Parental detention and psychosocial wellbeing of migrant children in Japan

Summary

Immigration policies in Japan and elsewhere have been toughened in recent years. To investigate the potential effects of parental detention on migrant children, psychosocial wellbeing of children from migrant families with and without parental detention was compared. In this cross-sectional study, adult asylum seekers and migrant workers with children staying in Japan were invited through non-governmental organizations to answer a self-administered anonymous questionnaire in June and July 2020. Children's psychosocial wellbeing was assessed based on the Strength and Difficulties Questionnaire. In the 49 participating families, including 28 where either parent had ever been detained in Japan, there were 85 children aged 4-17 years who were subject to the analyses. Psychosocial wellbeing of children in families with parental detention appeared to be worse than that of their counterparts, especially on the dimension of emotional problems. More attention should be paid to the wellbeing of migrant children in Japans immigration policy.

Introduction

Globally, the migrant population is growing substantially. In the past 20 years, the number of international migrants, refugees, and internally displaced persons increased from 150 to 272 million, from 14 to 26 million, and from 21 to 41 million, respectively (1). Even in Japan, there has been a large increase in the migrant population. In the past 10 years, the number of migrant workers and asylum seekers has increased about three-fold and nine-fold, respectively, with over 1.6 million migrant workers and over 10,000 asylum seekers in 2019 (2,3). In the same year, the number of unauthorized foreign nationals including migrant workers and asylum seekers was estimated to be nearly 80,000 (4).

In Japan, unauthorized foreign nationals are subject to detention and deportation. During the deportation process, they might be detained in immigration centers for an indeterminate period, even if they are seeking asylum or accompanied by children. Consequently, in violation of one of the rights endorsed in the United Nations Convention on the Rights of the Child, children are separated from their parents against their will (5).

With the increase in detention duration owing to stringent immigration policies in recent years, concerns have been raised about the harmful effects of parental detention on children's wellbeing (6). While the psychosocial impact of immigration policies has been well documented in the United States (7,8), this issue should be further highlighted for the sake of child protection in countries with stricter immigration policies. The present study, therefore, investigated the potential association of parental detention with psychosocial wellbeing of migrant children, including those seeking asylum in Japan.

Participants, Data collection, Measure, and Analysis

i) Participants. The participants were adult asylum seekers and migrant workers staying in Japan. The inclusion criteria were those accompanied by at least one child aged 4-17 and those who were literate, owing to the use of a self-administered questionnaire. They were recruited through non-governmental organizations supporting asylum seekers and migrant workers. This study was approved by the Research Ethics Committee of the Faculty of Medicine at the University of Tsukuba (No. 1505).

ii) Data collection. A self-administered anonymous questionnaire in the language the participants understand was sent to and collected from them through the non-governmental organizations. Since there was no definitive list of participants, approximately 100 individuals were invited, and 49 eligible participants answered the questionnaire in June and July 2020.

iii) Measures. Children's psychosocial wellbeing was assessed with the Strength and Difficulties Questionnaire (SDQ) (9). The SDQ consists of 25 items, with five each across five subscales (emotional problems, conduct disorders, hyperactivity, peer problems, and prosocial behavior) rated on a three-point scale from 0 (not true) to 2 (certainly true) by the parents or teachers of the children. Each subscale score ranges from 0 to 10. The sum of four subscale scores excluding prosocial behavior provides a "total difficulties" score ranging from 0 to 40. The SDQ has been translated, validated, and made available in more than 80 languages (https://sdqinfo. org/).

The characteristics of the participants and their families included age, sex, country of origin, marital status, length of stay in Japan, cohabitants, engagement in a paid job, refugee application, detention experience in Japan, perceived social capital and support in Japan, and children's age, sex, and schooling. Perceived social capital and support were assessed based on a total of four questions (10,11). Social capital was assessed through two questions concerning social trust and mutual aid rated on a five-point scale, with responses of "yes" and "somewhat" categorized as "yes". Social support was assessed through two yes/no questions concerning emotional and instrumental support. The questionnaire will be made available upon request to the author.

iv) Analysis. First, the characteristics of participants and families with and without parental detention in Japan were described. Then, children's psychosocial wellbeing scores were compared between families with and without parental detention to examine the potential association of parental detention with children's wellbeing. Mean scale score differences and 95% confidence intervals between the two groups of children were estimated in multilevel regression analyses controlling for the covariates of child's age and sex at the individual level and the family at the group level. Multilevel analyses were used to consider a clustered structure of the data (i.e., children within the family).

Key findings and discussion

There were 49 participants/families that met the inclusion criteria. Participants' mean age was 42 years, 29 (59%) were males, 37 (76%) came from Asian countries, 45 (92%) were married, 35 (71%) had spent 10 years or longer in Japan, and the median number of cohabitants was four. Of the 49 families, 28 (57%) had at least one parent engaged in a paid job in Japan, 30 (61%) had at least one parent in the process of refugee application, 28 (57%) had at least one parent who had ever been detained in Japan, and 13 of 28 reported that the length of detention was one year or longer. Regarding the participants' perception of social capital and support, 14 (29%) and 15 (31%) perceived social trust and mutual aid in the community, respectively, and 32 (65%) and 31 (63%) perceived emotional and instrumental support in their social network, respectively. In the 49 families, there were 85 children aged 4-17 years subject to the subsequent analyses. Their mean age was nine years, and 37 (44%) were males. All school-aged children were in school.

Table 1 compares the characteristics mentioned above between the families with and without parental detention, showing discernable differences by length of stay in Japan and refugee application. Families with parental detention had spent a longer time in Japan and had more refugee applications than their counterparts. Table 2 shows the mean scale score of psychosocial wellbeing among children in families with and without parental detention, and the estimated mean differences between these two groups. Children in families with parental detention had significantly higher scores, especially on the subscale of emotional problems, than their counterparts, while there was no significant difference on the subscale of prosocial behavior.

Table 1. Characteristics of participants and their households by parental detention in Japan^a.

Variables	Ever detained	Never detained
Participants	(n = 28)	(n = 21)
Age, mean (SD)	42.4 (5.7)	40.4 (5.9)
Sex (male)	20	9
Geographic region of origin
Asia	23	14
Africa	4	5
Others	1	2
Marital status
Married	27	18
Single	0	3
Missing	1
Length of stay in Japan (years)
< 10	2	11
10-19	20	7
≥ 20	6	2
Missing		1
Number of cohabitants, median (IQR)	4 (3, 5)	3 (3, 5)
Paid job (yes)b	12	16
Refugee application (yes)c	25	5
Missing		4
Perceived social capital
Social trust (yes)	8	6
Mutual aid (yes)	8	7
Perceived social support
Emotional support (yes)	17	15
Instrumental support (yes)	19	12
Children aged 4 to 17 years	(n = 49)	(n = 36)
Age, mean (SD)	9.1 (3.9)	8.8 (3.4)
Sex (male)	24	13
Missing	1	1

SD: standard deviation, IQR: interquartile range. ^aHouseholds were classified as "ever detained" if either parent had ever been detained in Japan. ^bYes if either parent had a paid job. ^cYes if either parent was applying for refugee status.

Table 2. Mean scores of children's psychosocial wellbeinga by parental detention, and mean difference estimated in multilevel regression analyses^b.

Variables	Ever detained Mean	Never detained Mean	Mean difference (95% CI)
Emotional problems	4.59	1.72	2.38 (0.68, 4.07)
Conduct disorders	3.00	1.44	1.36 (0.26, 2.48)
Hyperactivity	4.35	3.03	1.14 (0.31, 1.97)
Peer problems	3.27	2.31	0.85 (0.06, 1.65)
Prosocial behavior	7.31	8.14	-0.74 (-1.94, 0.46)
Total difficultiesc	15.20	8.50	5.46 (1.55, 9.38)

CI: confidence interval. ^aMeasured using the Strength and Difficulties Questionnaire. ^bControlling for child's age and sex at the individual level and household at the group level. The analyses excluded two children whose sex was not reported. ^cTotal difficulties score is the sum of subscale scores excluding prosocial behavior.

Psychosocial wellbeing of children in families with parental detention appeared to be worse than that of their counterparts among asylum seekers and migrant workers in Japan. This finding is consistent with previous studies (7,8). So, it is plausible that parental detention would potentially harm children's wellbeing in migrant populations in Japan, posing an unescapable question: is it still justifiable to let innocent children sacrifice their fundamental right to be with their parents? Japan's immigration policy should be considerate to the wellbeing of migrant children.

This study had several limitations. First, it is difficult to make causal inferences regarding the relationship between parental detention and children's wellbeing because changes in wellbeing before, during, and after parental detention were not examined; instead, the study involved a cross-sectional comparison of the wellbeing of children in families with and without parental detention. Yet, this does not entirely negate the potential effect of parental detention on children's wellbeing, given that parental detention was independent of children's wellbeing. Therefore, the wellbeing of children in families with and without parental detention could have been comparable before parental detention, unless children in families with parental detention tended to have any predisposing factors affecting their wellbeing. If wellbeing was comparable before parental detention, differences in the wellbeing of children in the two types of families might have been due, at least in part, to parental detention.

Second, children's psychosocial wellbeing was assessed by their parents. While this is the standard assessment method for the SDQ, one might assume that the evaluation of children's wellbeing was influenced by the wellbeing of the parents, and the magnitude of this influence was greater for those who had ever been detained than those who had never been detained; consequently, children's wellbeing scores were inflated in the former. However, the subscale scores rated by those who had ever been detained were not evenly greater than the scores rated by those never detained. This might imply that even if the subscale scores were somewhat inflated, they still reflected the potential effects of parental detention.

Third, the sample size was too small to analyze the relationship between length of parental detention and children's wellbeing. Moreover, as there were no families where a parent was in detention during the study period, the impact of current detention could not be examined. Such analyses will help establish causality. Finally, the participants were recruited through non-governmental organizations, so those not accessible were not included in this study. It is uncertain whether they were better off, not requiring any support, or hidden for legal reasons. Besides, about half of those contacted did not reply. It is also uncertain whether their characteristics were different from the participants'. In any case, the findings have important implications for immigration policy in light of child protection.

In conclusion, psychosocial wellbeing of children in families with parental detention was worse than that of their counterparts among asylum seekers and migrant workers in Japan. More attention should be paid to the wellbeing of migrant children in Japan's immigration policy.