light of the importance given to one's role in society; that is, the need to give back the benefits once received from society as a child. This idea is defended by the Resource Theory, which, as stated by Cox and colleagues (2003), is based on the principle of sharing resources with someone who has fewer resources. In other words, having adequate resources, such as time and space, is imperative when considering whether or not to become a foster parent (Doyle & Melville, 2013).
FFC requires foster families to be able to adapt their previous routines and functioning to receive children. Therefore, if difficulties are perceived at this level, some potential foster parents may consider that they do not have the adequate resources, which reduces their likelihood of fostering children. Furthermore, personal attributes (e.g., Keys et al., 2017) such as being empathic and flexible, having an internal locus of control, and having a higher perceived self-efficacy were identified as being important for the retention of these foster parents. As previously mentioned, foster families must have skills and specialized knowledge to take care of these children (Herczog et al., 2001;Marcellus, 2010) and to be able to address complex needs (Solomon et al., 2016). This finding suggests that foster parents need to have a specific profile to deal with the challenges of fostering (e.g., empathic skills, flexibility), and those people should be privileged in the recruitment processes. By contrast, some sociodemographic characteristics (e.g., Ahn et al., 2017), such as being employed, a single parent, retired, or chronically ill, had a negative influence on retention. This is because these people already have some challenges that might prevent them from continuing to foster. Work-family conflicts have also been studied over the years and it is known that being employed can have implications in family functioning (Judge et al., 2006). Depending on the flexibility, working hours, and work-related stress, employment can limit the parental involvement with children (Fraenkel, 2003). Finally, illness is challenging, and retirement might be too, especially considering those who are not financially stable and those for whom the retirement was undesired or forced (Walsh, 2016). These factors may undermine the willingness to continue fostering.
In this sense, the child protection system must be prepared to support families that are highly motivated and meet the necessary conditions to foster, so that retention can be enhanced. Retention seemed to be lower when foster parents experienced problems with the services/agencies (Denby & Rindfleisch, 1996) and professionals (e.g., poor communication and few contacts), whereas feelings of being supported by the agency or other foster parents (Blackburn, 2016) had a positive impact on foster family retention. Given that FFC is composed of a variety of challenges for the birth families, foster children, and foster parents, these supportive practices are crucial (Wolf et al., 2013) and are associated with greater foster parent satisfaction (Denlinger & Dorius, 2018).
Further, the choice to be a foster parent corresponds with holding certain values and beliefs about social responsibility. Findings from this review suggested that becoming a foster parent could derive from moral and social responsibility perceptions (e.g., "wanted to do something for the community" or "wanted to fulfill a societal need") (Cole, 2005). Theoretically, this sense of community includes the need to feel connected with others and doing something for the community facilitates one's personal growth, which is also related to lower levels of mental, social, and health problems (Hyde & Chavis, 2008). Evidence on social wellbeing proposes that individual wellbeing is also composed of feeling accepted by others, contributing to and feeling part of the community (Keyes, 1998). Furthermore, this finding is also consistent with the literature that suggests that a psychological sense of community is positively related to prosocial behaviors (Hackett et al., 2015).
Finally, being familiar with the child protection system (Ramsay, 1996), knowing a foster family, or being encouraged by a spouse or others (Doyle & Melville, 2013) were also recognized as important factors to become a foster family. This highlights the need to spread accurate knowledge about the FFC system, given that misconceptions or lack of information about the system can undermine the effort to recruit families (Leber & LeCroy, 2012). This dissemination of accurate information should include data on foster children's developmental trajectories and mental health difficulties. Foster children's characteristics were described as reducing the intention of foster parents to continue fostering (Rhodes et al., 2001), which points out the negative impact of children's behavioral and emotional problems on the fostering role (Sawyer et al., 2007). Due to their previous adverse experiences (e.g., child abuse and neglect), foster children are more prone to having developmental difficulties, such as mental health problems (Heim et al., 2010;Sinclair et al., 2005;Vasileva & Petermann, 2016). Specifically, externalizing problems are prevalent in this population , which leads to greater challenges for foster parents and subsequently can contribute to disruptions in placements (Chamberlain et al., 2006). Therefore, not only may this influence the decision to become a foster parent but could also reduce retention.
Lastly, placement challenges also influenced foster parent retention (e.g., Denby & Rindfleisch, 1996). Studies pointed to conflicting findings regarding the effect of reunification on foster family retention. Some evidence suggested that it undermined foster parents' retention, and others reported that it enhanced their willingness to continue fostering. This could be explained by inaccurate or unavailable information about permanency planning and reunification. When reunification occurs, it is often experienced as undesirable. Worries about children's development and the return to the same undesirable environment might lead to feelings of loss and grief. Nonetheless, foster parents need to be able to cope with these feelings (Wolf et al., 2013). Therefore, it is important for foster and birth families to have a close and supportive relationship, allowing foster families and birth families to have an insight into the child's permanency plan and consequent reunification. Moreover, this relationship may contribute to the diminishing of some challenging behaviors from the foster children when adjusting to the foster home.

Limitations and future directions
This literature review summarizes relevant findings from a range of studies; however, some limitations of the reviewed studies must be considered. Although most of the studies were quantitative, they were mostly descriptive in nature and did not provide meaningful data about what factors were strongly associated with intention and retention to be a foster parent. Forty-five of the studies were cross-sectional, so results should be interpreted carefully to avoid causal inferences. Moreover, additional longitudinal studies are required. Also, most of the studies focused on female foster parents; however, evidence suggests that the fathers' involvement is important for children's development (e.g., behavioral, social, and cognitive) and wellbeing (Heslop, 2019). This indicates that it is important for male foster parents to be involved in research and evaluative processes in order to obtain a reliable picture regarding motivations, feelings, and experiences from different foster parents in the FFC system. Moreover, most studies do not specify the type of foster families included in the sample or include mixed samples. As such, greater clarity is needed for future research regarding the description of foster families, given that recruitment, preparation, and training involves different challenges for kinship carers and non-kinship carers. It may also be possible to separately analyze kinship and nonkinship families if they are both included in the study and the results are discussed. Besides that, few studies included former foster parents, which may be an important population to consider when analyzing factors of retention. For that reason, future research should include foster parents that are no longer fostering and analyze their reasons to discontinue fostering, which might inform foster family programs. Also, for future research, further reviews should include terms like "breakdowns" or "failure", providing more consistent insight into factors associated with foster parent retention. Few studies focused on community samples; however, this could provide further innovative insight for the purposes of foster parent recruitment. Finally, this systematic review also had some limitations, namely, the exclusion of studies that focused on children with specific needs and characteristics. Considering that foster children have some developmental and health needs, it would be important to explore factors explaining the willingness to foster children with specific needs (e.g., with HIV or fetal alcohol spectrum disorder).

Implications for practice in the FFC system
This systematic review offers an important insight into the recruitment and retention of foster families. Recruitment campaigns should emphasize the intrinsic motivational factors and the resources needed to provide quality FFC. Also, strategies appealing to moral responsibility, as well as to the difference that individuals could make in children's lives, could be used in the context of recruitment. Considering that having adequate knowledge about the FFC system was important in decisions to foster, efforts must be made to disseminate accurate information. As such, providing reliable information allows people to acquire an indepth understanding of this public problem, which may enable them to make informed decisions. Further, this process of recruitment should be informed by the need to engage people who are empathic and flexible and who exhibit mostly an internal locus of control.
Regarding the retention of foster parents, support is needed from services and agencies. Not only close and warm relationships between professionals and foster families are relevant to help them adequately deal with those diverse challenges (e.g., children's behavioral problems, the relationship with birth family), but also specific training is needed. This should specifically include training on empathic relationships to prevent significant problems between foster families and agencies/services, and initial and continuous training to support foster parents so they can continue to develop their skills and acquire specialized knowledge. Receiving training throughout the fostering experience is important, particularly on educational strategies, expectations of the foster child and FFC system, and promoting positive attitudes towards the foster child's family and their life history (Amorós & Palacios, 2004;Schoemaker et al., 2020). For instance, PRIDE Model of Practice (from Child Welfare League of America and FosterParentCollege.com) has been adopted worldwide to support foster families. This model considers foster families as important agents in the child protection system, particularly agents who develop competencies related to children's protection and developmental needs, build quality relationship with birth families and work with the professional team as a member (for more information see https://www.cwla.org/pride-training/). This specific training opportunity may allow foster parents to feel more confident in their ability to deal with FFC challenges. Furthermore, an efficient participation processes should be adopted, allowing foster parents to be involved in the decisions related to the placement. They should be informed and engaged in permanency planning, should be aware of the reunification process, and should have adequate support to deal with their losses during this process. This support might increase their willingness to continue fostering.

Compliance with Ethical Standards
Conflict of Interest The authors declare no competing interests.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.