The findings from this study demonstrate that post-traumatic stress in mothers, but not fathers, was associated with increased use of psychiatric services by foreign-born refugee children as well as in Swedish-born children with refugee parents. Particularly high risks were seen in more recently arrived refugee adolescents and for diagnoses of introverted problems.

In the few previous studies on consequences of parental post-traumatic stress for psychiatric care contacts among refugee children, the findings have been conflicting. A recent Danish study reported no risk differences between children of parents who had been treated for war or torture trauma at specialist centres and non-exposed children of parents originating from the same region [13]. This apparent lack of association between parental trauma exposure and offspring psychiatric disorders may be explained by misclassification of exposure, diluting the association, a limitation also mentioned by the authors. Other potential explanations of these findings include important post-migration factors. Consistent with our findings [7] and previous findings [6, 21], children of foreign-born parents in this Danish study had fewer hospital contacts, indicating an underutilisation of mental health services, the existence of barriers to care, and an unmet need for healthcare, rather than lower levels of mental health problems. Another recent Danish study demonstrated an increased risk of psychiatric hospital contacts related to parental diagnosis of PTSD, defined by hospital contacts only [12]. This study demonstrated an increased risk associated with a diagnosis of PTSD in fathers as well as mothers, contrasting with our findings with regards to fathers. Our study, which includes information on maternal and paternal post-traumatic stress diagnosed on three levels of care and information from child and adolescent psychiatric services, provides further support for the significance of intergenerational effects of psychological trauma among refugees in exile.

The majority of previous studies on correlations between parental and children’s psychopathology has focused solely on mother–child dyads [22]. One previous study investigating quantity and quality of father involvement and the influence of post-traumatic stress in a population of refugees and asylum seekers in the Netherlands found fathers to be less involved than mothers in caregiving interactions, but no differences were seen in quality of parent–child involvement [23]. Even considering fathers increased involvement as caregivers, if the mother is still more often the primary caregiver this could potentially contribute to the explanation of the gender difference observed in our study. Furthermore, the authors of the Dutch study argue that fathers might have more opportunities to withdraw from family-interaction when symptoms of stress worsen and by doing so diminishing the negative impact of trauma-related stress on their children [23]. It can also be hypothesized, as a contributing explanation to this difference, that traumatic experiences that are shared by parent and child are more common for mothers and children than for fathers and children. Given the conflicting findings in existing studies and the important questions these findings raise, further studies are needed to clarify these patterns, preferably studies that include interviews with parents and children.

The diagnostic patterns described in Table 5 show a similar pattern with and without exposure to parental post-traumatic stress in the Swedish-born patients. In contrast, a very clear diagnostic pattern was found in the foreign-born patients exposed to maternal post-traumatic stress with a twofold elevated risk of having a diagnosis in the anxiety and depression chapters. A similar high burden of introverted psychiatric symptoms has been found in multiple Scandinavian studies of refugee children and adolescents during the first years after settlement [9]. These symptoms have been described to be associated with the children’s own exposure to organized violence and migration stress, with the exposure often being shared by children and their mothers. Thus, we could not detect any specific diagnostic pattern associated with exposure to parental post-traumatic stress in this study.

A particularly high use of psychiatric services in relation to maternal post-traumatic stress was seen among adolescents with a shorter duration of residence. Although previous studies have demonstrated an underutilisation of mental health services, particularly during the first years after settlement, longitudinal studies in Scandinavia have also shown very high prevalence of poor mental health shortly after settlement that tended to decrease with increasing duration of residence [9, 24, 25].

The increased use of psychiatric services among children of parents treated for post-traumatic stress was not limited to foreign-born children. Thus, an increased need for post-traumatic psychiatric treatment following their own experiences among children who were themselves refugees did not fully explain these associations. Although few studies have examined children of refugees, intergenerational transmission of trauma has been investigated at length in clinical studies with offspring of Holocaust survivors and among children of soldiers [26, 27]. Previous findings have suggested that psychological and social consequences of traumatic experiences may affect health and well-being of the children [9, 10]. Healthy development and psychological well-being of a child are closely linked to the health of the caregiver, and a positive relationship between child and caregiver is crucial. The literature suggests different mechanisms in the intergenerational transmission of trauma, mainly focusing on parental trauma-related mental health problems, caregiving behavior, and disrupted attachment [28, 29]. Caregiver mental health problems following traumatic experiences may have adverse effects on parents’ emotional and behavioral availability, undermine the parents’ ability to provide support and a sense of security for their children and have consequences for parent–child interaction and attachment [30, 31]. Well-being and mental health of their children may also be negatively affected as a result of poor family functioning and instability and parenting style [11, 31, 32]. Previous studies have also highlighted increased risks of harsh parenting and family violence in families affected by post-traumatic stress [33, 34]. Future research should focus on these mechanisms or pathways through which these associations are mediated.

A major strength of the present study is the availability of data from specialized treatment centres and primary care, in addition to the previously validated data on post-traumatic stress from specialist care as the use of hospitalization data only might underestimate psychiatric contacts. Other strengths include the use of register data and the large study population comprising all refugee children and adolescents residing in the county of Stockholm between 2011 and 2017. The data from reliable national and local registers allowed for adjustment for a wide range of potentially confounding or mediating variables, including socioeconomic factors and migration-related variables.

The prevalence of post-traumatic stress in refugee parents in this study was lower compared to what has been seen in many adult refugee populations [2]. Previous research has highlighted several barriers to psychiatric care [8, 35], suggesting that not everyone suffering from psychological consequences of trauma seeks medical attention, which could potentially lead to an underestimation of the affected parents.

Child psychiatric services in Stockholm, the largest of Sweden’s health care organizations, has considerable resources. These services are free and can be arranged through self-referral from parents as well as schools. Child psychiatric services are probably more accessible than in many other contexts, and as many as 10% of adolescents in Stockholm between 13 and 17 years old have at least one contact with these services annually [36].

One limitation of this study is that the register data do not include information on children in asylum-seeking families and children of undocumented parents. Other limitations are the lack of information on actual health status and health care need in the registries and the potential risk of detection bias (i.e., children with parents receiving care may be more likely to receive care themselves, resulting in a potential overestimation of the associations).

The findings from this study, based on data for more than 16000 children with a refugee background, demonstrated that refugee adolescents with a mother treated for post-traumatic stress were more than twice as likely to have been in contact with child and adolescent psychiatric services. These findings emphasize that efforts to improve mental health of refugee children and adolescents need to include mental health interventions across generations, ensuring that parents in need of post-traumatic psychiatric treatment and their children are offered correct treatment and support. As demonstrated previously [7, 21], refugee children and adolescents are less likely to use psychiatric healthcare services compared with their majority population peers. For refugee adolescents to access appropriate health care services when such services are needed, there is a need to address the different types of barriers to care. It is also important to acknowledge the important role of schools [37] in identifying mental health issues in refugee children and adolescences and facilitating access to mental health services.