Experimental Design
We build on an intervention that has been implemented as a survey experiment into the BiB-FReDA-Survey in November 2023 as part of the project AEARCTR-0013354. The intervention aims at destigmatizing mental health support and at providing information to increase uptake among Ukrainian Refugees in Germany. While the project AEARCTR-0013354 studies the effects of the intervention on adult outcomes, we focus on the effects of the intervention on the children of these adults. More precisely, we are interested in the help-seeking action and intentions of parents on behalf of their children, as well as outcomes measuring the children’s well-being, including their life satisfaction and overall health, sense of school belonging, and socio-emotional difficulties (SDQ). To analyze the parents’ help-seeking action and intentions, we narrow down the sample of adults to include only those who live together with at least one child in Germany.
The intervention consisted of two different treatment videos: the relatable peer video and the celebrity video. A control group did not see any video or other content. The relatable peer video shows a conversation between a patient and a Ukrainian psychiatrist who also came to Germany as a refugee. The conversation aimed at reducing stigma, counteracting myths (e.g., medical psychiatric treatments would be continued forever once started), and also provide information about the availability of free mental health services in Germany for Ukrainian refugees in their language under data protection guaranteed by German law.
In order to measure parent’s uptake of mental health support on behalf of their children, we ask parents two survey questions with respect to each of their children (translated from German):
-- Have you ever made use of the following help offers because of psychological problems/complaints (e.g., sleep disturbances, strong irritability, concentration problems, new anxieties) of [child name] since your arrival in Germany?
---- Answer possibilities:
1) Help by friends or acquaintances
2) Children’s doctor
3) Online psychological help (e.g., conversation or chat with a psychologist)
4) School psychologist
5) Counselling outside of school (e.g., child protective service, Caritas, Diakonie, youth counselling)
6) Social worker (e.g. at a youth center)
7) Helpline (e.g., Nummer gegen Kummer, telephone pastoral counselling)
8) Other
9) I have not made use of any of the above help offers
-- [If answer == 9] Have you ever considered making use of these help offers for [child name]?
---- Answer possibilities: Yes/No
Both survey questions are included in the anchor questionnaire of wave 6 of the BiB-FReDA-Survey, which is running from May to July 2025. Thus, parents’ reported help-seeking action and intentions on behalf of their children are observed at about 1,5 years after the intervention. The time of measurement for the other child outcomes depends on the child’s age. In wave 4 (March-June 2024), children aged 11-17 were first surveyed upon their parent’s consent. In wave 6, children aged 11-17 were again surveyed. From these child questionnaires, we observe the children’s self-reported SDQ, general life satisfaction, and sense of school belonging. The child outcomes of wave 4 and 6 are measured about 6 months and about 1,5 years after the intervention respectively. For younger children, we rely on the available data reported by their parents (including the SDQ). Additionally, we use outcomes that are available by parents’ reports for all children below age 18, including the general health status of the child. When analyzing child outcomes, we pool all children and cluster standard errors on the household level. For parents, only one parent per household participated in the survey. Additional variables about the parents come from previous waves of the panel survey, in particular wave 1, which includes socio-demographic variables.
Our outcomes include the help-seeking behavior of parents on behalf of their children, but also several outcomes that measure the well-being of children: socio-emotional difficulties (SDQ), school belonging, subjective life satisfaction, and general health. The intervention could increase mental health support uptake on the child level directly because parents seek support for their children. The intervention could also affect help-seeking on the child level more indirectly. If stigma is reduced on the level of parents, especially older children may be more open to help offers available to them, e.g. at school. Moreover, if the intervention increased mental health of parents, this increase would likely lead to better mental health and well-being of their children even in the absence of mental health support for the children. We hypothesize that the intervention increases parent’s help-seeking on behalf of their children and, more broadly, that the intervention increases children’s overall well-being.
We expect that attitudes towards mental health support change rather slowly, given the high stigma involved among Ukrainian refugees. Therefore, we consider including follow-up survey questions on the intervention in future waves of the BiB-FReDA-Survey (which will be integrated into FReDA – The German Family Demography Panel Study). The questionnaires for the future panel waves are currently under construction.
As agreed upon with infas Institut für angewandte Sozialwissenschaft, the Federal Institute for Population Research (BiB) will receive first data of wave 6 on 22 May 2025 the earliest (after having submitting this registration report).
We expect a potential sample size resulting from about 1500 adults who both participated in the intervention and have children. However, some of the outcomes are only available or meaningful for children of a certain age range which will reduce the sample of potential adults. The actual sample size depends on the participation and response rates in the waves and survey questions where outcomes are measured. Moreover, several children of the same parent could participate in the survey.
We analyze heterogeneities including the ex-ante health and psychological well-being of the child because the intervention is more likely to increase parents’ help-seeking on behalf of their children if the children exhibit low ex-ante health or well-being, i.e. are more likely in need of mental health support. Prior to the intervention, data on the children is only available through parent reports from the anchor questionnaires. To measure ex-ante psychological well-being of the child, we use the KINDLR-Subscore from parents’ questionnaire in wave 3, which was administered shortly before the intervention. To measure overall health of the child, we use the single item question from the parents’ questionnaire in wave 3. We also investigate heterogeneities by age, gender and family composition, taking into account that descriptive findings report a lower well-being among older children, especially older girls or those living together with only one parent.
Concerning IRB, we refer to the IRB approval for project AEARCTR-0013354, which includes the intervention we build upon.