Our second objective aimed to assess the use of mental health services. Our findings are the first to report on comparative rates of psychiatric emergency department use between hospitalised youth placed in FC relative to those who are not in alternative care. Patients in FC showed substantially higher rates of psychiatric emergency department use before and after the index hospitalisation and were more likely to reconsult the emergency department during the follow-up period and in a shorter time after discharge. Moreover, youth in FC were 1.5 times more likely to be readmitted to the hospital, although the time until readmission was not significantly different between groups. Our findings also suggest that hospitalisation is not equally effective for both groups. While hospitalisation was associated with more robust and long-lasting effects in the control group, it was less likely to substantially change the pattern of psychiatric emergency department use among foster children (see Fig. 3). Adjusting models for eating disorders did not change the results (see supplemental material). In fact, the length of hospitalisation was no longer significant and differences in both the frequency and the time to readmission were statistically significant. Our findings are in line with those of Stein et al.’s study of children investigated by child welfare [48]. Given the recognition of existing deficits in mental health services and the barriers to receiving community-based treatment, the increased use of acute health services is of particular concern as it suggests that the existing mental health service model is not appropriate and that the needs of this specific population are not being fully met. Addressing these disparities in service use is an important step toward building a more evidence-based and specialist community-based mental health system, as generic services do not do justice to the complexity of these children [49]. Additionally, improving the adequacy and quality of care of current services, both in the inpatient unit and in the emergency department, should be a priority.

Outpatient clinics are the most frequently used mental health service by youth in FC, with only a small proportion requiring hospital admission [50, 51]. However, standard child mental health treatments do not appear to have the same effect on behavioural or emotional difficulties in foster youth, either because the resources do not meet their needs or because formal mental health services may be insufficient for this population [52, 53]. Our findings indicate that children in FC would benefit from attending specialist outpatient mental health programmes that can provide comprehensive therapeutic approaches that focus on the individual as well as the systems around the child [38]. Given that a subgroup of youth in FC have a complex clinical presentation with multiple comorbidities and a high demand for inpatient and outpatient psychiatric resources, intensive and multidisciplinary care units may be greatly beneficial.

Another notable finding was the high rate of prescription of psychotropic medications among the FC sample, with 98.9% of these subjects taking one or more of these medications; this is consistent with findings from other studies on out-of-home care [49, 54]. These findings indicate that psychotropic medication may be prescribed at particularly elevated rates to youth in FC with a history of hospitalisation. A high risk of these prescribing patterns has been found among youth in FC, who have complex mental health needs and a history of placement changes or hospitalisations [55]. It could be argued that youths in FC who require admission to a psychiatric unit may be more likely to have severe and complex mental health disorders requiring more use of medication than general FC samples. Nevertheless, concerns have been raised that high rates of polypharmacy and off-label use of medications such as antipsychotics may show a tendency for clinicians to rely on these prescriptions instead of providing additional psychosocial interventions and long-term support [56]. Additionally, FC children’s more tenuous connections with caregivers might also result in less advocacy for psychosocial interventions for behavioural problems, less engagement with services, and more fragmented support. The study also points out the difficulty of institutions in facilitating the provision of mental health care and treatment to these children under the same conditions as those living in the household. All stakeholders should invest in evidence-based psychosocial interventions to address behavioural needs and potentially lessen psychotropic medication use and hospitalisation rates [49].

Our findings need to be interpreted in light of several limitations. First, our sample selection focused on more severely ill youth requiring hospitalisation may limit the generalisability of the results to all youth in FC since, arguably, the more severely ill the patients are, the greater the need for medication and intervention. A second limitation is that the control group presented a significantly higher prevalence of eating disorder diagnoses, which tend to be associated with longer lengths of hospital admission [57]. Finally, the retrospective nature of the study limits our capacity to assess causal associations and evaluate other potentially relevant variables or confounders that were missing in the medical records. It is also certainly possible that we are overrepresenting the impact of foster placement on well-being because of the inability to detect all family-level risk factors or other baseline factors. Understanding whether behavioural problems diminish, persist, or increase over time, and the underlying causal relationships, will require prospective, controlled studies.

However, our study also has certain strengths. For instance, the inclusion of a control group comparable in terms of severity grants weight to our findings regarding the clinical complexity observed in youth in FC. Another strength relates to generalisability since our subjects were recruited from the Hospital Clinic de Barcelona, which is an urban, tertiary care facility, that provides services to a representative ethnic and socio-demographic mix of eligible patients.

In summary, our findings provide evidence that children in FC who require admission to a psychiatric unit are a particularly vulnerable population with a complex presentation, who show higher rates of disruptive behaviours and drug misuse than their peers. The results also indicate that FC youth may require higher rates of pharmacological interventions and greater use of the psychiatric emergency department before and after admission. Our findings highlight that the mental health needs of children and adolescents in FC could differ from those never placed out-of-home, and they reinforce the need for prevention strategies and specialist outpatient mental health programmes. Further research is needed to investigate long-term outcomes and evaluate the impact and effectiveness of such interventions.