Psychosocial well-being of Flemish foster children residing in their foster homes during the COVID-19 lockdown.

The measures implemented to contain the COVID-19 pandemic profoundly affected the lives of children and families all around the world, probably affecting children's psychosocial well-being. The negative consequences of lockdowns are presumed to hit even harder on vulnerable groups such as foster children who already struggle with their psychosocial well-being in normal circumstances and who face specific challenges during lockdowns such as: additional help that is no longer available or only offered digitally and physical contact with birth parents that is forbidden. Nevertheless, some scholars point to the positive side of lockdowns (e.g.: relief due to closure of schools). This study aims to asses the psychosocial well-being of Flemish foster children residing in their foster homes during the COVID-19 lockdown and the factors that are associated with the change in their psychosocial well-being. 888 foster parents reported on the psychosocial well-being of just as many foster children through the Brief Assessment Checklist for Children and Adolescents. The COVID-19 lockdown was not associated with a decreased psychosocial well-being of Flemish foster children residing in their foster homes during the lockdown. Foster parents reported a slight improvement in their relationship with their foster child during the lockdown which points to a positive consequence of the lockdown. In addition, this improvement was positively associated with an increased psychosocial well-being during adverse circumstances and should therefore be enhanced. The type and amount of contact with birth parents is significantly associated with foster children's changed psychosocial well-being during the lockdown. Foster parents who reported an increase in alternative contact (e.g., (video)calls and messages) between their foster child and his/her birth parent(s) during the lockdown, also reported an increase in their foster child's psychosocial well-being during that period.

Introduction

The coronavirus disease 2019 (COVID-19) and the measures implemented to contain the pandemic profoundly affected the lives of children and families all around the world. Closing schools, teleworking, social distancing and cancelation of external leisure activities put a lot of pressure on children and their families. This possibly resulted in a decrease of children’s psychosocial well-being. In fact, studies during the early stages of the COVID-19 lockdown already reported worrisome results such as poor mental health outcomes in children (e.g., Achterberg et al., 2021, Jiao et al., 2020). The negative consequences of lockdowns are presumed to hit even harder on vulnerable groups such as foster children who already struggle with their psychosocial well-being in normal circumstances (e.g., Coyne et al., 2020, Goemans et al., 2016, Turney and Wildeman, 2017). Moreover, foster children also face specific challenges during lockdowns such as: additional help that is no longer available or only offered digitally and physical contact with birth parents that is forbidden (Agentschap Opgroeien, 2020). Nevertheless, some scholars point to the positive side of lockdowns. For example, the closure of schools may create a (temporary) sense of relief and alleviate some of the pressure in children who were struggling at school (Bruining et al., 2020, Hoekstra, 2020).

This study investigated the influence of the COVID-19 lockdown on the psychosocial well-being of foster children in Flanders (Dutch speaking part of Belgium). In what follows we first describe the possible impact of the lockdown due to the COVID-19 pandemic on the psychosocial well-being of children in general, followed by foster children specifically. We also review the evidence already at hand from recent studies and introduce our research questions. Next, a description of our method and results are given to finally end with a discussion and general conclusion.

Impact of lockdowns on children’s psychosocial well-being

Negative impact

Even though symptoms of COVID-19 are less severe in children and adolescents resulting in fewer hospitalizations and lower mortality rates (Liu et al., 2020, Ludvigsson, 2020), children suffer from the pandemic in an indirect way. Restrictive measures during lockdowns create numerous stressors on an individual and family level, leading to a decrease of children’s psychosocial well-being (e.g., Koller et al., 2010, Galea et al., 2020).

Individual level

On an individual level, the COVID-19 pandemic may lead to poor mental health outcomes as has been confirmed by research during prior pandemics (e.g., SARS) (Koller et al., 2010) and recent findings during the current one (e.g., Achterberg et al., 2021, Jiao et al., 2020). Emotional and behavioral problems such as anxiety and fears, post-traumatic stress symptoms, depression, irritability, stress, inattention, conduct problems and clinginess were reported during the early stages of the COVID-19 pandemic (Achterberg et al., 2021, Fegert et al., 2020, Jiao et al., 2020, Lee, 2020, Zhang et al., 2020). These outcomes can be linked to the drastic measures implemented to contain the coronavirus. Lockdowns lead to social isolation, as many occupations providing meaning, structure, routine and social interaction such as school and external leisure activities are canceled (UNICEF., 2020, Xiang and Zhang, 2020). This is especially hard for children and adolescents who need social contact to learn, grow, for their (identity) development and their well-being (Kroger et al., 2010, Orben et al., 2020, Ragelienė, 2016). Children also reported missing their friends and classmates, and noticed exacerbations of their (mental) problems (Achterberg et al., 2021, De Kinderombudsman, 2020).

Family-level

Families also struggle during lockdowns. The difficulties and challenges faced by children and their parents may destabilize families resulting in negative consequences for the well-being of children (Galea et al., 2020, Green, 2020). For example, children may be deprived of parental care due to illness-related absence (e.g., hospitalization, quarantine or death) or due to parents working from home who are not able to pay attention to their children (Sistovaris et al., 2020). Losing parental care is worrisome, as it is imperative for a healthy development and leads to negative long-term consequences for children’s psychological well-being when disrupted (Norredam et al., 2018, Santavirta et al., 2015).

Moreover, parents face additional stress and challenges as they cannot rely on their support system (i.e., extended family and friends, social services, schools and neighbors) anymore (Chung et al., 2020). For example, parents may struggle to combine the care for their children with home-schooling and teleworking (Courtney et al., 2020, Coyne et al., 2020, Sistovaris et al., 2020, UNICEF., 2020). The stress accompanying these parental challenges decreases parents’ psychosocial and mental well-being. Indeed, research found an increased risk for substance abuse, psychosocial and mental health problems among parents during the COVID-19 lockdown (Achterberg et al., 2021, Brown et al., 2020, Galea et al., 2020). Poor parental mental health is worrisome for the psychosocial well-being of children as parents help regulate their children’s emotions and protect them against stress. Research extensively showed an association between psychological symptoms in parents and both psychological symptoms and stress in children (Achterberg et al., 2021, Whittle, 2020, Zhang et al., 2020).

On a relational level, the increased mental, physical, financial and parental stress in combination with social isolation, living in forced close proximity and the lack of a support system, could contribute to strained parent–child relationships and cause conflicts. This, in turn, increases the risk of aggressive behavior, domestic violence, harsh parenting and child abuse and neglect (CAN) (Brown et al., 2020, Chung et al., 2020, Fegert et al., 2020, Hager, 2020, Lee, 2020, UNICEF., 2020). Meanwhile, the safety net around children weakens during lockdown, leading to an increase of undetected CAN as the most frequent reporters are teachers, sport coaches, pediatricians, social and youth workers. This has detrimental consequences for the children’s psychosocial well-being (Hager, 2020, Humphreys et al., 2020).

Positive impact

Notwithstanding the negative consequences of lockdowns, they may also lead to some advantages resulting in increased psychosocial well-being (e.g., Bruining et al., 2020, Isumi et al., 2020). First, the closure of schools may enhance a (temporary) sense of relief for children who struggle at school (Bruining et al., 2020, De Kinderombudsman, 2020, Fegert et al., 2020, Hoekstra, 2020, Vlaams Kinderrechtencommissariaat, 2020). Second, even though the lack of external leisure and group activities may be negative for the foster child; in combination with telework it also severely reduces the daily commute, which may decrease stress in families (Andresen et al., 2020). Third, children and parents may enjoy the additional time they have to spend together, improving their relationship and ultimately enhancing children’s resilience and mental health (Achterberg et al., 2021, Brown et al., 2020, Bruining et al., 2020, Courtney et al., 2020, De Kinderombudsman, 2020, Fegert et al., 2020, Hoekstra, 2020, Isumi et al., 2020, Vlaams Kinderrechtencommissariaat, 2020).

COVID-19 lockdown and foster children’s psychosocial well-being

Foster children are a vulnerable group. They are placed in out-of-home care because their safety was at risk and/or their birth parents were (temporarily) unable to care for them (Pleegzorg Vlaanderen, 2020). From 2018 to 2019, 7.567 children were placed in Flemish foster care (Pleegzorg Vlaanderen, 2019). In Flanders, a distinction is made between short-term, long-term and supportive foster care placements (Pleegzorg Vlaanderen, 2020, Vanderfaeillie et al., 2017). Short-term foster care aims at reunification within a year. If this cannot be realised, the result should be a long-term solution for the child (i.e., long-term foster care or residential care). Most children (65%) are placed in long-term foster care which aims at creating permanency. Although children can be placed in foster care until the age of 24, a return home can always be considered. In Flanders, adoption is not an option for creating permanency. Flemish child welfare is family-oriented and maintaining links between children with their birth parents is prioritized. Therefore, permanent severance of parental rights through adoption is rare (Vanderfaeillie & Van Holen, 2018). Supportive foster care is used to support the parenting situation at home. The child resides in the foster family either for a short continuous period or for several short periods alternating between the foster family and family of origin. Children can be placed voluntarily or by court order. In 2018, 37.3% of all placements were compulsory (Pleegzorg Vlaanderen, 2019). Moreover, two types of foster families can be distinguished based on the relationship between the foster parents and the foster child/birth parents prior to the placement. Kinship carers (61%) already know the child, whereas the child is a stranger to non-kinship carers (47%) (Pleegzorg Vlaanderen, 2019, Vanderfaeillie et al., 2017).

Foster children have been through adverse childhood experiences, such as CAN, and were separated from their birth parents (Stovall-McClough & Dozier, 2004). These experiences influence their development, resulting in decreased psychosocial functioning and well-being (Turney & Wildeman, 2017). Moreover, prior studies have reported extensive developmental issues such as behavioral, psychological and educational problems among foster children both before and during their placement (Goemans et al., 2016, Goemans et al., 2018, Sebba et al., 2015, Tarren-Sweeney, 2013, Van Holen et al., 2018, Vanderfaeillie et al., 2018).

Foster child-specific challenges and opportunities during lockdowns

Like children in the general population, foster children are subjected to the aforementioned challenges (see 1.1.1), possibly jeopardizing even more their psychosocial well-being during lockdowns. In addition, foster children face specific challenges, uncertainties and difficulties during lockdowns (Pisani-Jacques, 2020). However, they may also benefit from lockdowns in additional ways, compared to children in the regular population mentioned in 1.1.2.

First, both foster children and their foster parents are in need of additional help which they obtain through foster care or other child welfare agencies. However, due to the lockdown, the normal course of work was disrupted (e.g., cancellation of support programs and home visits), impacting the (quality of) provided care (Fegert et al., 2020, Hager, 2020). One example is the massive switch to providing virtual care which may result in a disadvantage for the most vulnerable families as they may not have appropriate equipment or accommodation to properly engage in virtual care (Conn, 2020, Lee, 2020, National Association of Counsel for Children, 2020). Additionally, developing a trustful relationship and continuity of care are essential to provide effective service (Sistovaris et al., 2020), which is more difficult through virtual interaction (Mairhofer et al., 2020). The loss of adequate support may be damaging for the foster children and the foster families, particularly during these challenging times (Green, 2020, Jentsch and Schnock, 2020, Lee, 2020).

Second, children are often placed with older relatives such as their grandparents (Van Holen et al., 2017). This can be problematic as older age is a risk factor for a more severe infection and increases the likelihood of hospitalization or even death (Liu et al., 2020). Foster children may therefore be more likely to lose parental care or experience losses impacting their well-being.

Third, foster children residing in their foster families are deprived of physical contact with their birth parents during lockdowns (e.g.: first lockdown in Flanders (Agentschap Opgroeien, 2020)). Contact with birth parents was previously associated with positive outcomes such as increased reunification and emotional well-being in both parents and children (McWey et al., 2010, McWey and Cui, 2017, Nixon et al., 2013). Therefore, the loss of contact or the change to an alternative (e.g.: text messages, (video)calls, letters) and maybe a less-satisfying way of communicating may have negatively impacted foster children’s psychosocial well-being (Burnson, 2020, De Kinderombudsman, 2020, Hager, 2020). Moreover, younger children are not able to interact adequately through telephone or videocalls (Burnson, 2020, Myers et al., 2017). Nevertheless, virtual contact may also be beneficial as it allows children and parents to maintain a more regular contact. This allows birth parents to be involved in the day-to-day activities such as family meals, helping with homework and putting children to bed (National Association of Counsel for Children, 2020, Pitzl, 2020). Furthermore, visits with birth parents are often emotionally demanding for foster children (Poehlmann-Tynan et al., 2015, Simms, 1991). Therefore, not having to see the parents (physically) could also be a relief for some of them.

Lastly, during the lockdown foster children residing in their foster homes had more time to spend with their foster families. As for children in the regular population, this could lead to improved relationships between foster children and foster parents (Bruining et al., 2020, Hoekstra, 2020). This can be positive for foster children’s psychosocial well-being as foster parents are providing a more stable and safe environment (Pleegzorg Vlaanderen, 2020).

Aim of study

Assessing the influence of the lockdown on the psychosocial well-being of foster children is necessary to set up effective interventions. Moreover, knowing what factors are associated with and may influence children’s psychosocial well-being during a pandemic, may help assess which children and families will need special attention during and after the lockdown.

The reviewed evidence shows that children’s psychological well-being was affected in the early stages of the COVID-19 lockdown. Moreover, it is assumed that vulnerable children and families struggle even more during these challenging times (Achterberg et al., 2021, Chung et al., 2020, Coyne et al., 2020, De Kinderombudsman, 2020, Fegert et al., 2020). On top of this, foster children residing in their foster homes are facing additional challenges which are specific to their situation, such as the loss of physical contact with their parents. Nevertheless, evidence also suggests that this period may have increased foster children’s psychosocial well-being by improving the relationship between foster child and foster parent, for example. Subsequently, we assessed the psychosocial well-being of Flemish foster children during the COVID-19 lockdown. Moreover, we also assessed the influence of the foster child – foster parent relationship and of contact with birth parents on the foster children’s psychosocial well-being. Finally, it was examined what other factors were associated with the change in psychosocial well-being of Flemish foster children during the COVID-19 lockdown.

Method

Procedure

On the 11th of May 2020, all Flemish foster families of the five provincial foster care agencies received an anonymous survey link by email from their foster care worker (in 2019, 5986 foster families were registered (Pleegzorg Vlaanderen, 2019)). Participants were informed that they could end the survey at any time and gave a written consent to the researchers to process and report their results in a confidential manner prior to filling in the online survey. All procedures were approved by the boards of the participating foster care agencies and the study was conducted in accordance with the Declaration of Helsinki Ethical Principles and Good Clinical Practices.

In a period of four weeks, 1490 foster parents started filling in the questionnaire. Four hundred and twenty-three (28.39%) participants were excluded from the dataset as they did not correspond to the inclusion criteria: (1) being a Dutch-speaking foster parent (2) of a foster child with a minimum age of 2.5 years (3) that was residing in the foster family at the time of the survey. When participants had more than one foster child over the age of 2.5 years living at home during lockdown, they were instructed to keep the child in mind whose name first occurs in the alphabet while filling in the survey. One hundred and seventy-nine participants were categorized as non-responders because of missing data. Therefore, the final data set comprised of 888 foster parents whose foster child resided at their home during the COVID-19 lockdown (estimated response rate is 21%).

A comparison between the responders and non-responders on the data collected from both groups revealed that responders did not significantly differ from non-responders regarding to the age and sex of their foster child, the duration of placement, if the placement was court-ordered and the type of foster care. However, there was a significant difference between both groups regarding the type of foster family. Non-responders were significantly more often kinship carers (χ2 = 8.994, p =.003, r = 0.092).

Participants

Eight hundred and eighty-eight foster parents, which were mostly Western (99%) foster mothers (68.5%) with a mean age of 50 years (SD = 10.84), filled in the survey. They reported on 459 boys (51.7%) and 428 girls (48.3%) with a mean age of 10 years (SD = 4.77). Most foster children lived in two-parents households (79%). The mean number of foster children and other children living at home was respectively 1 (SD = 0.56) and 0 (SD = 1.19). Almost 90% of placements was long-term with a mean duration of 70 months (SD = 53.93). About 43% of foster children stayed in kinship foster care and most placements (70.3%) were court-ordered. Although 24.2% of foster children had no contact with their birth parents before the lockdown, approximately 69% met at least every two weeks one or both parents and 45% was in contact with them through alternative ways (e.g., (video)calls, messages) (Table 1 ).

Table 1

Characteristics Foster Child, Foster Parent, Foster Family and Placement.

Variable		Range/Values	n(%)/M(SD)	N
Characteristics FC
	Age (years)	(2–23)	10.04 (4.77)	888
	Gender	Boy	459 (51.7)	887
		Girl	428 (48.3)
Characteristics FP
	Age (years)	(23–83)	50.34(10.84)	873
	Gender	Man	277 (31.5)	879
		Woman	602 (68.5)
	Ethnicity	Western	863 (99.0)	872
		Non-western	9 (1.0)
Characteristics FF
	Household composition	Single-parent	184 (21)	878
		Two-parent	694 (79)
	(Non–)Kinship	Non-Kinship	503 (56.6)	888
		Kinship	385 (43.4)
	Number of foster children at home	(1–3)	1.34 (0.56)	859
	Number of other children at home	(0–7)	0.94 (1.19)	879
Characteristics of placement
	Duration placement (months)	(1–259)	70.29 (53.93)	886
	Type placement	Supportive foster care	38 (4.4)	872
		Short-term	51 (5.8)
		Long-term	783 (89.8)
	Court-ordered	Yes	624 (70.3)	887
		No	263 (29.7)
	Contact with BPs before lockdown	No contact	206 (24.2)	852
		Only physical contact	255 (29.9)
		Only alternative contact	45 (5.3)
		Both	346 (39.0)
	Amount of physical contact with BPs before lockdown (times/month)	(0–32)	2.49 (4.19)	860
	Amount of alternative contact with BPs before lockdown (times/month)	(0–42)	2.89 (5.77)	853

Note. FC = foster child, FP = foster parent, FF = foster family, BP = birth parent.

Instruments

An online survey using Qualtrics was constructed, consisting of three parts: (1) characteristics of foster child, foster parent, foster family and foster placement, (2) characteristics of COVID-19 lockdown and (3) psychosocial well-being of the foster child, reported on by the foster parents.

Characteristics foster child, foster parent, foster family and foster placement

Demographic information of foster child collected was: age and gender. Demographic information of foster parents collected was: age, gender and ethnicity (western/other). Characteristics of foster family collected were: household composition (single-parent/two-parent), (non–)kinship care, number of foster children and other children at home during lockdown. Characteristics of foster placement collected were: duration placement (in months), type of placement (supportive/short-term/long-term) and if the placement was court-ordered.

Characteristics of COVID-19 lockdown

The negative aspects of the COVID-19 lockdown were measured with seven statements developed by the researchers and scored on a five-point scale (1 = totally disagree, 5 = totally agree). Six items investigated the suffering of foster children during lockdown for several reasons; not able to (1) see parents, (2) see siblings, (3) see friends, (4) go to school, (5) have leisure activities out of the foster home or (6) loss of additional help (e.g., therapy). When a foster child could not suffer from aforementioned changes as they were not applicable (for example, the child has no siblings), the score of 1 was given. A seventh statement inquired about the influence of the lockdown on the foster child – foster parent relationship, reported by the foster parents. Foster parents were asked if the relationship with their foster child deteriorated since the COVID-19 measures were instated (1 = relationship deteriorated a lot, 5 = relationship improved a lot).

Fear of COVID-19 contamination of the foster child (for (foster) family members and self) reported by the foster parent was measured with five questions developed by the researchers and scored on a five-point scale (1 = not worried at all, 5 = very worried; sample item: how worried is your foster child of getting contaminated with COVID-19?). The sum of all items represented the total fear of the foster child for COVID-19 contamination (α = 0.87).

Contact with birth parents before and during the COVID-19 lockdown was also examined. Foster parents were asked to report how many times a month the foster child had (1) physical and (2) alternative contact (e.g., (video)calls, messages) with its birth parents before and during the lockdown. Afterwards a difference score for both physical and alternative contact was computed by subtracting the total amount of physical and alternative contact during the lockdown from the total amount of respectively physical and alternative contact before the lockdown.

Psychosocial well-being foster child

Psychosocial well-being of foster children both before and during the lockdown was measured with the Brief Assessment Checklist for Children (BAC-C) and the Brief Assessment Checklist for Adolescents (BAC-A) (Tarren-Sweeny, 2012). Both caregiver-reports consist of 20 items scored on a 3-point scale (0 = not true/has not occurred, 1 = a little true/has occurred once, 2 = mostly true/has occurred more than once) measuring mental health problems in foster children. A total score of 7 or higher indicates clinical-level mental health problems. Good psychometric properties and validity were previously reported (Goemans et al., 2018, Tarren-Sweeney, 2013). Foster parents were instructed to fill in the BAC-C/A twice; once for the period prior to the lockdown and once for the present time (during the lockdown). Afterwards a difference in BAC-score was computed. Internal consistency in the actual study was good for both the present time (α = 0.84) and for the period prior to the lockdown (α = 0.83).

Data analysis

First, descriptive analysis was done. Second, several aspects of suffering due to the lockdown were compared using one-way repeated-measures ANOVA. A Greenhouse-Geisser correction was used as the assumption of sphericity was violated, and a pairwise comparison with Bonferroni correction for multiple comparisons was done. Third, paired samples t-tests, were used to assess the differences before and during the lockdown regarding contact and psychosocial well-being. Fourth, one-way ANOVA’s were performed to assess the influence of the type of contact with birth parents during the lockdown on the difference in psychosocial well-being. Pearson correlation was used to examine the association between the difference in both physical and alternative contact on one hand with the difference in psychosocial well-being on the other hand. Finally, associations between the difference in psychosocial well-being and the remaining characteristics of foster child, foster parent, foster family, foster placement and lockdown were examined by Pearson correlations, independent t-tests and one-way ANOVA’s. Effect sizes r were computed (Cohen, 1988). All characteristics significantly associated with the difference in psychosocial well-being at a p ≤ 0.10 level (Bursac et al., 2008) were entered in a multiple regression. Standard errors were bootstrapped as standardized residuals were not normally distributed. The remaining assumptions of multiple linear regression were met.

Results

Characteristics of COVID-19 lockdown

A repeated measures ANOVA showed that several negative aspects of the lockdown differed significantly from each other (Greenhouse-Geisser F(4.07, 5288.32) = 436.79, p < .001). The highest negative aspect of the lockdown for foster children according to their foster parents was suffering from not seeing friends (p < .001), followed by suffering from having no leisure activities outside of the foster home (p < .001, p < .001, p < .001, p = .028) and suffering from not going to school (p < .001, p < .001, p < .001), which were significantly higher than the remaining negative aspects of the lockdown. The mean score regarding the influence of the COVID-19 measures on the relationship between foster parent and foster child was 3.41 (SD = .96, min 1 and max 5) suggesting a small improvement during the lockdown reported by the foster parents. Only few foster parents (10.6%) reported that the relationship with their foster child deteriorated during lockdown, while 36.8% reported that their relationship improved and 52.6% reported a status quo. Compared to the anchor points (min 5 and max 25), foster children did not fear COVID-19 contaminations (M = 9.38, SD = 4.59) according to their foster parents (Table 2 ).

Table 2

Characteristics COVID-19 lockdown.

Variable	Range/Values	n(%)/M(SD)	N
Negative impact lockdown on FC
Not seeing BP as before	(1–5)	2.26 (1.35)	887
Not seeing brothers/sisters	(1–5)	2.26 (1.44)	887
Not seeing friends	(1–5)	3.82 (1.09)	887
Not going to school	(1–5)	3.53 (1.18)	887
No leisure activities outside of foster home	(1–5)	3.62 (1.14)	888
Loss of additional help	(1–5)	2.26 (1.33)	887
Improved relationship FC-FP	(1–5)	3.41 (0.96)	888
	Very deteriorated	25 (2.8)
	A little deteriorated	69 (7.8)
	Unchanged	467 (52.6)
	A little improved	170 (19.1)
	Very improved	157 (17.7)
Fear of COVID-19 contaminations	(5–25)	9.38 (4.59)	868
Self	(1–5)	1.89 (1.08)	877
Household members of FF	(1–5)	2.18 (1.29)	873
Other members of FF	(1–5)	1.97 (1.17)	875
Household members of BF	(1–5)	1.73 (1.05)	873
Other members of BF	(1–5)	1.60 (1.01)	870
Contact with BPs during lockdown	No contact	267 (30.7)	871
	Only physical contact	36 (4.1)
	Only alternative contact	406 (46.6)
	Both	162 (18.6)
Amount of physical contact with BPs during lockdown (times/month)	(0–31)	.85 (3.37)	855
Amount of alternative contact with BPs during lockdown (times/month)	(0–50)	4.90 (7.55)	881

Note. FC = foster child, FP = foster parent, FF = foster family, BF = birth family, BP = birth parent.

Contact with birth parents before and during lockdown

During lockdown, 267 or 30.7% of foster children had no contact with their birth parents, as reported by their foster parents, which is an increase of approximately 6% compared to the period prior to lockdown (24.2%). Families switched to alternative contact during lockdown which is reflected by the increase from 5.3% to 46.6% of foster children that exclusively had alternative contact with their birth parents. The amount of alternative contact with birth parents significantly increased during lockdown (t(8 5 2) = -11.86, p < .001, r = .38). Conversely, the amount of physical contact with birth parents significantly decreased during lockdown (t(8 4 2) = 13.44, p < .001, r = .42) (Table 1 and Table 2).

Psychosocial well-being of foster children

There was no significant difference (t(8 8 7) = -.44, p = .66, r = .015) between the BAC-scores before (M = 11.23, SD = 6.87) and during the lockdown (M = 11.28, SD = 7.19). Moreover, 44.9% of foster children had the same total BAC-score before and during the lockdown. The total BAC-score decreased according to 27.8% and increased according to 27.3% of the foster parents during lockdown.

Influence of contact with birth parents during lockdown on psychosocial well-being

The type of contact with birth parents during the lockdown was significantly associated with the difference score in BAC (F(3,867) = 3.833, p = .010), whereas the type of contact with birth parents before the lockdown was not. Post-hoc tests using Bonferroni correction showed that the BAC-score of foster children who had both alternative and physical contact increased significantly less, compared to foster children who had no (p = .005) or only physical contact (p = .009) during lockdown. Moreover, the difference in alternative contact was significantly associated with the difference in BAC (r = -.97, p = .004), whereas the difference in physical contact was not. When alternative contact increased during lockdown, the BAC-score decreased during lockdown.

Other determinants of changed psychosocial well-being during lockdown

As indicated in Table 3 , the following characteristics were significantly associated with an increased BAC-score during lockdown: a voluntary foster placement, a deteriorated relationship between foster child and foster parent, having only physical or no contact during lockdown, decreased alternative contact during lockdown, suffering from not seeing friends, from not going to school, from having no leisure activities outside of the foster home, and from losing additional help during lockdown. Except from the trivial association between difference in alternative contact with birth parents and the difference in total BAC-score (r < .10), all significant associations were small (.10 < r < .30).

Table 3

Associations difference in BAC of Foster Children and Characteristics.

Variable		Difference BACPearson r/t/F		Effect sizer	N
Foster child characteristics
	Age	r = -.022		.022	888
	Gender	t = .661		.044	887
Foster family characteristics
	Age	r = -.034		.034	873
	Gender	t = -.413		.030	879
	Ethnicity	t = .575		.193	872
	Household composition	t = 1.287		.107	878
	Duration placement (months)	r = -.019		.019
	(Non–)Kinship	t = .551		.036
	Number of foster children at home	r = .041		.041	859
	Number of other children at home	r = -.005		.005	879
Foster placement characteristics
	Type of placement	F = .167		.000	872
	Court-ordered	t = 2.128*		.168	887
	Contact with BPs before lockdown	F = .575		.090	852
Influence lockdown
	Negative impact lockdown on FC
	Not seeing BPs as before	r = .044		.106	884
	Not seeing brothers/sisters	r = .030		.053	885
	Not seeing friends	r = .165***		.165	887
	Not going to school	r = .207***		.207	886
	No leisure activities outside of foster home	r = .197***		.197	888
	Loss of additional help	r = .112***		.133	888
	Improved relationship FC-FP	r = -.292***		.292	888
	Fear of COVID-19 contaminations	r = .044		.044	868
	Contact with BPs during lockdown	F = 3.83**		.230	871
	Difference in physical contact	r = .019		.019	843
	Difference in alternative contact	r = -.097**		.097	853

Note. BP = birth parent, BAC = Brief Assessment Checklist, FC = foster child, FP = foster parent, gender (0 = man, 1 = woman), ethnicity (0 = western, 1 = non-western), household composition (0 = single-parent, 1 = two parents), type of placement (0 = supportive, 1 = short-term, 2 = long-term), (Non–)Kinship (0 = non-kinship, 1 = kinship), court-ordered (0 = no, 1 = yes), contact with BPs before/during lockdown (0 = no, 1 = only physical, 2 = only alternative, 3 = both).

*p ≤ 0.05, **p ≤ 0.01, ***p <.001.

Subsequently, all significant factors were entered in one multiple linear regression model, resulting in a significant model (F(10,841) = 14.621, p < .001) explaining about 15% of variance with seven significant variables (Table 4 ). Foster children whose placement was voluntary, whose amount of alternative contact with their birth parents decreased during lockdown, who had only physical, only alternative or no contact during lockdown in comparison to both physical and alternative contact, who suffered more from not being able to go to school and whose relationship with their foster parent deteriorated during lockdown, had a higher difference in their BAC-score. Improved relationship between foster child and foster parent contributed most to the model (β = -.255, p < .001).

Table 4

Multiple linear Regression Model with difference in BAC-score as Dependent Variable.

		β	B	C.I.
Characteristics foster placement
	Court-ordered	-.081*	-.556	-.990 – -.126
Influence lockdown
	Type of contact with BPs during lockdown
	No contact (ref: both)	.126**	.855	.258 – 1.452
	Only physical contact (ref: both)	.075*	1.209	.094 – 2.323
	Only alternative contact (ref: both)	.095*	.594	.059 – 1.128
	Difference in alternative contact	-.085*	-.051	-.091 – -.011
	Not seeing friends	.047	.136	-.098 – .370
	Not going to school	.134***	.360	.144 – .576
	No leisure activities outside of foster home	.043	.119	-.104 – .343
	Loss of additional help	.035	.082	-.070 – .234
	Improved relationship FC-FP	-.255***	-.833	−1.042 – -.624
R2		.148
F		14.621***
N		851

Note. BP = birth parent, FC = foster child, FP = foster parent, court-ordered (0 = no, 1 = yes).

*p ≤ 0.05, **p ≤ 0.01, ***p <.001.

Discussion

In this study, the psychosocial well-being of foster children residing with their foster parents during the COVID-19 lockdown in Flanders was examined. The psychosocial well-being of foster children whose foster parents completed this survey, was not negatively associated with the COVID-19 lockdown. This result is reassuring and points out that these foster children residing in their foster homes were resilient during the early stages of the pandemic and were able to cope with the challenges that were put upon them (Agentschap Opgroeien, 2020, UNICEF., 2020). However, it is important to take into account that this study was performed at the start of the COVID-19 pandemic in Flanders. At the time it was still unknown how long the pandemic would last and society was not prepared for the multiple and sustained restrictive measures that would follow. This implicates that the sustained stressors during the whole period may impact the foster children’s psychosocial well-being even more severely. Therefore, these results only reflect the psychosocial well-being of foster children at the initial stage of the pandemic. Moreover, it is important to emphasize that these results only reflect the psychosocial well-being of the foster children whose foster parents filled in the survey. As the response rate is approximately 21%, results cannot be generalized to all Flemish foster children residing in their foster homes during the COVID-19 lockdown. It is possible that foster parents who responded to the survey were the ones who care more about how to create a better environment for their foster children. Hence, this might explain why their psychosocial well-being was not negatively associated with the COVID-19-lockdown.

Another important side note is that the lack of decrease in psychosocial well-being of foster children residing in their foster homes during the lockdown is measured on a group-level but that individual differences exist. On the one hand, the psychosocial well-being of approximately one quarter of the foster children did decrease during lockdown, pointing out that, according to their foster parents, some children did have more difficulty coping and were suffering during the pandemic. Our results show that suffering from not being able to go to school was negatively associated with an increased psychosocial well-being during the lockdown. This proves that, as expected and suggested by previous studies (e.g., Achterberg et al., 2021, De Kinderombudsman, 2020, Jiao et al., 2020), the restrictive measures taken to control the pandemic did influence the psychosocial well-being of some Flemish foster children. Moreover foster children seem to suffer most from not being able to see their friends or go to school and not having leisure activities outside of the foster home. All of these activities include social interaction with peers which was heavily restricted during lockdown but is of importance for the development and well-being of children (Achterberg et al., 2021, De Kinderombudsman, 2020, Kroger et al., 2010, Orben et al., 2020, Ragelienė, 2016). Not being able to see their family of origin (parents and siblings) and the loss of additional help was reported as less impactful by the foster parents. Seeing birth parents and going to therapy can be characterized as emotionally stressful (Andresen et al., 2020, Poehlmann-Tynan et al., 2015, Simms, 1991) which may result in a (temporary) sense of relief when not taking place. Moreover, only 6% of foster children had no contact at all with their birth parents during lockdown which shows that most families did switch to alternative contact such as (video)calls and messaging, possibly explaining why the loss of physical contact was not experienced as very harmful. However, the loss of physical contact and switch to alternative contact may have a greater impact when it drags on longer during consecutive lockdowns.

On the other hand, foster parents of approximately one quarter of Flemish foster children residing in their foster homes during lockdown reported an increase in their foster child’s psychosocial well-being, confirming the idea of several scholars that the lockdown might have positive consequences (Bruining et al., 2020, Isumi et al., 2020). Our results show that the lockdown was positively associated with the relationship between foster child and foster parent, similar to what has been found in the general population (Achterberg et al., 2021, Brown et al., 2020, Bruining et al., 2020, Courtney et al., 2020, De Kinderombudsman, 2020, Fegert et al., 2020, Hoekstra, 2020, Isumi et al., 2020, Vlaams Kinderrechtencommissariaat, 2020). As for the general population, the additional time foster children and foster parents can spend together may explain this improved relationship. Moreover, this was the most important determinant of the changed psychosocial well-being during the lockdown. According to foster parents, the psychosocial well-being of foster children during lockdown was positively associated with the foster child – foster parent relationship. This result is not surprising as foster parents provide a more safe and stable environment for the foster child when their safety was at risk and/or their birth parents were unable to properly care for them (Pleegzorg Vlaanderen, 2020). Hence, a prolonged stay with them may be beneficial for one’s psychosocial well-being.

As expected, the amount of respectively physical and alternative contact that Flemish foster children residing in their foster homes had with their birth parents during lockdown significantly decreased and increased. The type and amount of contact during the lockdown was also related with the changed psychosocial well-being during the lockdown. A significant increase in psychosocial well-being was observed in foster children who had both types of contact in comparison with those who had no or only physical contact with their birth parents during lockdown. A marginal significant increase in psychosocial well-being was observed in foster children who had only alternative contact during lockdown, compared to those who had only physical contact. Additionally, an increase in alternative contact during the lockdown was significantly positively associated with an increase in psychosocial well-being of foster children, whereas the significant decrease in physical contact was not associated with the changed psychosocial well-being. These findings confirm that contact with birth parents is protective (McWey et al., 2010, McWey and Cui, 2017, Nixon et al., 2013) and that having contact is better than no contact at all. However, these findings also point out the importance of alternative contact. Even though it could be potentially stressful, it seems that alternative contact may be a good substitute for physical contact when the latter is prohibited. Physical contact can be emotionally very demanding for foster children (e.g., parents who cancel last minute or do not show at all, parents who are intoxicated during visits, etc.). The loss of physical contact may, therefore, actually be relieving, especially when it is substituted by a safer alternative. This may explain why the significant decrease of physical contact during the lockdown was not associated with the changed psychosocial well-being. Furthermore, alternative contact may in many ways be experienced as safer by foster children. It is often shorter in length, more predictable and can be sustained on a more regular base which makes it a good possible alternative when physical contacts are emotionally too demanding (National Association of Counsel for Children, 2020, Pitzl, 2020).

Finally, being placed by a juvenile judge was significantly associated with an increased psychosocial well-being during lockdown compared to voluntary placements. A possible explanation is the postponing of the annual court hearings due the lockdown and the (temporary) sense of relief that may accompany this postponement. These court hearings are very stressful for foster children as the juvenile judge decides on the (lack of) necessity of the foster care placement and therefore also the foster child’s living arrangement. Also, there is often more tension between birth parents and the foster child/family in court-ordered placements than in voluntary placements. The loss of physical contact may therefore be experienced as more relieving by foster children in court-ordered placements (Poehlmann-Tynan et al., 2015, Simms, 1991).

Implications for practice

This study has some important implications for practice. First, this study emphasizes the individual differences of foster children in handling challenging times. The lockdown and loss of physical contact with birth parents for example was experienced as positive by some and as negative by other foster children. Hence, it is very important to not be led by assumptions that apply for the whole group but to look at each case individually. Second, our results do reveal some positive aspects the lockdown for foster children residing in their foster homes, such as the improved relationship between foster child and foster parent. Moreover, an improved foster child – foster parent relationship was associated with increased rates of psychosocial well-being in foster children during adverse circumstances. Therefore it is important to focus on the foster parent – foster child relationship and commit to improving it by conducting further research into what helps to achieve a better relationship. At this moment, our results suggest that spending time with each other promotes a good relationship. Third, alternative contact with birth parents such as (video)calls and messages can be beneficial for foster children’s psychosocial well-being in situations where physical contact is not possible. Hence, it is important to explore further how alternative contact may be implemented more structurally into foster care placements. As alternative contact is less time consuming than physical contact, maybe both aims can be purchased. By promoting alternative contact with the birth parents, more time for foster parent – foster child activities can be realized.

Limitations

This study has some limitations. First, our results cannot be generalized to all Flemish foster children residing in their foster homes during the lockdown due to the fairly low response-rate (approximately 21%). Even if the response and non-response group did not differ from each other regarding the age and sex of their foster child, the duration of placement, if the placement was court-ordered and the type of foster care, there still may be important differences present. Second, this study was conducted at the start of the COVID-19 pandemic in Flanders. Therefore, the results only reflect the initial impact of the pandemic on the psychosocial well-being of foster children. Replicating this study at a later stage may reveal different results due to the sustained stressors of the consecutive lockdowns that followed. Third, only 15% of the variability was explained in our multiple linear regression model. However, only changes during the lockdown were measured and implemented in the model. Several other aspects of foster children’s life may impact their psychosocial well-being and were not included in this study. Moreover, it would be interesting to include the quality of the relationship between foster child and foster parent prior to the lockdown. This enables to examine the hypothesis that the less positive the prior relationship, the more children rely on outside relationships and thus the more they suffer from not seeing their friends in times of distress. Fourth, the results are based on the reports of foster parents which may not reflect the opinion of the foster children. As no past assessments of foster children’s well-being were at our disposal, foster parents were instructed to fill in one and the same instrument for two time periods. This may lead to a recall bias considering the behavior and mood of the foster child, as well as the contact with the birth parents in the period prior to the lockdown. Additionally, there is a possibility that the responses on both counts interact with each other which may impair the internal validity (Campbell & Stanley, 1963). Finally, the instruments used to measure the negative influence of the lockdown on the foster child and their fear of COVID-19 contaminations were developed by the researchers and the validity was not examined.

Conclusion

The COVID-19 lockdown was not associated with a decreased psychosocial well-being reported by foster parents on Flemish foster children residing in their foster homes during the lockdown. Moreover, foster parents reported a slight improvement in their relationship with their foster child during the lockdown which points to a positive consequence of the lockdown. In addition, this improvement was positively associated with an increased psychosocial well-being during adverse circumstances and should therefore be enhanced. The type and amount of contact is significantly associated with foster children’s changed psychosocial well-being during the lockdown. Foster parents who reported an increase in alternative contact between their foster child and his/her birth parents during the lockdown, also reported an increase in their foster child’s psychosocial well-being during that period.

CRediT authorship contribution statement

Camille Verheyden: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft. Frank Van Holen: Conceptualization, Methodology, Investigation, Writing – review & editing. Delphine West: Writing – review & editing. Johan Vanderfaeillie: Conceptualization, Methodology, Investigation, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.