Background: Foster care is a relatively common arrangement when parents are unable to meet the needs of their children. Specialized foster care is sometimes applied in cases when problems are more serious and complex. More knowledge is needed about the effects of such specialized foster care.

Aims: To explore trajectories of change associated with specialized foster care in a treatment model with a relational and mentalization-based orientation, and to develop the understanding of foster children’s and their foster parents’ experiences of their relationship living in a treatment foster family.

Methods: Children and young people between the ages of 5 and 20 years who received treatment within a specialized foster care model, Treatment By Foster care (TBF), participated in this study. Longitudinal data collected in a naturalistic setting were analyzed quantitatively. In Studies 1 and 2, the number of participants at baseline varied for different instruments between 76 – 105. The Achenbach System of Empirically Based Assessment (ASEBA) was used to measure how psychiatric symptoms change from the perspectives of the foster children, the foster parents, and teachers. The Adaptive Behavior Assessment System – second edition (ABAS-II) was used to measure adaptive functioning from the foster parents’ perspective. Self-ratings by the children and young people of their emotional and social problems were measured with the Beck Youth Inventories of Emotional and Social Impairment (BYI). Data about experiences of the relationship between child and foster parent were collected through repeated individual short interviews/speeches with both children and their foster parents according to Five Minute Speech Sample (FMSS) (n = 14). Interviews/speeches were analyzed using Thematic Analysis (TA).

Results: The ratings of foster parents and foster children differed. The analyses showed a significant reduction in psychiatric symptoms, emotional and social problems according to the self-ratings by the children and young people. According to foster parents and teachers, psychiatric symptoms did not decrease. The baseline ratings of adaptive functioning by foster parents showed that adaptive functioning was considerably below peers from the Swedish non-clinical norm group. Adaptive functioning did improve but not enough to approach or catch up with peers.

Analysis of the interviews/speeches generated three main themes containing seven subthemes. Main themes were: No 'real' family, A co-created relationship, and Time. Participants related to a norm for what a 'real' family is and seemed to presuppose that the biological family is the 'real' family. A co-created relationship related to No 'real' family as an answer or a solution. The challenges in the foster family constellation could be overcome by a mutual ambition to build a relationship and by liking each other. Time appeared as a common theme and both as an opportunity and a threat to the relationship. Despite the fact that no interview question concerned the duration of the relationship, the participants described their relationship based on how long they had known each other.

Conclusions: According to the foster children’s and young people’s self-ratings, their mental health improved, and their social problems decreased. It is likely that the TBF-model contributed to this improved psychological well-being, although causal relationships could not be established without any comparison group. However, the model did not seem to contribute to the foster parents experiencing improvement in the foster children’s psychological well-being or adaptive functioning.

Based on the results of this thesis, it may be effective to place children and young people in specialized foster care with a relational and mentalization-oriented focus, but the results are not clear-cut. Practice and policies should take greater account of the time aspect in foster care, and work with the aim of increasing clarity and security, and thereby enabling a more stable upbringing for some of society’s most vulnerable children. Also, this may make foster parents want to continue their mission.

More studies are needed to gain knowledge about how specialized foster care should be applied. Future studies also need to focus on creating knowledge about which aspects of the treatment are decisive.

Knowledge about the development of mental health in young people in foster care is limited. This naturalistic study examined the effects of a relational and mentalization-focused treatment in foster families in Sweden on the placed young peoples mental health. The Achenbach System of Empirically Based Assessment (ASEBA) was used to measure change in psychiatric symptoms. Self-ratings showed significant improvements and medium to strong effects after 24 months in both boys and girls. No significant changes were found in the foster parents ratings or in the school staffs ratings. Foster parents ratings suggested that girls behavioral problems decreased, but not the boys. Based on these findings, we want to emphasize the importance of evaluating treatment effects using self-ratings by the young people in addition to parents and parent substitutes ratings.

Various models of specialized foster care have been developed, but research on them is limited. This longitudinal, exploratory study analysed data on adaptive functioning, emotional and social problems and self-concept in a specialized foster care service in Sweden. The focus of the study was on the development of the children and young people in placement. The Adaptive Behaviour Assessment System (ABAS-II) was used to measure adaptive functioning, and the Beck Youth Inventories of Emotional and Social Impairment (BYI) was used to measure self-rated emotional and social problems and self-concept. Self-ratings showed significant improvements in disruptive behaviour, anger, anxiety and depression. Adaptive functioning as rated by foster parents improved but not enough to catch up with the non-clinical norm group. The average adaptive functioning among the participants at baseline was considerably below the Swedish norm group. Similar to the results of a previous study of the same treatment model, children and young people rated improvement while their foster parents did not do so to the same extent. Possible explanations for this are discussed in the paper. The study is limited by the lack of a control group and by data attrition.