Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 22.
Published in final edited form as: J Affect Disord. 2023 Aug 15;340:649–657. doi: 10.1016/j.jad.2023.08.088

Examining the relationship between genetic risk for depression and youth episodic stress exposure

Thomas J Harrison a,*, Anna R Docherty b, Megan C Finsaas c, Roman Kotov d, Andrey A Shabalin b, Monika A Waszczuk e, Benjamin A Katz a, Joanne Davila a, Daniel N Klein a
PMCID: PMC10958668  NIHMSID: NIHMS1970590  PMID: 37591353

Abstract

Background:

Offspring of depressed mothers have elevated risk of developing depression because they are exposed to greater stress. While generally assumed that youth’s increased exposure to stress is due to the environmental effects of living with a depressed parent, youth’s genes may influence stress exposure through gene-environment correlations (rGEs). To understand the relationship between risk for depression and stress, we examined the effects of polygenic risk for depression on youth stress exposure.

Methods:

We examined the relations of a polygenic risk score (PRS) for depression (DEP-PRS), as well as PRSs for 5 other disorders, with youth stress exposure. Data were from a longitudinal study of a community sample of youth and their parents (n = 377) focusing on data collected at youth’s aged 12 and 15 assessments.

Results:

Elevated youth DEP-PRS was robustly associated with increased dependent stress, particularly interpersonal events. Exploratory analyses indicated that findings were driven by major stress and were not moderated by maternal nor paternal history of depression, and of the 5 additional PRSs tested, only elevated genetic liability for bipolar I was associated with increased dependent stress—particularly non-interpersonal events.

Limitations:

Like other PRS studies, we focused on those of European ancestry thus, generalizability of findings is limited.

Conclusion:

Polygenic risk contributes to youth experiencing stressful life events which are dependent on their behavior. This rGE appears to be specific to genetic risk for mood disorders.

Keywords: Polygenic risk score, rGE, Depression, Stress exposure, Genetic liability

1. Introduction

Depression is one of the most pressing mental disorders affecting the physical and mental well-being of over 264 million people worldwide (WHO, 2021). One robust risk factor for depressive episode onset is stressful life events (Bahji et al., 2021; Brown and Harris, 1978; Kendler et al., 1999; Kessler, 1997). Importantly, exposure to stressful life events is not entirely random. Some stressful life events are independent of an individual’s behavior or characteristics (i.e., fateful), while others are at least partially dependent (i.e., controllable) on an individual’s behavior or characteristics (Brown and Harris, 1978).

Offspring of depressed parents experience higher levels of both dependent (e.g., break-up of a romantic relationship) and independent life events (e.g., death of a relative) compared to offspring of nondepressed parents (Adrian and Hammen, 1993; Feurer et al., 2016; Goodman, 2020). This raises the question of whether genetic influences on depression play a role in the higher levels of stress experienced by offspring of depressed parents—an example of gene-environment correlation (rGE).

Genetic influences can lead to stressors in three ways: passive, active, and evocative rGE (Kendler and Karkowski-Shuman, 1997; Plomin et al., 1977). In passive rGE, parents’ genes influence both the parenting environment and the individual’s phenotype. In active rGE, offspring’s genetically influenced traits lead to the selection of environments that fit, and often reinforce, those traits. Finally, evocative rGE refers to heritable phenotypes that shape an individual’s environmental experiences by directly evoking specific responses. These three forms of rGEs may contribute to our understanding of the connection between familial depression risk and stress exposure in youth. More specifically, parental genes may contribute to parents engaging in behaviors or making decisions that create stressors for offspring (passive rGE), while youth genes may contribute to youth self-selecting into environments in which there is a higher risk for stressful life events (i.e., active rGE) and engaging in behavior that elicits negative responses, such as interpersonal rejection from others (i.e., evocative rGE).

rGE could be a mechanism involved in Hammen’s (1991, 2006) stress generation hypothesis, which proposes that depression-prone individuals experience greater self-generated, dependent stress. Active or evocative rGE may contribute to stress generation given that the individual’s behavior or other phenotypic characteristics contribute to the occurrence of dependent life stressors. In contrast, youth’s exposure to independent life events may reflect a mix of truly fateful events and events influenced by their parents’ behavior and decisions (i.e., passive rGE). Thus, youth at elevated risk for depression may have increased exposure to both dependent and independent stress, but genetic effects may account for a larger proportion of the former.

Behavioral and molecular genetics research has illuminated the role of genes on stress exposure. Indeed, behavioral geneticists have demonstrated that the heritability of dependent life events is greater than that of independent life events (Bemmels et al., 2008; Boardman et al., 2011; Kendler and Baker, 2007). Similarly, molecular genetic studies have reported that genetic variants associated with liability for depression may be related to increased risk for dependent stress (see Bahji et al., 2021). For instance, genetic variation in the serotonin transporter gene (5-HTTLPR; Harkness et al., 2015), an elevated multilocus genetic profile score for hypothalamic-pituitary-adrenal (HPA) axis dysfunction (Huang and Starr, 2020), and a single polymorphism of the oxytocin receptor gene (Ebbert et al., 2019) each predicted exposure to dependent, but not independent, stress.

Importantly, genetic risk for depression is highly polygenic, with no polymorphisms having more than small effects (Howard et al., 2019; Hyde et al., 2016; Levey et al., 2020). Genome-wide association studies (GWAS) of depression can be used to generate polygenic risk scores (PRS) indexing depression risk (DEP-PRS, e.g., Howard et al., 2019; Levey et al., 2020). For example, Howard et al. (2019) created a DEP-PRS based on the aggregation of three GWAS studies totaling >1.3 million adults. In doing so, Howard et al. (2019) identified 102 genetic variants contributing to the depression phenotype. This same DEP-PRS was recently used to prospectively predict depressive symptoms following elevated stress exposure in adults (Fang et al., 2019). Despite derivation from adult samples, this DEP-PRS also distinguished youth with steeper trajectories of depressive symptoms across development into adulthood (Kwong et al., 2021). Most recently, Feurer et al. (2022) used the Howard et al. (2019) DEP-PRS to demonstrate that elevated polygenic risk for depression was associated with greater youth exposure to both dependent and independent stress (n = 180). These associations were also moderated by maternal history of major depressive disorder (MDD) and driven by major, as opposed to minor life events.

The present study aimed to extend Feurer et al. (2022) in four respects: 1) examining if genetic liability for depression, indexed by the same GWAS-derived DEP-PRS (Howard et al., 2019) was associated with youth stress assessed twice (at ages 12 and 15) using the same gold-standard interview-based measure in a larger sample of youth (n = 377); 2) considering the implications of both maternal and paternal depression, 3) employing a more conservative statistical approach, and 4) examining the specificity of the association of genetic risk for depression with youth stress exposure by also exploring PRSs for generalized anxiety disorder (GAD; Levey et al., 2020), post-traumatic stress disorder (PTSD; Stein et al., 2021), bipolar I disorder (Mullins et al., 2020), schizophrenia (Trubetskoy et al., 2022), and attention-deficit/hyperactivity disorder (ADHD; Demontis et al., 2019). Like prior behavioral and molecular genetic research (Bahji et al. 2021; Boardman et al., 2011), we hypothesized that youth with higher DEP-PRS would experience greater exposure to dependent stressors. We also hypothesized that this effect would exist over and above the influence of both maternal and paternal history of MDD. In controlling for parental history of MDD we ensure that environmental influences are not explaining what could appear to be rGE. In subsequent analyses, we controlled for the opposite form of stressor (e.g., independent vs. dependent) to remove any shared covariance between them in examining the unique effects of youth DEP-PRS on stressor type. We also controlled for youth lifetime depression diagnosis (Krackow and Rudolph, 2008) and both youth depressive (Liu and Alloy, 2010) and anxiety symptoms (Harrison et al., 2022) given their associations with stress generation that could account for the link between DEP-PRS and stress if their influence was not removed.

Finally, we conducted three exploratory analyses. First, we followed up effects of PRS on dependent stress by distinguishing between interpersonal and non-interpersonal stressors. Second, we examined if the relationship between elevated DEP-PRS and youth stress exposure was driven by major and not minor stressors, as major life events are stronger predictors of depression (Vrshek-Schallhorn et al., 2015). Finally, we explored interaction effects between DEP-PRS and parental history of MDD to determine whether genetic and environmental risk have multiplicative effects on youth stress exposure. In all analyses, we adjusted for the first 10 genetic ancestral principal components of analysis (PCAs) to control for population stratification.

2. Method

2.1. Participants

Participants were drawn from the Stony Brook Temperament Study (Klein and Finsaas, 2017) which is a longitudinal study examining early antecedents and pathways to psychopathology from preschool through adolescence. Three-year-old children and their families (N = 559) who lived within 20 miles of Stony Brook, New York were invited to participate via commercial mailing lists and included if: (1) at least one biological parent spoke English and (2) if the child did not have any medical or developmental disabilities. The sample has been followed at three-year intervals. Parents provided consent and youth provided assent to participate. All study procedures were approved by the Stony Brook University Institutional Review Board (IRB).

In the current study, we examined youth participants who had: (1) age 12 (M = 12.67, SD = 0.42) or 15 (M = 15.22, SD = 0.34) assessment data, (2) a minimum of 80 % European ancestral history (estimated from DNA) to match the ancestry of the original sample that the DEP-PRS was derived (i.e., Howard et al., 2019), and (3) data on parental history of MDD (maternal and paternal). This resulted in an analysis sample of 377 youth and their parents (Table 1).

Table 1.

Means, standard deviations, caseness and percentages for primary variables of interest.

Variable M (SD) n (%)
Maternal MDD (present) – 144 (38.2)
Paternal MDD (present) – 69 (18.3)
Youth MDD (present) – 24 (6.4 %)
Youth depressive symptoms 5.48 (4.96) –
Youth anxiety symptoms 16.31 (9.61) –
Youth sex (% female) – 178 (47.2)
Family INCOME (median) $120,000–149,999 –
Youth DEP-PRS −0.268 (0.006) –
Dependent stress 2.37 (1.85) –
Dependent interpersonal stress 2.45 (1.76) –
Dependent non-interpersonal stress 2.43 (1.45) –
Independent stress 2.50 (1.46) –
Open in a new tab

Note: MDD = Major depressive disorder; Youth DEP-PRS = Youth polygenic risk score for depression. Youth DEP-PRS was multiplied by a constant of 1000 to assist in data interpretation and readability and alternatively to mirror descriptive statistics of previous research and is presented here alongside untransformed data.

2.2. Measures

2.2.1. Parental and youth depression diagnoses

The Structured Clinical Interview for DSM-IV, Nonpatient version (SCID-NP: First et al., 1997) was used to assess maternal and paternal lifetime diagnoses at youth’s initial (age 3) assessment and for the intervening period at youth’s age 9 assessment. Interviews were conducted by telephone and in-person by master’s and doctoral level raters who were supervised by a senior licensed clinical psychologist. Excellent interrater reliability was observed for lifetime maternal and paternal mood disorder diagnoses at youth’s initial (κ = 0.93) and age 9 (κ = 0.91) assessments. The two assessments were then combined to reflect parental lifetime history through the age 9 assessment.

The Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children, Present and Lifetime Version (K-SADS-PL: Kaufman et al., 1997) was used to assess youth depressive disorders. The K-SADS-PL is a widely used semi-structured diagnostic interview that has demonstrated good convergent validity across different measures and reporters (see Dougherty et al., 2018). In the present study the DSM-IV version was administered to a parent (generally mothers) and youth (parent first) at the age 9, 12 and 15 assessments. Interviewers combined information from both reporters to make final ratings. At the age 9 assessment, youth depression was assessed using a lifetime reporting period. The age 12 and 15 assessments focused on the time since the prior interview. The final variable collapsed timepoints to account for child MDD at any wave. All K-SADS-PL interviews were conducted by trained evaluators and supervised by a licensed clinical psychologist and child and adolescent psychiatrist. Substantial interrater reliability was observed on K-SADS-PL depressive disorder diagnoses (κ across waves ranged from 0.72 to 0.79).

2.2.2. Youth symptoms

The Children’s Depression Inventory-Child (CDI-C; Kovacs, 2003) was used to assess depressive symptoms at youth’s aged 12 and 15 assessments. The CDI-C includes 27-items that assess cognitive, affective, and behavioral symptoms of child and adolescent depression. For each item, the child is asked which statement best describes how they have been thinking or feeling over the past 2-weeks on a 3-point Likert scale ranging from 0 (e.g., I am sad once in a while) to 2 (e.g., I am sad all of the time). For analyses, CDI-C scores were averaged across the 2 assessments. The CDI-C demonstrates excellent internal consistency, adequate test-retest reliability, and good convergent validity (Dougherty et al., 2018). Excellent internal consistencies were observed at the age 12 (α = 0.90) and 15 (α = 0.93) assessments in the current study.

The Screen for Child Anxiety Related Emotional Disorders-Child (SCARED-C; Birhamer et al., 1997) was used to assess youth anxiety symptoms at the age 12 and 15 assessments. The SCARED-C is a 41-item questionnaire completed by youth (ages 7–17) assessing a comprehensive range of anxiety symptoms on a 3-point Likert scale ranging from 0 (not true or hardly true) to 2 (very true or often true). Scores were averaged across the 2 assessments. Previous studies have demonstrated strong internal consistency, test-retest reliability, and convergent validity of the SCARED-C (e.g., Behrens et al., 2019; Birmaher et al., 1999; Rappaport et al., 2017). In the current study, good internal consistencies were observed for the SCARED-C at the age 12 (α = 0.82) and 15 assessments (α = 0.83).

2.2.3. Youth episodic life stress

Youth experiences of episodic life stress were assessed at the age 12 and 15 assessments using the UCLA Life Stress Interview (LSI; Adrian and Hammen, 1993). The LSI is a widely used semi-structured interview based on Brown and Harris’ (1978) contextual threat method. It probes episodic life stressors by content-specific domains including social life, friendships, family relationships and work/school. At the age 12 assessment both youth and a parent (generally mothers) were interviewed separately about any stressful life events that occurred in the year prior to that follow-up wave. Following recommendations from the measure’s developers, at the age 15 assessment only youth were interviewed, as parents often have limited knowledge of their child’s experiences by that age. The interviewer presented each event and its context, without including information about the participant’s response to the event to a team of at least 3 experienced raters. The team derived consensus ratings for negative impact (1 = minimal or no effect, 5 = great effect), event dependency (i.e., the degree to which the occurrence of an event was due to the actions of the individual’s actions or fateful/independent of the individual’s actions; 1 = completely dependent, 2 = mixed or intermediate dependence, 3 = completely independent) and interpersonal/non-interpersonal nature (0 = non-interpersonal, 1 = interpersonal) of the event. Episodic stressors with a negative impact score of 3 or greater were considered as major life events (e.g., fight with peer that led to end of friendship), as opposed to minor events (e.g., minor argument with peer). We derived 12 episodic stress variables from these ratings: negative impact scores for total independent and dependent events, total dependent interpersonal and non-interpersonal events, and major and minor independent, dependent, dependent interpersonal and dependent non-interpersonal events. The dependent interpersonal and non-interpersonal stress variables were examined only if the effect for total dependent stress was significant to hone in on the effect. These 12 youth episodic stress variables were computed for the age 12 and 15 assessments and scores were aggregated across the 2 assessments. Previous studies have reported excellent inter-rater reliability regarding the impact (i.e., major vs minor; r = 0.85) and behavioral dependence (r = 0.97) of events using the LSI (Rudolph and Hammen, 1999).

2.2.4. Genotyping

DNA for GWAS was obtained using standard DNA saliva-based collection kits (Genotek’s Oragene). Genotyping of saliva samples was performed in a single batch at the Genomics Shared Resource at Roswell Park Cancer Institute, using the Infinium Global Screening Array (Illumina, San Diego, CA, USA), according to the manufacturer’s protocols. Data were imputed on the Michigan Imputation Server pipeline v1.2.4, using the Haplotype Reference Consortium reference panel (McCarthy et al., 2016). Before imputation, genotypes were filtered for ambiguous strand orientation, missingness rate > 5 % (by marker exclusion, then by individual), Hardy-Weinberg equilibrium violation (p < 10–6), sex mismatch (“sex check” function for X chromosome homozygosity estimate), and non-European ancestry (estimated via principal component analysis combined with the 1000 Genomes Project reference panel). After imputation, the SNPs were excluded for imputation R2 < 0.5, average call rate below 90 % and minor allele frequency below 0.1 %. PLINK was used to handle genetic data and perform quality control (Purcell et al., 2007).

2.2.5. Genetic scores

The primary PRS of interest—youth DEP-PRS, was calculated using GWAS summary statistics from a recent, large GWAS of MDD (i.e., Howard et al., 2019). Exploratory PRSs for generalized anxiety disorder (GAD, i.e., Levey et al., 2020), posttraumatic stress disorder (PTSD, i.e., Stein et al., 2021), bipolar disorder (i.e., Mullins et al., 2020), schizophrenia (i.e., Trubetskoy et al., 2022), and attention-deficit/hyperactivity disorder (ADHD, i.e., Demontis et al., 2019) were also calculated using GWAS summary statistics from large disorder-specific discovery GWASs. Polygenic scores were computed using the PRSice 2.0 software (Euesden et al., 2015), using a complete list of SNPs after clumping and their corresponding weights from the GWAS discovery samples (p-threshold of 1) to incorporate more of the genome (Wray et al., 2014). To partial out the effects of population stratification in all analyses, the first 10 genetic ancestral PCAs were used. Analyses focused on individuals with an at least 80 % European ancestry admixture to match the original sample from which the DEP-PRS was derived (i.e., Howard et al., 2019).

2.3. Data analytic plan

Hierarchical multiple regression analysis was employed to test hypotheses of interest using IBM SPSS Statistics (Version 26). Analyses focused on the main effect of youth’s DEP-PRS and the interactions of youth’s DEP-PRS with both maternal and paternal history of MDD (separate interaction models) on aggregate measures of independent and dependent stress. We did not examine changes in stress exposure over time, but rather averaged episodic life stress scores across the age 12 and 15 assessments to increase the reliability of the stress assessment. Using youth stress as the dependent variable, the first 10 PCAs and both maternal and paternal history of MDD (yes vs. no) were entered as covariates in the first and second block, respectively, and youth DEP-PRS was entered in the third block. Analyses were conducted separately regressing total, and major and minor dependent and independent stress as separate dependent variables. Primary analyses focused on total stress categories, while analyses of major and minor stress were exploratory in nature. In the event that youth DEP-PRS significantly predicted total major or minor dependent stress, we examined whether this effect was driven by dependent interpersonal and/or non-interpersonal stress.

Two follow-up analyses were conducted to test the robustness of significant findings. First, we examined if the effect of youth DEP-PRS on youth stress exposure still held when controlling for the opposite form of stress (e.g., independent stress in a model predicting dependent stress). Second, in addition to the first 10 PCAs in block 1 and maternal and paternal history of MDD in block 2, we also added age, sex, family income, lifetime youth MDD diagnosis, and CDI-C and SCARED-C scores as covariates in block 2. Thus, this model contained 18 covariates and is a substantially more conservative approach than Feurer et al. (2022) who covaried for each of these variables iteratively. The main effect of youth DEP-PRS was entered in block 3.

Next, a set of exploratory analyses were conducted separately regressing minor and major dependent and independent stress on the first 10 PCAs (step 1), paternal and maternal history of MDD (block 2), and youth DEP-PRS (block 3). A final set of exploratory analyses separately examined if maternal or paternal history of MDD moderated the impact of youth DEP-PRS on exposure to total and major dependent and independent stress.

Lastly, to examine the specificity of GWAS-derived mental disorder PRSs in predicting youth stress exposure, we conducted separate hierarchical multiple regressions for each PRS (GAD, PTSD, bipolar I disorder, schizophrenia, and ADHD) with dependent and independent stress, adjusting for the first 10 principal components to control for population stratification. If any PRS predicted dependent stress, we examined whether that effect was driven by dependent interpersonal and/or non-interpersonal stress. The tests of robustness described above were conducted for models with significant PRSs.

3. Results

3.1. Preliminary analyses

Multiple variables of interest (e.g., CDI-C, SCARED-C, LSI scores) were significantly skewed (z > 3.29; see Tabachnick and Fidell, 2007). To meet the assumptions of normality, variables were either square root or inverse transformed preceding all analyses.1 Missing data were observed for <30 cases (<8 %) for covariates and dependent variables of interest. There were no missing data on the independent variable of interest—youth DEP-PRS. Little’s missing completely at random test (Little and Rublin, 1987) was nonsignificant, χ2 (137) = 126.83, p = .722. Thus, maximum likelihood estimation was used to impute missing data in all analyses. Descriptive statistics of untransformed data are presented in Table 1.

3.2. Primary analyses

First, we examined the role of youth DEP-PRS on exposure to total dependent and independent stress. After controlling for the first 10 PCAs and both paternal and maternal history of MDD (SCID-NP) there was a significant main effect of youth DEP-PRS on dependent stress (β = 0.12, p = .017) such that elevated DEP-PRS was associated with higher levels of dependent stress (Table 2). Follow-up analyses demonstrated that the effect of youth DEP-PRS on dependent stress was driven by dependent interpersonal stress (β = 0.11, p = .030; Table 2).

Table 2.

Primary regression analysis.

Block Variable Independent Dependent Dependent interpersonal Dependent non-interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.02 – – – – 0.04 – – – – 0.05 – – – – 0.02
2 Maternal MDD 0.08 0.10 1.44 0.152 0.01 0.06 0.12 1.09 0.277 <0.01 0.06 0.11 1.07 0.287 <0.01 0.10 0.10 1.90 0.058 0.01
Paternal MDD 0.06 0.05 1.18 0.241 −0.02 0.06 −0.46 0.645 <−0.01 0.06 −0.01 0.989 0.05 0.05 0.90 0.371
3 Youth DEP-PRS 0.03 3.48 0.50 0.616 <0.01 0.12 4.28 2.39 0.017 0.02 0.11 4.06 2.18 0.030 0.01 0.05 3.51 0.93 0.356 <0.01
Open in a new tab

Note: PCAs 1–10 = The first ten principal components of analysis to control for population stratification; MDD = Major Depressive Disorder (yes = 1, no = 0); DEP-PRS = Youth polygenic risk score for depression.

A series of analyses then examined the robustness of these significant findings. First, we examined if the main effect of youth DEP-PRS on youth dependent and dependent interpersonal stress exposure held when covarying for the opposite form of stress along with the first 10 PCAs and maternal and paternal history of MDD. Results demonstrated that the significant main effect of youth DEP-PRS held in predicting dependent (β = 0.12, p = .020) and dependent interpersonal stress (β = 0.10, p = .049; Table 3). Second, we examined if these results still held after controlling for the first 10 PCAs, maternal and paternal history of MDD, youth age, sex, and family income, and youth history of MDD and depression and anxiety symptoms. The significant main effect of youth DEP-PRS on dependent (β = 0.10, p = .043) and dependent interpersonal episodic stress (β = 0.10, p = .036) was maintained after controlling for these 18 covariates (Table 4).

Table 3.

Summary of the first test of robustness of primary findings statistically controlling for parental MDD history and opposite episodic stress form.

Block Variable Dependent Dependent interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.04 – – – – 0.05
2 Maternal MDD 0.04 0.12 0.75 0.457 0.07 0.02 0.11 0.40 0.643 0.11
Paternal MDD −0.04 0.06 −0.78 0.434 −0.02 0.06 −0.33 0.744
Opposite stress form 0.25 0.06 5.02 <0.001 0.33 0.06 6.65 <0.001
3 Youth DEP-PRS 0.12 4.15 2.33 0.020 0.02 0.10 3.85 1.98 0.049 0.01
Open in a new tab

Note: PCAs 1–10 = The first ten principal components of analysis to control for population stratification; MDD = Major Depressive Disorder (yes = 1, no = 0); DEP-PRS = Youth polygenic risk score for depression.

Table 4.

Summary of the Second Test of Robustness of Primary Findings Statistically Controlling for Various Different Covariates.

Block Variable Dependent Dependent interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.04 – – – – 0.06
2 Youth MDD status 0.06 0.24 1.08 0.282 0.11 0.03 0.23 0.51 0.612 0.14
Youth depressive Sx 0.28 0.04 4.11 <0.001 0.29 0.03 4.40 <0.001
Youth anxiety Sx −0.07 0.03 −1.12 0.263 −0.04 0.03 −0.63 0.533
Age −0.17 9.18 −3.36 0.001 −0.04 8.53 −0.98 0.326
Sex 0.04 0.05 0.85 0.395 0.10 0.04 2.06 0.040
Family income −0.03 0.05 −0.67 0.501 −0.16 0.05 −3.21 0.001
Maternal MDD 0.04 0.12 0.83 0.405 0.02 0.11 0.46 0.643
Paternal MDD −0.04 0.06 −0.82 0.412 −0.04 0.06 −0.72 0.472
3 Youth DEP-PRS 0.10 4.15 2.04 0.042 0.01 0.10 3.86 2.12 0.034 0.01
Open in a new tab

Note: PCAs 1–10 = The first ten principal components of analysis to control for population stratification; MDD = Major Depressive Disorder (yes = 1, no = 0); Sx: Symptoms; DEP-PRS = Youth polygenic risk score for depression.

3.3. Minor and major episodic stress

Next, we examined if results differed as a function of the impact of the episodic stressors. Paralleling the primary analyses, the first 10 PCAs and both paternal and maternal history of MDD were entered in the first and second blocks as covariates and youth DEP-PRS was entered in the third block. Youth DEP-PRS was a significant predictor of major dependent stress (β = 0.17, p = .002). Follow-up analyses demonstrated that the effect of youth DEP-PRS on dependent stress was driven by both dependent interpersonal and dependent non-interpersonal stress (βs ≥ 0.11, ps ≤ .045; Table 5). However, youth DEP-PRS was not a significant predictor of minor dependent or independent stress.

Table 5.

Summary of regression analyses examining minor versus major youth stress exposure.

Block Variable Minor independent Minor dependent Minor dependent interpersonal Minor dependent non-interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.03 – – – – 0.05 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
2 Maternal MDD −0.01 0.06 −0.12 0.905 <0.01 0.04 0.07 0.77 0.433 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Paternal MDD −0.02 0.03 −0.36 0.722 0.03 0.03 0.49 0.625 ~ ~ ~ ~ ~ ~ ~ ~
3 Youth DEP-PRS 0.06 2.08 1.04 0.298 <0.01 0.02 2.35 0.39 0.695 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Block Variable Major independent Major dependent Major dependent interpersonal Major dependent non-interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.03 – – – – 0.02 – – – – 0.04 – – – – 0.03
2 Maternal MDD 0.11 0.09 2.19 0.029 <0.01 0.09 0.10 1.65 0.100 0.01 0.09 0.10 1.65 0.102 0.01 0.15 0.08 2.80 0.005 0.02
Paternal MDD 0.05 0.05 0.88 0.377 −0.01 0.05 −0.17 0.861 <−0.01 0.05 −0.06 0.950 0.04 0.04 0.77 0.439
3 Youth DEP-PRS 0.06 3.19 1.18 0.238 <0.01 0.17 3.59 3.18 0.002 0.03 0.11 3.58 2.01 0.045 0.02 0.13 3.03 2.55 0.011 0.02
Open in a new tab

Note: ~ = untested model; PCAs 1–10 = The first ten principal components of analysis to control for population stratification; MDD = Major Depressive Disorder (yes = 1, no = 0); DEP-PRS = Youth polygenic risk score for depression.

3.4. Parental MDD history by youth DEP-PRS interaction

We then examined if maternal or paternal history of MDD moderated the relationships between youth DEP-PRS and exposure to both total and major dependent and independent stress. The maternal history of MDD X youth DEP-PRS interaction was not statistically significant in any of the analyses, although a trend for a maternal MDD X youth DEP-PRS interaction for major dependent stress emerged (β = 0.14, p = .054). No paternal history of MDD X youth DEP-PRS interaction models tested were significant (see Supplemental Material).

3.5. Specificity of mental disorder PRSs in predicting youth episodic stress exposure

Lastly, to examine the specificity of GWAS-derived mental disorder PRSs in predicting youth stress exposure, we conducted separate hierarchical multiple regressions for each PRS (GAD, PTSD, bipolar I disorder, schizophrenia, and ADHD) with independent and dependent stress. In doing so, we simultaneously partialed out the first 10 principal components to control for population stratification. If any PRS predicted dependent stress we examined if that effect was driven by dependent interpersonal and/or non-interpersonal stress. Results demonstrated a significant main effect of the bipolar I disorder PRS on dependent (β = 0.12, p = .022) and dependent non-interpersonal stress (β = 0.13, p = .011) such that higher polygenic risk for bipolar I disorder was associated with increased exposure to both stressor forms. The effect of bipolar I PRS on dependent (β = 0.12, p = .032) and dependent non-interpersonal (β = 0.13, p = .013) stress remained significant after controlling for youth DEP-PRS (see Supplemental Material). When additionally controlling for youth and parental history of depression, the effects of bipolar I PRS on dependent stress attenuated to a trend (β = 0.10, p = .055), while dependent non-interpersonal stress remained significant (β = 0.12, p = .027; see Supplemental Material). No significant effects were observed for the PRSs for GAD, PTSD, schizophrenia, and ADHD and total independent or dependent stress (Table 6). Thus, we did not examine their effect on dependent interpersonal nor non-interpersonal stress.

Table 6.

Summary of regression analyses examining the specificity of additional GWAS-derived mental disorder polygenic risk scores.

Block Variable Independent Dependent Dependent interpersonal Dependent non-interpersonal
β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2 β SE (β) t p ΔR2
1 PCAs 1–10 – – – – 0.02 – – – – 0.04 – – – – 0.05 – – – – 0.02
2 Bipolar I PRS 0.08 1.25 1.56 0.119 0.01 0.12 1.54 2.29 0.022 0.01 0.09 1.47 1.70 0.091 0.01 0.13 1.25 2.55 0.011 0.02
2 Schizophrenia PRS 0.03 1.59 0.64 0.523 <0.01 0.11 1.96 0.21 0.830 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
2 ADHD PRS −0.01 0.84 −0.25 0.800 <0.01 0.06 1.03 1.12 0.265 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
2 GAD PRS 0.02 2.79 0.29 0.769 <0.01 0.02 3.44 0.37 0.711 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
2 PTSD PRS 0.03 1.63 0.61 0.609 <0.01 0.06 2.01 1.05 0.296 <0.01 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Open in a new tab

Note: ~ = untested model; PCAs 1–10 = The first ten principal components of analysis to control for population stratification; PRS = Polygenic Risk Score; ADHD: Attention-Deficit Hyperactivity Disorder; GAD = Generalized Anxiety Disorder; PTSD = Post-Traumatic Stress Disorder.

4. Discussion

We extended previous examinations of genetic risk for depression and stress generation in youth (Feurer et al., 2022) by using a larger sample (n = 377 vs n = 180), employing a more conservative statistical approach, and assessing both paternal and maternal history of MDD. Moreover, we tested the specificity of the findings using GWAS-derived PRSs for 5 other mental disorders. Consistent with our primary hypothesis, elevated youth DEP-PRS was associated with increased exposure to dependent stress, particularly of an interpersonal nature, but not to independent stress. These findings held after statistically controlling for the opposite stressor form and youth history of MDD and depression and anxiety symptoms, youth age, sex, and family income, the first 10 PCAs, and maternal and paternal MDD history. Examination of associations of youth DEP-PRS with minor and major life events demonstrated that elevated youth DEP-PRS was associated with greater exposure to major dependent stress—both interpersonal and non-interpersonal - but not minor stress. Moreover, parental history of MDD (maternal nor paternal) did not significantly interact to predict youth stress exposure. Lastly, exploratory analyses examining the specificity of mental disorder PRSs in predicting youth stress exposure, demonstrated a high degree of specificity as only increased genetic liability for bipolar I disorder was associated with increased youth exposure to dependent stress—particularly non-interpersonal in nature. Interestingly, this association persisted after controlling for youth DEP-PRS and later, youth and parental history of depression. Overall, our findings largely replicated previous findings, were impressively robust to potential confounders, and extended the literature by demonstrating a high degree of specificity of GWAS-derived PRSs for mental disorders in predicting youth stress exposure.

Our primary findings are consistent with both behavioral (Boardman et al., 2011; Bemmels et al., 2008) and molecular (Clarke et al., 2019; Harkness et al., 2015; Huang and Starr, 2020) genetic research that has observed a greater degree of heritability for dependent than independent stress. Like Feurer et al. (2022), we found that youth DEP-PRS was associated with elevated exposure to dependent stress—particularly interpersonal events. However, despite using similar methods as Feurer et al. (2022), an association between youth DEP-PRS and independent stress was not observed. This may reflect the challenge of determining the dependence versus independence of stressful life events (Connolly et al., 2010; Harkness and Stewart, 2009; Harrison et al., 2022). Passive rGE may be particularly likely to contribute to independent stress in youth, as events that are independent of the youth’s behavior may still be the result of parental behavior and choices (e.g., parental divorce), and may, in part, reflect genetic influences that are transmitted inter-generationally. However, as we only found effects of youth DEP-PRS on dependent life events, our results suggest that active and evocative rGE play a larger role in depression-related stress generation.

Previous studies have reported that maternal history of MDD is associated with greater youth stress exposure, and particularly increased interpersonal stress (e.g., Adrian and Hammen, 1993; Carter and Garber, 2011; Feurer et al., 2016, 2022). However, the role of paternal MDD has rarely been examined. In the present study, maternal MDD was associated only with youth dependent non-interpersonal stress and paternal MDD was not associated with any form of youth stress exposure. However, some studies in this area have focused exclusively on dependent interpersonal stress (e.g., Bouchard and Shih, 2013; Siegel et al., 2018), suggesting that future work should also examine the ways in which parental depression influences offspring exposure to non-interpersonal stressors.

Feurer et al. (2022) reported a significant interaction in which offspring with depressed mothers and an elevated DEP-PRS experienced the greatest number of dependent major life events. Despite doubling the sample size, we did not observe interactions between parental MDD and youth DEP-PRS on youth stress exposure. Rather, genetic liability to depression had more consistent effects on offspring stress exposure than parental history of depression. This suggests that the DEP-PRS may index youth disposition for stress generation more directly than proxies such as parental MDD.

To broaden the scope of this body of research, we examined the main effect of 5 additional GWAS-derived mental disorder PRSs (bipolar 1 disorder, GAD, PTSD, schizophrenia, and ADHD) on youth stress exposure. Only the PRS for bipolar I disorder significantly predicted elevated levels of youth stress—specifically dependent non-interpersonal events. This effect persisted after controlling for youth DEP-PRS and both youth and paternal history of MDD. The ability of polygenic risk for bipolar 1 disorder to predict elevated dependent noninterpersonal stress parallels literature demonstrating the role of stress generation in bipolar illness (Bender et al., 2010; Grandin et al., 2007), Moreover, offspring of parents with bipolar disorder have reported similar levels of dependent interpersonal and non-interpersonal stress as offspring of parents with MDD (Adrian and Hammen, 1993; Ostiguy et al., 2009). Behavior genetic and GWAS studies have demonstrated considerable genetic overlap between bipolar I disorder and MDD (Kendler et al., 2020; Mullins et al., 2020). However, since PRSs for MDD and bipolar I disorder were both independently associated with dependent life events in youth it suggests that aspects of both their shared and unique liabilities may contribute to stress generation. Given the substantial genetic pleiotropy among all mental disorders (The Brainstorm Consortium, 2018), it is striking that associations with youth stress were specific to the mood disorders and were not evident for any of the other PRSs examined (GAD, PTSD, schizophrenia, and ADHD).

The current study had several strengths such as use of up-to-date polygenic risk scores derived from large discovery samples, and repeated assessment of youth stress and both youth and parental psychopathology, and a gold-standard interview-based measure of youth stress. Moreover, to our knowledge, this is the first study to consider paternal psychopathology—particularly history of MDD in the context of rGE and stress generation. Lastly, we used a much more conservative data analytic approach than previous examinations by controlling for numerous covariates simultaneously rather than one at a time.

However, the current study also had several limitations. First, although much larger than previous studies (e.g., Feurer et al., 2022), our sample was small by standards of contemporary genetic research. Second, we used a DEP-PRS derived from an adult sample. Unfortunately, no large GWAS study of youth depression exists. Finally, like most studies using PRSs, we focused on participants with European ancestry to match the original GWAS sample (Howard et al., 2019). Thus, our findings cannot be generalized to more diverse samples of youth. Fortunately, GWAS studies using samples with non-European ancestry are emerging and more are underway (Choudhury et al., 2020; Graham et al., 2021; Peterson et al., 2019).

In summary, our findings replicate prior work by demonstrating associations between genetic risk for depression as indexed by a GWAS-derived DEP-PRS and youth exposure to dependent stress, suggesting that rGE contributes to stress generation. We extend previous work by demonstrating specificity in PRSs predicting youth stress as only mood disorder PRSs —depression and bipolar I - predicted youth stress. Future research is needed to test the full stress generation model by determining whether dependent stress mediates the association between genetic liability to MDD and the subsequent maintenance or recurrence of major depressive episodes.

Supplementary Material

supplemental tables

Acknowledgements

Support for this research was provided through NIMH R01 MH069942 (Klein).

Role of the sponsor

The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript, and decision to submit the manuscript for publication.

Footnotes

Ethical standards

The authors assert that all procedures contributing to this work comply with the ethical standard of the relevant and institutional committees on human experimentation and with the Helsinki Declaration of 1975. We obtained written consent from subjects.

Location of work

Department of Psychology, Stony Brook University.

CRediT authorship contribution statement

Conception and design of study: Daniel Klein, Benjamin Katz, Thomas Harrison acquisition of data: Daniel Klein, Roman Kotov, Monika Wazczuk, Joanne Davila, Megan Finsaas analysis and/or interpretation of data: Thomas Harrison, Daniel Klein, Monika Wazczuk, Anna Docherty, Andrey Shabalin.

Drafting the manuscript: Thomas Harrison revising the manuscript critically for important intellectual content: Daniel Klein, Roman Kotov, Monika Wazczuk Joanne Davila, Anna Docherty, Andrey Shabalin, Benjamin Katz.

Approval of the version of the manuscript to be published (the names of all authors must be listed): Thomas Harrison, Anna Docherty, Megan Finsaas, Roman Kotov, Andrey Shabalin, Monika Wazczuk, Benjamin Katz, Joanne Davila, Daniel Klein.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jad.2023.08.088.

1

A few variables did not reach the criteria for normality (z < 3.29) defined by Tabachnick and Fidell (2007). Following transformation, whichever transformation type (i.e., square root or inverse) produced the lowest skew statistic was used for those variables.

Data availability

Data can be made available upon formal request.

References

  1. Adrian C, Hammen C, 1993. Stress exposure and stress generation in children of depressed mothers. J. Consult. Clin. Psychol 61 (2), 354–359. 10.1037/0022-006X.61.2.354. [DOI] [PubMed] [Google Scholar]
  2. Bahji A, Forth E, Hargreaves T, Harkness K, 2021. Genetic markers of the stress generation model: A systematic review. Psychiatry Res. 304, 1–11. 10.1016/j.psychres.2021.114139. [DOI] [PubMed] [Google Scholar]
  3. Behrens B, Swetlitz C, Pine DS, Pagliaccio D, 2019. The Screen for Child Anxiety Related Emotional Disorders (SCARED): informant discrepancy, measurement invariance, and test–retest reliability. Child Psychiatry Hum. Dev 50 (3), 473–482. 10.1007/s10578-018-0854-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bemmels HR, Burt SA, Legrand LN, Iacono WG, McGue M, 2008. The heritability of life events: an adolescent twin and adoption study. Twin Res. Hum. Genet 11 (3), 257–265. 10.1375/twin.11.3.257. [DOI] [PubMed] [Google Scholar]
  5. Bender RE, Alloy LB, Sylvia LG, Uroševic S, Abramson LY, 2010. Generation of life events in bipolar spectrum disorders: a re-examination and extension of the stress generation theory. J. Clin. Psychol 66, 1–20. 10.1002/jclp.20705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Birhamer B, Khetarpal S, Brent D, Cully M, Balach L, Kaufman J, 1997. The Screen for Child Anxiety Related Emotional disorders (SCARED): scale construction and psychometric characteristics. J. Am. Acad. Child Adolesc. Psychiatry 36 (4), 545–553. 10.1097/00004583-199704000-00018. [DOI] [PubMed] [Google Scholar]
  7. Birmaher B, Brent DA, Chiappetta L, Bridge J, Monga S, Baugher M, 1999. Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): a replication study. J. Am. Acad. Child Adolesc. Psychiatry 38 (10), 1230–1236. 10.1097/00004583-199910000-00011. [DOI] [PubMed] [Google Scholar]
  8. Boardman JD, Alexander KB, Stallings MC, 2011. Stressful life events and depression among adolescent twin pairs. Biodemogr. Soc. Biol 57 (1), 53–66. 10.1080/19485565.2011.574565. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bouchard LC, Shih JH, 2013. Gender differences in stress generation: examination of interpersonal predictors. J. Soc. Clin. Psychol 32 (4), 424–445. 10.1521/jscp.2013.32.4.424. [DOI] [Google Scholar]
  10. Brown GW, Harris TO, 1978. Social Origins of Depression: A Study of Psychiatric Disorder in Women. Free Press, New York. [Google Scholar]
  11. Carter JS, Garber J, 2011. Predictors of the first onset of a major depressive episode and changes in depressive symptoms across adolescence: stress and negative cognitions. J. Abnorm. Psychol 120 (4), 779–796. 10.1037/a0025441. [DOI] [PubMed] [Google Scholar]
  12. Choudhury A, Aron S, Botigúe LR, Sengupta D, Botha G, Bensellak T, Hanchard NA, 2020. High-depth African genomes inform human migration and health. Nature 586, 741–748. 10.1038/s41586-020-2859-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Clarke T, Zeng Y, Navrady L, Xia C, Haley C, Campbell A, McIntosh A, 2019. Genetic and environmental determinants of stressful life events and their overlap with depression and neuroticism. Wellcome Open Res. 3 (11), 1–17. 10.12688/wellcomeopenres.13893.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Connolly NP, Eberhart NK, Hammen CL, 2010. Specificity of stress generation: a comparison of adolescents with depressive, anxiety, and comorbid diagnoses. Int. J. Cogn. Ther 3 (4), 368–379. 10.1521/ijct.2010.3.4.368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Demontis D, Walters RK, Martin J, Mattheisen M, Als TD, Agerbo E, Neale BM, 2019. Discovery of the first genome-wide significant risk loci for attention deficit/hyperactivity disorder. Nat. Genet 51, 63–75. 10.1038/s41588-018-0269-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dougherty LR, Klein DK, Olino T, 2018. Chapter 6: depression in children and adolescents. In: Hundlsey J, Mash EJ (Eds.), A Guide to Assessments That Work, 2nd ed. Oxford Academic, pp. 99–130. 10.1093/med-psych/9780190492243.003.0006. [DOI] [Google Scholar]
  17. Ebbert AM, Infurna FJ, Luthar SS, Lemery-Chalfant K, Corbin WR, 2019. Examining the link between emotional childhood abuse and social relationships in midlife: the moderating role of the oxytocin receptor gene. Child Abuse Negl. 98, 104–151. 10.1016/j.chiabu.2019.104151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Euesden J, Lewis CM, O’Reilly PF, 2015. PRSice: polygenic risk score software. Bioinformatics 31, 1466–1469. 10.1093/bioinformatics/btu848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fang Y, Scott L, Song P, Burmeister M, Sen S, 2019. Genomic prediction of depression risk and resilience under stress. Nat. Hum. Behav 4 (1), 111–118. 10.1038/s41562-019-0759-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Feurer C, Hammen CL, Gibb BE, 2016. Chronic and episodic stress in children of depressed mothers. J. Clin. Child Adolesc. Psychol 45 (3), 270–278. 10.1080/15374416.2014.963859. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Feurer C, McGeary JE, Brick LA, Knopik VS, Carper MM, Palmer RHC, Gibb BE, 2022. Associations between depression-relevant genetic risk and youth stress exposure: evidence of gene–environment correlations. J. Psychopathol. Clin. Sci 131 (5), 457–466. 10.1037/abn0000757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. First MB, Gibbon M, Spitzer RL, Williams JBW, Benjamin LS, 1997. Structured Clinical Interview of DSM-IV Axis II Personality Disorders, SCID-II. American Psychiatric Association, Washington, DC. [Google Scholar]
  23. Goodman SH, 2020. Intergenerational transmission of depression. Annu. Rev. Clin. Psychol 16, 213–238. 10.1146/annurev-clinpsy-071519-113915. [DOI] [PubMed] [Google Scholar]
  24. Graham SE, Clarke SL, Wu KH, Kanoni S, Zajac GJM, Ramdas, … Willer C, 2021. The power of genetic diversity in genome-wide association studies of lipids. Nature 600, 675–679. 10.1038/s41586-021-0406403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Grandin LB, Alloy LB, Abramson LY, 2007. Childhood stressful life events and bipolar spectrum disorders. J. Soc. Clin. Psychol 26 (4), 460–478. 10.1521/jscp.2007.26.4.460. [DOI] [Google Scholar]
  26. Hammen C, 1991. Generation of stress in the course of unipolar depression. J. Abnorm. Psychol 100 (4), 555–561. 10.1037/0021-843X-100.4.555. [DOI] [PubMed] [Google Scholar]
  27. Hammen C, 2006. Stress generation in depression: reflections on origins, research, and future directions. J. Clin. Psychol 62 (9), 1065–1082. 10.1002/jclp.20293. [DOI] [PubMed] [Google Scholar]
  28. Harkness KL, Stewart JG, 2009. Symptom specificity and the prospective generation of life events in adolescence. J. Abnorm. Psychol 118 (2), 278–287. 10.1037/a0015749. [DOI] [PubMed] [Google Scholar]
  29. Harkness KL, Bagby RM, Stewart JG, Larocque CL, Mazurka R, Strauss JS, Ravindran A, Rector NA, Wynne-Edwards KE, Kennedy JL, 2015. Childhood emotional and sexual maltreatment moderate the relation of the serotonin transporter gene to stress generation. J. Abnorm. Psychol 124 (2), 275–287. 10.1037/abn0000034. [DOI] [PubMed] [Google Scholar]
  30. Harrison TJ, Ginsburg GS, Smith IC, Orlando CM, 2022. Youth stress generation: an examination of the role of anxiety, anxiety symptoms and cognitive distortions. Anxiety Stress Coping 1–16. 10.1080/10615806.2022.2076083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Howard DM, Adams MJ, Clark T-K, Hafferty JD, Gibson J, Shirali M, Ward J, 2019. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the important of the prefrontal brain regions. Nat. Nueorsci 22 (3), 343–352. 10.1038/s41593-018-0326-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Huang M, Starr LR, 2020. Interpersonal childhood adversity and stress generation in adolescence: moderation by HPA axis multilocus genetic variation. Dev. Psychopathol 32, 865–878. 10.1017/S0954579419001123. [DOI] [PubMed] [Google Scholar]
  33. Hyde CL, Nagle MW, Tian C, Chen X, Paciga SA, Wendland JR, Tung JY, Hinds DA, Perlis RH, Winslow AR, 2016. Identification of 15 genetic loci associated with risk of major depression in individuals of European descent. Nat. Genet 48 (9), 1031–1036. 10.1038/ng.3623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, Williamson D, Ryan N, 1997. 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 36 (7), 980–988. 10.1097/00004583-199707000-00021. [DOI] [PubMed] [Google Scholar]
  35. Kendler KS, Baker JH, 2007. Genetic influences on measures of the environment: a systematic review. Psychol. Med 37 (5), 615–626. 10.1017/S0033291706009524. [DOI] [PubMed] [Google Scholar]
  36. Kendler KS, Karkowski-Shuman L, 1997. Stressful life events and genetic liability to major depression: genetic control of exposure to the environment. Psychol. Med 27 (3), 539–547. 10.1017/s0033291797004716. [DOI] [PubMed] [Google Scholar]
  37. Kendler KS, Karkowski LM, Prescott CA, 1999. Causal relationship between stressful life events and the onset of major depression. Am. J. Psychiatr 156 (6), 837–841. 10.1176/ajp.156.6.837. [DOI] [PubMed] [Google Scholar]
  38. Kendler KS, Ohlsson H, Sundquist J, Sundquist K, 2020. An extended Swedish national adoption study of bipolar disorder illness and cross-generational familial association with schizophrenia and major depression. JAMA Psychiatry 77 (8), 814–822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kessler RC, 1997. The effects of stressful life events on depression. Annu. Rev. Psychol 48, 191–214. 10.1146/annurev.psych.48.1.191. [DOI] [PubMed] [Google Scholar]
  40. Klein DN, Finsaas MC, 2017. The stony brook temperament study: early antecedents and pathways to emotional disorders. Child Dev. Perspect 11 (4), 257–263. 10.1111/cdep.12242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Kovacs M, 2003. Children’s Depression Inventory: Technical Manual Update. Multi-Health Systems Inc, North Tonawanda, NY. [Google Scholar]
  42. Krackow E, Rudolph KD, 2008. Life stress and the accuracy of cognitive appraisals in depressed youth. J. Clin. Child Adolesc. Psychol 37 (2), 376–385. 10.1080/15374410801955797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Kwong ASF, Morris TT, Pearson RM, Timpson NJ, Rice F, Stergiakouli E, Tiling K, 2021. Polygenic risk for depression, anxiety, and neuroticism are associated with the severity and rate of change in depressive symptoms across adolescence. J. Child. Psychol. Psychiatry 62 (12), 1462–1474. 10.1111/jcpp.13422. [DOI] [PubMed] [Google Scholar]
  44. Levey DF, Gelernter J, Polimanti R, Zhou H, Cheng Z, Aslan M, Quaden R, Concato J, Radhakrishnan K, Bryois J, Sullivan PF, the Million Veteran Program, Stein MB, 2020. Reproducible genetic risk loci for anxiety: results from ~200,000 participants in the Million Veteran Program. Am. J. Psychiatr 177 (3), 223–232. 10.1176/appi.ajp.2019.19030256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Little RJA, Rublin DB, 1987. Statistical Analysis with Missing Data. Wiley. [Google Scholar]
  46. Liu RT, Alloy LB, 2010. Stress generation in depression: a systematic review of the empirical literature and recommendations for future study. Clin. Psychol. Rev 30 (5), 582–593. 10.1016/j.cpr.2010.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. McCarthy S, Das S, Kretzschmar W, Delaneau O, Wood AR, Teumer A, Haplotype Reference Consortium, 2016. A reference panel of 64,976 haplotypes for genotype imputation. Nat. Genet 48 (10), 1279–1283. 10.1038/ng.3643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mullins N, Forstner AJ, O’Connell KS, Coombes B, Coleman JRI, Qiao Z, Als TD, Bigdeli TB, Børte S, Bryois J, Charney AW, Drange OK, Gandal MJ, Hagenaars SP, Ikeda M, Kamitaki N, Kim M, Krebs K, Panagiotaropoulou G, Andreassen OA, 2020. Genome-wide association study of over 40,000 bipolar disorder cases provides new insights into the underlying biology [preprint]. Genet. Genom. Med 10.1101/2020.09.17.20187054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Ostiguy CS, Ellenbogen MA, Linnen A-M, Walker EF, Hammen C, Hodgins S, 2009. Chronic stress and stressful life events in the offspring of parents with bipolar disorder. J. Affect. Disord 114 (1–3), 74–84. 10.1016/j.jad.2008.08.006. [DOI] [PubMed] [Google Scholar]
  50. Peterson RE, Kuchenbaecker K, Walters RK, Chen C-Y, Popejoy AB, Periyasamy S, Duncan LE, 2019. Genome-wide association studies in ancestrally diverse populations: Opportunities, methods, pitfalls, and recommendations. Cell 179 (3), 589–603. 10.1016/j.cell.2019.08.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Plomin R, DeFries JC, Loehlin JC, 1977. Genotype-environment interaction and correlation in the analysis of human behavior. Psychol. Bull 84 (2), 309–322. 10.1037/0033-2909.84.2.309. [DOI] [PubMed] [Google Scholar]
  52. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Sham PC, 2007. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet 81, 559–575. 10.1086/519795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Rappaport BI, Pagliaccio D, Pine DS, Klein DN, Jarcho JM, 2017. Discriminant validity, diagnostic utility, and parent-child agreement on the Screen for Child Anxiety Related Emotional Disorders (SCARED) in treatment- and non-treatment-seeking youth. J. Anxiety Disord 51, 22–31. 10.1016/j.janxdis/2017.08.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Rudolph KD, Hammen C, 1999. Age and gender as determinants of stress exposure, generation, and reactions in youngsters: a traditional perspective. Child Dev. 70 (3), 660–677. 10.1111/1467-8624.00048. [DOI] [PubMed] [Google Scholar]
  55. Siegel DM, Burke TA, Hamilton JL, Piccirillo ML, Scharff A, Alloy LB, 2018. Social anxiety and interpersonal stress generation: the moderating role of interpersonal distress. Anxiety Stress Coping 31 (5), 526–538. 10.1080/10615806.2018.1482723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Stein MB, Levey DF, Cheng Z, Wendt FR, Harrington K, Pathak GA, Gelernter J, 2021. Genome-wide association analyses of post-traumatic stress disorder and its symptom subdomains in the Million Veteran Program. Nat. Genet 53 (2), 174–184. 10.1038/s41588-020-00767-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Tabachnick BG, Fidell LS, 2007. Using Multivariate Statistics, 5th ed. Pearson. 10.1037/02267. [DOI] [Google Scholar]
  58. The Brainstorm Consortium, Anttila V, Bulik-Sullivan B, Finucane HK, Walters RK, Bras J, Neale BM, 2018. Analysis of shared heritability in common disorders of the brain. Science 360 (6395), eaap8757. 10.1126/science.aap8757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Trubetskoy V, Pardiñas AF, Qi T, Panagiotaropoulou G, Awasthi S, Bigdeli TB, van Os J, 2022. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature 604 (7906), 502–508. 10.1038/s41586-022-04434-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Vrshek-Schallhorn S, Stroud CB, Mineka S, Hammen C, Zinbarg RE, Wolitzky-Taylor K, Craske MG, 2015. Chronic and episodic interpersonal stress as statistically unique predictors of depression in two samples of emerging adults. J. Abnorm. Psychol 124 (4), 918–932. 10.1037/abn0000088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. WHO (World Health Organization), 2021. Depression. Fact Sheet. WHO, Geneva. http://www.who.int/news-room/fact-sheets/detail/depression. [Google Scholar]
  62. Wray NR, Lee SH, Mehta D, Vinkhuyzen A, Dudbridge F, Middledorp CM, 2014. Research review: polygenic methods and their application to psychiatric traits. J. Child Psychol. Psychiatry 55 (10), 1068–1087 doi:10.1111.jcpp.12295. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplemental tables
NIHMS1970590-supplement-supplemental_tables.docx (25.5KB, docx)

Data Availability Statement

Data can be made available upon formal request.

RESOURCES