Skip to main content
NCBI home page
JAACAP Open logoLink to JAACAP Open
. 2026 Feb 18;4(3):522–529. doi: 10.1016/j.jaacop.2026.02.002

Childhood Maltreatment in the Context of Familial Bipolar I Disorder Risk Predicts Major Depressive Disorder in Adolescents

Savannah D Layfield a, Jennifer E Siegel-Ramsay a, Lucy Chibib a, Amy Bichlmeier a, Patricia L Nick b, Corey R Jones b, Caleb M Adler b, David E Fleck b, Luis R Patino b, Gregory A Fonzo a, Elizabeth Lippard a, Charles B Nemeroff a, Melissa DelBello b, Jorge RC Almeida a, Stephen M Strakowski a,c,∗
PMCID: PMC13221821  PMID: 42220634

Abstract

Objective

To quantify associations between childhood maltreatment and emerging major depressive disorder (MDD) in adolescents/young adults with and without a first-degree family history of bipolar I disorder.

Method

A total of 116 youths 14 to 21 years old were classified into a high-risk group (n = 58) based on at least 1 first-degree relative diagnosed with bipolar I disorder and a low-risk group (n = 58) based on no first- or second-degree family history of bipolar disorder. Childhood maltreatment was assessed using the Childhood Trauma Questionnaire–Short Form (CTQ-SF). Psychiatric diagnoses were established with semistructured interviews. Logistic regression tested interactions between childhood maltreatment and familial risk, controlling for age and sex, as predictors of MDD.

Results

Half of participants in the high-risk group exhibited current or past MDD compared with 24% of the low-risk group. Regression models showed that each 1-point increase in CTQ-SF total score was associated with a 7% increase in the odds of MDD (odds ratio [OR] = 1.07, 95% CI 1.01-1.17, p = .004). High-risk status conferred approximately 3 times greater odds of MDD vs low-risk status (OR = 3.04, 95% CI 1.14–8.43, p = .03). Each year of age increased risk of MDD by 34% (OR = 1.34, 95% CI 1.09–1.67, p = .01). The childhood maltreatment × familial risk interaction was not significant (p > .6).

Conclusion

Childhood maltreatment and familial risk for bipolar I disorder independently increase the risk of MDD. Mitigating either risk factor independently could meaningfully reduce depression risk in at-risk youth.

Diversity & Inclusion Statement

We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We worked to ensure that the study questionnaires were prepared in an inclusive way. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented sexual and/or gender groups in science. We actively worked to promote sex and gender balance in our author group. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group.

Key words: abuse, bipolar I disorder, childhood maltreatment, major depressive disorder, trauma

Plain language summary

This study investigated associations between childhood maltreatment and development of major depressive disorder (MDD) symptoms in 116 adolescents (ages 14-21 years) with (high-risk [HR]) and without (low-risk [LR]) a first-degree family member diagnosed with bipolar I disorder. Half of HR participants exhibited current or past MDD, compared with 24% of the LR group. The authors also found that higher total trauma questionnaire score, HR status, and older age increased the odds of having MDD. Findings suggest that childhood maltreatment and familial risk for bipolar disorder are associated with major depression risk, highlighting opportunities for screening and early intervention efforts in offspring of adults living with bipolar 1 disorder.

Bipolar I disorder is highly heritable, with twin studies demonstrating genetic contributions between 60% and 90% and prospective studies demonstrating that offspring of parents with bipolar disorder have approximately a 13% lifetime risk of developing the disorder themselves.1,2 Although a family history of bipolar I disorder is strongly linked with the emergence of mania, an elevated risk of unipolar major depression is more commonly observed in affected families, with approximately 18% of children of a parent with bipolar disorder meeting lifetime criteria for major depressive disorder (MDD).2, 3, 4 Patients with bipolar I disorder also report high rates of childhood maltreatment compared with the general population.5, 6, 7

Childhood maltreatment, encompassing sexual, emotional, or physical abuse in addition to or apart from physical or emotional neglect, is a well-established risk factor for adverse physical and mental health outcomes, including psychiatric illnesses.8,9 Maltreatment has been linked to neurodevelopmental and epigenetic changes that heighten vulnerability for MDD, as well as other psychiatric and medical conditions, particularly when coupled with familial risk.1,10,11 Despite its high prevalence among patients with bipolar I disorder, the specific mechanisms through which childhood maltreatment interacts with familial risk to influence adolescent mental health remain unclear.10,12 Understanding these interactions may prove essential to creating targeted prevention and intervention strategies to mitigate long-term adverse outcomes.4,8,9,12, 13, 14

A recent study using the Adolescent Brain Cognitive Development (ABCD) dataset affirmed previous work indicating an increased prevalence of depressive symptoms reported by children who have a first-degree family member with bipolar disorder compared with healthy controls. However, this investigation found no significant difference between a subgroup with exposure to interpersonal trauma and an unexposed subgroup in the development of any psychopathologies, countering much of the existing literature linking trauma exposure to the development of bipolar I disorder.15 This discrepancy highlights ongoing uncertainty regarding how childhood maltreatment may contribute to early development of mood disorders in children at risk for bipolar I disorder.

Guided by evidence linking early adversity to neurodevelopmental and epigenetic mechanisms of bipolar disorder inheritance, we investigated associations between cumulative childhood maltreatment and the development of MDD in adolescents and emerging adults with and without a familial risk for bipolar I disorder. We hypothesized that a history of childhood maltreatment would be associated with increased risk of occurrence of MDD in both groups, with an additional synergistic increase in participants with a family history of bipolar I disorder.1,10,11 Here we use the term synergistic to describe a statistical interaction in which the joint presence of familial bipolar I disorder risk and degree of reported childhood maltreatment confers a greater increase in the likelihood of MDD development than would be expected from either factor alone.

Method

Participants

Participants (N = 116) were youths ages 14 to 21 years with or without a first-degree family member diagnosed with bipolar I disorder. Recruitment for both high-risk (with an affected first-degree relative) and low-risk (without an affected first- or second-degree relative with bipolar or a psychotic disorder) groups began in January 2022 at the University of Cincinnati and Dell Medical School at the University of Texas at Austin. The study was approved by the University of Cincinnati Institutional Review Board, with the University of Texas at Austin Institutional Review Board providing approval as the relying board. All participants older than 18 years provided written informed consent, whereas legal guardian consent and minor assent were obtained for participants younger than 18 years before study procedures. Participants for both groups were recruited from similar sources at both sites, including word of mouth, paper flyers around local communities, social media advertisements, and referrals from clinical venues. Exclusion criteria for both groups included evidence of an unstable medical or psychiatric disorder requiring immediate hospitalization or emergency medical intervention (eg, active suicidality); a history of psychiatric hospitalization combined with more than 2 months of lifetime antipsychotic or mood stabilizer treatment; a history of autism spectrum disorder, psychosis, bipolar I or II disorder (mania, mixed episodes, or hypomania with major depression), moderate and severe cannabis or alcohol use disorders, other substance use disorders, or developmental delay; major neurological disorders (eg, seizure disorder or traumatic brain injury with loss of consciousness); and Tanner stage less than 4 (self-reported using a validated scale, categorizing physical development into 5 stages) or an IQ score <80 (assessed by the Kaufman Brief Intelligence Test Second Edition to approximate verbal and nonverbal intelligence).16,17 Additionally, although not part of this report, participants also underwent magnetic resonance imaging examinations and were therefore excluded by a positive urine pregnancy test or contraindications to magnetic resonance imaging (eg, metallic implants, claustrophobia).

The low-risk group (n = 58) comprised participants without a familial history of bipolar or a psychotic disorder (type I or II) in any first- or second-degree relative. The high-risk group (n = 58) consisted of participants with at least 1 parent diagnosed with bipolar I disorder, verified by Structured Clinical Interview for DSM-V–Research Version (SCID-RV), or a biological sibling age 14 or older with bipolar I disorder as assessed by SCID-RV or the Schedule for Affective Disorders and Schizophrenia in School-Age Children–Present and Lifetime version (K-SADS-PL) when their shared parent also met DSM-5 criteria for a bipolar spectrum disorder (eg, bipolar I, bipolar II, cyclothymia, other specified bipolar disorder) or recurrent MDD.18,19

Measures

Demographic Measures

Basic demographic information, including sex, age, race, and ethnicity, was obtained via self-report. Ethnicity was categorized as Hispanic or non-Hispanic and race as Black, Asian, White, Native Hawaiian or Pacific Islander, Native American or Alaskan Native, or other/mixed race. For this analysis, racial data were simplified into 4 groups—Black, White, Asian, and other/mixed race—due to the low frequency of individuals in the groups included in the last-mentioned designation. Socioeconomic status (SES) was assessed using the Hollingshead Four-Factor Index, a survey measuring social status via marital status, employment status, educational attainment, and occupational prestige.20

Childhood Trauma

Childhood trauma was assessed using the Childhood Trauma Questionnaire–Short Form (CTQ-SF), a widely used and psychometrically validated retrospective self-report scale measuring the extent of exposure to abuse and neglect before age 18. Each item on the CTQ-SF is rated on a 5-point Likert scale (“never true” to “very often true”). CTQ-SF total scores (range 25–125) were used to index the degree or severity of cumulative maltreatment exposure, and validated subscales (emotional abuse, emotional neglect, physical abuse, physical neglect, and sexual abuse) were analyzed to characterize specific types of maltreatment exposure.21

Psychopathology Assessment

Psychiatric diagnoses were obtained using the K-SADS-PL for participants and the SCID-RV for adult relatives to confirm a bipolar I diagnosis.18,19 The presence of MDD was defined by DSM-5 criteria as identified by the K-SADS-PL.18,19 To improve specificity, only “definite” presentations were considered to meet MDD criteria for this analysis.

Socioeconomic Status

Responses to the Hollingshead Four-Factor Index, a survey providing a measure of social status by 4 domains, including marital status, employment status, educational attainment, and occupational prestige, were collected by research staff from participants’ head of household and spouse when available.20 A popular, previously unpublished tool to measure SES was made widely available in 2011 and continues to demonstrate good validity.22

Statistical Analysis

We specified a conceptually guided logistic regression to evaluate associations between degree of childhood maltreatment, the presence of a familial risk of developing bipolar I disorder, age, and sex and the probability of having current or previously met diagnostic criteria for MDD.23 To test whether degree of childhood maltreatment exposure and the presence of a familial risk of developing bipolar disorder I combined multiplicatively, we included a CTQ-SF total score × familial risk interaction term. Continuous predictors age and CTQ-SF total scores were mean centered before forming the CTQ-SF total score × familial risk interaction term to reduce multicollinearity and improve interpretability. The predictive performance of this model was assessed using a confusion matrix, along with evaluations of accuracy, sensitivity, specificity, and the area under the curve of the receiver operating characteristic curve, which were provided for transparency but were not emphasized as inferential outcomes.24 Additionally, the variance inflation factor was computed to assess multicollinearity among predictors.25 As a sensitivity analysis, we repeated model selection using stepwise Akaike information criterion (AIC)26 starting from the full model to confirm that excluding the interaction did not improve model fit or change the main effect pattern.26

Group comparisons of CTQ-SF total and subscale scores across risk groups were performed using analysis of covariance controlling for age, sex, race, ethnicity, and SES scores. For skewed variables, such as sexual abuse, we confirmed findings using the Scheirer-Ray-Hare test.27 For instances in which assumptions of normality or homoscedasticity were violated, nonparametric Wilcoxon rank-sum tests were applied to explore age and sex differences. These tests were selected due to their robustness to non-normal distributions and unequal variances, ensuring accurate comparisons when parametric assumptions were not met.28

Results

Participant Characteristics

Participant demographic characteristics are provided in Table 1. The high-risk group was younger (t114 = 2.50, p = .01) than the low-risk group. High-risk participants’ SES scores were lower than their low-risk counterparts (t114 = 2.30, p = .02). Both high- and low-risk groups were primarily White. Racial composition between groups was evaluated using Fisher’s exact test, which found no significant difference between the groups (p = .82).

Table 1.

Demographic Characteristics of Youth

Characteristics High risk (n = 58)
 Low risk (n = 58)
Mean (SD) Mean (SD)
Age, y 17.8 (2.2) 18.8 (2.3)
n (%) n (%)
Sex
 Male 19 (33) 22 (38)
 Female 39 (67) 36 (62)
Race
 Asian 3 (5) 11 (19)
 Black 6 (10) 3 (5)
 Other/mixed race 5 (9) 4 (7)
 White 44 (76) 40 (69)
Hispanic/Latino ethnicity 12 (21) 11 (19)
Mean (SD) Mean (SD)
SES scorea 49 (13) 54 (11)
Open in a new tab

Note: SES = socioeconomic status.

a

SES was assessed by Hollingshead Four-Factor Index.20

Psychiatric Assessments

Participants across groups endorsed current and previous psychiatric conditions, with only 28% of high-risk participants and 59% of low-risk participants having no previous or current presentations of any psychiatric condition. Central to this investigation, MDD was well represented across groups, with 50% of high-risk participants and 24% of low-risk participants endorsing MDD either at the time of enrollment or previously. Of participants who met criteria for MDD, approximately 42% were currently experiencing a depressive episode, and approximately 58% had a past episode but were not currently depressed.

Childhood Maltreatment Reporting

Total CTQ-SF scores of high-risk participants were, on average, 24% higher than those of low-risk participants. High-risk subscores were significantly higher for emotional neglect (F1,107 = 11.5, p < .01, η2 = 0.10), physical neglect (F1,107 = 25.8, p < .01, η2 = 0.20), and emotional abuse (F1,107 = 11.7, p < .01, η2 = 0.10), in which η2 represents the proportion of variance explaining group differences, indicating medium to large effect sizes for these subscales.29 High-risk participants had significantly higher rates of moderate exposure to at least 1 of the 5 trauma subtypes (55%) compared with low-risk participants (21%) (F1,107 = 25.4, p < .01, η2 = 0.20). All results persisted after controlling for age, sex, race, ethnicity, and SES scores.

A detailed account of maltreatment reporting is presented in Table 2. The greatest difference in average CTQ-SF subscores occurred in reports of emotional abuse, in which scores of high-risk participants were, on average, 2.6 points higher than scores of low-risk participants. Physical neglect scores differed by an average of 2.4 points and emotional neglect scores differed by 1.6 points. Although the mean difference in sexual abuse scores was relatively small (1 point), the distribution of scores was highly skewed due to a floor effect, with most participants endorsing the lowest possible score and a few outliers reporting more severe experiences. To account for this, group comparisons for sexual abuse were confirmed using the Scheirer-Ray-Hare nonparametric test (H1 = 6.4, p = .01).27 Reporting of physical abuse did not differ significantly between groups (F1,107 = 3.1, p = .08, η2 = 0.02).21

Table 2.

Childhood Maltreatment Reports

Maltreatment reports High risk (n = 58) Low risk (n = 58) p
Mean (SD) Mean (SD)
Total score 42.6 (12.9) 34.4 (8.7) <.001∗∗∗
Physical abuse 6.4 (2.1) 6.0 (1.8) .097ns
n (%) n (%)
 ≥ Moderate abuse 3 (5.2) 4 (6.9) 1.000ns
Mean (SD) Mean (SD)
Emotional abuse 10.7 (5.4) 8.1 (3.4) .002∗∗
n (%) n (%)
 ≥ Moderate abuse 17 (29.3) 5 (8.6) .013∗
Mean (SD) Mean (SD)
Sexual abusea 6.2 (2.8) 5.2 (0.6) .013∗
n (%) n (%)
 ≥ Moderate abuse 9 (15.5) 1 (1.7) .022∗
Mean (SD) Mean (SD)
Physical neglect 8.8 (3.4) 6.4 (1.5) <.001∗∗∗
n (%) n (%)
 ≥ Moderate abuse 18 (43.1) 1 (1.7) <.001∗∗∗
Mean (SD) Mean (SD)
Emotional neglect 10.4 (3.8) 8.8 (3.6) .002∗∗
n (%) n (%)
 ≥ Moderate abuse 10 (17.2) 6 (10.3) .462ns
Open in a new tab

Note: Moderate cutoff scores on Childhood Trauma Questionnaire–Short Form (CTQ-SF)21 were emotional abuse ≥13, physical abuse ≥10, sexual abuse ≥8, emotional neglect ≥15, and physical neglect ≥10. Analysis of covariance was used for continuous variables; Fisher’s exact test was used for categorical variables. All p values were adjusted using the Benjamini-Hochberg method. ns = nonsignificant p value.

a

Sexual abuse subscale verified by Scheirer-Ray-Hare nonparametric test (H = 6.4, p = .01).

∗

p < .05; ∗∗p < .01; ∗∗∗p < .001.

Multivariable Binary Logistic Regression Analysis

A conceptually guided binary logistic regression evaluated associations between the degree of childhood maltreatment, familial risk of bipolar I disorder, age, and sex and the probability of having current or previously met diagnostic criteria for MDD. The coefficient for CTQ-SF-reported trauma exposure (total score) was 0.07 (p = .004), indicating a significant effect of trauma exposure on the probability of youth in our sample having had MDD. The corresponding odds ratio (OR) was 1.07 (95% CI 1.02-1.12), indicating that for every 1-unit increase (1 point) in the CTQ-SF total trauma score, the odds of MDD increased by approximately 7%. The OR for high-risk youth relative to low-risk youth was 3.06 (95% CI 1.13-8.28, p = .03) suggesting that youth with a familial risk of developing bipolar I disorder have approximately 3 times higher odds of meeting the diagnostic criteria for MDD.

The CTQ-SF × familial risk interaction term was not statistically significant (β = −.02, p = .68), indicating that no multiplicative (synergistic) effect was detected between childhood maltreatment and familial risk. A likelihood ratio test comparing the interaction model with the main effects–only model confirmed that adding the interaction term did not improve fit (ΔAIC = 3.12, p = .41).

The variance inflation factor values for CTQ-SF total score (1.08), risk status (1.28), age (1.21), and sex (1.01) also indicate no significant multicollinearity among these predictors. Predictive performance indices—accuracy (77%), sensitivity (56%), specificity (89%), and area under the curve (0.80)—are reported for transparency but were not used for inferential interpretation. As a sensitivity analysis, we repeated model selection using stepwise AIC selection starting from the full model to confirm that excluding the interaction term did not improve the model fit or change the main effect pattern.

Demographics

Age was a significant predictor of MDD, with each additional year of age within the observed range of 14 to 21 years increasing the relative risk of MDD by 34% compared with the previous year (OR = 1.34, 95% CI 1.07-1.67, p = .01). Male participants had lower odds than female participants of meeting diagnostic criteria for MDD, though this difference did not reach statistical significance (OR = 0.39, 95% CI 0.16-1.17, p = .05). None of the four race categories showed a significant effect on the probability of MDD (Asian: OR = 0.52, 95% CI 0.04-7.72, p = .62; other/mixed race: OR = 2.54, 95% CI 0.21-37.12, p = .47; White: OR = 1.77, 95% CI 0.24-17.36, p = .59). Self-identification as Hispanic ethnicity did not affect the probability of MDD (OR = 0.97, 95% CI 0.31-3.13, p = .96), nor did SES (OR = 1.01, 95% CI 0.97-1.05, p = .76).

Discussion

Consistent with a large existing body of literature, we found that both a history of childhood maltreatment and a family history of bipolar I disorder were associated with an increased risk of MDD in youth.2,8, 9, 10,12,30 The unique contribution of this study was to test the hypothesis that these factors synergistically increased risk, but this hypothesis was not supported by our findings.9, 10, 11 Nonetheless, as with many studies, we found that exposure to childhood maltreatment was associated with increased risk of developing MDD in adolescents.8,9,12 In particular, adolescents with MDD reported significantly greater emotional abuse, emotional neglect, physical neglect, and sexual abuse, but not physical abuse, compared with their nondepressed peers.12,31 This distinction may support a hypothesis that the impact of maltreatment on MDD risk is not only dependent on its severity or presence, but also on its emotional and psychological nature. Emotional and neglect-based maltreatment were associated with MDD in our sample, whereas physical abuse was not; however, given the high comorbidity among maltreatment types, this finding should be interpreted cautiously, as larger samples may detect broader patterns of association.12,30,31

Our observation that childhood maltreatment and familial risk do not synergistically increase risk for emergent depression in youth contrasts with some prior research suggesting gene × environment interactions in mood disorder risk.1,11,32 One possible explanation for this may lie in our unique methodology: we assessed participants at an early stage of potential illness, drawing from a community sample rather than adults or clinical populations.10,32 It is therefore possible that our finding may capture a period of development before gene × environment interactions are fully expressed.14

The high-risk group in our sample was significantly younger, on average, by approximately 1 year, compared with the low-risk group (t114 = 2.50, p = .01). Because the prevalence of MDD typically increases with age, this age difference would be expected to attenuate between-group contrasts in MDD prevalence. Notably, however, we still observed significantly higher rates of depression among the younger high-risk participants, suggesting that the association between familial bipolar risk and depression is robust despite this age difference.

This investigation diverged from previous work in its diagnostic rigor. Diagnostic assessments were conducted using both the SCID and K-SADS, administered by licensed mental health professionals rather than relying on self-report or symptom checklists.1,11,19,30 This approach enhanced diagnostic accuracy for participants and ensured that familial bipolar disorder diagnoses were confirmed through structured interviews rather than inferred from reported family history alone.

If family history of bipolar disorder and childhood maltreatment are independent contributors to risk of MDD in adolescents, then addressing each factor separately during adolescence may offer multiple avenues for intervention.1,9,12,33,34 For example, childhood screening for and early intervention in instances of ongoing maltreatment may significantly reduce exposure to maltreatment and, in turn, lower the risk of developing mood disorders such as MDD.8,9,30 Similarly, familial risk–stratified early childhood depression screening and subsequent preventive interventions aimed at bolstering resilience may reduce the likelihood of mood disorder development in children with increased familial risk.33, 34, 35, 36 Together, these findings support the value of parallel prevention efforts targeting both inherited and environmental vulnerabilities in children and adolescents.8,9,12,30

Childhood maltreatment encompasses a broad spectrum of adversities, making it challenging to develop a universal model for its study.37 Although the CTQ-SF is efficient and widely validated, its retrospective nature limits our ability to confirm the timing of childhood maltreatment relative to the onset of MDD and inherently limits the investigation to an aggregate view of childhood maltreatment exposure.21,30 The use of CTQ-SF coupled with our sample size restricted our ability to delve deeply into the specific roles that different forms of or timing of childhood maltreatment might play in the developmental trajectories associated with MDD.14

This aggregate approach may oversimplify the diverse experiences and effects of different types of maltreatment, potentially missing nuanced impacts specific to emotional, physical, or sexual abuse and neglect.14,30 Though dimensional models show promise in enhancing our understanding of specific pathways through which childhood maltreatment impacts outcomes, operationalizing such an approach in a retrospective analysis remains challenging due to the complexity of childhood maltreatment phenomena, in which adversities often co-occur and overlap.37

Further, the retrospective nature of the CTQ-SF, even when administered to adolescents and emerging adults and therefore temporally closer to the reported experiences, introduces potential recall and response bias. Although retrospective self-report is a pragmatic and widely used method, it may be influenced by mood-congruent cognitive distortions, as individuals currently experiencing a depressive episode tend to recall past experiences in a more negative light. As noted, only a minority of our participants were experiencing current depressive episodes at the time of assessment, reducing the likelihood of such biases substantively affecting reporting. Nevertheless, recent evidence demonstrates good long-term reliability of retrospective reporting of maltreatment using the CTQ-SF in participants with bipolar disorder and consistent associations between depressive symptoms and abuse reporting.38

Identity factors such as sex, gender, and race-specific factors can influence the development and manifestation of MDD and bipolar I disorder, potentially leading to differences in how patients with these disorders present and respond to environmental stressors such as childhood maltreatment.1,39,40 The overrepresentation of White female participants in our sample limits our capacity to fully examine these potential differences. Additionally, our sample size lacks the statistical power to detect subtle influences that race and ethnicity might have on the development of MDD among individuals at familial risk for bipolar I disorder.

We did not independently confirm the hereditary relationship between study participants and their parents with genetic testing. Without genotyping, we cannot confirm whether an affected father was really the proband’s parent. Although the likelihood of this discrepancy is relatively low, it is not zero in many genetic studies. Such uncertainty may have diluted the interaction effect, as some participants labeled as high risk may not be truly at genetic risk. Further, our modest sample size limits power to discern small multiplicative effects between familial risk and degree of childhood maltreatment exposure; therefore, the absence of a significant interaction in our study should be interpreted cautiously.

This study predominantly relied on self-reported data, supplemented by clinical assessments conducted by study physicians. Despite thorough reporting and assessments, the possibility remains that unrecognized bipolar I disorder within families of low-risk participants could influence the familial risk profile of these individuals.2,19

Prospective longitudinal studies that follow adolescents and emerging adults at high risk for developing bipolar I disorder or chronic mood disorders could elucidate the role of adolescent MDD as a precursor or indicator of subsequent bipolar I disorder risk.41 Tools such as risk calculators, which incorporate genetic, environmental, and clinical factors, could further refine early identification efforts, enabling personalized prevention and intervention strategies.34 Moreover, there is a compelling need to explore the specific influences of MDD on bipolar I disorder development within the context of childhood maltreatment.9,30 This focus could provide valuable insights into the interplay between genetic, environmental, and psychological factors in the etiology of mood disorders.

CRediT authorship contribution statement

Savannah D. Layfield: Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jennifer E. Siegel-Ramsay: Writing – review & editing, Methodology, Formal analysis. Lucy Chibib: Writing – review & editing, Investigation, Data curation. Amy Bichlmeier: Writing – review & editing, Project administration, Methodology, Investigation. Patricia L. Nick: Writing – review & editing. Corey R. Jones: Investigation. Caleb M. Adler: Writing – review & editing, Investigation. David E. Fleck: Investigation. Luis R. Patino: Investigation. Gregory A. Fonzo: Writing – review & editing. Elizabeth Lippard: Writing – review & editing, Investigation. Charles B. Nemeroff: Writing – review & editing, Funding acquisition. Melissa DelBello: Writing – review & editing, Supervision, Project administration, Funding acquisition. Jorge R.C. Almeida: Writing – review & editing, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization. Stephen M. Strakowski: Writing – review & editing, Writing – original draft, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.

Footnotes

This article was received under and accepted by Ad Hoc Editor Benjamin I. Goldstein, MD, PhD.

This research was funded by National Institutes of Health Grants: R01MH125886 and R01MH125867 and 5F32MH135706.

Consent has been provided for descriptions of specific patient information.

Data Sharing: The shared dataset will consist of baseline data collected as part of a longitudinal study.

The research team would like to thank the participants and families actively participating in the NECTARY study for their continued contribution to advancing our understanding of the inheritance and early development of bipolar I disorder.

Disclosure: Charles B. Nemeroff is supported by the National Institutes of Health, the National Institute of Mental Health, the Texas Child Mental Health Consortium, and the National Institute of Alcohol Abuse and Alcoholism. Charles B. Nemeroff is a consultant for Abbott Laboratories, Engrail Therapeutics, Clexio Biosciences LTD, Sero (previously Galen Mental Health LLC), Sage Therapeutics, Senseye Inc, Precisement Health, Autobahn Therapeutics Inc, EMA Wellness, Denovo Biopharma LLC, Alvogen, Acadia Pharmaceuticals, Inc, Reunion Neuroscience, Kivira Health, Inc, Wave Neuroscience, Patient Square Capital LP, Invisalert Solutions Inc., Bluecarta Inc., Incannex Healthcare Inc., Supernus, Concierge Clinical Trials Inc., Otsuka, Sensorium, and Neurocrine Biosciences, LLC. Charles B. Nemeroff owns the following patents: Method and devices for transdermal delivery of lithium (US 6,375,990B1), Method of assessing antidepressant drug therapy via transport inhibition of monoamine neurotransmitters by ex vivo assay (US 7,148,027B2), Compounds, compositions, methods of synthesis, and methods of treatment (CRF receptor binding ligand) (US 8,551,996B2). Charles B. Nemeroff owns stock in Corcept Therapeutics Company, EMA Wellness, Precisement Health, Relmada Therapeutics Inc, Signant Health, Galen Mental Health LLC, Kivira Health, Inc., Denovo Biopharma LLC, Headlamp Health Inc., and Senseye Inc. Gregory A. Fonzo has received funds for consulting work for SynapseBio AI and Alto Neuroscience and owns equity in Alto Neuroscience. Jorge R.C. Almeida had research support from NRx Pharmaceutics, Janssen, Alto Neuroscience, PCORI, Magnus Medical, and Baszucki Brain Research Fund. Stephen M. Strakowski has consulted with Sunovion/Sumatomo, WebMD/Medscape, and Oxford University Press, and had research support from Janssen. Savannah D. Layfield, Jennifer E. Siegel-Ramsay, Lucy Chibib, Amy Bichlmeier, Patricia L. Nick, Corey R. Jones, Caleb M. Adler, David E. Fleck, Luis R. Patino, Elizabeth Lippard, and Melissa DelBello have reported no biomedical financial interests or potential conflicts of interest.

References

  • 1.Heim C., Binder E.B. Current research trends in early life stress and depression: review of human studies on sensitive periods, gene-environment interactions, and epigenetics. Exp Neurol. 2012;233(1):102–111. doi: 10.1016/j.expneurol.2011.10.032. [DOI] [PubMed] [Google Scholar]
  • 2.Uher R., Pavlova B., Radua J., et al. Transdiagnostic risk of mental disorders in offspring of affected parents: a meta-analysis of family high-risk and registry studies. World Psychiatry. 2023;22(3):433–448. doi: 10.1002/wps.21147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.O’Connell K.S., Coombes B.J. Genetic contributions to bipolar disorder: current status and future directions. Psychol Med. 2021;51(13):2156–2167. doi: 10.1017/s0033291721001252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Hakamata Y., Suzuki Y., Kobashikawa H., Hori H. Neurobiology of early life adversity: a systematic review of meta-analyses towards an integrative account of its neurobiological trajectories to mental disorders. Front Neuroendocrinol. 2022;65 doi: 10.1016/j.yfrne.2022.100994. [DOI] [PubMed] [Google Scholar]
  • 5.Hosang G.M., Fisher H.L., Hodgson K., Maughan B., Farmer A.E. Childhood maltreatment and adult medical morbidity in mood disorders: comparison of unipolar depression with bipolar disorder. Br J Psychiatry. 2018;213(5):645–653. doi: 10.1192/bjp.2018.178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Etain B., Aas M., Andreassen O.A., et al. Childhood trauma is associated with severe clinical characteristics of bipolar disorders. J Clin Psychiatry. 2013;74(10):991–998. doi: 10.4088/JCP.13m08353. [DOI] [PubMed] [Google Scholar]
  • 7.Garno J.L., Goldberg J.F., Ramirez P.M., Ritzler B.A. Impact of childhood abuse on the clinical course of bipolar disorder. Br J Psychiatry. 2005;186:121–125. doi: 10.1192/bjp.186.2.121. [DOI] [PubMed] [Google Scholar]
  • 8.Lippard E.T.C., Nemeroff C.B. Going beyond risk factor: childhood maltreatment and associated modifiable targets to improve life-long outcomes in mood disorders. Pharmacol Biochem Behav. 2022;215 doi: 10.1016/j.pbb.2022.173361. [DOI] [PubMed] [Google Scholar]
  • 9.Teicher M.H., Gordon J.B., Nemeroff C.B. Recognizing the importance of childhood maltreatment as a critical factor in psychiatric diagnoses, treatment, research, prevention, and education. Mol Psychiatry. 2022;27(3):1331–1338. doi: 10.1038/s41380-021-01367-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Vandeleur C., Rothen S., Gholam-Rezaee M., et al. Mental disorders in offspring of parents with bipolar and major depressive disorders. Bipolar Disord. 2012;14(6):641–653. doi: 10.1111/j.1399-5618.2012.01048.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Coombes B.J., Markota M., Mann J.J., et al. Dissecting clinical heterogeneity of bipolar disorder using multiple polygenic risk scores. Transl Psychiatry. 2020;10(1):314. doi: 10.1038/s41398-020-00996-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Teicher M.H., Samson J.A., Anderson C.M., Ohashi K. The effects of childhood maltreatment on brain structure, function and connectivity. Nat Rev Neurosci. 2016;17(10):652–666. doi: 10.1038/nrn.2016.111. [DOI] [PubMed] [Google Scholar]
  • 13.Bethell C.D., Carle A., Hudziak J., et al. Methods to assess adverse childhood experiences of children and families: toward approaches to promote child well-being in policy and practice. Acad Pediatr. 2017;17(7S):S51–S69. doi: 10.1016/j.acap.2017.04.161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Diaz A., Petersen A.C. Institute of medicine report: new directions in child abuse and neglect research. JAMA Pediatr. 2014;168(2):101–102. doi: 10.1001/jamapediatrics.2013.4560. [DOI] [PubMed] [Google Scholar]
  • 15.Ghaleb C., Penney D., Lavigne K.M., Raucher-Chéné D. Exploring behavioural patterns in youth predisposed to bipolar disorder and the role of interpersonal trauma using the Adolescent Brain Cognitive Development (ABCD) dataset. Early Interv Psychiatry. 2025;19(6) doi: 10.1111/eip.70058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Tanner J.M. 2nd ed. Thomas; Springfield, IL: 1962. Growth at Adolescence. [Google Scholar]
  • 17.Kaufman A.S., Kaufman N.L. In: Encyclopedia of Special Education: A Reference for the Education of Children, Adolescents, and Adults With Disabilities and Other Exceptional Individuals. 4th ed. Reynolds C.R., Vannest K.J., Fletcher-Janzen E., editors. Wiley; Hoboken, NJ: 2014. Kaufman Brief Intelligence Test, Second Edition. [Google Scholar]
  • 18.First M.B. Diagnostic and Statistical Manual of Mental Disorders, 5th Edition, and clinical utility. J Nerv Ment Dis. 2013;201(9):727–729. doi: 10.1097/NMD.0b013e3182a2168a. [DOI] [PubMed] [Google Scholar]
  • 19.Kaufman J., Birmaher B., Brent D., et al. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. J Am Acad Child Adolesc Psychiatry. 1997;36(7):980–988. doi: 10.1097/00004583-199707000-00021. [DOI] [PubMed] [Google Scholar]
  • 20.Hollingshead A.B. Four factor index of social status. Yale Journal of Sociology. 2011;8(11):21–51. [Google Scholar]
  • 21.Bernstein D.P., Stein J.A., Newcomb M.D., et al. Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse Negl. 2003;27(2):169–190. doi: 10.1016/s0145-2134(02)00541-0. [DOI] [PubMed] [Google Scholar]
  • 22.Cirino P.T., Chin C.E., Sevcik R.A., Wolf M., Lovett M., Morris R.D. Measuring socioeconomic status. Assessment. 2002;9(2):145–155. doi: 10.1177/10791102009002005. [DOI] [PubMed] [Google Scholar]
  • 23.Hosmer D.W., Lemeshow S., Sturdivant R.X. Wiley; Hoboken, NJ: 2013. Applied Logistic Regression. [Google Scholar]
  • 24.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters. 2006;27:861–874. doi: 10.1016/j.patrec.2005.10.010. [DOI] [Google Scholar]
  • 25.O’Brien R. A caution regarding rules of thumb for variance inflation factors. Quality & Quantity. 2007;41:673–690. doi: 10.1007/s11135-006-9018-6. [DOI] [Google Scholar]
  • 26.Akaike H. In: Selected Papers of Hirotugu Akaike. Parzen E., Tanabe K., Kitagawa G., editors. Springer; New York, NY: 1998. A new look at the statistical model identification; pp. 215–222. [Google Scholar]
  • 27.Scheirer C.J., Ray W.S., Hare N. The analysis of ranked data derived from completely randomized factorial designs. Biometrics. 1976;32(2):429–434. [PubMed] [Google Scholar]
  • 28.Wilcoxon F. In: Breakthroughs in Statistics: Methodology and Distribution. Kotz S., Johnson N.L., editors. Springer; New York, NY: 1992. Individual comparisons by ranking methods; pp. 196–202. [Google Scholar]
  • 29.Cohen J. Statistical power analysis. Curr Dir Psychol Sci. 1992;1(3):98–101. [Google Scholar]
  • 30.Wade M., Wright L., Finegold K.E. The effects of early life adversity on children’s mental health and cognitive functioning. Transl Psychiatry. 2022;12(1):244. doi: 10.1038/s41398-022-02001-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bijlsma A.M.E., Assink M., Overbeek G., van Geffen M., van der Put C.E. Differences in developmental problems between victims of different types of child maltreatment. J Public Child Welf. 2023;17(2):408–429. doi: 10.1080/15548732.2022.2044429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Aas M., Bellivier F., Bettella F., et al. Childhood maltreatment and polygenic risk in bipolar disorders. Bipolar Disord. 2020;22(2):174–181. doi: 10.1111/bdi.12851. [DOI] [PubMed] [Google Scholar]
  • 33.Ratheesh A., Hett D., Ramain J., et al. A systematic review of interventions in the early course of bipolar disorder I or II: a report of the International Society for Bipolar Disorders Taskforce on early intervention. Int J Bipolar Disord. 2023;11(1):1. doi: 10.1186/s40345-022-00275-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.DelBello M.P. A risk calculator for bipolar disorder in youth: improving the odds for personalized prevention and early intervention? J Am Acad Child Adolesc Psychiatry. 2018;57(10):725–727. doi: 10.1016/j.jaac.2018.07.871. [DOI] [PubMed] [Google Scholar]
  • 35.Van Schoors M., Van Lierde E., Steeman K., Verhofstadt L.L., Lemmens G.M.D. Protective factors enhancing resilience in children of parents with a mental illness: a systematic review. Front Psychol. 2023;14 doi: 10.3389/fpsyg.2023.1243784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ahlborg M.G., Nygren J.M., Svedberg P., Regber S. Resilience in children of parents with mental illness, alcohol or substance misuse—an integrative review. Nurs Open. 2024;11(6) doi: 10.1002/nop2.2219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Catalano R.F., Kellogg E. Fostering healthy mental, emotional, and behavioral development in children and youth: a national agenda. J Adolesc Health. 2020;66(3):265–267. doi: 10.1016/j.jadohealth.2019.12.003. [DOI] [PubMed] [Google Scholar]
  • 38.Hosang G.M., Manoli A., Shakoor S., Fisher H.L., Parker C. Reliability and convergent validity of retrospective reports of childhood maltreatment by individuals with bipolar disorder. Psychiatry Res. 2023;321 doi: 10.1016/j.psychres.2023.115105. [DOI] [PubMed] [Google Scholar]
  • 39.Kuehner C. Why is depression more common among women than among men? Lancet Psychiatry. 2017;4(2):146–158. doi: 10.1016/s2215-0366(16)30263-2. [DOI] [PubMed] [Google Scholar]
  • 40.Piccinelli M., Wilkinson G. Gender differences in depression. Critical review. Br J Psychiatry. 2000;177:486–492. doi: 10.1192/bjp.177.6.486. [DOI] [PubMed] [Google Scholar]
  • 41.Gotlib I.H., Goodman S.H., Humphreys K.L. Studying the intergenerational transmission of risk for depression: current status and future directions. Curr Dir Psychol Sci. 2020;29(2):174–179. doi: 10.1177/0963721420901590. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from JAACAP Open are provided here courtesy of Elsevier

RESOURCES