Association between adverse childhood experience and unintended pregnancy among Japanese women: a large-scale cross-sectional study

Abstract

Unintended pregnancy (UP) can negatively impact the health of mothers, children, and families. While Adverse Childhood Experiences (ACEs) are increasingly known to affect sexual health, the influence on pregnancy intention is not fully understood. This study examines the relationship between ACEs and UP and explores other related factors, using 5049 pregnant and postpartum women data from the Japan COVID-19 and Society Internet Survey (JACSIS). We measured participants’ pregnancy intentions, ACEs, family functioning, and social network size. Logistic regression analysis provided odds ratios and 95% confidence intervals (CI). The prevalence of UP was approximately 16.5% (n = 893). Cumulative ACEs were consistently associated with UP, even after adjusting for intermediate variables in adulthood. The odds ratio for UP with a single ACE was 1.00 (CI: 0.82–1.21) but rose significantly with multiple ACEs: 1.39 (CI: 1.10–1.76) with double, 1.38 (CI: 1.02–2.86) with triple, and 1.81 (CI: 1.37–2.39) with more. Additionally, bad family functioning and lack of social networks emerged as contributors to UP. In conclusion, this study showed that ACEs are potentially correlated with UP. A deeper understanding of the transition from childhood experiences to UP is important for health interventions, necessitating further investigation.

Keywords:

• Adverse childhood experiences
• unintended pregnancy
• unwanted/mistimed pregnancy
• reproductive health

Introduction

Unintended pregnancy is a global public health problem, which negatively affects the health of mothers, children and their families [1]. Unintended pregnancy is measured as the sum of those who reported ‘unwanted’ and ‘mistimed’ among pregnancy intentions (intended, unwanted, and mistimed) [2]. It is estimated that 121 million unintended pregnancies will occur all over the world annually between 2015 and 2019. In Japan, unintended pregnancy was reported by 46.2% in a 2002 cross-sectional survey [3] and by 41% in a study using 2011 data [4]. Several studies have shown that unintended pregnancies have negative effects on appropriate maternal behavior during pregnancy (e.g. alcohol, illegal drug use and smoking) [5–7], access to antenatal and postnatal care [8–15], and maternal mental health (e.g. depression [16, 17], anxiety [18], decreased psychological well-being and psychosocial conditions [19, 20]). A study examining the relationship between birth decisions and long-term psychological distress among women with unintended pregnancies found that psychological distress increased for wanted births, abortions, adoptions, and unwanted births, suggesting the magnitude of the long-term psychological impact of unintended pregnancy [21]. There are also some stigmatizing attitudes in health care settings and communities that blame women’s sexual behavior for unintended pregnancies and impart mental hardship [22].

Adverse Childhood Experiences (ACEs) are one of the most important factors associated with unintended pregnancy. ACEs have been shown to be associated with mental and physical health in adulthood [23]. They are moderately associated with physical inactivity, being overweight or obesity, and diabetes, smoking, heavy alcohol use, poor self-rated health, cancer, heart disease, and respiratory disease. ACEs are strongly associated with mental illness, problematic alcohol use, or higher. Lastly, ACEs have the strongest association with problematic drug use and interpersonal and self-initiated violence. There is also concern that health outcomes that are strongly associated with ACEs may also lead to ACEs in the next generation (e.g. violence, mental illness, substance use) [23]. ACEs are also increasingly being found to be associated with sexual health. Exposure to most ACEs subtypes has been associated with an increased risk of adolescent pregnancy [24–28]. Women exposed to household dysfunction, as subtypes of ACEs, during childhood were significantly more likely to experience psychosocial difficulties during pregnancy [29].

To our knowledge, research on the association between ACEs and unintended pregnancy is limited. Two studies found that having more than two or three ACEs was associated with a higher likelihood of unwanted pregnancy [30, 31]. However, neither study was able to adequately control for factors other than ACEs which may impact pregnancy intentions, such as the stability of current family relationships. In addition, we believe that no study has examined the association between ACEs and unintended pregnancy in Japan.

The purpose of this study is to examine the association between ACEs and unintended pregnancy and to explore other related factors of unintended pregnancy. The main hypothesis was that ACEs and unintended pregnancy are associated even after controlling for intermediate variables factors in adulthood such as current family functioning.

Materials and methods

Study design, participants, and data setting

The present study used data from an internet-based survey conducted as a part of the Japan COVID-19 and Society Internet Survey (JACSIS). The JACSIS study consisted of three surveys with the following targets: the general population, pregnant and postpartum women, and single mothers/fathers. The study sample for each survey was retrieved from the pooled panels of an Internet research agency (Rakuten Insight, Inc.), which had approximately 2.2 million panelists in 2019. The pregnant and postpartum women survey was performed between 24 July 24 and 30 August 2021, based on the recruitment of 14,080 pregnant women who were anticipated to give birth by December 2021 or who had given birth after July 2019 and 3434 of their partners, who had signed up to be included in the online panel. We used the data of 8047 women from the first 10,000 responders.

Measurements

Unintended pregnancy

We defined the dependent variable for the present study as unintended pregnancy (0 = No, 1 = Yes). The participants are required to answer which of the following applies most to their recent pregnancy: (1) Spontaneous (wanted and planned pregnancy), (2) Spontaneous (unexpected or unwanted pregnancy), (3) Infertility treatment (timing method), (4) Infertility treatment (fertility drugs), (5) Infertility Treatment (Artificial Insemination), and (6) Advanced Reproductive Medicine (IVF/ICSI). We defined the response (2) as unintended pregnancy, and the others as intended pregnancy.

Adverse childhood experiences

We asked the participants if they had any Adverse Childhood Experiences (ACEs). The original ACEs study described ACEs as child abuse (physical, psychological, and sexual) and household dysfunction (substance abuse, mental illness, mother treated violently, and criminal behavior in household), while a further ACEs study added parental separation or divorce as an item of household dysfunction [26]. It has been proposed to expand the concept of ACEs from the original ACEs study to other ACEs, such as peer victimization (i.e. bullying), exposure to community violence (i.e. unsafe neighborhood), witnessed violence, childhood poverty, someone close had a bad accident or illness, below-average grades, parents always arguing, having no good friend, felt discrimination, or lived in foster care [32, 33].

In this study, ACEs were measured thru the following 10 items: (1) Parental separation, (2) Parental divorce, (3) Parental mental illness, (4) Mother treated violently, (5) Physical abuse, (6) Neglect, (7) Emotional abuse, (8) Childhood poverty, (9) Bullying, and (10) Sexual abuse.

Family adaptation partnership growth and resolved (APGAR) score

To evaluate the current functioning and support of the family, we calculated the Family Adaptation Partnership Growth and Resolved (APGAR) Score. Smilkstein created the Family APGAR score as a quick screening questionnaire aimed at eliciting five aspects of family function [34]. These five concepts are represented by the acronym APGAR: adaptability, partnership, growth, affliction, and resolve. A five-item survey is used with a score of 0 and ≦3 is classified as severe family dysfunction, ≧4 and ≦6 as moderate family dysfunction, and ≧7 as good family functioning. No reliable and validated questionnaire has been published in Japanese.

These items were: I am satisfied with the help that I receive from my family when something is troubling me; I am satisfied with the way my family discusses items of common interest and shares problem solving with me; I find that my family accepts my wishes to take on new activities or make changes in my lifestyle; I am satisfied with the way my family expresses affection and responds to my feelings such as anger, sorrow, and love; and I am satisfied with the amount of time my family and I spend together. Originally, the choices of answers and their scores were: Almost always (2 points), Some of the time (1 point), or Hardly ever (0 points). In our survey, choices of answers were set as: Almost always (2 points); Some of the time (1 point); Hardly ever (0); and Never (0), which is consistent with the original choices.

Social network

As an indicator of the social network size when you have a problem, we asked participants how many people they could talk to if they really needed help. Participants selected their answers from the following options: (1) 0, (2) 1–2, (3) 3–5, (4) 6–10, (5) 11–15, (6) 16–20, (7) 21 or more. We sorted cases with no one (that is option 1)) and any person (option 2) to (7)) in the analysis, with the concept that social network size, particularly the risk of social disconnection, has a negative impact on sexual health outcomes [35, 36].

Other covariables

We included the following items as basic and socioeconomic confounding variables: Age (< 24 years, 25–30 years, 31–34 years, 35–40 years, or ≥ 41 years), household income (< 3 million JPY [approximately 24,000 USD], 3 to < 10 million JPY, ≥ 10 million JPY), and education level (high school or college/university). We also included the following items as perinatal status: Parity (0, 1, 2, ≥ 3) or marital status (married, other).

Data analyses

To assess the association of ACEs or other factors with unintended pregnancy, we estimated odds ratios (ORs) and 95% confidence interval (CI) using logistic regression analysis. The reference group was women whose most recent pregnancy was intended. For this analysis, we used the JACSIS study data of pregnant and postpartum women (N = 8047). We excluded 2638 women who did not answer about her education level or income, which are covariables in this study. Finally, 5409 women were eligible for the analysis.

The association between unintended pregnancy and the above factors was first examined in univariate logistic regression and then analyzed in three models. In model 1, the number of ACEs and age were entered as independent variables. In model 2, multiple logistic regression was performed with the addition of demographics as a possible mediator. In Model 3, we added the Family Adaptation Partnership Growth and Resolved (APGAR) score, as an indicator of family functioning, and an indicator of the availability of someone to ask for help in times of need. For each categorical variable, we chose the normative or minimum group as the reference category.

We measured the impact of multicollinearity among the independent variables in our regression model using the variance inflation factor (VIF), with a cutoff of 5 (i.e. VIF > 5 was considered to indicate high multicollinearity).

As sensitivity analyses, we calculated E-values, which assess how strongly unmeasured and residual confounding would need to be associated with the exposure and the outcomes on the OR scale to reduce the point estimates to null [37, 38].

Statistical analyses were performed using R version 15.1. The significance level was set at 5 and 1% for all analyses.

Ethics approval

The Osaka International Cancer Institute’s Research Ethics Committee evaluated and approved the study protocol (approved on 19 June 2020; approval number 20084). It was also approved by the Research Ethics Committee of Graduate School of Medicine/Faculty of Medicine, The University of Tokyo (no. 2020336NI). Prior to doing the online survey, each participant gave their informed consent online. The participants were given "Epoints," which were credit points that could be converted into cash and used for online shopping.

Results

Sample characteristics and bivariate analysis

Table 1 shows the characteristics of 5409 pregnant women. Their mean age was 32.0 years (SD = 5.2). Among them, 40.1% were 30–34 years old, 50.5% had university degrees, and 78.0% earned 3-10 million JPY annually. 99.3% were married, 44.6% were experiencing their first pregnancy, and 16.5% had unintended pregnancies. While 52.6% had no ACEs, 6.3% had four or more. 67.8% scored good on the Family Adaptation Partnership Growth and Resolved (APGAR) Score, and 1.1% lacked someone to call for help in times of need.

Table 1. Basic participant demographics included in the analysis (N = 5409).

Mean ± SD or number (%)
Maternal age	32.0 ± 5.2
18 − 24	196 (3.6)
25 − 29	1565 (28.9)
30 − 34	2169 (40.1)
35 − 39	1256 (23.2)
40+	223 (4.1)
Education
High school graduate	2680 (49.5)
University graduate	2729 (50.5)
Household income (million JPY)
<3	189 (3.5)
3–10	4220 (78.0)
≧10	1000 (18.5)
Occupation
Unemployed	992 (18.3)
Employed	4417 (81.7)
Marital Status
Married	5369 (99.3)
Others	40 (0.7)
Parity
0	2415 (44.6)
1	2053 (38.0)
2	754 (13.9)
≧3	187 (3.5)
Pregnancy intention
Intended	4516 (83.5)
Unintended	893 (16.5)
ACEs
0	2847 (52.6)
1	1276 (23.6)
2	623 (11.5)
3	324 (6.0)
4+	339 (6.3)
Family APGAR
Good (≦3)	3668 (67.8)
Fair (≧4, ≦6)	1260 (23.3)
Bad (≧7)	481 (8.9)
Social network (Number of people you can call for help in times of need)
Any	5348 (98.9)
None	61 (1.1)

Multivariate analysis model

Table 2 shows the results of crude and multivariate analyses of ACEs and other risk factors with unintended pregnancy. Model 1 confirmed that the number of ACEs was significantly associated. In Model 2, intermediate variables between ACEs and unintended pregnancy were also entered as covariates, but the significant association of cumulative ACEs remained. Educational level, more than $10 million household income, marital status, and two or more births were also predictors of unintended pregnancy. In Model 3, in addition to the results from Model 2, family functioning and social network size were also found to be factors associated with unintended pregnancy. For estimates with a 95% CI that excluded the null, the E-values suggested that unmeasured confounders would need to have an odds ratio greater than 1.64 for two ACEs, 1.63 for three ACEs, and 2.03 for four or more ACEs, to move the point estimates to null.

Table 2. Crude and multivariate analysis of ACEs and other risk factors for unintended pregnancy among Japanese women.

	Crude analysis OR (95% CI)	p-value	Model 1 OR (95% CI)	p-value	Model 2 OR (95% CI)	p-value	Model 3 OR (95% CI)	p-value
ACEs
0 (Ref)	−	−	−	−	−	−	−	−
1	1.21 (1.04 − 1.40)	0.012	1.08 (0.89 − 1.30)	0.436	1.04 (0.86 − 1.26)	0.694	1.00 (0.82 − 1.21)	0.960
2	1.60 (1.34 − 1.91)	< 0.001	1.54 (1.23 − 1.93)	< 0.001	1.51 (1.20 − 1.89)	< 0.001	1.39 (1.10 − 1.76)	0.005
3	1.67 (1.32 − 2.12)	< 0.001	1.59 (1.20 − 2.12)	0.001	1.52 (1.13 − 2.04)	0.005	1.38 (1.02 − 1.86)	0.037
4+	2.18 (1.76 − 2.69)	< 0.001	2.20 (1.69 − 2.86)	< 0.001	2.07 (1.57 − 2.72)	< 0.001	1.81 (1.37 − 2.39)	< 0.001
Maternal age
18 − 24 (Ref)	−	−	−	−	−	−	−	−
25 − 29	0.39 (0.31 − 0.48)	< 0.001	0.43 (0.31 − 0.59)	< 0.001	0.43 (0.31 − 0.60)	< 0.001	0.41 (0.30 − 0.57)	< 0.001
30 − 34	0.28 (0.23 − 0.34)	< 0.001	0.28 (0.21 − 0.39)	< 0.001	0.26 (0.18 − 0.35)	< 0.001	0.24 (0.18 − 0.34)	< 0.001
35 − 39	0.31 (0.25 − 0.39)	< 0.001	0.29 (0.21 − 0.41)	< 0.001	0.23 (0.16 − 0.33)	< 0.001	0.22 (0.16 − 0.32)	< 0.001
40+	0.35 (0.22 − 0.57)	< 0.001	0.36 (0.23 − 0.56)	< 0.001	0.28 (0.17 − 0.45)	< 0.001	0.26 (0.16 − 0.42)	< 0.001
Education
High school graduate (Ref)	−	−			−	−	−	−
University graduate	0.71 (0.62 − 0.81)	< 0.001			0.83 (0.71 − 0.96)	0.015	0.83 (0.71 − 0.97)	0.022
Household income (million JPY)
<3 (Ref)	−	−			−	−	−	−
3–10	0.61 (0.46 − 0.80)	< 0.001			0.72 (0.50 − 1.04)	0.083	0.76 (0.52 − 1.10)	0.141
≧10	0.35 (0.25 − 0.49)	< 0.001			0.50 (0.33 − 0.77)	0.001	0.53 (0.35 − 0.81)	0.003
Occupation
Unemployed (Ref)	−	−			−	−	−	−
Employed					1.04 (0.85 − 1.26)	0.718	1.03 (0.85 − 1.26)	0.730
Marital Status
Married (Ref)	−	−			−	−	−	−
Other	5.97 (4.20 − 8.49)	< 0.001			2.56 (1.28 − 5.13)	0.008	2.41 (1.20 − 4.87)	0.014
Parity
0 (Ref)	−	−			−	−	−	−
1	1.09 (0.95 − 1.25)	0.242			1.17 (0.99 − 1.40)	0.072	1.15 (0.96 − 1.36)	0.128
2	1.64 (1.39 − 1.93)	< 0.001			2.06 (1.65 − 2.57)	< 0.001	2.02 (1.62 − 2.53)	< 0.001
≧3	5.21 (4.12 − 6.59)	< 0.001			6.64 (4.78 − 9.21)	< 0.001	6.61 (4.76 − 9.19)	< 0.001
Family APGAR
Good (≦3)	−	−					−	−
Fair (≧4, ≦6)	1.34 (1.17 − 1.54)	< 0.001					1.36 (1.14 − 1.62)	0.001
Bad (≧7)	1.83 (1.53 − 2.19)	< 0.001					1.43 (1.11 − 1.85)	0.006
Social Network (Number of people you can ask for help if needed)
Any (Ref)	−	−					−	−
None	2.38 (1.59 − 3.56)	< 0.001					2.23 (1.27 − 3.90)	0.005
AIC			4742.790		4568.588		4548.269

OR: odds ratio, CI: confidence interval, Ref: reference, AIC: Akaike’s Information Criterion.

Table 3 shows the association between specific ACE domains and unintended pregnancy. Adjusted for age, parental divorce and childhood poverty are significant as associated factors.

Table 3. Crude and multivariate analysis of specific ACE domains and other risk factors for unintended pregnancy among Japanese women.

	N (%)	Crude analysis OR (95% CI)	p-value	Model 1 OR (95% CI)	p-value
ACEs
Parental separation	257 (4.8)	1.10 (0.85 − 1.44)	0.472	1.10 (0.84 − 1.44)	0.476
Parental divorce	680 (12.6)	1.44 (1.22 − 1.71)	< 0.001	1.27 (1.07 − 1.52)	0.006
Parental mental illness	334 (6.2)	1.17 (0.93 − 1.48)	0.177	1.15 (0.91 − 1.45)	0.245
Mother treated violently	388 (7.2)	0.96 (0.76 − 1.21)	0.726	0.97 (0.77 − 1.22)	0.806
Physical abuse	258 (4.8)	1.28 (0.97 − 1.69)	0.081	1.21 (0.91 − 1.60)	0.183
Neglect	84 (1.6)	1.17 (0.78 − 1.75)	0.454	1.10 (0.73 − 1.67)	0.640
Emotional abuse	837 (15.5)	1.07 (0.90 − 1.28)	0.456	1.11 (0.93 − 1.33)	0.240
Childhood poverty	682 (12.6)	1.38 (1.16 − 1.64)	< 0.001	1.38 (1.16 − 1.64)	< 0.001
Bullying	1400 (25.9)	1.10 (0.96 − 1.26)	0.176	1.11 (0.96 − 1.27)	0.146
Sexual abuse	218 (4.0)	0.95 (0.71 − 1.27)	0.702	0.98 (0.74 − 1.31)	0.908
Maternal age
18 − 24 (Ref)				−	−
25 − 29				0.37 (0.28 − 0.47)	< 0.001
30 − 34				0.25 (0.19 − 0.32)	< 0.001
35 − 39				0.27 (0.21 − 0.36)	< 0.001
40+				0.33 (0.24 − 0.48)	< 0.001
AIC		7324.8		7217.4

OR: odds ratio, CI: confidence interval, Ref: reference, AIC: Akaike’s Information Criterion.

In model 1, maternal age was entered between the specific ACE domains and unintended pregnancy.

In model 1, maternal age was entered between ACEs and unintended pregnancy. In Model 2, demographics as a possible mediator were added to Model 1. In Model 3, the Family APGAR score, an indicator of family functioning, and an indicator of the availability of someone to ask for help in times of need were added into Model 2. E-values were calculated for each model to evaluate unmeasured confounding factors.

Discussion

The present study provides two findings, which are shown for the first time regarding Japan. First, cumulative ACEs were consistently associated with unintended pregnancy even after accounting for the effects of intermediate variables in adulthood. Second, we found an association for unintended pregnancy with other related factors such as current family functioning and social network.

The incidence of unintended pregnancy in this study was about 16.5%, which was low compared with previous studies from Japanese and other countries. This may be related to potential underestimate, from the missing data that did not respond to education history and income. The marriage rate was 99.3%, which is extremely high compared with previous Japanese and foreign studies. The population in this study may be a relatively low-risk group, considering the low percentage of married women in other studies and the research that marital status is also associated with unintended pregnancy [39–41].

The finding that ACEs were associated with unintended pregnancy is consistent with two previous studies that showed an association between the number of ACEs and unintended pregnancy [30, 31]. Similar to the two previous studies in the U.S., the results of this study showed a dose-response relationship between ACEs and unintended pregnancy. The present study further examines the association between ACEs and unintended pregnancy even after controlling for current family functioning and social network size, which we believe provides additional robust evidence. Some previous studies have investigated whether cumulative ACE is associated with negotiation with one’s partner in adulthood [42, 43], which may lead to an increased risk of unintended pregnancy, but so far the results are inconsistent. The results of this study relate to the mechanism of the association between ACEs and unintended pregnancy, and is meaningful to clarify a hypothesis regarding the association of ACEs and interpersonal skills.

Among the specific ACE domains, parental divorce and childhood poverty were associated with unintended pregnancy. Childhood poverty can lead to additional adversities experience with greater frequency and intensity [44], and be significantly associated with physical and mental health, even after adjusting for other ACEs [45]. The present results support that hypothesis. To determine how the association between parental divorce and unintended pregnancy is demonstrated, more studies are needed, with the previous studies showing that parental divorce does not independently predict depressive symptoms during pregnancy [46].

In this study, we added evidence that current family functioning and social networks are also associated with unintended pregnancy. These findings are consistent with previous research showing that social support reduces unintended pregnancy [47]. We believe these findings are important for public health practice, including outreach by public health nurse practitioners and other care workers. It has been noted in a previous study that those who lack secure emotional attachments and a sense of belonging may engage in sexual risk-taking to fill a perceived void as a means of compensating for social exclusion [48]. The results of this suggest that a social network may also be important for people from whom help can be sought in addition to the social support that is helpful for maternal and infant health [49]. The importance of trauma-informed care should also be emphasized, as cumulative ACEs can affect interpersonal difficulties and may have an impact on pregnancy intentions.

The current study has some limitations. The first concerns the variables used in the analysis. In this study, we were not able to distinguish between unintended and mistimed pregnancies. It should be noted that unintended pregnancies should be qualitatively distinguished from mistimed pregnancies [50, 51]. Indeed, in the two previous studies examining the association between ACEs and unintended pregnancy [30, 31], a significant association was found only in unintended pregnancies and not in mistimed pregnancies. In the present results, unintended pregnancy was defined as a combination of unintended and mistimed pregnancies, which may have underestimated the results. 2638 (32.8%) participants did not respond on education and income and were explored from the analysis. The authors believe that these missing are not random and that the lower education and income groups are missing [52]. Model 1, adjusted for just age, may underestimate ACE odds. Models 2 and 3, also adjusted for adult intermediate variables, may overestimate ACE odds in this study. In addition, we were not able to adjust for acceptance of pregnancy [53] and partner’s position [54, 55] on pregnancy, which also affect perinatal mental health. Second, we were unable to measure important confounding factors. A possible confounder of ACEs between unintended pregnancy is parental demographics, such as parental history of mental illness, which we were unable to measure and control for in this study. According to the E-value analysis, unmeasured and residual confounding could potentially explain away the observed relationships. However, any potential confounding would have to be modestly related to exposures and outcomes, independent of all measured covariates. Third, this study uses data from a cross-sectional survey and cannot show causality. A pre- and post-temporal relationship can be assumed with regard to between ACEs and unintended pregnancy, but especially with regard to current family functioning and social networks, a causal reversal cannot be ruled out for whether the experience of an unintended pregnancy has caused a change in the relationship with family or others. Fourth, the survey from which we analyzed data in this study was an online survey, which limits its external validity and the bias associated with the self-administered form could not be controlled for. Finally, it is crucial to stress that, while pregnancy intentions are valuable for understanding women’s pregnancy preferences, the construct of pregnancy intentions may not be applicable to all women [56, 57].

Conclusions

This study revealed that ACEs might be associated with unintended pregnancy and also found other related factors, which can be improved by a public health approach in adulthood such as current family functioning and social networks. To clarify the mechanisms from childhood to an unintended pregnancy and reduce the negative health effects, further research is needed, including separate measures of unintended and mistimed pregnancy, and use of validated ACEs items.

Disclosure Statement

DN reports personal fees from Startia, Inc., en-power, Inc., MD.net and Takeda Pharmaceutical Company Ltd. outside the submitted work. The other authors declare no conflict of interest.

Data availability statement

Due to the presence of personally identifiable or potentially sensitive information, the data used in the present study are not deposited in a public repository. The Research Ethics Committee of the Osaka International Cancer Institute has restricted data sharing in compliance with Japanese ethical guidelines laws. Any questions on data use should be directed to Dr. Takahiro Tabuchi at [email protected]. More information about data availability can be obtained on the JACSIS website (https://jacsis-study.jp/howtouse/).

Additional information

Funding

This study was supported by the Japan Society for the Promotion of Science KAKENHI Grants [grant numbers JP 21H04856]; the Japan Science and Technology Agency [grant number JPMJSC21U6]; Intramural fund of the National Institute for Environmental Studies; Innovative Research Program on Suicide Countermeasures [grant number: R3-2-2]; The Ministry of Health Labor and Welfare of Health Labor Sciences Research Grant [grant number: 23GC1019]; and Ready for COVID-19 Relief Fund [grant number: 5th period 2nd term 001]. The findings and conclusions of this article are the unique responsibility of the authors and do not represent the views of the research funding.