research paper about parental separation

• Open access
• Published: 29 November 2023

Parental separation, negative life events and mental health problems in adolescence

• Kateryna Karhina   ORCID: orcid.org/0000-0002-5067-1609 1 ,
• Tormod Bøe 1 , 2 ,
• Mari Hysing 1 , 2 &
• Sondre Aasen Nilsen 2 , 3  

BMC Public Health volume  23 , Article number:  2364 ( 2023 ) Cite this article

5349 Accesses

Metrics details

Parental separation is associated with mental health problems in adolescence. One suggested pathway for this association is through the accumulated exposure to stress and other negative life events. This study aimed to document the distribution of negative life events among adolescents with separated compared to non-separated parents, and to assess the direct and interactive associations between parental separation, negative life events, and mental health problems in adolescence.

Data stem from the cross-sectional population-based youth@hordaland study of adolescents (aged 16–19) conducted in Norway in 2012, providing self-reported information about parental separation, negative life events, and depression-, anxiety-, conduct-, and ADHD symptoms. Regression analyses were used to assess the direct and interactive associations between parental separation, negative life events, and mental health problems.

Adolescents with separated parents had more mental health problems across all symptom scales compared to peers with non-separated parents, with standardized mean differences [SMDs] ranging from 0.15 to 0.20. Negative life events moderately attenuated these differences (reduced the SMDs with about 0.04–0.08, depending on the outcome). However, none of the interactions between parental separation and negative life events on mental health problems were statistically significant.

Conclusions

Higher exposure to negative life events explains parts of the association between parental separation and mental health problems in adolescence. However, a parental separation does not seem to increase the vulnerability of the effects of negative life events on adolescents’ mental health. Assessing exposure to negative life events is important when providing mental health services to adolescents, particularly to those who have parents separated.

Peer Review reports

Introduction

Parental separation is associated with more mental health problems in children and adolescents [ 1 ]. For example, studies have shown that adolescents with separated parents have a higher risk of emotional and behavioural problems and have a higher prevalence of psychiatric disorders than peers living in non-separated two-parent families [ 2 , 3 ]. Moreover, these differences are not only confined to childhood but tend to persist into adulthood [ 4 ].

One pathway through which parental separation influences children’s mental health is through adverse negative life events. As noted by the divorce-stress-adjustment perspective [ 5 ], a parental separation may set into motion several family stressors, such as parental mental health issues, financial problems and family conflicts. Such stressors may also put children at risk of serious negative life events, such as accidents, violence, and death of close ones [ 6 ], which are the focus of the current investigation.

A growing body of research has documented that exposure to negative life events is a robust predictor of mental health outcomes extending from childhood through adulthood [ 7 , 8 , 9 , 10 ]. According to the cumulative risk hypothesis, it is the accumulation of risk, more than the specific types of risk factors, that are important for children and adolescents’ mental health [ 11 , 12 ]. For example, an early longitudinal study from the U.S. of at-risk children ( N  = 171) found that the number of risk exposures at age 7 strongly predicted both externalizing and internalizing mental health problems at age 16 [ 13 ]. More recent work has provided similar findings. Indeed, a recent study drawing on data from the national representative National Survey of Children’s Health in the U.S. ( N  = 29,617), found a graded association between number of negative life events and outcomes such as depression, anxiety, conduct disorder, and ADHD in adolescence [ 14 ].

Although the separate associations between parental separation and negative life events with mental health outcomes in adolescence have been well-documented, there is a lack of studies that have examined how parental separation and negative life events jointly influence adolescents’ mental health, and whether negative life events disproportionately affect children with separated parents. Studies often include parental divorce or separation within a cumulative risk framework [ 13 , 15 ], but few studies have considered whether these factors interact. However, experiencing a parental divorce or separation may not only increase the risk of experiencing negative life events, but may also lower the tolerance levels of future negative events, making it more difficult for children and youth to cope with such events later in life [ 16 ]. This may occur due to the emotional distress children may experience, as well as other family issues such as family conflicts and lower parental support and supervision in the post-separation period [ 17 ].

Although very little empirical work has considered such interaction effects, one study found a significant interaction effect between parental marital status and childhood abuse, whereby the association between childhood abuse and lifetime ADHD diagnosis was stronger for adolescents who also had experienced parental separation [ 12 ]. Another study found that children who had experienced divorce were generally more likely to rate negative life events as more difficult to adjust to, compared to those without divorce experiences [ 6 ]. We are, however, unaware of more population-based studies that have examined whether cumulative exposure to negative life events disproportionately affects adolescents with separated parents.

The Norwegian welfare state is characterized by an elaborate social safety net, including free access to health care and family related benefits. Highly subsidized childcare and schools, combined with generous parental leave rights, have led to the dual-earner family being the norm [ 18 ]. Moreover, in the case of divorce, custodians are provided support through tax deductions, cash allowances, and child support. Despite these benefits, a parental divorce or separation have been associated with more mental health problems also in Norway, with similar effect sizes to those found in the US [ 19 ]. Previous investigations drawing on the same data as the current have also shown that those in separated or single-parent families have more sleep problems [ 20 ] and poorer academic outcomes than peers from non-divorced two-parent families [ 21 ]. We are, however, unaware of previous Norwegian studies that have investigated the interrelationship between parental separation, negative life events, and mental health outcomes in adolescence.

Given these considerations, this study sought to document the distribution of negative life events among adolescents based on whether their parents had divorced/separated or were living together. We further sought to examine the degree to which exposure to negative life events may account for some of the association between parental separation and mental health outcomes in adolescence, and whether the association between negative life events and mental health problems is moderated by parental separation. Based on existing theory and empirical research, we expected that adolescents with separated parents had a higher exposure to negative life events and more mental health problems than peers with non-separated parents, and that the higher exposure to negative life events would account for some of the association between parental separation and adolescent mental health problems. Due to few previous empirical studies, we had no directional hypothesis regarding potential interaction effects between negative life events and parental separation.

Design and procedure

Data stem from the youth@hordaland (y@h) study, conducted in former Hordaland County in Western Norway in 2012. The y@h study was a population-based study of adolescents with the main aim of assessing mental health and health service use during adolescence. All adolescents born between 1993 and 1995 were invited to participate (aged 16 to 19), whereby 10,257 agreed, yielding a participation rate of 53% for the entire study. Information about the study was given by email, and one school hour was allocated to complete the survey by answering an electronic questionnaire. A teacher organized the data collection and ensured confidentiality. Information about the study was sent by post to those not in school, and alternative solutions for participation were made for students in hospitals or institutions. The present study is based on a subsample of adolescents who had valid information about parental separation, negative life experiences, sociodemographic characteristics, and mental health outcomes ( N  = 7953).

Representativeness of the sample

Previous investigations have shown that the grade point average of participants in the youth@hordaland study was approximately equal to national and county-level statistics [ 22 ]. However, the proportion of parents with higher education was higher than observed in official statistics, although differences in methodology do not allow for direct comparison by numbers. Similarily, the proportion of adolescents categorized as living in a non-separated two-parent family was slightly higher (70% vs. 63%), when compared to the closest possible age group (15–17 years) presented in official statistics in 2012 [ 23 ]. Based on these considerations, we note that our sample was skewed toward higher socioeconomic status but representative regarding academic achievement and fairly well captured the distribution of youth in separated families based on official statistics.

Conduct problems were measured by the Youth Conduct Disorder (YCD) scale [ 24 ]. The YCD consists of eight items measured on a binary scale with the answers “yes” and “no”. YCD is a part of the Diagnostic Interview Schedule for Children Predictive Scales [ 24 ]. The coefficient omega for ordinal data [ω 𝑢 ] was 0.75.

Attention deficit hyperactivity disorder (ADHD) symptoms were measured using the Adult ADHD self-report scale (ASRS) from the World Health Organization [ 25 ]. The ASRS consists of 24 items rated on a five-point scale, ranging from “never” to “very often”. Previous studies on the ASRS have shown high construct validity and internal consistency in adolescent populations [ 26 ]. The ω u for ordinal data for the sum score was 0.90.

Depressive symptoms were measured by the short version of the Mood and Feelings Questionnaire (SMFQ) [ 27 ]. SMFQ consists of thirteen items rated on a 3-point Likert scale (“not true”, “sometimes true”, “true”). The SMFQ have been found to confirm well to a unidimensional measurement scale in a previous study from the youth@hordaland sample [ 28 ]. The ω u for the sum score was 0.92.

Anxiety symptoms were assessed by the short version of the Screen for Child Anxiety Related Emotional Disorders (SCARED). The SCARED consists of five items measured on a 3-point Likert scale (“not true”, “sometimes true” and “often true”). This short scale has shown similar psychometric properties as the full 41- item version [ 29 ]. The ω u for the sum score was 0.69.

Negative life events were measured by five items detailing whether the adolescents had ever experienced (i) ‘a catastrophe or serious accident’, (ii) ‘violence from an adult’, (iii) ‘witnessed someone you care about being exposed to violence from an adult’ and (iv) ‘unwanted sexual actions. In addition, a fifth item detailed whether the adolescents had experienced ‘death of someone close to you’. In case adolescents had experienced death of someone close, they were asked to specify their relationship with the person(s). Death of a (v) parent/guardian, (vi) sibling, (vii) close friend and (viii) girlfriend/boyfriend were included as separate negative events. A cumulative indices of negative life events were created by adding up the number of positive responses. The index scale was coded into 0,1,2,3, and 4 or more, due to few respondents reporting to have experienced five or more negative life events.

Perceived economic well-being was measured by asking adolescents to rate how they perceive their economic well-being compared to others. The 3-point Likert scale was used for rating, i.e., “poorer than others”, “equal to others” or “better than others”.

Parental education

Maternal and paternal education were reported separately, using the options “primary school,” “high school vocational,” “high school general,” “college/university less than four years,” “college/university four years or more,” and “do not know.” We combined the two high school alternatives into one category (i.e., “high school”).

Sex was derived from the personal identity number in the Norwegian National Population Register.

Age was taken through the personal identity number in the National Population Register of Norway. Exact age was calculated using the date of participation in the study.

Parental separation was defined according to the adolescents’ answer to the questions: “Do your parents live together?” and “Have your parents divorced or separated?” Adolescents confirming that their biological parents did not live together and that they had divorced or separated were defined as having separated parents.

Statistical analyses

The distribution of the negative life events between adolescents with and without separated parents was examined using Fisher’s exact test due to low number of expected frequencies in some cells. To visualize the distribution of the index scale of number of negative life events between the two groups, a violin plot with a boxplot embedded inside was created.

A series of regression models were conducted to examine the associations between parental separation and negative life events on mental health outcomes. For each outcome, we first tested a crude model with parental separation as the focal independent variable. Gender was added to this model due to the unbalanced gender distribution across the two groups and the well-known association between gender and mental health problems in adolescence. In the next model, the cumulative index of number of negative life events was added to assess whether it attenuated the association between parental separation and mental health problems. In the third and fourth models, we added perceived economic-wellbeing and parental education, to examine whether the association between parental separation and mental health problems would further attenuate when accounting for socioeconomic differences between the two groups. Finally, we tested an interaction model to assess whether the association between negative life events and mental health was moderated by parental separation. The interaction models were tested both without and with adjustments for sociodemographic characteristics. Linear models are presented, as preliminary analyses did not suggest any strong nonlinear associations between negative life events and the mental health outcomes. However, we also present robustness analyses using negative life events as a factor variable (represented by a vector of dummy coded variables) to account for any nonlinear patterns in the associations. The results of the interaction analyses are presented visually. All data preparations, main analyses, and visualizations were conducted using R version 4.1.3 for Windows [ 30 ] using functions from the “tidyverse” package [ 31 ]. The reliability for each mental health scale was assessed by the coefficient omega for categorical/ordinal data [ 32 ], using the lavaan [ 33 ] and semTools [ 34 ] R-packages.

Sociodemographic characteristics by parental separation are presented in Table  1 . Among those with separated parents, 59% were female participants. Overall, there was a tendency towards lower parental education and perceived economic well-being among those with separated compared to non-separated parents.

The frequency of experiencing negative life events stratified by parental separation is presented in Table  2 . Adolescents with separated parents were more likely to have experienced a serious accident, having been exposed to or witnessed violence from a grownup, and unwanted sexual acts than adolescents with non-separated parents. They were also more likely to having experienced death of a close friend or family member but were not more likely to have experienced death of a girlfriend/boyfriend. Adolescents with separated parents were more likely to experience multiple negative life events than those from non-separated families.

Regression results

In the regression analyses, parental separation was associated with significantly more symptoms of ADHD, conduct problems, depression, and anxiety, with similar standardized mean differences [SMDs] across the symptom scores (range 0.15 to 0.19). Accounting for number of negative life events (Model 2) partly attenuated and reduced the strength of the associations by about 0.04 to 0.08 standard deviation units but all associations between parental separation and mental health outcomes remained significant. Further adjustments for perceived economic well-being (Model 3) and parental education (Model 4), slightly attenuated the associations between parental separation and mental health problems. However, all associations between parental separation and mental health problems remained statistically significant in the fully adjusted models. The association between negative life events and mental health problems hardly changed when adjusting for perceived economic-well-being and parental education (See Table  3 for details).

None of the interaction terms between parental separation and negative life events on the mental health outcomes were statistically significant, neither before nor after adjustments for sociodemographic characteristics, suggesting that the slopes of the associations between number of negative life events on all outcomes were similar and not statistically significantly different between adolescents with and without separated parents (see Fig.  1 ).

Predicted symptom scores by NLE and parental separation. (Note: This figure shows the predictions from the interaction analyses between parental separation and negative life events (NLE), adjusted by gender (predicted estimates for girls shown). The y-axis is scaled in standard deviation units, where 0 represent the predicted mean of the entire sample. The error bars including the shaded areas represent 95% confidence intervals)

Robustness analyses using negative life events as a vector of dummy coded variables to better account for any nonlinear associations yielded a highly similar pattern of results as reported above, both in analysis that adjusted for negative life events and in analysis including an interaction term between parental separation and negative life events (see Supplementary Fig.  1 ). There were some indications that the association between negative life events and anxiety and depressive symptoms became somewhat stronger from 4 or more negative life events (i.e., a curvilinear relationship). However, we consider this trend uncertain as shown by the wide confidence intervals at 4 or more negative life events.

The aims of the current study were to investigate the distribution of negative life events by parental separation and to assess the direct and interactive association between parental separation and negative life events on adolescents’ mental health problems. For all negative life events except death of a girl/boyfriend, we found that adolescents with separated parents had a higher exposure than peers with non-separated parents. The associations between parental separation and mental health problems were moderately attenuated when adjusting for negative life events, suggesting that negative life events partially accounted for the higher risk of mental health problems among adolescents with separated parents. However, no interaction effects between parental separation and negative life events were detected. Thus, the idea that parental separation amplifies the consequences of negative life events on adolescents’ mental health problems was not supported in the current study.

Adolescents exposed to parental separation had more symptoms of mental health problems relative to their nonexposed peers. The associations were consistent across all mental health outcomes, with adolescents with separated parents scoring about 0.15 to 0.20 standard deviation units higher than nonexposed peers. Overall, these findings align with many previous studies [ 3 , 35 , 36 ], and highlight that adolescents with separated parents are at a higher risk of mental health problems across several domains of mental health. Considering the prevalent occurrence of childhood exposure to parental divorce or separation, the modest effect sizes identified in this study may still be important, as scaled to the population level and across time, they affect many children.

Exposure to negative life events was consistently more common among adolescents with separated parents. For some of these events, such as violence from a grownup and unwanted sexual act, it was twice as likely to have experienced them for adolescents from separated families than their peers with non-separated parents. Adolescents with separated parents were also more likely to have experienced multiple negative life events. Although few studies have focused on more serious negative life events exposure among adolescents with separated parents, our results generally corroborate the findings of existing work suggesting higher exposure of family stress and negative life events among youth with separated parents [ 5 , 6 ].

When negative life events were accounted for, the strength of the association between parental separation and mental health problems was reduced for about 30 − 40%, depending on the outcome measure. Although we cannot determine whether they experienced these negative life events before or after the parental separation took place, these findings highlight that exposure to multiple negative life events may be one important factor in understanding the association between parental separation and mental health problems in adolescence. Further adjustments of PEW and parental education slightly attenuated the association between parental separation and the mental health outcomes. However, parental separation remained a significant predictor of all mental health problems even after such adjustments, suggesting that socioeconomic differences between those with and without separated parents cannot fully account for the differences in symptom scores. It has been suggested that a parental separation may not only be a cause of future negative life events and stressors but may also act as a source of vulnerability to the effects of such events [ 17 ]. We did not find support for this hypothesis, as none of the interaction effects between parental separation and negative life events on adolescents’ mental health problems were statistically significant. Thus, from our results, parental separation and negative life events appear to have more of an additive rather than interactive association with mental health problems in adolescence. These results diverge somewhat from a previous study suggesting that the association between physical abuse and the likelihood of ADHD diagnosis were stronger for adolescents with separated compared to non-separated parents [ 12 ]. However, that study focused on a sample of adolescents reported to be abused by the Child Protective services and did not measure cumulative exposure to negative life events, and hence, do not necessarily compare nor generalize to a sample of adolescents drawn from the general population.

Strengths and limitations

A strength of this study was the large sample size of a well-defined cohort of older adolescents, combined with the assessment of multiple negative life events and mental health problems in adolescents using established measures. The findings should nonetheless be interpreted considering some methodological limitations. First, as many other studies, this study is cross-sectional and we cannot disentangle the direction of the associations between parental separation, negative life events, and mental health outcomes. For example, we cannot establish the degree to which a parental separation leads to higher exposure to negative life events, whether a particular life event was related to the separation of parents, or if these events were already experienced before the separation. Similarly, we cannot ascertain that more mental health problems among youth with separated parents are due to tparental separation or other unmeasured factors may select adolescents into experiencing parental separation and mental health problems. For example, parental mental health problems and interparental conflicts may both increase the risk of parental separation and mental health problems among adolescents [ 35 ]. Longitudinal studies tracking these events from before to after the separation are needed to better disentangle the direction of the associations reported in this study. A further potential limitation was the use of self-reported experiences of negative life events, as there is a risk of recollection bias [ 37 ]. Thus, some caution should be applied when interpreting the rates of negative life events in the present study. However, we see no reason to assume that recollection bias is more pronounced among adolescents with separated parents compared to their peers with non-separated parents, and that recollection bias would invalidate the relative magnitude of exposure to such events between the two groups, or their associations to mental health outcomes. Moreover, although our measure of negative life events consisted of a parsimonious set of serious events that adolescents may be exposed to, no follow-up information regarding the adolescents´ perception of the severity of the events was collected, which could have further nuanced our findings. Moreover, we cannot exclude the possibility that a more exhaustive measure, also including less severe events, could have yielded different findings. Finally, with a response rate of 53% for the entire study and that the sample size was further reduced due to item non-response, caution should be applied when generalizing the results to the population level.

Our results demonstrate that adolescents with separated parents are more exposed to negative life events than peers with non-separated parents, and that this added burden may contribute to the observed association between parental separation and mental health problems in adolescence. However, parental separation does not appear to increase the vulnerability to mental health problems following exposure to negative life events among older adolescents.

Implications and contributions

Given the robust association between negative life events and mental health problems in adolescence, our results highlight the importance of assessing exposure to such events when providing mental health services to adolescents, and particularly when giving services to adolescents with separated parents.

Data availability

The Norwegian Health research legislation and the Norwegian Ethics committees require explicit consent from the participants to transfer health research data outside of Norway. For the Bergen Child study, which constitutes the data for the current analyses, ethics approval was also contingent on storing the research data on secure storage facilities located in our research institution, which prevents us from providing the data as supplementary information or to transfer it to data repositories. Individual requests for data access should be sent to [email protected] contacting the last author of this paper.

Härkönen J, Bernardi F, Boertien D. Family dynamics and child outcomes: an overview of research and open questions. Eur J Popul = Revue europeenne de demographie. 2017;33(2):163.

PubMed   Google Scholar  

Chase-Lansdale PL, Cherlin AJ, Kiernan KE. The long‐term effects of parental divorce on the mental health of young adults: a developmental perspective. Child Dev. 1995;66(6):1614–34.

Article   CAS   PubMed   Google Scholar  

Perales F, et al. Family structure and childhood mental disorders: new findings from Australia. Soc Psychiatry Psychiatr Epidemiol. 2017;52(4):423–33.

Article   PubMed   Google Scholar  

Tullius JM, et al. Adolescents’ mental health problems increase after parental divorce, not before, and persist until adulthood: a longitudinal TRAILS study. Eur Child Adolesc Psychiatry. 2022;31(6):969–78.

Amato PR. The consequences of divorce for adults and children. J Marriage Family. 2000;62(4):1269–87.

Article   Google Scholar  

Ängarne-Lindberg T, Wadsby M. Fifteen years after parental divorce: mental health and experienced life-events. Nord J Psychiatry. 2009;63(1):32–43.

Atkinson L, et al. Cumulative risk, cumulative outcome: a 20-Year longitudinal study. PLoS ONE. 2015;10(6):e0127650–0.

Article   PubMed   PubMed Central   Google Scholar  

Evans GW, Li D, Whipple SS. Cumulative risk and child development. Psychol Bull. 2013;139(6):1342–96.

Sahle BW, et al. The association between adverse childhood experiences and common mental disorders and suicidality: an umbrella review of systematic reviews and meta-analyses. Eur Child Adolesc Psychiatry. 2022;31(10):1489–99.

Chapman DP, et al. Adverse childhood experiences and the risk of depressive disorders in adulthood. J Affect Disord. 2004;82(2):217–25.

Rutter M. Protective factors in children’s responses to stress and disadvantage. Ann Acad Med Singapore. 1979;8(3):324–38.

CAS   PubMed   Google Scholar  

Cohen AJ, et al. Interactional effects of marital status and physical abuse on adolescent psychopathology. Child Abuse Negl. 2002;26(3):277–88.

Appleyard K, et al. When more is not better: the role of cumulative risk in child behavior outcomes. J Child Psychol Psychiatry. 2005;46(3):235–45.

Bomysoad RN, Francis LA. Adverse childhood experiences and Mental Health conditions among adolescents. J Adolesc Health. 2020;67(6):868–70.

Bøe T, et al. Cumulative effects of negative life events and family stress on children’s mental health: the Bergen Child Study. Soc Psychiatry Psychiatr Epidemiol. 2018;53:1–9.

Ge X, Natsuaki MN, Conger RD. Trajectories of depressive symptoms and stressful life events among male and female adolescents in divorced and nondivorced families. Dev Psychopathol. 2006;18(1):253–73.

Aseltine RH Jr. Pathways linking parental divorce with adolescent depression. J Health Soc Behav, 1996: p. 133–48.

Kitterød RH, Wiik KA. Shared residence among parents living apart in Norway. Family Court Review. 2017;55(4):556–71.

Breivik K, Olweus D. Children of divorce in a scandinavian welfare state: are they less affected than US children? Scand J Psychol. 2006;47(1):61–74.

Nilsen SA, et al. Sleep in adolescence: considering family structure and family complexity. J Marriage Family. 2022;84(4):1152–74.

Nilsen SA, et al. Divorce and adolescent academic achievement: heterogeneity in the associations by parental education. PLoS ONE. 2020;15(3):e0229183.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Hysing M, et al. Sleep and academic performance in later adolescence: results from a large population-based study. J Sleep Res. 2016;25(3):318–24.

Statistics Norway. Families and households. 06204: Children 0–17 Years, by Number of Parents in the Family, the Parents’ Cohabitation Arrangements and the Child’s Age 2001–2021. 2021.

Lucas CP, et al. The DISC predictive scales (DPS): efficiently screening for diagnoses. J Am Acad Child Adolesc Psychiatry. 2001;40(4):443–9.

Kessler RC, et al. The World Health Organization adult ADHD self-report scale (ASRS): a short screening scale for use in the general population. Psychol Med. 2005;35(2):245–56.

Adler LA, et al. Preliminary examination of the reliability and concurrent validity of the Attention-Deficit/Hyperactivity disorder self-report scale v1.1 Symptom Checklist to rate symptoms of Attention-Deficit/Hyperactivity disorder in adolescents. J Child Adolesc Psychopharmacol. 2012;22(3):238–44.

Thapar A, McGuffin P. Validity of the shortened Mood and feelings Questionnaire in a community sample of children and adolescents: a preliminary research note. Psychiatry Res. 1998;81(2):259–68.

Lundervold AJ, et al. Symptoms of depression as reported by Norwegian adolescents on the short Mood and feelings Questionnaire. Front Psychol. 2013;4:613.

Birmaher B, et al. Psychometric properties of the screen for child anxiety related Emotional disorders (SCARED): a replication study. J Am Acad Child Adolesc Psychiatry. 1999;38(10):1230–6.

R_Core_Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. 2022.

Wickham H, et al. Welcome to the Tidyverse. J open Source Softw. 2019;4(43):1686.

Flora DB. Your coefficient alpha is probably wrong, but which coefficient omega is right? A tutorial on using R to obtain better reliability estimates. Adv Methods Practices Psychol Sci. 2020;3(4):484–501.

Rosseel Y. Lavaan: an R package for structural equation modeling. J Stat Softw. 2012;48:1–36.

Jorgensen TD, et al. semTools: useful tools for structural equation modeling. R Package Version. 2018;0(5):1.

Google Scholar  

Amato PR. Research on Divorce: Continuing trends and New Developments. J Marriage Family. 2010;72(3):650–66.

Nilsen SA, et al. Divorce and family structure in Norway: associations with adolescent mental health. J Divorce Remarriage. 2018;59(3):175–94.

Dohrenwend BP et al. Life events and other possible psychosocial risk factors for episodes of schizophrenia and major depression: a case-control study. 1995.

Download references

Acknowledgements

The authors are grateful to all participants who made this study possible and would also like to thank the Bergen Child Study group.

Open access funding provided by University of Bergen. The study has been funded by the Norwegian Research Council grant # 302225.

Open access funding provided by University of Bergen.

Author information

Authors and affiliations.

Department of Psychosocial Science, Faculty of Psychology, University of Bergen, Bergen, Norway

Kateryna Karhina, Tormod Bøe & Mari Hysing

Regional Centre for Child and Youth Mental Health and Child Welfare, NORCE Norwegian Research Centre, Bergen, Norway

Tormod Bøe, Mari Hysing & Sondre Aasen Nilsen

Department of Health Promotion and Development, Faculty of Psychology, University of Bergen, Bergen, Norway

Sondre Aasen Nilsen

You can also search for this author in PubMed   Google Scholar

Contributions

Kateryna Karhina and Sondre Aasen Nilsen analyzed data, wrote the main manuscript text and prepared figures. All the authors discussed the findings and reviewed the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Kateryna Karhina .

Ethics declarations

Ethics approval and consent to participate.

Following the regulations from the Reginal Committee for Medical and Health Research Ethics (REC) and Norwegian Health authorities, adolescents aged 16 and older can make own decisions about their health, including participation in health studies. Thus, adolescents informed consent was obtained to participate in the study. Their parents/guardians received information about the study in advance. All the methods were performed in accordance with Declaration of Helsinki. The study was approved by REC in the Western Norway (2011/811/REK Vest).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Karhina, K., Bøe, T., Hysing, M. et al. Parental separation, negative life events and mental health problems in adolescence. BMC Public Health 23 , 2364 (2023). https://doi.org/10.1186/s12889-023-17307-x

Download citation

Received : 23 March 2023

Accepted : 22 November 2023

Published : 29 November 2023

DOI : https://doi.org/10.1186/s12889-023-17307-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Parental separation
• Mental health
• Negative life events
• Adverse childhood conditions
• Adolescence

BMC Public Health

ISSN: 1471-2458

• Submission enquiries: [email protected]
• General enquiries: [email protected]

Parental separation and children's behavioral/emotional problems: the impact of parental representations and family conflict

Affiliation.

• 1 Department of Child & Adolescent Psychiatry, University of Basel, Switzerland. [email protected]
• PMID: 20377637
• DOI: 10.1111/j.1545-5300.2010.01310.x

In this longitudinal study, we examine whether the effect of parental separation on kindergarten children's behavioral/emotional problems varies according to the level of family conflict, and children's parental representations. One hundred and eighty seven children were assessed at ages 5 and 6. Family conflict was assessed using parents' ratings. Children's parental representations were assessed using a story-stem task. A multiinformant approach (parent, teacher, child) was employed to assess children's behavioral/emotional problems. Bivariate results showed that separation, family conflict, and negative parental representations were associated with children's behavioral/emotional problems. However, in multivariate analyses, when controlling for gender and symptoms at age 5, we found that children of separated parents who showed negative parental representations had a significantly greater increase in conduct problems between 5 and 6 than all other children. In terms of emotional symptoms and hyperactivity, symptoms at 5 and (for hyperactivity only) gender were the only predictors for symptoms 1 year later. Our results suggest that kindergarten children's representations of parent-child relationships moderate the impact of parental separation on the development of conduct problems, and underline play and narration as a possible route to access the thoughts and feelings of young children faced with parental separation.

• Adaptation, Psychological*
• Affective Symptoms / etiology*
• Attention Deficit Disorder with Hyperactivity
• Child, Preschool
• Conflict, Psychological*
• Longitudinal Studies
• Mental Disorders / etiology*
• Models, Psychological
• Multivariate Analysis
• Parent-Child Relations*
• Psychometrics
• Stress, Psychological*
• Surveys and Questionnaires

The Long-Term Effects of Parental Separation on Childhood Multidimensional Deprivation: A Lifecourse Approach

• Original Research
• Published: 18 January 2019
• Volume 144 , pages 921–954, ( 2019 )

Cite this article

• Marion Leturcq   ORCID: orcid.org/0000-0003-2243-1760 1 &
• Lidia Panico   ORCID: orcid.org/0000-0001-9291-7622 1  

1603 Accesses

10 Citations

79 Altmetric

10 Mentions

Explore all metrics

A large literature has documented the impact of parental separation on children’s financial poverty. However, income has been increasingly criticized as an indicator of childhood living conditions and deprivation. In this paper, we propose a conceptual framework and adapt existing measures of adult multi-domain deprivation to produce childhood deprivation indicators that are age-specific and child-centred. These new indicators allow within-individual, longitudinal analyses to measure the impact of a shock on childrens living conditions. We apply this method to explore the long term effects of parental separation on childhood deprivation, considering four dimensions of children’s lives: leisure; material conditions; parenting behaviours and routines; and basic material goods. We track children over the first decade of life by using a nationally representative UK cohort of over 18,000 children. Using a fixed-effects framework, we find that, while the increase in income poverty after parental separation is large, the impact on childhood deprivation was more mixed. Our results suggest that, while facing strong financial constraints, separated parents cut back on normative but costly activities such as holidays and outings, but attempt to maintain children’s basic material circumstances and their day-to-day parenting and routines, at least around separation. However, heterogeneous effects exist, suggesting that parents’ pre-separation social and economic capital may play an important role. This approach therefore adds more precision and nuance to our understanding of the processes around parental separation and its impacts on children.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Does Household Income Affect children’s Outcomes? A Systematic Review of the Evidence

The Causal Effects of Parental Divorce and Parental Temporary Separation on Children’s Cognitive Abilities and Psychological Well-being According to Parental Relationship Quality

The long shadow of child labour on adolescent mental health: a quantile approach

The survey weights have been corrected to be representative of all children, including twins and triplets. As we keep only one Millennium child per household, we multiplied by 2 children born with a twin, and we multiply by 3 children born with triplets. There are 240 pairs of twins in the sample and 10 households with triplets. We observe parental separation for 38 households with twins, and 2 parental separations for households with triplets.

We have tested the robustness of our results to the different definitions of the equivalence scale and to different poverty thresholds. While the magnitude of the impact is slightly affected by the choice of the equivalence scale and the poverty line, the overall evolution of the impact is not affected. The relevant table can be found in the Online Appendix, table 3.

We prefer to construct a series of dummies indicating whether the current child observation is 0–1 years after separation, 2–3 years, 4–7 years, and 8 to 11 years. We do not use a linear variable of time (in months) since separation to avoid making linear assumptions between time and the impact of separation, and to be able to measure short versus long term effects.

Briefly, the National Vocational Qualifications is a system of competence-based education and training that aims to record individual levels of competence. Briefly, an NVQ5 is equivalent to a graduate degree; an NVQ3 is equivalent to two A-levels, a high-school qualification. NVQ levels correspond to: NVQ5—Higher degree, NVQ4—First degree/diploma, NVQ 3—A/AS levels, NVQ 2—GCSE grades A-C, NVQ 1—GCSE grades D-G

These variables are included in the descriptive statistics but not in the analysis because we used a fixed-effect framework.

For the same proportions item by item, see Online Resource 1.

We tested the robustness of our results to different equivalence scales and poverty lines. We found that a complete recovery in terms of income poverty occurred 8–11 years after separation when more stringent measures of poverty are used (the Oxford equivalent scales or poverty line defined as 50% of the median income). However, whatever the measure of monetary poverty, we always find that the risk of monetary poverty is higher than the risk of deprivation after parental separation.

Aassve, A., Betti, G., Mazzuco, S., & Mencarini, L. (2007). Marital disruption and economic well-being: A comparative analysis. Journal of the Royal Statistical Society: Series A (Statistics in Society) , 170 (3), 781–799.

Article   Google Scholar  

Alkire, S., Apablaza, M., Chakravarty, S., Yalonetzky, G., et al. (2014). Measuring chronic multidimensional poverty: A counting approach. Technical report , Queen Elizabeth House, University of Oxford.

Alkire, S., & Foster, J. (2011). Counting and multidimensional poverty measurement. Journal of Public Economics , 95 (7), 476–487.

Amato, P. R. (2001). Children of divorce in the 1990s: An update of the Amato and Keith (1991) meta-analysis. Journal of Family Psychology , 15 (3), 355.

Amato, P. R. (2005). The impact of family formation change on the cognitive, social, and emotional well-being of the next generation. The Future of Children , 15 , 75–96.

Amato, P. R., & Anthony, C. J. (2014). Estimating the effects of parental divorce and death with fixed effects models. Journal of Marriage and Family , 76 (2), 370–386.

Attanasio, O. P., & Rios-Rull, J. V. (2003). Consumption smoothing and extended families. Econometric Society Monographs , 37 , 209–242.

Google Scholar  

Becker, G. S. (1981). A treatise on the family . Cambridge: Harvard University Press.

Ben-Shlomo, Y., & Kuh, D. (2002). A life course approach to chronic disease epidemiology: Conceptual models, empirical challenges and interdisciplinary perspectives. International Journal of Epidemiology , 31 (2), 285–293.

Blau, D. M. (1999). The effect of income on child development. Review of Economics and Statistics , 81 (2), 261–276.

Bourguignon, F., & Chakravarty, S. R. (2003). The measurement of multidimensional poverty. The Journal of Economic Inequality , 1 (1), 25–49.

Bradshaw, J., Hoelscher, P., & Richardson, D. (2007). An index of child well-being in the European Union. Social Indicators Research , 80 (1), 133–177.

Brewer, M., Nandi, A., et al. (2014). Partnership dissolution: How does it affect income, employment and well-being? . Colchester: Institute for Social and Economic Research, University of Essex.

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology , 22 (6), 723.

Carlson, M. J., & Corcoran, M. E. (2001). Family structure and children’s behavioral and cognitive outcomes. Journal of Marriage and Family , 63 (3), 779–792.

Cherlin, A. J., Chase-Lansdale, P. L., & McRae, C. (1998). Effects of parental divorce on mental health throughout the life course. American Sociological Review , 63 (2), 239–249.

Conger, R. D., Conger, K. J., Elder, G. H., Lorenz, F. O., Simons, R. L., & Whitbeck, L. B. (1992). A family process model of economic hardship and adjustment of early adolescent boys. Child Development , 63 (3), 526–541.

De Neubourg, C., Chai, J., de Milliano, M., & Plavgo, ID. (2013). Cross-country MODA study: Multiple overlapping deprivation analysis (MODA). Technical note. Technical report .

Dewilde, C., & Uunk, W. (2008). Remarriage as a way to overcome the financial consequences of divorcea test of the economic need hypothesis for European women. European Sociological Review , 24 (3), 393–407.

Dex, S., & Joshi, H. (2005). Children of the 21st century: From birth to nine months . Bristol: Policy Press.

Book   Google Scholar  

Dickerson, A. P., & Popli, G. (2015). The many dimensions of child poverty: Evidence from the UK Millennium Cohort Study. The Sheffield economic research paper series (SERPS) .

Duncan, G., & Magnuson, K. (2013). The importance of poverty early in childhood. Policy Quarterly , 9 (2), 12–17.

Duncan, G. J., & Murnane, R. J. (2011). Whither opportunity? Rising inequality, schools, and children’s life chances . New York: Russell Sage Foundation.

DWP. (2013). Percentage of children living with both birth parents, by age of child and household income; and estimated happiness of parental relationships. Technical report, DWP available at: https://www.gov.uk/government/statistics/percentage-of-children-living-with-both-birth-parents . Accessed 16 Jan 2018.

Ferraro, K. F., & Shippee, T. P. (2009). Aging and cumulative inequality: How does inequality get under the skin? The Gerontologist , 49 (3), 333–343.

Fisher, H., & Low, H. (2015). Financial implications of relationship breakdown: Does marriage matter? Review of Economics of the Household , 13 (4), 735–769.

Fitzgerald, J., Gottschalk, P., & Moffitt, R. (1998). An analysis of sample attrition in panel data: The Michigan panel study of income dynamics. Journal of Human Resources , 33 (2), 251–299.

Fusco, A., Guio, A.-C., & Marlier, E. (2011). Income poverty and material deprivation in european countries. Technical report, LISER .

Gambaro, L., & Joshi, H. (2016). Moving home in the early years: What happens to children in the UK? Longitudinal and Life Course Studies , 7 (3), 265–287.

Guio, A.-C., Fusco, A., & Marlier, E., et al. (2009). A European Union approach to material deprivation using eu-silc and eurobarometer data. Integrated research infrastructure in the socio-economic sciences (IRISS) working paper series (Vol. 19).

Hansen, K. (2012). Millennium cohort study first, second, third and fourth surveys: A guide to the datasets (6th ed.). London: Centre for Longitudinal Studies, University of London.

Jackson, M., Kiernan, K., & McLanahan, S. (2017). Maternal education, changing family circumstances, and children’s skill development in the United States and UK. The Annals of the American Academy of Political and Social Science , 674 (1), 59–84.

Jansen, M., Mortelmans, D., & Snoeckx, L. (2009). Repartnering and (re) employment: Strategies to cope with the economic consequences of partnership dissolution. Journal of Marriage and Family , 71 (5), 1271–1293.

Jenkins, S. P. (2012). Marital splits and income changes over the longer term. In J. F. Ermisch & M. Bryning (Eds.), Changing relationships . London: Routledge.

Ketende, S. (2008). Millennium Cohort Study: Technical report on response. second edition. Technical report . London: Centre for Longitudinal Studies.

Kiernan, K., & Mensah, F. (2010). Partnership trajectories, parent and child well-being. Children of the 21st century , 2 , 77–94.

Kiernan, K. E., & Mensah, F. K. (2009). Poverty, maternal depression, family status and children’s cognitive and behavioural development in early childhood: A longitudinal study. Journal of Social Policy , 38 (04), 569–588.

Masarik, A. S., & Conger, R. D. (2017). Stress and child development: A review of the family stress model. Current Opinion in Psychology , 13 (Supplement C), 85–90. Relationships and stress.

Mayer, S. E. (1997). What money can’t buy: Family income and children’s life chances . Cambridge: Harvard University Press.

McKeever, M., & Wolfinger, N. H. (2001). Reexamining the economic costs of marital disruption for women. Social Science Quarterly , 82 (1), 202–217.

McLanahan, S. (2004). Diverging destinies: How children are faring under the second demographic transition. Demography , 41 (4), 607–627.

McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent. What hurts, what helps . Cambridge: ERIC.

McLanahan, S., Tach, L., & Schneider, D. (2013). The causal effects of father absence. Annual Review of Sociology , 39 , 399–427.

McMunn, A. M., Nazroo, J. Y., Marmot, M. G., Boreham, R., & Goodman, R. (2001). Children’s emotional and behavioural well-being and the family environment: Findings from the health survey for England. Social Science & Medicine , 53 (4), 423–440.

Moore, J. C., Stinson, L. L., & Welniak, E. J. (2000). Income measurement error in surveys: A review. Journal of Ofiicial Statistics-Stockholm , 16 (4), 331–362.

Mooyaart, J. E., & Liefbroer, A. C. (2016). The influence of parental education on timing and type of union formation: Changes over the life course and over time in the Netherlands. Demography , 53 (4), 885–919.

Narayan, D., et al. (2000). Poverty is powerlessness and voicelessness. Finance & Development , 37 (4), 18–21.

Nolan, B., Whelan, C. T., et al. (2011). Poverty and deprivation in Europe . Oxford: OUP Catalogue.

OECD. (2009). What are equivalence scales? Technical report, organisation for Eeconomic co-operation and development available at: http://www.oecd.org/els/soc/OECD-Note-EquivalenceScales.pdf . Accessed 16 Jan 2018.

Oroyemi, P. (2009). Understanding the risks of social exclusion across the life course—Families with children: A research report.

Osborne, C., Berger, L. M., & Magnuson, K. (2012). Family structure transitions and changes in maternal resources and well-being. Demography , 49 (1), 23–47.

Panico, L., Bartley, M., Kelly, Y., McMunn, A., & Sacker, A. (2010). Changes in family structure in early childhood in the Millennium Cohort Study. Population Trends , 142 , 78.

Phoenix, A. (1996). Social constructions of lone motherhood. In E. B. Silva (Ed.), Good enough mothering? Feminist perspectives on lone motherhood (p. 175). London: Routledge.

Plewis, I., Calderwood, L., Hawkes, D., Hughes, G., & Joshi, H.(2007). Millennium Cohort Study: Technical report on sampling. Technical report . London: Centre for Longitudinal Studies.

Ravallion, M. (1996). Issues in measuring and modeling poverty . Washington: World Bank Publications.

Rowlingson, K., & McKay, S. (2005). Lone motherhood and socio-economic disadvantage: Insights from quantitative and qualitative evidence. The Sociological Review , 53 (1), 30–49.

Russ, S. A., Larson, K., Tullis, E., & Halfon, N. (2014). A lifecourse approach to health development: Implications for the maternal and child health research agenda. Maternal and Child Health Journal , 18 (2), 497–510.

Ryan, R. M., & Claessens, A. (2013). Associations between family structure changes and childrens behavior problems: The moderating effects of timing and marital birth. Developmental Psychology , 49 (7), 1219.

Ryan, R. M., Claessens, A., & Markowitz, A. J. (2015). Associations between family structure change and child behavior problems: The moderating effect of family income. Child Development , 86 (1), 112–127.

Schoon, I., Hope, S., Ross, A., & Duckworth, K. (2010). Family hardship and children’s development: The early years. Longitudinal and Life Course Studies , 1 (3), 209–222.

Sen, A. (1999). Commodities and capabilities . Oxford: OUP Catalogue.

Steele, F., Sigle-Rushton, W., & Kravdal, Ø. (2009). Consequences of family disruption on childrens educational outcomes in Norway. Demography , 46 (3), 553–574.

Stiglitz, J., Sen, A. K., & Fitoussi, J.-P. (2009). The measurement of economic performance and social progress revisited: Reflections and overview.

Thomson, E., Hanson, T. L., & McLanahan, S. S. (1994). Family structure and child well-being: Economic resources vs. parental behaviors. Social Forces , 73 (1), 221–242.

Thomson, E., & McLanahan, S. S. (2012). Reflections on family structure and child well-being: Economic resources vs. parental socialization. Social Forces , 91 (1), 45–53.

Tomlinson, M., Walker, R., & Williams, G. (2008). Measuring poverty in Britain as a multi-dimensional concept, 1991 to 2003. Journal of Social Policy , 37 (04), 597–620.

Townsend, P. (1979). Poverty in the United Kingdom: A survey of household resources and standards of living . Berkeley: University of California Press.

Unicef et al. (2013). Child well being in rich countries: A comparative overview . Florence: UNICEF Office of Research-Innocenti.

Van de Kaa, D. J. (2001). Postmodern fertility preferences: From changing value orientation to new behavior. Population and Development Review , 27 , 290–331.

Waldfogel, J., Craigie, T.-A., & Brooks-Gunn, J. (2010). Fragile families and child wellbeing. The Future of Children/Center for the Future of Children, the David and Lucile Packard Foundation , 20 (2), 87.

Wooldridge, J. (2002). Econometric analysis of cross-sectional and panel data . Cambridge: MIT Press.

Download references

Acknowledgements

We thank the Millennium Cohort Study (MCS) families for their time and cooperation, as well as the MCS team at the Institute of Education. The MCS is funded by the Economic and Social Research Council, UK. We would also like to thank Elizabeth Thompson for helpful comments on an earlier draft, as well as participants to the several seminars, including the Paris Seminar of Economic Demography; the Laboratoire de Sociologie Quantitative, CREST Paris; and the Duke University Population Research Institute seminar.

Author information

Authors and affiliations.

Institut National d’Etudes Demographiques (INED), 133 boulevard Davout, 75980, Paris Cedex 20, France

Marion Leturcq & Lidia Panico

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Lidia Panico .

Ethics declarations

Conflict of interest.

The authors declare that they have no conflict of interest.

This work was partly funded by an IReSP (Institut de Recherche en Santé Publique) Grant to Lidia Panico (AAP-2015-10).

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 0 KB)

Rights and permissions.

Reprints and permissions

About this article

Leturcq, M., Panico, L. The Long-Term Effects of Parental Separation on Childhood Multidimensional Deprivation: A Lifecourse Approach. Soc Indic Res 144 , 921–954 (2019). https://doi.org/10.1007/s11205-018-02060-1

Download citation

Accepted : 27 December 2018

Published : 18 January 2019

Issue Date : 30 July 2019

DOI : https://doi.org/10.1007/s11205-018-02060-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Child poverty
• Deprivation indicators
• Family instability
• Longitudinal methods
• United Kingdom

JEL Classification

• Find a journal
• Publish with us
• Track your research

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Impact of parental separation or divorce on school performance in preterm children: A population-based study

Roles Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft

* E-mail: [email protected]

Affiliation INSERM, CIC 1413, Nantes University Hospital, Nantes, France

Roles Writing – review & editing

Affiliations Department of Child and Adolescent Psychiatry, Centre Hospitalier Esquirol, Limoges, France, INSERM, UMR1094, Tropical Neuroepidemiology, Limoges, France

Roles Conceptualization, Data curation, Writing – review & editing

Affiliations INSERM, CIC 1413, Nantes University Hospital, Nantes, France, Department of Neonatal Medicine, Nantes University Hospital, Nantes, France

Roles Data curation, Writing – review & editing

Affiliation Department of Neonatal Medicine, Nantes University Hospital, Nantes, France

Affiliation Réseau “Grandir Ensemble”, Nantes University Hospital, Nantes, France

Affiliation Department of Neonatal Medicine, Angers University Hospital, Angers, France

Affiliation Department of Neonatal Medicine, Le Mans Hospital, Le Mans, France

Roles Conceptualization, Writing – review & editing

Affiliation SOS Prema (parents of French preterm children organization), Boulogne-Billancourt, France

Affiliations INSERM, CIC 1413, Nantes University Hospital, Nantes, France, Department of Neonatal Medicine, Nantes University Hospital, Nantes, France, Réseau “Grandir Ensemble”, Nantes University Hospital, Nantes, France

Roles Conceptualization, Methodology, Supervision, Writing – review & editing

• Simon Nusinovici, 
• Bertrand Olliac, 
• Cyril Flamant, 
• Jean-Baptiste Müller, 
• Marion Olivier, 
• Valérie Rouger, 
• Géraldine Gascoin, 
• Hélène Basset, 
• Charlotte Bouvard, 

• Published: September 7, 2018
• https://doi.org/10.1371/journal.pone.0202080
• Reader Comments

The objective of this study was to quantify the possible decrease in school performance at five years of age in preterm children associated with parental separation or divorce, and to test whether this effect varies according to the child’s age at the time of the separation. This study included 3,308 infants delivered at < 35 weeks of gestation born between 2003 and 2011 who were enrolled in the population-based LIFT cohort and who had an optimal neurodevelopmental outcome at two years of age. These infants were evaluated by their teachers to assess their abilities and behavior when they had reached five years of age, using the Global School Adaptation (GSA) questionnaire. The mean GSA score was 50.8 points. Parental separations (assessed as parents either living together or living separately) were associated with a decrease in school performance at five years of age, although this was only the case for children who exhibited difficulties at school (3.7 points, p < 0.01). A decrease in school performance only occurred when parental separations took place between 3 and 5 years after the child’s birth. Parental separation was associated with a decrease in these children’s levels of motivation, autonomy, and manual dexterity. This study indicates that preterm infants of parents who had separated are particularly at risk of a lower scholar performance.

Citation: Nusinovici S, Olliac B, Flamant C, Müller J-B, Olivier M, Rouger V, et al. (2018) Impact of parental separation or divorce on school performance in preterm children: A population-based study. PLoS ONE 13(9): e0202080. https://doi.org/10.1371/journal.pone.0202080

Editor: Harald Ehrhardt, Center of Pediatrics, GERMANY

Received: July 21, 2017; Accepted: July 27, 2018; Published: September 7, 2018

Copyright: © 2018 Nusinovici et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data are available from the scientific Committee of the LIFT cohort for researchers who meet the criteria for access to confidential health data. Interested researchers have to comply with the French legislation i.e. require the advice of the “Comité consultatif sur le traitement de l’information en matière de recherché sur le domaine de la santé” (CCTIRS) as well as the authorization of the “Commission nationale de l’information et des libertés” (CNIL) for the treatment of personal health data. Research projects have also to be approved by an independent Ethics Committee. Contact information is as follows: Phone number: 00 332 40 48 55 81, Email: [email protected] .

Funding: This work was supported by grants from the Regional Health Agency of Pays de la Loire. SOS Prema provided support in the form of salaries for author CB, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific role of this author is articulated in the ‘author contributions’ section.

Competing interests: We have the following interest. Charlotte Bouvard is the founder of SOS Prema. There are no patents, products in development or marketed products to declare. This does not alter our adherence to all the PLOS ONE policies on sharing data and materials, as detailed online in the guide for authors.

Introduction

Parental separation has been associated with reduced cognitive development and educational performance[ 1 – 6 ]. Since parental separation can result in multiple negative effects, including perceived guilt, blame, stressors, and diminished resources for the children, it is not surprising that parental separation has also been reported to negatively affect a child’s motivation, engagement, and learning-related behavior in the classroom[ 6 , 7 ].

Although several studies have investigated the impact of the timing of the parental separation on scholastic performance, they have yielded discordant results. Most studies found that for younger children (i.e. those in elementary school) the adverse impact was more pronounced than for older children (i.e. those in high school)[ 4 , 8 , 9 ], suggesting that younger children may feel more anxious about abandonment and that they may be more likely to blame themselves[ 10 ]. However, others studies have found that divorce may adversely affect adolescents more than elementary school-age children[ 11 ], or that the child’s age did not alter the effect of divorce on academic achievement[ 12 ]. Lastly, very few studies have focused on the effect of parental separation on very young children.

One would expect that a negative effect of parental separation on school performance should also be apparent in children who were delivered preterm, in light of the vulnerability of this population. Preterm births are indeed associated with lower educational performances[ 13 – 15 ]. Preterm children are at risk of achieving lower cognitive test scores, and their level of immaturity is directly proportional to the mean cognitive scores at school age[ 13 ]. To our knowledge, the effect of parental separation on the school performance of children who were delivered preterm has not been investigated to date. The objective of this study was (i) to quantify the possible decrease in school performance at five years of age that is associated with parental separation or divorce in preterm children and (ii) to test whether this effect varies according to the child’s age at the time of the separation.

Materials and methods

Study population.

The study population comprised surviving preterm infants enrolled in the Loire Infant Follow-up Team (LIFT), born at less than 35 weeks of gestation between January 2003 and December 2010. The LIFT network includes 24 maternity clinics in the Pays-de-la-Loire region (one of the 13 administrative regions of France) with the objective of screening for early clinical anomalies associated with preterm births and to provide specifically adapted care. The follow-up consisted of standardized visits by trained physicians at 3, 6, 9, 18, and 24 months as well as at 3, 4, 5, 6, and 7 years after the birth of the child. The children were evaluated at two and five years of corrected age to assess their neurodevelopmental outcomes. Written consent was obtained at the time of enrollment in the study. The cohort was registered with the French Data Protection Authority (“Commission National de l’Informatique et des Libertés”–CNIL) (n o 851117).

Selection of optimal infants at two years of age

At two years of age, the neurodevelopmental evaluation was based on a physical examination by a trained pediatrician, a psychomotor evaluation by a psychologist and/or a questionnaire completed by the parents. This evaluation aimed to assess whether the infants had an optimal or a non-optimal neurodevelopment. Only infants that were considered to be optimal at two years of age were included.

Abilities and behavior of the children at five years of age

At five years of age, the evaluation was based on a questionnaire for the children’s teachers, the “Global School Adaptation” (GSA) score, which allows for assessment of the children’s abilities and behaviors in the classroom. This questionnaire has been used previously by the French ministry of education to investigate the performance of children in the French public school system. It has been validated as a screening tool for adverse neurodevelopmental outcomes in preterm children[ 16 ]. Indeed, for more than 89% of the children in our cohort, the GSA score was highly or moderately consistent with the IQ score. At the age of five years ± two months, the kindergarten teachers were asked by the parents to complete this questionnaire. It comprises 20 items, exploring five areas: language, transferable skills, socialization, motor skills, and number processing[ 16 ]. Each item corresponded to a score between 1 and 3 ( S1 Fig ). The total score was calculated by adding the points for the 20 items, and it hence ranged between 20 and 60. The higher the GSA score, the higher the school performance level.

Parental separation

Information regarding marital status was binary, i.e. it was rated either as parents living together or as parents living separately. For parents who had separated, the earliest date at which they were reported to have separated was used. To investigate whether the effect of parental separation varied according to the time of the separation, three groups of infants were considered: infants with intact families, infants whose parents had undergone separation prior to the 24-month visit (included), and those whose parents had separated between the 36 and the 60-month visits (included). Parental separations that occurred after the 60-month visit were excluded.

Adjustment variables

Adjustment variables comprised perinatal characteristics, socioeconomic data, and urbanicity of the residential municipality. The following perinatal characteristics were taken into account in the analyses: gestational age (three classes: 24 to 27 weeks, 28 to 31 weeks, and 32 to 34 weeks GA), gender, twinship, and birth weight. The birth weight Z-score was computed according to the Olsen standards[ 17 ], and it was considered as a four-class-categorical variable (< -1, between -1 and 0, between 0 and 1, and > 1). The socioeconomic data consisted of the socioeconomic level and eligibility for social security benefits for those with low incomes. The socioeconomic level took into account the parent with the more highly rated job according to a scale based on the official classification developed by the French Institute for Statistics and Economic Studies (INSEE). Lastly, the residential municipality was considered to be either urban or rural based on definitions developed by the INSEE. Municipalities were considered to be urban if there no buildings separated by a horizontal distance of more than 200 meters (i.e. if it was a continuously built-up area) and there were at least 2,000 inhabitants. All other municipalities were considered to be rural.

Statistical analyses

Infants whose parents had separated and infants whose parents were living together were matched using an exact matching procedure. Indeed, the possible differences between characteristics of infants whose parents had separated and infants whose parents were living together could bias the estimation of the effect of the separation on their school performance (school performance could be highly influenced by the gestational age, birth weight, and socioeconomic status in particular). The matching was based on the following covariates: their gestational age and birth weight Z-scores (as categorical variables), gender, twinship, socioeconomic level, eligibility for social security benefits for those with low incomes, and urbanicity. Exact matching resulted in groups with different numbers of infants that had the same characteristics. Therefore, weights were used in the following models to ensure that the matched infants were weighted up to be similar. This matching procedure allows the possible bias due to the differences between infants with parents living together and those who had separated to be minimized.

The effect of separation on the GSA score was quantified using weighted quantile regressions. Quantile regression allowed for quantification of the effect of separation on conditional quantiles of the GSA score, thus providing a complete view of possible relationships between parental separation and the GSA score. By contrast, a linear regression only allows an overall effect of parental separation on the mean of the GSA score to be discerned. Moreover, the effect of parental separation was tested on each of the 20 items of the GSA questionnaire. To do so, a weighted generalized estimated equation (GEE) was considered for each item (response 1 versus responses 2 or 3). All of the models were adjusted for the GA and Z-scores for the birth weights as continuous variables. All of the statistical analyses were performed with R software[ 18 ]. Finally, a sensitivity analysis was performed with three additional adjustment variables to test the robustness of the results. These variables were: breast feeding only, severe neurological anomalies, and the duration of oxygen supply.

Ethics approval

The LIFT cohort (Grandir ensemble en Pays de la Loire) is registered at the French data protection authority in clinical research (Commission Nationale de l’Informatique et des Libertés or CNIL, No. 851117). This study received the favorable opinion of an ethics committee (GNEDS, Groupe nantais d’éthique en santé). Written consent was obtained from the parents of each child before inclusion.

Between January 2003 and December 2011, 7,119 infants born at less than 35 weeks of gestation in the Pays-de-la-Loire region were enrolled in the LIFT cohort ( Fig 1 ). Infants who were deemed to have a non-optimal neurodevelopmental outcome at two years (n = 1,230) were excluded from the study population. Of these, 70% (n = 3,399) underwent assessment for a GSA score when they reached five years of age. The mean GSA score was 50.8 (median = 53.0) ( Table 1 ). The proportion of parents reported to have undergone separation within five years after their child’s birth was 8.4% (n = 280), for 3.3% (n = 110) this occurred before or at two years, and for 5.1% (n = 170) this occurred between three and five years. The median timing of the visit at which the separation occurred was 36 months following the child’s birth. Of these children, 2,707 were considered further after being matched. The number of children in each group that had the same characteristics varied between 2 and 94, with a median of 15 infants.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0202080.g001

https://doi.org/10.1371/journal.pone.0202080.t001

Parental separations were associated with a lower school performance. This lower performance only occurred for children with a low GSA score (corresponding to the 20 th percentile) whose parents had undergone separation when the child was between 3 and 5 years of age. For these children, the decrease in performance amounted to 3.7 points (standard error = 1.54, p = 0.01) ( Fig 2B ). Moreover, parental separations were associated with a decrease in their ability to pay attention in the classroom (OR = 2.46 [1.39 to 4.34], p < 0.01), being able to properly engage in school conversation (OR = 1.81 [1.18 to 2.77], p = 0.01), and in terms of independence when confronted with a task (OR = 1.86 [1.07 to 3.24], p = 0.03), as well as an increase in serious difficulties with activities requiring manual dexterity (OR = 2.37 [1.27 to 4.43], p = 0.01) ( Fig 3 ).

( A) Separation that occurred prior to two years of age (included) (n = 104) and (B) separation that occurred between three and five years of age (n = 164), compared to children whose parents who were still living together at five years (n = 2,439). The variations were quantified based on the quantiles of the GSA (i.e. 0.2, 0.5, and 0.8).

https://doi.org/10.1371/journal.pone.0202080.g002

Results were expressed as Odds Ratios (OR) with their 95% Confidence Intervals (CI).

https://doi.org/10.1371/journal.pone.0202080.g003

Our results indicated that, for preterm infants that had an optimal neurodevelopment at two years, parental separation was associated with a decrease in school performance at five years of age that was independent of their socioeconomic background. This decrease was only noted for parental separations that occurred between three and five years after the child’s birth when the children exhibited difficulties at school. Furthermore, parental separations were associated with a decrease in the child’s motivation, engagement, autonomy, and manual dexterity.

Parental separation only impacted children who exhibited low scholastic performances. These children faced a stressful situation at school as their poor results were exacerbated as a result of their parents being separated. Their poor scholar performances may be related to factors prior to the parental separation, such as parental conflict or lack of parental commitment to the child’s education. This would be consistent with results of previous studies that revealed a decrease in educational performance[ 19 , 20 ], or an increase in academic, psychological, and behavioral problems[ 21 ] prior to the parents becoming separated. Additionally, this supports the notion that divorce is a process rather than merely an event[ 6 , 22 ].

Moreover, only separations that occurred between three and five years were associated with decreased school performance of children at five years of age. For term infants, the same result was found regarding the age at separation. Cognitive development was found to not be affected by parental separation during the first three years of life[ 23 ]. An explanation for this could be that parental separation is unlikely to have negative consequences for children who have not yet formed a mental representation of their relationship with their parents. Moreover, these infants were likely to be too young to remember this stressful event. Furthermore, parental separation that occurred when a child is very young could have effects that were not investigated in the present study. Indeed, younger children are much more dependent on their parents for socioemotional and physical care than older children. In terms of the effect of parental separation on attachment to parents, it has been suggested that the impact appears to be more pronounced in children who are younger at the time of the family dissolution[ 24 ].

The negative consequences of parental separation comprised effects on the children’s behavior and abilities. Children whose parents had undergone a separation had more difficulty with being able to continue to pay attention, and they more readily became overwhelmed as compared to children from intact families. Consequently, their conversations at school were considered to be less appropriate by their teachers, and they had serious difficulties with activities requiring manual dexterity such as drawing or handicrafts. This result is in accordance with studies performed with term infants that have provided evidence that parental separation negatively impacts their children’s academic motivation and engagement[ 6 ].

The GSA score was used to evaluate the children’s school performances. This score, which correlates well with the IQ score, is used to detect adverse neurodevelopmental outcomes in preterm children [ 16 ]. This decrease in the GSA score corresponds with a standard deviation (SD) of 0.5. A meta-analysis carried out in regard to term infants has estimated a median decrease corresponding to 0.14 SD [ 5 ]. The more pronounced effect found in our study might be due to the vulnerability of the preterm infant population. Preterm infants could indeed be more sensitive to stressful situations such as parental separation. The difference might also be due to our statistical approach that allows an effect in a specific population to be quantified. Finally, because the median GSA score was around 40 for children with low GSA scores, we roughly estimate that parental separation is associated with a 10% diminution of the GSA score, after adjusting for children and parent/family characteristics.

As a follow-up to this study, it would be of considerable interest to investigate long-term effects of parental separation. Several studies have reported that there may be more of an impact on long-term consequences in regard to achievements and quality of life as adults than on the short-term emotional effects in children[ 3 , 25 ]. Another interesting aspect would be to investigate whether the effect of parental separation on school performance varies according to the parental custody and possible remarriage after separation.

A strength of this study was the large number of infants included, which allowed a high statistical power to be attained. Moreover, the exact matching allowed a very high comparability to be reached between the infants whose parents were living together and those for whom they had separated. In terms of categorical variables, the infants in these two groups had exactly the same characteristics. All of the standardized mean differences were < 0.01 after matching ( S2 Fig ), it can be concluded that there was an absence of bias in regard to the observed characteristics of the infants. Lastly, the use of quantile regressions allowed evidence to be obtained of a specific negative effect of parental separation for children with low GSAs. A sensitivity analysis was performed with three additional adjustment variables that can influence children’s neurodevelopment: breast fed only (yes/no), severe abnormality (affirmative, if one of the following pathologies was diagnosed: stage 3 or 4 intraventricular hemorrhaging, ventriculomegaly, periventricular leukomalacia), and the duration of oxygen supply. As the results were similar, it can be concluded that there was an absence of bias in regard to these variables ( S3 Fig ).

The characteristics of the children who were excluded from the study population were not comparable to those who were included ( S1 Table ). The excluded infants comprised those without GSA scores at five years (n = 1,461), infants whose parents separated after the 60-month visit (n = 59), or infants with incomplete medical records (n = 32). These infants were more likely to come from disrupted families with a lower socioeconomic level and who were living in rural areas, as compared to the included children. This suggests that parental separation may be a cause for non-inclusion. Children who moved to another area after their parents separated may be the most affected, as moving to another school is generally a stressful situation[ 26 ] that may be associated with a decrease in scholastic performance. The exclusion of these children might therefore lead to an underestimation of the effect of separation. Moreover, some information related to the context of the separation were not available in this study and were thus not accounted for, such as the level of parental stress, the financial sequelae, a change in residence, or the parenting roles after separation. These factors may have affected the children. Further research should thus be carried out while accounting for these factors in order to assess the causal pathway between parental separation and the cognitive performance of children.

Conclusions

Parental separations that occurred when preterm children were between three and five years of age were associated with a decrease in the school performance when the children exhibited difficulties at school. Our results suggest that this effect could be due to a decrease in the children’s level of motivation, autonomy, and manual dexterity. In light of our findings, specific support could be given to young children who were delivered preterm who exhibit scholastic difficulties and whose parents separated when they were of elementary school-age.

Supporting information

S1 table. comparison of the characteristics of included and excluded children..

https://doi.org/10.1371/journal.pone.0202080.s001

S1 Fig. Global School Adaptation questionnaire.

Drafted by Agnes Florin, Director of the Education, Cognition, and Development laboratory and Professor of Developmental Psychology at the University of Nantes (reproduced with permission). As an example, for Question 1, the answer “just a little or not at all” received a score of 1, “moderately” received a score of 2, and “quite a lot” received a score of 3.

https://doi.org/10.1371/journal.pone.0202080.s002

S2 Fig. Standardized mean differences, before and after matching, between children whose parents were still living together at five years of age and children whose parents had undergone a separation.

The right part of the figure presents the differences in percentages or means (with standard deviations) between the two populations before matching.

https://doi.org/10.1371/journal.pone.0202080.s003

S3 Fig. Variations in “Global School Adaptation” (GSA) score associated with parental separation that occurred (A) prior to two years of age (included) (n = 104) and (B) between three and five years of age (n = 164), compared to children whose parents were still living together at five years (n = 2,439).

The variations were quantified according to the quantiles of the GSA (i.e. 0.2, 0.5, and 0.8). The same model as the one used for Fig 2 with three additional adjustment variables: breast fed only (yes/no), severe abnormality (affirmative, if one of the following pathologies was diagnosed: stage 3 or 4 intraventricular hemorrhage, ventriculomegaly, periventricular leukomalacia), and the duration of oxygen supply (four classes: no oxygen supply, < 28 d, between 28 and 36 d, > 36 d).

https://doi.org/10.1371/journal.pone.0202080.s004

• View Article
• Google Scholar
• PubMed/NCBI
• 9. Emery RE. Marriage, divorce and children’s adjustment. Beverly Hills; 1988.
• 18. R Core Team (2016). R: A language and environment for statistical computing. [Internet]. R Foundation for Statistical Computing, Vienna, Austria.; Available: http://www.R-project.org/ .

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Taylor & Francis Open Select

A qualitative analysis of parental loss and family separation among youth in post-conflict Liberia

Elizabeth j. levey.

a Division of Global Psychiatry , Massachusetts General Hospital , Boston, MA, USA

b Department of Psychiatry , Harvard Medical School , Boston, MA, USA

c Department of Psychiatry , University of Illinois College of Medicine , Chicago, IL, USA

Claire E. Oppenheim

d Department of Psychiatry , Boston Medical Center , Boston, MA, USA

Brittany C. L. Lange

e Department of Social Policy and Intervention , University of Oxford , Oxford, UK

Naomi S. Plasky

f Bloomberg School of Public Health , Johns Hopkins University , Baltimore, MD, USA

Benjamin L. Harris

g A.M. Dogliotti College of Medicine , University of Liberia , Monrovia, Liberia

G. Gondah Lekpeh

Isaac kekulah, david c. henderson.

h Department of Psychiatry , Boston University School of Medicine , Boston, MA, USA

Christina P. C. Borba

Associated data.

Between 1989 and 2003, the Republic of Liberia experienced a brutal civil war. In 2008, the population was approximately 3.5 million people, and there were an estimated 340,000 orphans. Nearly 6000 more children were orphaned by the Ebola epidemic from 2014–2015. The goal of this research was to explore the impact of parental loss, identify moderating factors, and consider interventions that could help vulnerable youth in post-conflict societies following the loss of a parent. Seventy-five young people (age 13–18 years) in Monrovia, the capital city of Liberia, were recruited in 2012. Semi-structured interviews were conducted, and demographic data were collected. Interviews were then transcribed and coded thematically. The loss of a parent or other primary caregiver had a significant impact on psychosocial and emotional health. The timing of the loss, strength of connection with the deceased parent, and relationship with surviving parent or substitute caregiver were all relevant factors. Children separated from living parents were functioning better than those whose parents were deceased. The case of Liberian children underscores the importance of early caregiver relationships and the difficulties children face when such relationships are disrupted. Children who did not experience stable early relationships suffered disconnection from their families and communities.

Introduction

There are approximately 132 million orphans worldwide (UNICEF, 2015 ); sub-Saharan Africa (SSA) alone has an estimated 56 million orphans (UNICEF, 2013 ), nearly half the global total. There is a large body of literature describing the needs of orphans in SSA, the majority of which has focused on children orphaned by the HIV epidemic. Children who lost one parent before the age of 18 are considered orphans, and those who have lost both parents are referred to as ‘double orphans’. A recent review of qualitative studies found that orphans in SSA experience neglect, abuse and exploitation at the hands of substitute caregivers (Morantz et al., 2013 ). These exposures put orphans at risk for mental health problems, including depression and suicidal ideation (L. Cluver, Gardner, & Operario, 2007 ; L. D. Cluver, Gardner, & Operario, 2008 ; L. Cluver & Orkin, 2009 ). Orphans were negatively impacted when they perceived a lack of caring, particularly when they received inferior treatment compared with the caregivers’ biological children (Ansell & Young, 2004 ; L. Cluver & Gardner, 2007 ). A study of children orphaned by HIV in Uganda found that orphanhood was associated with psychological ill health among the males (Kaggwa & Hindin, 2010 ). Male double orphans and male maternal orphans had a significantly higher level of hopelessness than their non-orphaned counterparts. Lower parent/guardian connectedness, having a chronically ill adult in the household and ill treatment in residence, were associated with a higher level of depression, especially among the males. School attendance was found to be protective.

The study of attachment has illuminated the critical role of early caregiving relationships in fostering healthy development and forming a basis for future relationships (Ainsworth, Blehar, Waters, & Wall, 2015 ; Bowlby, 1988 ; Freud & Burlingham, 1943 ; Lyons-Ruth, 1996 ; Lyons-Ruth, Zoll, Connell, & Grunebaum, 1986 ). The loss of a parent is a hardship for any child, but the availability of care from other sources can have a meaningful impact upon recovery and ongoing development. Studies conducted in Romanian orphanages demonstrated a significant lag in intellectual and social development in institutionalized children compared with those reared in families (Smyke et al., 2007 ). Those who formed secure attachments with their foster families were protected from developing anxiety and depression (McLaughlin, Zeanah, Fox, & Nelson, 2012 ). Relationships with peers and other caring adults outside the home have also been found to be protective in children who face early adversity and disruptions in their caregiving relationships (Betancourt & Khan, 2008 ; Werner, 1989 ).

Post-conflict societies struggle to reestablish order while also addressing the consequences of the trauma experienced by their populations. Orphans in these settings face challenges that are unique from those orphaned by HIV, and these have not been well-characterized. One study of war orphans in South Sudan found that they were concerned with survival, stigma, the psychological impact of trauma exposure and access to education. They were attending an orphan school, which they valued for the educational opportunities and the social support it provided (Muller, Munslow, & O’Dempsey, 2015 ). Factors related to the relationship with the deceased caregiver and the relationship with the substitute caregiver were not explored.

None of the studies referenced earlier included data from Liberia. Between 1989 and 2003, the Republic of Liberia experienced a brutal civil war characterized by ethnic killings, sexual violence and the use of child soldiers. The war displaced more than 800,000 people, of a total population of 3.5 million (LISGIS, 2009 ). Moreover, it destroyed the productive capacity and physical infrastructure of the country and eroded family and community ties (UNDP, 2006 ). In 2008, five years after the war ended, there were an estimated 340,000 orphans in Liberia, 18% of the total child population of the country (LISGIS, 2009 ). It is estimated that nearly 6000 more children were orphaned by the Ebola epidemic from 2014 to 2015 (Collins, 2015 ).

This study was undertaken as a follow-up to the Liberia Needs Assessment Survey, which was conducted in 2009 to characterize and quantify the mental health burden of the war on the children of Liberia (Borba et al., 2016 ). Key informants were asked to describe the most emotionally disturbing events or experiences to have affected young Liberians in the past 20 years and how best to help specific vulnerable groups, including orphans, homeless children, former child soldiers, criminals and children with severe mental illness. Youth were perceived to be experiencing significant adverse emotional, behavioral and functional outcomes related to exposure to war and its aftermath (Levey et al., 2013 ).

The findings presented in this article were obtained from in-depth interviews with Liberian youth designed to assess their postwar experience and identify factors impacting resilience. Resilience was defined as evidence of daily functioning, ability to make realistic plans for the future and presentation of an overall consistent and coherent narrative. A number of individual characteristics were found to be protective in the overall study population, including emotion regulation, cognitive flexibility, agency, social intelligence and, in some cases, meaning-making (Levey et al., 2016 ). Here, we focus specifically on resilience and recovery following the loss of a parent, identify factors that moderate the impact of the loss and consider interventions that could help vulnerable children following the loss of a parent.

Study procedures and participants

A qualitative research design was used with a purposive sample of 75 participants. In-depth interviewing was chosen as the data collection method in order to best capture the personal narratives of individual children (Hennink, Hutter, & Bailey, 2011 ). The interview guide was designed and implemented by the researchers using both a deductive and an inductive conceptual framework. The initial interview guide was developed deductively and was rooted in the existing literature. Prompts were developed to assess factors that had already been identified to be associated with resilience, including agency, self-esteem, altruism, hope, spirituality and meaning-making. Revisions were made after each interview to refine and deepen the questions. Oversight, guidance and approval were provided by the University of Liberia Institutional Review Board (ULIRB) and the Partners Human Research Committee.

Participants were recruited from 13 February to 1 March 2012 and 4 to 21 December 2012. The initial group was recruited from three schools and the subsequent group of children not attending school was recruited from the neighborhoods in Monrovia, the capital city of Liberia, and the surrounding area. All children aged 13–18 years were eligible to participate. Children in this age range were 4–9 years old when the war ended, so they were old enough to have some memory of it, and 10 years later, they were adolescents approaching adulthood, when the consequences of childhood trauma can begin to manifest as mental illness (Chapman et al., 2004 ; Colman et al., 2013 ; Schilling, Aseltine, & Gore, 2007 ). An effort was made to recruit both males and females.

The Republic of Liberia’s Ministry of Education identified three large, well-attended schools in three distinct areas within Monrovia. The participating schools included two government schools (a junior high school and a senior high school) and a private senior high school. These schools were chosen in order to capture distinct age and socioeconomic groups. At each school, the principal notified the students of the opportunity to participate in the research study. Children under 18 years of age had to obtain consent from a parent or guardian and then give their assent to participate. Those who were 18 consented for themselves. Those students who returned signed consent forms first were interviewed. Interviews were conducted over a 3–4-day period at each school.

The research team then recruited and interviewed children not attending school. They were identified in two different communities in Monrovia, with the help of community leaders. Both communities had high rates of poverty, crime, substance use and low rates of school attendance. The leaders went to a few areas in their communities where youth commonly gathered and explained that there was an opportunity to participate in a research study. They distributed printed information and consent forms and told the young people when and where the interviews would take place. They helped to organize a loose schedule with approximate days and times for each child to come. The leaders were present outside the building while the interviews were conducted but were not present in the room during the interviews.

The initial compensation for study participants was a 5 USD phone card, which was approved by the ULIRB. After the initial phase of recruitment, the research team observed that this was a significant inducement to participate. With the approval of the IRB, the research team changed the compensation to 2 USD. Any child identified as having acute psychiatric needs was referred to the study physician, a Liberian child psychiatrist.

Data collection

Primary data were collected using semi-structured interviews averaging 1 hour in length. The interview guide was developed to capture basic demographic information, family composition, adverse experiences, response to adversity and future outlook (see Supplementary Material). Interviews were conducted by the principal investigator together with a Liberian medical student, who acted as both a linguistic and cultural interpreter. The interviews were conducted in a private area. Student participants were interviewed on school grounds. The participants not attending school were interviewed in an office where the community leader worked. Interviews were digitally recorded, and notes were taken by the interviewers regarding participants’ affect and nonverbal communication during the interview. The interviews were transcribed verbatim by Liberians living in the US.

A directed content analysis was used because there was an existing body of literature about both resilience and orphanhood, but there was insufficient data about resilience among post-conflict Liberian orphans (Hsieh & Shannon, 2005 ). Coding was performed by four coders who worked separately and sequentially over the course of two years, beginning six months after data collection was initiated. While coding the first five interviews, the coders met after each interview to compare their findings. There was a discussion about topics that were raised by multiple participants and whether they were saying the same thing or articulating distinct themes. Codes developed and used by multiple coders were then added to the codebook. When there was a discrepancy, coders explored it in an effort to uncover the deeper meaning at the heart of the theme. Discussion continued until a consensus was reached. Coders then met after every five interviews to further refine the codebook. Once there was a consensus as to the codes and definitions being used, all interviews were recoded. Then the themes and definitions were further refined.

Both inductive and deductive approaches were used (Hennink et al., 2011 ). First, manifest content was grouped thematically. Thematic groupings were then labeled, and these group labels were used to generate broad themes. These broad, overarching themes were divided into subthemes. Within each theme and subtheme, the researchers drew comparisons, looking for overlap and differences, as well as newly emerging topics and patterns. Themes identified included violence, death, sadness, fear, uncertainty and family and community relationships. Responses were reviewed to identify each theme and subtheme. The analysis was conducted using grounded theory, memoing and coding (Hennink et al., 2011 ). NVivo was employed for data management (QSR, 2015 ).

Descriptive data

A total of 75 children were interviewed (see Table 1 ). They ranged in age from 13 to 18 years, with an average age of 16.4 years. Females comprised 51% of the sample. The rate of school attendance was 61.4%. Many children were orphaned by the war or lost one of their parents. Many more were sent to live with substitute caregivers. Sometimes this was done for their protection during the war or to give them greater educational opportunities. It was also done to decrease the burden of caring for many children in large families while providing household help to relatives without children.

Overall demographics of all study participants.

SD = standard deviation.

Of the 75 children interviewed, none were living in institutions. They were either with their families, with other caregivers, living alone or homeless. If one parent was living but the child did not stay with the parent, the child was more likely to be with another caregiver, whereas if both parents died, the child was more likely to be on his/her own. Twenty-four of 26 children living with a caregiver had a living parent; just 8 of 14 children living on their own had a living parent (see Table 2 ). Fourteen children had no caregiver; eight of them had one or two living parents. One was male. He was living alone but attending school with some support from his uncle. Seven of the eight were females, and only one was attending school (see Table 3 ).

Current living situation and parental status for participants, aggregated and by gender ( N ).

N/A = not applicable.

School attendance by living situation and parental status ( N , %).

Twenty participants lost one parent. Eleven lost fathers, and nine lost mothers. Of the 11 who lost their fathers, five were living with their mothers and four were with their caregivers. Nine were attending school (see Table 3 ). By contrast, just two of the nine participants who lost their mothers were living with their fathers, and four were attending school. Eight participants lost both parents. Six of them were on their own and had no caregiver. None of those six were attending school. Of the two participants living with caregivers, only one was attending school. Of the 11 participants interviewed who were attending private school, 10 lived with one or both parents, and one lived with a caregiver. All had at least one parent living, and none were on their own.

Analyzed data

In our analysis, we considered the themes of the impact of the loss of a parent, the importance of understanding the loss, the timing of the loss, the relationship with the deceased parent, the relationship with the surviving parent or substitute caregiver and the difference between death and separation. We found that the loss of a parent or other primary caregiver had a significant impact on psychosocial and emotional health. Understanding the cause of death did not mitigate the negative impact of the loss for these study participants. The timing of the loss did have an impact; children who lost their parents at a younger age (10 years old or younger) generally struggled more than those whose parents died when they were older in the postwar context. Experiencing a strong connection with the deceased parent and the surviving parent or substitute caregiver were both important factors in supporting recovery. Many children were separated from their parents and sent to live with other relatives; they typically found themselves in more supportive situations than children who were living with relatives after the death of their parents.

Impact of the loss of a primary caregiver

The impact of the death of a parent or other primary caregiver was experienced emotionally and also impacted daily functioning and future opportunities. Some children had ongoing difficulty coping with the loss and turned to substance use. Many also lost support for their education and experienced uncertainty about the future.

A 17-year-old girl who had been attending school and living with her sister dropped out and started using drugs after her sister died. She had lived with her sister since she was young. She could have gone back to live with her parents, but she did not, she explained ‘because I was used to her’. Other children expressed this feeling of being ‘used to’ someone, usually about their parents, as compared to other caregivers, but for this girl, she felt very close to her sister and was used to her life with her in the city. Rather than go back to her family, she sought comfort from other children around her. She described her path of following them into drug use:

[My sister] was taken to the sick bush and died there, so my mind got disturbed. I started to follow behind friends, and that’s how friends involved me into smoking. When I think about my sister … she used to send me to a very good school. So after she died, it played on me. First I involved into drinking. After that I started to smoke the grass. When I smoke the grass it makes me to eat and sleep, and then I don’t think too much.

Children who had lost a caregiver also expressed uncertainty about their future. Some were forced to leave school because they had no one to sponsor their education. Others remained in school but worried that they might have to leave at any time and abandon their career plans. One boy who lost his father explained:

He knew about [my plan to study engineering]. He was going to encourage me because, as I learned, he loved me when I was small. But now, I come home from school, I will study. Nobody tell me to study. I don’t have, yet some other people have their fathers, telling them to do this, telling them to do that other thing… I don’t have a father to tell me to do this or that. My mother is old now.

The child’s understanding of the cause of death

While participants clearly tried to explain the deaths of their loved ones, in most cases they were left without a satisfactory explanation; the narratives they did construct did not seem to help them recover from these losses. Children described the concept of ‘African science,’ a curse placed by one person on another to cause misfortune and sometimes death. Seventeen reported that a family member had been poisoned by a jealous relative or neighbor, which reinforced their belief that their parent was envied by others and gave them a way of understanding their death.

One girl lost her father and sister following gastrointestinal illnesses, so she believed they were poisoned by African science. For her, this was less an explanation than a label for something unexplainable. She wanted to study Western medicine and become a nurse, which she viewed as a path to greater understanding:

For me I prefer to do the hospital medicine. For the native medicine I don’t have idea on it. For people who are wise they the one who have idea on it. Some people will put medicine in your food so the native doctor will know what kind of leaf to use for that food to come out, but for me I don’t know it … If I feeling sick I will choose to go to the hospital. Because if I have headache, I will go to the hospital then the people will check me and tell what is happening to me.

African science was used to explain death by a mysterious illness but not death due to war violence. Superficially, those whose parents were killed in the war had an understanding of why their parents had died because they witnessed them being brutally killed. And yet it was clear that losing their parents in this way was still very disorganizing. There was a need to make meaning out of it and, in some cases, take back control by seeking revenge. An 18-year-old boy lost both parents when rebel forces attacked his village. When the government troops arrived, he joined them. He wanted to avenge his parents’ death. This led him on a path to violence and drug use:

I did not go anywhere. I lived in the town until the men captured the town. They wanted to kill me, and one of them said no. So when they went in the house and saw my parents’ bodies they decided to take me with them. So when I started to grow up I just feel that I don’t have any family in my life. I can feel like paying back and doing bad things to other for what they did to my parents. I don’t have mercy on people because at the times I did not know who kill my parents so anyone I see, I executed them.

Having an explanation was not helpful, but believing that they could have done something to save their parents made the grieving process even more complicated. A 13-year-old boy lost his father after the war from an asthma attack that this boy witnessed when he was about 6 years old: ‘That day he fell down inside, and they carried him to the hospital. They said he stopped breathing … I feel bad when I think that I couldn’t do anything [to save him]’. Every year on the anniversary of his father’s death, he was overcome by this feeling of helplessness. At multiple points during the interview, he said he was doing ‘nothing’. He conveyed a lack of self-worth and a sense that he did not have control over his life.

Timing of the loss

Some children lost both parents together. Some lost one parent during the war and the other years later. Some children were so young when their parent died that they had no memory of that parent, while for others it was much more recent. In the worst cases, children lost both parents amid the chaos of the war, without the presence of extended family or community support to step in, and they were left with nowhere to go. A 17-year-old boy who lost both parents during the war dropped out of school after 7th grade and supported himself by stealing. He was addicted to drugs. He could identify no sources of support and saw no way to stop using:

So the rocket killed my father and my mother instantly right in the house along with other two or three families. After that, my sister did not have anything to cater to me. So I started to go from place to place and stealing. I was shocked, because I used to love my father and mother, to hear that they were dead. After the war, the drugs and cocaine and everything just polluted the market and at that time to get the drugs, it was very easy.

A girl who lost her father as an adolescent, after the war, explained her experience. He was ill, and her mother was caring for him and the rest of the family when he died. This left her with a sense of order and security:

He was sick. His sickness was somebody that witched him. Sitting by himself all day, not doing nothing, somebody that is not old and yet … he is not cripple, he can’t walk. Yeah, … my mother was there, she used to take care of him, feed him, and that’s how he died.

Relationship with deceased caregiver

As discussed, participants who lost parents at younger ages were generally more negatively impacted. The timing of the death was significant because these children were too young when their parents died to have developed an enduring connection with them. They were unable to hold on to the memory of their parents. The quality of the relationship was also significant. Children who had a strong connection with their deceased parent were more likely to continue on a productive path and recruit other supportive adults.

A 17-year-old girl with two children of her own was already an adolescent when her father died. She suffered from the loss, but she was working and providing for her children and had connected with a neighbor who was able to help her with childcare:

When I delivered, my father did not give me chance to sit home for long, and he paid my school fees. He sent me to one school in the area and one big-sized girl was at the house so when I come from school at 12:00, I will stay with my son and then the girl will go to school. That was how we used to do it but since he died, things turned upside down.

Children who did not have strong attachments to their parents while they were living had more difficulty after they died. One girl had a mother who was mentally ill, and three of her siblings died as infants. Then her mother died, and she was not able to make sense of the loss: ‘She was sick. She used to talk by herself, like to say the person going mad. You [can]not hold her, she goes by herself. So one morning we just see her body lying down in the house’. She was sent to stay with an older brother, who then sent her to live with his mother-in-law. The mother-in-law died, and the girl was on her own. She had no consistent relationship with any reliable caregiver. Her brother was pressuring her to marry for financial reasons, but she wanted to finish school first. She broke down crying as she explained this: ‘My mother not living. So I live with my brother … He ain’t care about me … Everything I can do it for myself’.

Relationship with surviving caregiver

After the death of a parent, children turned to the other parent, if they could, or to another caregiver. A consistent, supportive relationship with a surviving parent allowed a child to continue to grow and develop and to plan for the future. Under the best circumstances, this parent shared with the child memories of the deceased parent. The surviving parent saw the best qualities of their deceased spouse in their child and spoke openly about their pride in and affection for the child. Of the 26 children living with a nonparent caregiver, 11 were able to make use of another caregiver, but the rest did not feel cared about or supported by the other caregiver. Children living with other caregivers were more likely to report neglect, inferior treatment to the caregiver’s biological children, and exploitation.

A 13-year-old boy not attending school was living with his aunt after the death of his mother. He explained that his aunt treated him like her own child. She wanted to send him to school but did not have the money, and she was not sending her own children to school either: ‘When she goes out and see anything, she will buy for her son and buy something for me’. Feeling cared for by his aunt, he was able to develop a coherent narrative and plan for the future. He maintained an interest in returning to school and wanted to become a social studies teacher.

Some families fell apart following the loss of a parent, even if the other parent remained alive. This was particularly true after the loss of the mother. Of the nine children who lost their mothers but not their fathers, five were sent away because their fathers felt incapable of caring for them, and two had no caregivers. After marital separation or spousal death, single fathers were more likely to take another wife, which meant the child was faced with the complexity of dealing with a stepmother. Children reported that they were treated differently than the stepmother’s biological children. They were spoken to harshly, punished more severely, and sometimes abused and exploited. In contrast, children did not report the presence of a stepfather after the death of their father.

A 15-year-old boy attending government school was living with his father and stepmother after his mother died. Initially, he described them both as caring, but then he spoke about the uncertainty he felt about whether he would be able to continue his education. After his mother died, he began to worry about his father dying, although he was healthy, which was a fear that many children reported after the loss of one parent:

[My father] just behind me to put strong because I don’t know when he will be leaving, so I should put strong and get what I want from him because anytime God can take him away, so I should put strong when he spend more time behind me to go to school.

He also spoke about the differences between his mother and his stepmother: ‘We can’t eat together and also she can’t wash my clothes. And when my mother was living she do all, she wash my clothes, draw water, do all the other work’. There was no longer anyone looking out for him; he had to take care of himself in a way that left him alone.

A 15-year-old girl, whose mother died near the end of the war, was sent to live with her mother’s sister, who treated her brutally. She had sporadic contact with her father, who visited when he could and provided financial support until he died two years prior to the interview. Her aunt had always been dismissive and harsh, but after her father died, she began encouraging the girl to engage in prostitution:

She can want me to follow my friends on the road to go and hustle. I can tell her I am not able to do that. So when I say I am not able to do that sometimes she gets mad … She forces me. She cut my hair, and she put the dye there. And then when the girls were going on the road at night she told me to follow behind them.

Family separation

It might be expected that having living parents and not living with them would signal problems, and sometimes that was the case, but having a living parent usually meant that children fared better, even if they were living with another caregiver. Having a living parent meant that a child was more likely to find a caregiver and that the caregiver was more likely to take better care of the child. In some cases, their parents had made arrangements for them to live with another adult. Others did not have the benefit of their parents having made specific arrangements in this way, but they described secure connections with their parents and were able to form other connections themselves. Once a child was with a caregiver, the parents’ ongoing involvement effected how the child was treated by the caregiver.

A 16-year-old girl attending government school was sent to Monrovia by her parents, who were in the interior, to live with her uncle and attend school. Her parents made these arrangements for her and prepared her for the trip. While some children complained about restrictions from their caregivers, she liked the restrictions placed by her uncle because she felt that he would protect her from getting into trouble or getting pregnant before she was ready:

My parents advised me that when I come here, I should be good. I should respect the people I live with … When you living with your born people, you can get [bad] idea, you can put [bad] something in your head, but when you living with different person, you will be scary. All the one you used to do you will leave it, you change, you put different way in you.

Although she rarely saw her parents, she retained a sense of connection to them: ‘I can resemble [my father] … The only thing I bring that my mom’s way is she like plenty people, she loves to see people’. With support from her uncle and connection to her parents, she conveyed confidence and a clear, balanced sense of herself and ability to connect with others: ‘When I get vexed, I get vexed, but I like playing fun, I like laughing, I like joking, and I love people’.

While having living parents was generally better, it was not sufficiently protective in all cases. There were children with two living parents who were living on their own without a caregiver. One boy was attending school, working to support himself, maintaining a relationship with his family and making supportive connections in the community. The rest were not attending school, and all appeared to be struggling. Although both of their parents were alive, they had lost an important caregiving relationship or experienced a trauma from which their caregiver could not protect them.

An 18-year-old girl began engaging in prostitution after she was raped during the war. Her parents were unable to protect her from this event, and she found no use for their rules after the war: ‘The soldier boys started to beat us, and I was raped. That’s how I got in life, and them there we came back. Since we came back, that’s the life I’m living [prostitution]’. After the war, she left home because she did not like the rules imposed by her parents. She explained that living with her family was ‘nothing bad, except that we used to go to bed [early]’.

The loss of a parent posed particular challenges to youth in this post-conflict setting. There were a number of factors that moderated the impact of the loss, including timing, relationship with the deceased parent and the relationship with the surviving parent or caregiver. Understanding the cause of death did not have a meaningful impact on the participants’ ability to recover from the loss. Among orphans in Uganda, being orphaned by HIV was worse than losing their parents due to another cause (Kaggwa & Hindin, 2010 ). In our study population, however, participants seemed to recognize that they were facing circumstances beyond their control, whether they watched their parents be killed by gunfire or believed they were poisoned by a jealous neighbor. When participants believed they could have done something to save their parent, this complicated their grieving process.

From the attachment literature, we know that children learn to believe in their own capacities in the context of their relationship with parents who believe in them (Bowlby, 1988 ; Freud & Burlingham, 1943 ), and these early relationships in the first few years of life form the template for future relationships (Ainsworth et al., 2015 ; Lyons-Ruth, 1996 ; Lyons-Ruth et al., 1986 ). Our findings were consistent with this insofar as children who lost their parents later generally fared better. This differs, however, from another study which suggested that children who lost their fathers during the war in Croatia were less negatively impacted if they were very young (under two years) when the loss occurred (Plasc, Poljarevic, Loncar, & Henigsberg, 2011 ). This may be explained by the fact that we were comparing children who lost their parent in early or middle childhood (10 years old or younger) with those whose parent died during adolescence. It may be that early and middle childhood are the most difficult times for such a loss to occur. Also, when parents died after the war, there was more support available through the extended family and the community.

The study had eight participants who lost both parents; only one was attending school and all were struggling significantly. Among participants who lost one parent, there was more variation. When the relationship with the deceased parent had been positive, children had confidence in themselves and their ability to relate to others, which continued to support their functioning even after that parent died. The relationship with the surviving parent or caregiver was also important. When the mother was the surviving parent, children were more likely to live with her and be enrolled in school. This is consistent with the finding that maternal orphans typically fare worse, experiencing more depression and hopelessness than paternal orphans (Kaggwa & Hindin, 2010 ). Children living with nonparent caregivers often received inferior treatment, compared to the caregiver’s biological children. They were forced to do more work and their education was not prioritized. In some cases, they were overtly exploited to earn money for the caregiver, including forced prostitution. This differential treatment is similar to what have been described among orphans in other sub-Saharan African countries (Morantz et al., 2013 ).

For those children who were living on their own, forming a connection with an adult in their community was extremely valuable. These relationships were typically formed with a neighbor who was an elderly woman without children, or whose children were grown. Sometimes children were able to make extra money by helping these women with chores. What was emphasized most, though, was being able to talk with these women about their problems and rely on them for emotional support. Other research on high-risk youth has found that such extra-familial relationships can be crucial (Werner, 1989 ), but this has not been well described in orphans. Some studies of orphans have found that access to schools and other institutions is helpful because it creates opportunities for supportive relationships with teachers and staff, but the focus of these studies has not been on the relationships themselves (Hermenau, Eggert, Landolt, & Hecker, 2015 ; Morantz et al., 2013 ; Muller et al., 2015 ).

In the literature, male orphans in SSA have been reported to suffer greater depression and hopelessness than female orphans (Kaggwa & Hindin, 2010 ). Our findings suggest that while male orphans do indeed suffer, particularly double orphans, female children in the low-income post-conflict setting of Liberia face unique risks even if their parents are living. We found that while there were both boys and girls who were living on the streets without caregivers, boys were in this situation only if both of their parents were deceased. In contrast, several girls had left home despite having one or two living parents. This finding was a surprise but has multiple potential causes. Girls described restrictions imposed on them for their protection; some left home because they were angry about these restrictions over where they could go, when and with whom. On the other hand, girls living with other caregivers were sometimes pressured to engage in sexual relationships with older men, or outright prostitution, to offset their expenses. Some girls also experienced unwanted sexual advances and sometimes frank abuse from older men living in the same household.

The prevailing view is that orphans are best served in the community setting, and not in institutions (Communique, 2008 ). Recent research has challenged this view. A qualitative study of war orphans in South Sudan found that those attending an orphan school felt that their needs were being better met there than with their families (Muller et al., 2015 ). A study of 2837 orphans across five low- and middle-income countries found that those living in institutions fared as well or better than those living in the community, based on measures of physical health as well as cognitive and emotional functioning (Whetten et al., 2009 ). None of the participants in our study were living in institutions, but many found that their environments were not meeting their needs. Orphans in particular had unmet needs including basic physical care, stable and safe home environment, education and emotional care. The type of setting that could best meet those needs and is feasible in Liberia will need to be carefully considered.

There were a number of limitations of this study. Participants were recruited from Monrovia only. Children in other parts of Liberia were not included, and the data may not be representative of their experiences. Interviews were conducted at a single time point, so resilience could not be assessed over time. No collateral information was obtained from teachers or family members. This allowed the children’s voices to come through, but it also meant that we were not able to obtain a complete picture of their circumstances. Some may have chosen, for a variety of reasons, to minimize the difficulties they were facing, while others may have chosen to magnify them. There is also the possibility of selection bias, as children who chose to participate may be different from those who did not.

The case of Liberian children underscores the importance of early caregiver relationships and the difficulties children face when such relationships are disrupted. In a country that endured 14 years of brutal civil war, followed by a devastating Ebola outbreak, many children have experienced deprivation, violence and uncertainty. Orphans in particular suffered painful losses, and their supports were compromised. Those who were connected to their communities fared better. There was a divide such that children were either connected to multiple community institutions, like school and church, or they were totally disconnected from the larger community and instead were part of a subculture of substance use and criminal activity.

Further research is needed to evaluate interventions to support orphans in post-conflict settings. Our findings suggest that helping orphans to reintegrate into existing community institutions would be an important next step. This would require financial assistance, ongoing emotional support and personal contact. Orphans who are not reached by this type of effort may have more significant needs that would be better addressed by an institution designed specifically for them. Some children who were attending school suffered excessively at the hands of harsh caregivers, and they might also benefit from access to an institutional setting where they could interact with multiple caregivers and find protection from exploitation and abuse.

Supplementary Material

Funding statement.

This work was supported by the National Institutes of Health [T32-MH-093310] and Partners Healthcare Center of Expertise in Global Health Travel Award.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental meterial

Supplemental data for this article can be accessed here .

• Ainsworth M. D., Blehar M. C., Waters E., Wall S. Patterns of attachment: A psychological study of the strange situation . New York, NY: Psychology Press; 2015. [ Google Scholar ]
• Ansell N., Young L. Enabling households to support successful migration of AIDS orphans in southern Africa. AIDS Care . 2004; 16 (1):3–10. doi: 10.1080/09540120310001633921. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Betancourt T. S., Khan K. T. The mental health of children affected by armed conflict: Protective processes and pathways to resilience. International Review of Psychiatry . 2008; 20 (3):317–328. doi: 10.1080/09540260802090363. pii:793972454. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Borba C., Vesga Lopez O., Harris B., Henderson D., Carney J., Dominguez S., Henderson D. A mental health needs assessment of children and adolescents in post-conflict Liberia: Results from a quantitative key informant survey. International Journal of Culture and Mental Health . 2016; 9 :56–70. doi: 10.1080/17542863.2015.1106569. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bowlby J. A secure base: Parent-child attachment and healthy human development . London: Basic Books; 1988. [ Google Scholar ]
• Chapman D. P., Whitfield C. L., Felitti V. J., Dube S. R., Edwards V. J., Anda R. F. Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of Affective Disorders . 2004; 82 (2):217–225. doi: 10.1016/j.jad.2003.12.013. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cluver L., Gardner F. Risk and protective factors for psychological well-being of children orphaned by AIDS in Cape Town: A qualitative study of children and caregivers’ perspectives. AIDS Care . 2007; 19 (3):318–325. doi: 10.1080/09540120600986578. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cluver L., Gardner F., Operario D. Psychological distress amongst AIDS-orphaned children in urban South Africa. Journal of Child Psychology and Psychiatry, and Allied Disciplines . 2007; 48 (8):755–763. doi: 10.1111/j.1469-7610.2007.01757.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cluver L., Orkin M. Cumulative risk and AIDS-orphanhood: Interactions of stigma, bullying and poverty on child mental health in South Africa. Social Science & Medicine . 2009; 69 (8):1186–1193. doi: 10.1016/j.socscimed.2009.07.033. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cluver L. D., Gardner F., Operario D. Effects of stigma on the mental health of adolescents orphaned by AIDS. Journal of Adolescent Health . 2008; 42 (4):410–417. doi: 10.1016/j.jadohealth.2007.09.022. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Collins P. What happened to Liberia’s Ebola orphans? IRIN News, 2016 . 2015 http://www.irinnews.org/analysis/2015/10/07/what-happened-liberia’s-ebola-orphans Retrieved from IRIN: The inside story on emergencies website.
• Colman I., Garad Y., Zeng Y., Naicker K., Weeks M., Patten S. B., Wild T. C. Stress and development of depression and heavy drinking in adulthood: Moderating effects of childhood trauma. Social Psychiatry and Psychiatric Epidemiology . 2013; 48 (2):265–274. doi: 10.1007/s00127-012-0531-8. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Communique . Paper presented at the fourth global partners forum on children affected by HIV and AIDS . Dublin: The Royal Hospital Kilmainham; 2008. [ Google Scholar ]
• Freud A. B., Burlingham D. T. War and children . New York, NY: Ernst Willard; 1943. (P. R. Lehrman, Ed.) [ Google Scholar ]
• Hennink M., Hutter I., Bailey A. Qualitative research methods . Los Angeles, CA: SAGE Publications; 2011. [ Google Scholar ]
• Hermenau K., Eggert I., Landolt M. A., Hecker T. Neglect and perceived stigmatization impact psychological distress of orphans in Tanzania. European Journal of Psychotraumatology . 2015; 6 :28617. doi: 10.3402/ejpt.v6.28617. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hsieh H.-F., Shannon S. E. Three approaches to qualitative content analysis. Qualitative Health Research . 2005; 15 (9):1277–1288. doi: 10.1177/1049732305276687. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kaggwa E. B., Hindin M. J. The psychological effect of orphanhood in a matured HIV epidemic: An analysis of young people in Mukono, Uganda. Social Science & Medicine . 2010; 70 (7):1002–1010. doi: 10.1016/j.socscimed.2009.12.002. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Levey E. J., Borba C. P., Harris B. L., Carney J. R., Dominguez S., Wang E. K., Henderson D. C. Assessment of the needs of vulnerable youth populations in post-conflict Liberia. African Journal of Psychiatry . 2013; 16 (5):349–355. doi: 10.4314/ajpsy.v16i5.47. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Levey E. J., Oppenheim C. E., Lange B. C., Plasky N. S., Harris B. L., Lekpeh G. G., Borba C. P. A qualitative analysis of factors impacting resilience among youth in post-conflict Liberia. Child and Adolescent Psychiatry and Mental Health . 2016; 10 :26. doi: 10.1186/s13034-016-0114-7. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• LISGIS 2008 Population and housing census . 2009 http://www.lisgis.net/page_info.php?7d5f44532cbfc489b8db9e12e44eb820=MzQy Retrieved from.
• Lyons-Ruth K. Attachment relationships among children with aggressive behavior problems: The role of disorganized early attachment patterns. Journal of Consulting and Clinical Psychology . 1996; 64 (1):64–73. doi: 10.1037/0022-006X.64.1.64. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lyons-Ruth K., Zoll D., Connell D., Grunebaum H. U. The depressed mother and her one-year-old infant: Environment, interaction, attachment, and infant development. New Directions for Child and Adolescent Development . 1986; 1986 (34):61–82. doi: 10.1002/(ISSN)1534-8687. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• McLaughlin K. A., Zeanah C. H., Fox N. A., Nelson C. A. Attachment security as a mechanism linking foster care placement to improved mental health outcomes in previously institutionalized children. Journal of Child Psychology and Psychiatry, and Allied Disciplines . 2012; 53 (1):46–55. doi: 10.1111/j.1469-7610.2011.02437.x. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Morantz G., Cole D., Vreeman R., Ayaya S., Ayuku D., Braitstein P. Child abuse and neglect among orphaned children and youth living in extended families in sub-Saharan Africa: What have we learned from qualitative inquiry? Vulnerable Children and Youth Studies . 2013; 8 (4):338–352. doi: 10.1080/17450128.2013.764476. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Muller B., Munslow B., O’Dempsey T. When community reintegration is not the best option: Interethnic violence and the trauma of parental loss in South Sudan. The International Journal of Health Planning and Management . 2015 doi: 10.1002/hpm.2311. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plasc I., Poljarevic S., Loncar M., Henigsberg N. Age-developmental stage and severity of trauma-related symptoms, anxiety and depressive symptoms in participants who lost their fathers during the war in Croatia. Collegium Antropologicum . 2011; 35 (Supplement 1):139–144. [ PubMed ] [ Google Scholar ]
• QSR . NVivo qualitative data analysis software (version 10) Cambridge, MA: Author; 2015. [ Google Scholar ]
• Schilling E. A., Aseltine R. H., Jr., Gore S. Adverse childhood experiences and mental health in young adults: A longitudinal survey. BMC Public Health . 2007; 7 :30. doi: 10.1186/1471-2458-7-30. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Smyke A. T., Koga S. F., Johnson D. E., Fox N. A., Marshall P. J., Nelson C. A., Zeanah C. H. The caregiving context in institution-reared and family-reared infants and toddlers in Romania. Journal of Child Psychology and Psychiatry, and Allied Disciplines . 2007; 48 (2):210–218. doi: 10.1111/j.1469-7610.2006.01694.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• UNDP National human development report of Liberia . 2006 http://www.undp.org/content/dam/liberia/docs/docs/NHDR_LIBERIA_2006.pdf Retrieved from.
• UNICEF Children and AIDS: Sixth stocktaking report. 2013 http://www.unicef.org/publications/files/Children_and_AIDS_Sixth_Stocktaking_Report_EN.pdf Retrieved from.
• UNICEF Press centre. 2015 http://www.unicef.org/media/media_45279.html Retrieved from.
• Werner E. E. High-risk children in young adulthood: A longitudinal study from birth to 32 years. The American Journal of Orthopsychiatry . 1989; 59 (1):72–81. doi: 10.1111/j.1939-0025.1989.tb01636.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Whetten K., Ostermann J., Whetten R. A., Pence B. W., O’Donnell K., Messer L. C., Positive Outcomes for Orphans Research Team A comparison of the wellbeing of orphans and abandoned children ages 6–12 in institutional and community-based care settings in 5 less wealthy nations. PLoS One . 2009; 4 (12):e8169. [ PMC free article ] [ PubMed ] [ Google Scholar ]