You are here

  • Home
  • Archive
  • Volume 77, Issue 4
  • Associations between type of childhood adversities and labour market participation and employment conditions in young adults

Article Text

Article menu

Download PDFPDF

Original research
Associations between type of childhood adversities and labour market participation and employment conditions in young adults
  1. Tjeerd Rudmer de Vries1,
  2. Iris Arends1,2,
  3. Albertine J. Oldehinkel3,
  4. Ute Bültmann1
  1. 1Department of Health Sciences, Community and Occupational Medicine, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
  2. 2Arbo Unie BV, Utrecht, The Netherlands
  3. 3Department of Psychiatry, Interdisciplinary Center Psychopathology and Emotion Regulation, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands
  1. Correspondence to Tjeerd Rudmer de Vries, Department of Health Sciences, Community & Occupational Medicine, University Medical Center Groningen, University of Groningen, Groningen 9700 RB, The Netherlands; t.r.de.vries{at}umcg.nl

Abstract

Background Cumulative exposure to childhood adversity is associated with a variety of labour market outcomes in young adulthood. It remains largely unclear whether the type of adversity matters in this association. This prospective study examined the differences in exposure to 14 adverse experiences among groups of young adults aged 22 characterised by distinct labour market participation states and employment conditions.

Methods We used data from the TRacking Adolescents’ Individual Lives Survey, a Dutch prospective cohort study with 15 years of follow-up (N=1524). We included 14 adverse experiences (ages 0–16) across five domains: peer influences, loss or threat of loss, material deprivation, family dynamics and maltreatment. Labour market participation states and employment conditions were assessed at age 22. We used latent class analysis to derive labour market outcome groups, which we subsequently compared on exposure to adverse experiences using pairwise comparisons.

Results Inactive individuals (n=85, 5.6%), often neither in education (77.4%) nor employment (98.6%) and on benefits (94.4%), were more likely to be exposed to many distinct types of adverse experiences (eg, parental addiction, bullying victimisation) as compared with all other groups. Early workers (n=413, 27.1%), often on temporary contracts and low monthly incomes, were more likely to be exposed to parental divorce (22.7%) compared with students with side jobs (12.9%).

Conclusions Different adverse experiences are not equally associated with labour market outcomes. Researchers and stakeholders in policy and practice should be aware of the differences between adverse experiences in their importance for labour market outcomes in young adults.

  • public health
  • employment
  • life course epidemiology
  • education
  • unemployment

Data availability statement

Data may be obtained from a third party and are not publicly available. TRAILS data of the T1, T2, T3, T4 and T5 measurement waves are deposited in the Data Archiving and Networked Services of the Royal Dutch Academy of Sciences (DANS-KNAW) and access can be requested at http://www.dans.knaw.nl.

https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi.org/10.1136/jech-2022-219574

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What is already known on this topic

    • Current evidence suggests that cumulative exposure to childhood adversity is associated with early labour market entry and labour market inactivity in young adulthood.

    • It remains unclear whether the type of childhood adversity matters (eg, experiences of maltreatment vs experiences of material deprivation) and whether childhood adversity exposure matters for labour market outcomes beyond labour market participation, namely employment conditions such as contract types, work hours and monthly income.

    What this study adds

    • Our study reveals important nuances in the association between childhood adversity and labour market outcomes in early working life.

    • We found that different groups of young adults, characterised by distinct labour market participation states and employment conditions, vary markedly in their exposure to types of adverse experiences.

    How this study might affect research, practice or policy

    • Future research on childhood adversity and labour market outcomes should continue disentangling the different types of adversity to facilitate translating findings to policy and practice.

    • Stakeholders in policy and practice may benefit from an increased awareness of the differences between adverse experiences in their importance for labour market outcomes in young adults to improve labour market outcomes at the population level.

    Introduction

    Childhood adversity, which includes adverse experiences (AEs) such as financial difficulties, parental mental illness and bullying victimisation, is common around the world.1–3 AEs are defined as experiences that require significant adaptation by the developing child in terms of psychological, social and neurodevelopmental systems, and are outside the normal expected environment.4 Childhood adversity is an important determinant of health inequalities throughout the life course.5 6 Employment in young adulthood has been suggested as an important driver of these health inequalities,7–9 but only a few studies have examined the association between childhood adversity and employment outcomes in young adulthood thus far.

    Studies to date have shown that cumulative exposure to childhood adversity is associated with a variety of outcomes relevant to young adults’ working lives: unemployment,10 11 benefit recipiency,12 income,13 NEET (not in employment, education or training) and labour market participation trajectories.14–17 However, cumulative exposure conceptualisations of childhood adversity have been criticised for not providing evidence that is required to better understand the mechanisms through which adversity is associated with outcomes of interest. Developing mechanistic insight requires knowledge on the role of individual AEs.18 Most recently, Hansen et al15 showed that young adults whose working lives were characterised by early full-time employment throughout young adulthood after obtaining a vocational degree, or by reduced labour market participation (eg, frequent or long-term unemployment spells), largely differed in the type of exposure to adversity compared with young adults who were studying for most of their early 20s. Especially those AEs that relate to behaviours of parents, such as abuse and parental addiction, were strongly associated with frequent or long-term unemployment spells; AEs such as parental divorce and witnessing violence were strongly associated with early labour market entry.15 AEs that more or less befall parents, such as a physical illness or a death of a parent, were not associated with labour market participation in young adulthood.12 15 Only few other studies investigated the importance of individual AEs for labour market outcomes, and there is marked between-study heterogeneity with regard to assessments of AEs and labour market outcomes. As a result, current evidence on the importance of individual AEs for labour market outcomes is fragmented and remains limited.12 14–16

    Currently available evidence suggests that it is possible to distinguish between groups of young adults on the labour market, which may be differentiated in terms of labour market participation states, employment conditions and adversity exposure patterns.15 These groups may require diverging policy strategies tailored to differences in exposure to adversity, as well as to the mechanisms through which adversities are associated with labour market participation and employment conditions. For example, young adults who experience maltreatment might be more often unemployed and on benefits due to ongoing mental health problems,16 whereas those who experienced material deprivation might enter the labour market earlier due to lower educational attainment.19 Providing policy makers with the necessary information to develop tailored strategies requires empirical evidence of labour market outcomes groups among young adults and how these groups differ in exposure to AEs.

    In this study, we aimed to examine the associations between individual AEs and labour market outcomes in young adults aged 22. We extend the literature in two ways. First, we derived labour market outcome groups in young adults aged 22 through latent class analysis (LCA). These labour market outcome groups were based on labour market participation states, that is, employment, educational status, benefit status, and employment conditions, that is, monthly income, contract type and work hours. Second, we included a broad array of AEs in both the family and peer context to provide more conclusive evidence on the importance of individual AEs for labour market outcomes. Given the explorative nature of this study, we have not specified a priori hypotheses. The findings of this study may help identify groups of young adults who require tailored support in their early working lives, conditional on the type of childhood adversity exposure in childhood.

    Methods

    Sample and design

    We analysed data from 1524 individuals included in the TRacking Adolescents’ Individual Lives Survey (TRAILS) cohort (68.4% of the baseline cohort), a Dutch prospective cohort study with biennial measurement waves. In March 2001, a representative sample of 2935 children born in one of the three northern provinces of the Netherlands between 1 October 1989 and 30 September 1991 were approached, of whom 2229 children (76%) enrolled in the study at baseline. We used data from the first, second, fourth and fifth waves of TRAILS. The mean ages of the respondents were 11.08 years (SD=0.56), 13.53 years (SD=0.52), 19.02 years (SD=0.58) and 22.26 years (SD=0.65) across waves, respectively. Individuals who were lost to follow-up (n=705, 31.6%) were more likely to be male (Z=−7.28, p<0.001) and to have parents with low educational level (Z=−10.41, p<0.001). More indepth information about TRAILS has been presented elsewhere.20–22 Informed consent was obtained from all participants and their parents.

    Measures

    AEs between ages 0 and 16

    We included 14 AEs across five domains, akin to Rod et al23: peer influences (bullying victimisation, peer rejection), loss or threat of loss (familial death, illness of a sibling, physical illness of a parent), family dynamics (parental divorce, familial conflicts, parental mental health problems, parental addiction), material deprivation (parental unemployment, financial difficulties) and maltreatment (sexual abuse, physical abuse, emotional abuse). Answer categories for all AEs were 0: not exposed and 1: exposed. Exposure to most of the AEs was reported by parents (apart from bullying victimisation, peer rejection, and abuse). A detailed description of the AEs with information on measurement instruments, the timing of measurement, the reporter and the observation period is provided in online supplemental data 1 and table S1.

    Supplemental material

    Labour market participation states and employment conditions at age 22

    Three labour market participation states were considered: educational status (not currently studying, following a high school education, currently pursuing a vocational degree, currently pursuing a (applied) university degree), employment status (unemployed, employed, self-employed) and benefit status (receiving benefits yes/no). Three employment conditions were considered: contract type (permanent contract, temporary contract, temporary employment agency contract), work hours (12–35 hours, >35 hours) and monthly income (€300 or less, between €301 and €600, between €601 and €900, between €901 and €1200, and €1201 or higher). A detailed description of the labour market participation states and employment conditions, including rationales for specific categorisations, is provided in online supplemental data 2.

    Control variables

    Parental educational level at baseline was included to adjust for confounding.12 15 24 25 Parental educational level was classified into low (primary, lower vocational and lower secondary education), medium (intermediate vocational, intermediate secondary education and higher secondary education) and high (higher vocational education and university).26 The highest educational level across both parents was considered.

    Demographic characteristics

    We included sex (male/female) at baseline to further describe our study sample.

    Statistical analysis

    Labour market outcome groups were derived using LCA,27 using the following variables: educational status, a combined employment status and contract type variable (unemployed, employed with a permanent contract, employed with a temporary contract, employed with a temporary employment agency contract, self-employed), benefit status, work hours and monthly income. For the purpose of the analysis, the work hours variable was complemented with two additional categories (side jobs, reflecting 1–12 hours work and 0 hours to align with the unemployment category of the employment status variable). Five different class solutions were estimated (ranging from two to six classes). Model fit (global) was assessed through the Bayesian information criterion and Akaike information criterion metrics. Model interpretability and entropy (class separability) were also considered. The optimal class solution was (parametrically) bootstrapped to assess local model fit through an investigation of bivariate residuals between variables included in the LCA. Direct effects were included to address any violations of the local independence assumption.28 The four-class solution was found to be most optimal. This decision was mainly based on class separation and interpretability, as the fit statistics of the four-class solution were similar to the five-class and six-class solutions (see online supplemental table S2).

    To investigate whether exposure to the AEs significantly differed across the labour market outcome groups (overall effect), we included each AE to the latent class solutions separately. In case of a significant overall effect (alpha of 0.05), pairwise comparisons (Wald tests) were conducted to ascertain group-specific differences (alpha of 0.05). We present conditional probabilities of exposure to each AE given class membership from these multinomial logistic regression models as well as gamma coefficients (due to the use of effects coding as opposed to reference category coding) and their SEs. Pairwise comparisons were adjusted for multiple testing with the Benjamini-Hochberg procedure (false discovery rate (FDR) procedure).29 We set the fixed FDR to 5%, indicating that 5% of all significant findings were allowed to be false-positive. All analyses included parental education to adjust for confounding. We used proportional class assignment with maximum-likelihood-based (ML) adjustments to avoid downwards-biased estimates (ie, the bias-adjusted three-step approach).30 Missing data were considered missing at random and handled using full information maximum likelihood in all analyses (see online supplemental table S5 for an overview of missing data across all study variables). All analyses were conducted with the Latent GOLD 6.0 software.31

    Sensitivity analyses

    We re-examined the associations between AEs and labour market outcome groups using the six-class solution and found relatively few differences as compared with the four-class solution (see online supplemental tables S3 and S4). Comparisons with the three-class and five-class solutions were not performed due to multiple violations of the local independence assumptions in these solutions.

    Results

    Most respondents were female (56.3%) and had parents with a medium educational level (62.9%). The prevalence rates of AEs ranged from 2.6% (parental addiction) to 25.9% (parental physical illness) (see table 1).

    View this table:
    Table 1

    Sample characteristics

    Labour market outcome groups at age 22

    Table 2 contains the conditional probabilities of the labour market outcome variables across each of the four labour market outcome groups (latent classes). The four groups were labelled ‘students with side jobs’ (n=640, 41.2%), ‘early workers’ (n=413, 27.1%), ‘non-working students’ (n=386, 25.3%) and ‘inactive individuals’ (n=85, 5.6%), respectively.

    View this table:
    Table 2

    Labour market outcome variables at age 22, total and conditional on labour market outcome groups

    Type of AEs between ages 0 and 16 and labour market outcome groups at age 22

    The labour market outcome groups differed significantly in exposure to nine AEs across domains: bullying victimisation, peer rejection, parental divorce, familial conflicts, financial difficulties, parental addiction, sexual abuse, physical abuse and emotional abuse. A total of 29 pairwise comparisons (out of 54) were significant (table 3). The majority of the significant pairwise comparisons pertain to differences between the inactive individuals group and the other groups, with the inactive individuals group being significantly more likely to be exposed to each of the nine AEs compared with the other groups (with the exception of financial difficulties, for which the inactive group only differed from the students with side jobs). The early workers group had significantly higher exposure rates to parental divorce than students with side jobs (22.7% vs 12.9%).

    View this table:
    Table 3

    Distribution of adverse experiences between ages 0 and 16, conditional on labour market outcome groups at age 22

    The four groups did not differ in exposure to any of the AEs belonging to the domain of loss or threat of loss, parental unemployment and parental mental health problems. Table 4 provides the estimates for the magnitude of the associations between AEs and the four labour market outcome groups.

    View this table:
    Table 4

    Parameter estimates (gamma coefficients and their SEs) for the associations between adverse experiences between ages 0 and 16 across the four labour market outcome groups at age 22

    Discussion

    Our results show four distinct labour market outcome groups in young adults aged 22, students with side jobs, early workers, non-working students and inactive individuals, which differ in type of childhood adversity exposure. Exposure across several domains of childhood adversity was significantly more likely among the inactive individuals group as compared with all other groups. We also found that exposure to parental divorce was significantly more likely among the early workers group compared with students with side jobs. We found no evidence that the labour market outcome groups differed in exposure to any of the AEs belonging to the domain of loss or threat of loss, parental unemployment and parental mental health problems. Our results suggest that inactive individuals differ from the other groups in exposure to a wide variety of AEs, both in the family (eg, parental addiction) and peer contexts (eg, bullying victimisation). Differences in exposure to AEs among the other groups are less pronounced and limited to parental divorce.

    Across studies, it becomes apparent that exposure to especially AEs that pertain to direct psychological (eg, bullying victimisation, peer rejection, emotional abuse) or physical harm (physical abuse), or increase the likelihood thereof (eg, parental addiction, conflicts in the family, financial difficulties), is very high among individuals whom we characterised as inactive individuals in this study.12 14–16 One likely explanation for this finding is that exposure to the aforementioned AEs is strongly associated with increased mental health problems,32 33 which in turn have been demonstrated to be associated with unemployment in young adulthood.34

    Our findings suggest that students with side jobs, early workers and non-working students largely do not differ in exposure to childhood adversity exposure, despite differences in labour market participation states and employment conditions. Similar to Hansen et al,15 we found that exposure to parental divorce was more likely among young adults with an early labour market entry than among students with side jobs. Parental divorce has been shown to be associated with school dropout and lower educational attainment, which might explain why the exposure to parental divorce was higher among the early workers than among students.35 Unlike Hansen et al,15 we did not find that the early workers were also more exposed to abuse and parental addiction. It is possible that this difference is attributable to differences in assessment since we used parental report, whereas Hansen et al15 used adult self-report; adult self-report of childhood adversity has been shown to be more strongly associated with outcomes than parental report.36 37 The impact of differences in reporters (eg, caregiver, self-report, register-based assessment) warrants attention in future research.

    This study has several strengths. TRAILS’ longitudinal design made it possible to investigate the association between types of AEs and labour market outcomes in young adults in a prospective fashion. The richness of the TRAILS cohort also allowed us to include a broad array of AEs with relatively short recall periods in the family and peer contexts, including often-ignored AEs such as financial difficulties and bullying victimisation.38 39 This study is also the first to confirm and extend previous findings from primarily Scandinavian studies with data from the Netherlands.12 14 15 Nevertheless, the findings should be interpreted considering several limitations. First, although TRAILS has only limited loss to follow-up, the study is subject to selection bias due to attrition. Missing data analyses showed that men and participants with parents with lower educational level were more likely to not participate over time. Because exposure to childhood adversity was more likely to occur in families of lower socioeconomic background (results not shown), the findings of this study may underestimate the differences in exposure to childhood adversity among the four labour market outcome groups (see also Walsh et al25). Studies with data on more individuals from a lower socioeconomic background may provide more insight into the generalisability of our findings. Second, due to the relatively small sample size of this study, it is possible that smaller differences in exposure to AEs between the four labour market outcome groups were not uncovered due to low statistical power (although these differences might be negligible and thus irrelevant). We are confident however that most differences between the inactive individuals group and the other groups were uncovered because these groups differed drastically in their exposure to the included AEs.

    Our findings have several implications for research, policy and practice. Our study highlights that especially maltreatment and peer influences, and family dynamics and material deprivation to some extent, are important for labour market outcomes in young adulthood. We recommend practitioners and policy makers to focus specifically on those AEs that may be regarded as psychologically or physically harmful (eg, abuse, peer victimisation), or AEs that increase the likelihood thereof (eg, parental addiction, conflicts in the family, financial difficulties), as those AEs are particularly important for labour market inactivity early in young adults’ working careers. Nevertheless, it is also important to not overlook individuals who are exposed to a few specific AEs (eg, parental divorce) and who are likely to be employed in young adulthood. These individuals require different support from individuals who are inactive on the labour market. We have three important recommendations for future research. First, future studies should explore the mechanisms that explain the associations between specific AEs and labour market outcomes. It is likely that the different labour market outcome groups are affected by childhood adversity through different mechanisms. Elevated and ongoing mental health problems and school dropout are likely to play a prominent role for especially inactive individuals.16 32 34 35 Divorced parents of early workers might have fewer resources to support their children in their educational careers, which may lead to young adults to discontinue studying after obtaining a low-level qualification and start working thereafter.19 Understanding through which mechanics AEs, both individually and conjunctively, affect labour market outcomes may be vital to reduce the long-term impact of childhood adversity. Second, future studies may investigate the role of co-occurrence of AEs, preferably including information on the timing, frequency or severity,18 that were predictive of labour market outcomes. Doing so may provide more fine-grained insight into exposure patterns of the various labour market outcome groups described in this study. Third and last, future studies should further investigate to what extent labour market outcomes in young adulthood mediate the association between childhood adversity and health inequalities throughout the life course, taking into account the differences in exposure patterns among different groups of young adults.7–9 To conclude, our findings show that the type of AEs matters in labour market outcomes in young adults. Stakeholders in policy and practice may benefit from an increased awareness of the differences between AEs in their importance for labour market outcomes in young adults when aiming to reduce the occurrence of childhood adversity and its potential consequences for labour market outcomes at the population level.

    Data availability statement

    Data may be obtained from a third party and are not publicly available. TRAILS data of the T1, T2, T3, T4 and T5 measurement waves are deposited in the Data Archiving and Networked Services of the Royal Dutch Academy of Sciences (DANS-KNAW) and access can be requested at http://www.dans.knaw.nl.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants and was approved by the Dutch Central Committee on Research Involving Human Subjects (CCMO, NL67411.042.18) and conforms to the principles laid out in the Declaration of Helsinki. Participants and their parents gave informed consent to participate in the study before taking part.

    Acknowledgments

    This research is part of the TRacking Adolescents’ Individual Lives Survey (TRAILS). Participating centres of TRAILS include various departments of the University Medical Center and University of Groningen; the Erasmus University Medical Center, Rotterdam; the University of Utrecht; the Radboud Medical Center, Nijmegen; and the Parnassia Psychiatric Institute, all in the Netherlands. The authors are grateful to all adolescents, their parents and teachers who participated in this research.

    References

    1. ↵
      1. Giano Z,
      2. Wheeler DL,
      3. Hubach RD
      . The frequencies and disparities of adverse childhood experiences in the U.S. BMC Public Health 2020;20:1327. doi:10.1186/s12889-020-09411-z
    2. ↵
      1. Bussemakers C,
      2. Kraaykamp G,
      3. Tolsma J
      . Co-occurrence of adverse childhood experiences and its association with family characteristics. A latent class analysis with dutch population data. Child Abuse Negl 2019;98:104185. doi:10.1016/j.chiabu.2019.104185
      OpenUrlPubMed
    3. ↵
      1. Felitti VJ,
      2. Anda RF,
      3. Nordenberg D, et al
      . Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The adverse childhood experiences (ACE) study. Am J Prev Med 1998;14:245–58. doi:10.1016/s0749-3797(98)00017-8
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. McLaughlin KA,
      2. Sheridan MA
      . Beyond cumulative risk: a dimensional approach to childhood adversity. Curr Dir Psychol Sci 2016;25:239–45. doi:10.1177/0963721416655883
      OpenUrlCrossRefPubMed
    5. ↵
      1. Hughes K,
      2. Bellis MA,
      3. Hardcastle KA, et al
      . The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. Lancet Public Health 2017;2:e356–66. doi:10.1016/S2468-2667(17)30118-4
      OpenUrlCrossRefPubMed
    6. ↵
      1. Nelson CA,
      2. Scott RD,
      3. Bhutta ZA, et al
      . Adversity in childhood is linked to mental and physical health throughout life. BMJ 2020;371:m3048. doi:10.1136/bmj.m3048
    7. ↵
      1. Metzler M,
      2. Merrick MT,
      3. Klevens J, et al
      . Adverse childhood experiences and life opportunities: shifting the narrative. Children and Youth Services Review 2017;72:141–9. doi:10.1016/j.childyouth.2016.10.021
      OpenUrlCrossRef
    8. ↵
      1. Font SA,
      2. Maguire-Jack K
      . Pathways from childhood abuse and other adversities to adult health risks: the role of adult socioeconomic conditions. Child Abuse Negl 2016;51:390–9. doi:10.1016/j.chiabu.2015.05.013
      OpenUrlCrossRef
    9. ↵
      1. Schurer S,
      2. Trajkovski K,
      3. Hariharan T
      . Understanding the mechanisms through which adverse childhood experiences affect lifetime economic outcomes. Labour Economics 2019;61:101743. doi:10.1016/j.labeco.2019.06.007
      OpenUrl
    10. ↵
      1. Liu Y,
      2. Croft JB,
      3. Chapman DP, et al
      . Relationship between adverse childhood experiences and unemployment among adults from five U.S. states. Soc Psychiatry Psychiatr Epidemiol 2013;48:357–69. doi:10.1007/s00127-012-0554-1
      OpenUrlCrossRefPubMed
    11. ↵
      1. Hardcastle K,
      2. Bellis MA,
      3. Ford K, et al
      . Relationship between adverse childhood experiences and unemployment among adults from five US states. Eur J Health Econ 2013;20:357–73. doi:10.1371/journal.pone.0075860
      OpenUrl
    12. ↵
      1. Lund T,
      2. Andersen JH,
      3. Winding TN, et al
      . Negative life events in childhood as risk indicators of labour market participation in young adulthood: a prospective birth cohort study. PLoS One 2013;8:e75860. doi:10.1371/journal.pone.0075860
    13. ↵
      1. Mersky JP,
      2. Topitzes J
      . Comparing early adult outcomes of maltreated and non-maltreated children: a prospective longitudinal investigation. Child Youth Serv Rev 2010;32:1086–96. doi:10.1016/j.childyouth.2009.10.018
      OpenUrlCrossRefWeb of Science
    14. ↵
      1. Pitkänen J,
      2. Remes H,
      3. Moustgaard H, et al
      . Parental socioeconomic resources and adverse childhood experiences as predictors of not in education, employment, or training: a Finnish register-based longitudinal study. Journal of Youth Studies 2021;24:1–18. doi:10.1080/13676261.2019.1679745
      OpenUrl
    15. ↵
      1. Hansen CD,
      2. Kirkeby MJ,
      3. Kjelmann KG, et al
      . The importance of adverse childhood experiences for labour market trajectories over the life course: a longitudinal study. BMC Public Health 2021;21:2044. doi:10.1186/s12889-021-12060-5
    16. ↵
      1. Jaffee SR,
      2. Ambler A,
      3. Merrick M, et al
      . Childhood maltreatment predicts poor economic and educational outcomes in the transition to adulthood. Am J Public Health 2018;108:1142–7. doi:10.2105/AJPH.2018.304587
      OpenUrl
    17. ↵
      1. Strøm IF,
      2. Thoresen S,
      3. Wentzel-Larsen T, et al
      . Violence, bullying and academic achievement: a study of 15-year-old adolescents and their school environment. Child Abuse Negl 2013;37:243–51. doi:10.1016/j.chiabu.2012.10.010
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Lacey RE,
      2. Minnis H
      . Practitioner review: twenty years of research with adverse childhood experience scores-advantages, disadvantages and applications to practice. J Child Psychol Psychiatry 2020;61:116–30. doi:10.1111/jcpp.13135
      OpenUrlCrossRefPubMed
    19. ↵
      1. Bussemakers C,
      2. Kraaykamp G
      . Youth adversity, parental resources and educational attainment: contrasting a resilience and a reproduction perspective. Research in Social Stratification and Mobility 2020;67:100505. doi:10.1016/j.rssm.2020.100505
      OpenUrl
    20. ↵
      1. de Winter AF,
      2. Oldehinkel AJ,
      3. Veenstra R, et al
      . Evaluation of non-response bias in mental health determinants and outcomes in a large sample of pre-adolescents. Eur J Epidemiol 2005;20:173–81. doi:10.1007/s10654-004-4948-6
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Huisman M,
      2. Oldehinkel AJ,
      3. de Winter A, et al
      . Cohort profile: the Dutch “ tracking adolescents ” individual liv’s' surv’y'; trails. Int J Epidemiol 2008;37:1227–35. doi:10.1093/ije/dym273
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Oldehinkel AJ,
      2. Rosmalen JG,
      3. Buitelaar JK, et al
      . Cohort profile update: the tracking adolescents’ individual lives survey (trails). Int J Epidemiol 2015;44:76–76n. doi:10.1093/ije/dyu225
      OpenUrlCrossRefPubMed
    23. ↵
      1. Rod NH,
      2. Bengtsson J,
      3. Budtz-Jørgensen E, et al
      . Trajectories of childhood adversity and mortality in early adulthood: a population-based cohort study. Lancet 2020;396:489–97. doi:10.1016/S0140-6736(20)30621-8
      OpenUrlPubMed
    24. ↵
      1. Hardcastle K,
      2. Bellis MA,
      3. Ford K, et al
      . Measuring the relationships between adverse childhood experiences and educational and employment success in england and wales: findings from a retrospective study. Public Health 2018;165:106–16. doi:10.1016/j.puhe.2018.09.014
      OpenUrlPubMed
    25. ↵
      1. Walsh D,
      2. McCartney G,
      3. Smith M, et al
      . Relationship between childhood socioeconomic position and adverse childhood experiences (aces): a systematic review. J Epidemiol Community Health 2019;73:1087–93. doi:10.1136/jech-2019-212738
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Veldman K,
      2. Bültmann U,
      3. Stewart RE, et al
      . Mental health problems and educational attainment in adolescence: 9-year follow-up of the trails study. PLoS One 2014;9:e101751. doi:10.1371/journal.pone.0101751
    27. ↵
      1. Porcu M,
      2. Giambona F
      . Introduction to latent class analysis with applications. The Journal of Early Adolescence 2017;37:129–58. doi:10.1177/0272431616648452
      OpenUrlCrossRef
    28. ↵
      1. Hagenaars JA
      . Latent structure models with direct effects between indicators. Sociological Methods & Research 1988;16:379–405. doi:10.1177/0049124188016003002
      OpenUrlCrossRefWeb of Science
    29. ↵
      1. Benjamini Y,
      2. Hochberg Y
      . Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological) 1995;57:289–300. doi:10.1111/j.2517-6161.1995.tb02031.x
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Bakk Z,
      2. Tekle FB,
      3. Vermunt JK
      . Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological Methodology 2013;43:272–311. doi:10.1177/0081175012470644
      OpenUrlCrossRefWeb of Science
    31. ↵
      1. Vermunt JK,
      2. Magidson J
      . Technical guide for latent GOLD 5. 1: basic, advanced, and syntax. Stat Innov Inc, 2013.
    32. ↵
      1. Bevilacqua L,
      2. Kelly Y,
      3. Heilmann A, et al
      . Adverse childhood experiences and trajectories of internalizing, externalizing, and prosocial behaviors from childhood to adolescence. Child Abuse Negl 2021;112:104890. doi:10.1016/j.chiabu.2020.104890
      OpenUrlPubMed
    33. ↵
      1. Ettekal I,
      2. Li H,
      3. Chaudhary A, et al
      . Chronic, increasing, and decreasing peer victimization trajectories and the development of externalizing and internalizing problems in middle childhood. Dev Psychopathol 2022:1–19. doi:10.1017/S0954579422000426
    34. ↵
      1. de Groot S,
      2. Veldman K,
      3. Amick Iii BC, et al
      . Does the timing and duration of mental health problems during childhood and adolescence matter for labour market participation of young adults? J Epidemiol Community Health 2021;75:896–902. doi:10.1136/jech-2020-215994
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Marlow SA,
      2. Rehman N
      . The relationship between family processes and school absenteeism and dropout: a meta-analysis. The Educational and Developmental Psychologist 2021;38:3–23. doi:10.1080/20590776.2020.1834842
      OpenUrl
    36. ↵
      1. Baldwin JR,
      2. Reuben A,
      3. Newbury JB, et al
      . Agreement between prospective and retrospective measures of childhood maltreatment: a systematic review and meta-analysis. JAMA Psychiatry 2019;76:584–93. doi:10.1001/jamapsychiatry.2019.0097
      OpenUrl
    37. ↵
      1. Batty GD,
      2. Kivimaki M
      . Adverse childhood experiences and adult health: the need for stronger study designs to evaluate impact. J Epidemiol Community Health 2021;75:485–8. doi:10.1136/jech-2020-215870
      OpenUrlFREE Full Text
    38. ↵
      1. Braveman P,
      2. Heck K,
      3. Egerter S, et al
      . Economic hardship in childhood: A neglected issue in ACE studies? Matern Child Health J 2018;22:308–17. doi:10.1007/s10995-017-2368-y
      OpenUrlCrossRef
    39. ↵
      1. Troop-Gordon W
      . Peer victimization in adolescence: the nature, progression, and consequences of being bullied within a developmental context. J Adolesc 2017;55:116–28. doi:10.1016/j.adolescence.2016.12.012
      OpenUrlCrossRef

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Twitter @vriesTR

    • Correction notice This article has been corrected since it first published online. Specifically, Table 3 of the article (column 'Inactive individuals', row 'Sexual Abuse') contained a wrong number. This has been corrected to 12.7.

    • Contributors TRdV: conceptualisation, methodology, formal analysis, data curation, writing—original draft, visualisation, project administration. IA: conceptualisation, writing—original draft, supervision. AJO: conceptualisation, writing—review and editing. UB: conceptualisation, writing—original draft, supervision, funding acquisition. TRdV acts as guarantor for this study.

    • Funding TRdV and IA were funded through the Netherlands Organisation for Scientific Research (NWO) Vici Project (‘Today’s youth is tomorrow’s workforce: Generation Y at work’; NWO Vici 453-16-007/2735), which was granted to UB. TRAILS has been financially supported by various grants from the NWO (Medical Research Council Program Grant GB-MW 940-38-011; ZonMw Brainpower Grant 100-001-004; ZonMw Risk Behavior and Dependence Grant 60-60600-97-118; ZonMw Culture and Health Grant 261-98-710; Social Sciences Council Medium-Sized Investment Grants GB-MaGW 480-01-006 and GB-MaGW 480-07-001; Social Sciences Council Project Grants GB-MaGW 452-04-314 and GB-MaGW 452-06-004; ZonMw Longitudinal Cohort Research on Early Development and Treatment in Mental Health Care Grant 636340002; NWO Large-Sized Investment Grant 175.010.2003.005; NWO Longitudinal Survey and Panel Funding 481-08-013 and 481-11-001; NWO Vici 016.130.002, 453-16-007/2735 and VICI 191.021; NWO Gravitation 024.001.003), the Dutch Ministry of Justice (WODC), the European Science Foundation (EuroSTRESS Project FP-006), the European Research Council (ERC-2017-STG-757364 and ERC-CoG-2015-681466), Biobanking and Biomolecular Resources Research Infrastructure (BBMRI-NL) (CP 32), the Gratama Foundation, the Jan Dekker Foundation, the participating universities, and Accare Centre for Child and Adolescent Psychiatry.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.