Childhood Adversity and Mental Health Outcomes Among University Students: A Longitudinal Study

Abstract

French

Background

Mental health concerns are common among university students and maybe elevated among those with specific risk exposures. The study examined the association between childhood adversities and mental health outcomes among undergraduate university students and assessed whether psychosocial and behavioral factors mediate those associations.

Methods

The Queen’s University Student Well-Being and Academic Success Survey identified two large cohorts of first-year undergraduate students entering university in Fall 2018 and 2019 (n = 5,943). At baseline, students reported sociodemographic information, family-related mental health history, childhood physical abuse, sexual abuse, peer bullying, and parental separation or divorce. Baseline and follow-up surveys in Spring 2019, Fall 2019, and Spring 2020 included validated measures of anxiety (7-item Generalized Anxiety Disorder) and depressive symptoms (9-item Patient Health Questionnaire ), non-suicidal self-harm, and suicidality, along with psychological processes and lifestyle variables. Repeated measures logistic regression using Generalized Estimating Equations was used to characterize the associations between childhood adversities and mental health outcomes and examine potential mediation.

Results

Adjusting for age, gender, ethnicity, familial mental illness, and parental education, any childhood abuse (odds ratio: 2.89; 95% confidence interval, 2.58 to 3.23) and parental separation or divorce (odds ratio: 1.29; 95% confidence interval, 1.12 to 1.50) were significantly associated with a composite indicator of mental health outcomes (either 9-item Patient Health Questionnaire score ≥10 or 7-item Generalized Anxiety Disorderscore ≥10 or suicidality or self-harm). The association with childhood abuse weakened when adjusted for perceived stress, self-esteem, and insomnia (odds ratio: 2.05; 95% confidence interval, 1.80 to 2.34), and that with parental divorce weakened when adjusted for self-esteem (odds ratio: 1.17; 95% confidence interval, 1.00 to 1.36).

Conclusion

Childhood abuse and parental separation or divorce were associated with mental health concerns among university students. Childhood adversities may impact later mental health through an association with stress sensitivity, self-esteem, and sleep problems. The findings suggest that prevention and early intervention focusing on improving sleep, self-esteem, and coping with stress while considering the individual risk profile of help-seeking students may help support student mental health.

Keywords

childhood adversity childhood abuse parental separation or divorce student mental health prevention and early intervention

Introduction

University students represent a significant and increasing sector of the young adult population,¹ and mental health concerns are common among university students.² In a recent study of university (majority undergraduate) students in Canada, 36.1% and 31.9% of students reported symptoms of anxiety and depression, respectively.³ A survey of multiple postsecondary institutions across Canada reported the prevalence of past-year self-injury, serious suicidal thoughts, and suicidal attempts to be 8.7%, 13%, and 2.1%, respectively.⁴ Entry to university may be an important time for identifying students at risk of developing distressing symptoms and mental disorders, so that timely interventions can be planned and implemented to address varied student needs.^1,5

Various sociodemographic factors (such as young age, gender, ethnicity, and socioeconomic status) are consistently reported to be associated with poor mental health; however, socio-contextual factors such as childhood adversities are underexplored in university student populations.² The existing childhood adversity literature focusing in the general population has reported an association between childhood adversities such as abuse, neglect, and early loss of attachment figures^6,7 and problems with emotional responses, cognition, and self-concept, with persistent vulnerability to poor mental health into adulthood.^8–10 Compared to children who have not experienced such adversities, those who report higher levels of exposure are four times more likely to develop a mental illness in adulthood.¹¹ Specifically, people with childhood exposure to abuse, neglect, or early loss have a higher risk of depression, anxiety, substance use disorders, and suicidality.^11,12

Some studies have reported an association between childhood adversities and poor mental health among university students. However, these studies have some methodological limitations, such as their reliance on cross-sectional designs, the use of composite scores that merely count the number of adversities, and their focus on a limited set of mental health outcomes.^2,13–15 In addition, entry to university is associated with several new stressful challenges (i.e., more autonomy, financial concerns) and other specific risk factors, including maladaptive coping (such as smoking, substance use), separation from family and other support systems, changes to one's lifestyle (sleep disruption, lack of routine), and increased academic load—all potentially predisposing students to poor mental health.^1,5 Also, the application of the life-course perspective suggests that childhood adversities may influence mental health trajectories throughout life ^16,17 through mechanisms such as perceived stress, health risk behaviors, and lifestyle factors.^2,13,18 We hypothesized that such psychosocial and behavioral factors might mediate the association between childhood adversities and poor mental health. Identification of the mediating pathways might inform early intervention efforts aimed at improving student mental health. The conceptualized causal mechanism can be illustrated using a novel epidemiological tool, that is, Directed Acyclic Graphs (DAGs).¹⁹

Hence, the objective of this analysis was to assess the association between childhood adversity and common mental health concerns and assess the mediating effects of psychosocial and behavioral factors among university students.

Methods

Data Source

The study analyzed data from the U-Flourish Student Well-Being and Academic Success Survey as part of an ongoing longitudinal study of successive cohorts of first-year undergraduate students recruited from Queen’s University in Ontario, Canada, starting in Fall 2018. The survey intended to enroll all first-year undergraduate students who were invited to complete an online survey following a student-led awareness and engagement campaign at the entry to university (Fall: Time 1). Consenting students completed the survey, which collected sociodemographic data, psychological and lifestyle information, family and personal mental health history, history of childhood adversity, and current mental health status. Those students further completed a follow-up survey at the end of the academic year prior to final exams (Spring: Time 2), followed similarly by repeat surveys at the subsequent start of Fall Term (Time 3) and end of next Winter Term (Time 4). Full details of the U-Flourish survey study methodology have been published elsewhere.²⁰ This analysis included data from the two cohorts of first-year students entering Queen’s University in 2018 and 2019. Data from Fall 2018 to Spring 2020 for Cohort 1 (four time points) and Fall 2019 to Spring 2020 for Cohort 2 (two time points) were analyzed.

Out of 5,242 eligible students in Fall 2018 (Cohort 1), 2,994 (57.1%) completed the baseline survey, 1914 completed at Time 2, 914 completed at Time 3, and 619 completed at Time 4. Similarly, out of 5,123 students in Fall 2019 (Cohort 2), 2,949 (57.6%) completed the baseline survey, and 1,119 completed at Time 2. There were less than 10% missing data in each of the exposure and outcome variables at baseline.

Measures

Childhood Adversity Exposure

Students were asked about traumatic experiences from childhood and adolescence in the baseline survey, using questions derived from the Childhood Experience of Care and Abuse (CECA) questionnaire.²¹ The adversity variables were defined based on a “yes” response to the following binary response questions:

Physical abuse: “When you were a child or teenager were you ever hit repeatedly with an implement (such as a belt or stick) or punched, kicked, or burnt by someone in the household?”

Sexual abuse: “When you were a child or teenager did you ever have any unwanted sexual experiences?”

Peer bullying: “When you were a child, or a teenager were you physically or verbally bullied or teased very badly by peers?”

Parental separation/divorce: “Were your parents ever separated or divorced?”. Additionally, to be considered a childhood adversity, respondents had to be under 16 years of age when their parents got separated or divorced.

Mental Health Outcomes

Mental health outcomes were assessed in the baseline and all follow-up surveys. Depressive and anxiety symptoms were measured using the 9-item Patient Health Questionnaire (PHQ-9)²² and the 7-item Generalized Anxiety Disorder (GAD-7) scale, respectively.²³ Suicidal ideation, suicidal attempts, and non-suicidal self-harm were measured using questions derived from the Columbia Suicide Severity Rating Scale.²⁴ The specific outcome variables were defined as follows:

Clinically significant depression symptoms: Present if the score on PHQ-9 scale was 10 or more (standard cut-off).²²

Clinically significant anxiety symptoms: Present if the score on GAD-7 scale was 10 or more (standard cut-off).²³

Non-suicidal self-harm: Responding yes to the question, “Have you ever harmed yourself without the intent of ending your life?”

Suicidality: Having suicidal thoughts (any thoughts of ending your life) and/or suicidal attempts.

Functional impairment related to symptoms of depression and anxiety was considered present if the symptoms in PHQ-9 and GADS-7 scales were rated as making it “very difficult” or “extremely difficult” to work, take care of things at home or get along with other people.²⁵ The reference period for suicidality and non-suicidal self-harm was lifetime at baseline and past 6 months for the follow-up surveys.

Possible Modifying and/or Confounding Variables

Age in years, gender (female vs. male), ethnicity (White vs. non-White), history of reported mental illness in first-degree family members (such as mood, anxiety, psychotic, and substance misuse disorder), and maternal and paternal education (high school or less vs. postsecondary or university) have been reported to be associated with childhood adversity^26,27 and adult mental health.^8,16,18,28 These variables may not fall on the causal pathway and thus were examined as confounders (i.e., the association may be distorted due to intermixing of the effects of exposure with these variables) or effect modifiers (i.e., the magnitude of the effect may differ depending on the level of these variables)²⁹ in the associations under study.

Possible Mediating Variables

Various psychosocial and behavioral variables, which have been reported to be associated with both childhood adversities and poor mental health and may fall on the causal pathway, were examined as potential mediators.^2,13,18 The variables were defined as follows:

Past-month nicotine use: Using 1–10 times or more a day.

Binge drinking: Having ≥4 drinks on one occasion at least once a week in cohort 1 and having ≥5 drinks on one occasion in cohort 2.

Regular cannabis use: Weekly or more frequent use in the past month.

High caffeine use (past month): Consumption more than the amount generally recommended by Health Canada (i.e., ≥4 drinks a day).³⁰

Regular exercise: Exercising once a week or more.

Adiposity subgroups: Body mass index (BMI) is categorized into normal weight, underweight, overweight, and obese, with age- and sex-adjusted thresholds for students aged <18 years.

Clinically significant insomnia or poor sleep quality: Present if the total score on the Sleep Condition Indicator (range: 0–32) was 16 or less.³¹

Stress and psychological traits, namely perceived stress, self-esteem, social support, and social competence, were treated as continuous variables and measured using the following brief validated scales: the Brief Perceived Stress scale,³² the Rosenberg Self-Esteem scale,³³ and the Social Support and Social Competence subscales of the Resiliency Scale for Adolescence, respectively.³⁴

All these variables (except BMI) were measured repeatedly across surveys, and therefore multiple time points were used in the analyses. Refer to Table 1 for information about reference categories for all the variables.

Table 1.

Sample Characteristics at Entry to University (n = 5,943).

Characteristics	% (95% CI)^a
Age in years (mean/SD)	18.1 (1.6)
Gender
Female	67.1 (65.9 to 68.3)
Male	32.9 (31.7 to 34.1)
Ethnicity
Non-White	25.1 (24.0 to 26.2)
White	74.9 (73.8 to 76.0)
Maternal education, highest level completed
High school or below	9.4 (8.6 to 10.2)
Postsecondary or university	90.6 (89.8 to 91.4)
Paternal education, highest level completed
High school or below	10.8 (10.0 to 11.6)
Postsecondary or university	89.2 (88.4 to 90.0)
Mental disorder in family	45.5 (44.2 to 46.8)
Past-month tobacco or vaping use
1–10 times or more	12.6 (11.8 to 13.5)
Never or rarely	87.4 (86.5 to 88.2)
Past-month binge drinking ^c
Weekly or more frequently	65.7 (64.5 to 67.0)
Less than weekly or never	34.3 (33.0 to 35.5)
Past-month cannabis use
Weekly or more frequently	16.8 (15.9 to 17.9)
Less than weekly or never	83.2 (82.1 to 84.1)
Past-month caffeine use
4 or more times a day	2.0 (1.6 to 2.4)
2–3 times a day or less	98.0 (97.6 to 98.4)
Past-month exercise
Once a week or more	61.2 (59.9 to 62.5)
Less than once a week or never	38.8 (37.5 to 40.1)
Clinically significant insomnia/poor sleep quality	23.9 (22.8 to 25.1)
Adiposity subgroups (BMI categories)Normal weight	70.9 (69.7 to 72.1)
Underweight	8.3 (7.7 to 9.1)
Overweight	14.6 (13.7 to 15.5)
Obese	6.2 (5.6 to 6.8)
Childhood adversity
Physical abuse	7.6 (6.9 to 8.3)
Sexual abuse	14.9 (14.0 to 15.8)
Bullying	20.9 (19.8 to 22.0)
Any childhood abuse^b	32.6 (31.3 to 33.8)
Parental separation/divorce before age 16	15.9 (15.0 to16.9)
	Median (IQR)
Perceived stress (range = 0–16)	7.0 (5.0–9.0)
Self-esteem (range = 0–30)	19.0 (15.0–23.0)
Social competence (range = 0–20)	14.0 (10.0–16.0)
Social support (range = 0–20)	17.0 (15.0–19.0)

Notes. BMI = body mass index; CI = confidence interval; IQR = interquartile range.

Descriptive statistics is calculated excluding missing data.

Includes physical abuse, sexual abuse, or bullying.

The 2018 U-Flourish cohort used a definition of binge drinking based on ≥ 4 drinks on a single occasion, but this was subsequently revised to ≥ 5 on one occasion to align more closely with national guidelines in most countries.⁶⁴

Statistical Analysis

The distribution of sample characteristics and prevalence of mental health outcomes were similar across Cohort 1 and Cohort 2 (Supplementary Tables 1 and 2) at the baseline survey. Therefore, for subsequent analyses, data from the two cohorts were combined. Cohort 2 had “unsure” as one of the response options on the sexual abuse item, and the unsure category was significantly associated with mental health outcomes (i.e., appeared a similar response to the “yes” category). Hence, in the combined sample, the unsure category for sexual abuse was collapsed with the “yes” category. At baseline (Supplementary Table 5) and consistently across time points (results not shown), the estimates for the association between physical abuse, sexual abuse, and bullying with all the mental health outcomes were similar, while parental separation/divorce had a smaller strength of association.

Table 2.

Unadjusted and Adjusted Associations Between Childhood Adversity and any Adverse Mental Health Outcome^a Among University Students.

	Unadjusted	Adjusted 1^b	Adjusted 2^c
	OR (95% CI)	OR (95% CI)	OR (95% CI)
Any childhood abuse
Yes	3.17 (2.85 to 3.53)	2.89 (2.58 to 3.23)	2.87 (2.56 to 3.21)
No	1.00	1.00	1.00
Parental separation/divorce
Yes	1.54 (1.35 to 1.75)	1.29 (1.12 to 1.50)	1.19 (1.03 to 1.38)
No	1.00	1.00	1.00

Notes. CI = confidence interval; OR = odds ratio.

Any mental health outcome includes clinically significant symptoms of anxiety and depression, self-harm, and suicidality.

Model 1 adjusted for age, gender, ethnicity, mental disorder in the family, and maternal and paternal education status.

Model 2 included any abuse, parental separation/divorce, age, gender, ethnicity, mental disorder in family, and maternal and paternal education status.

Based on consideration of homogeneity and improvement of precision, we created composite indicators of adversity exposure and mental health outcomes. We analyzed two separate indicators of childhood adversity in further primary analysis: any childhood abuse and parental separation or divorce. “Any abuse” was defined as reported exposure to physical abuse, sexual abuse, or bullying. A composite dichotomous variable, “any mental health outcome,” was created, representing screening positive for at least one of the following mental health outcomes: anxiety and depression symptoms over the screening threshold, self-reported suicidal thoughts/attempts, and non-suicidal self-harm. The pattern of association between specific childhood adversities and mental health outcomes was similar across all the time points. Therefore, repeated measures logistic regression, using Generalized Estimating Equations (GEE), was used to produce population-averaged estimates of effect, accounting for within-subject correlations.³⁵ Exchangeable working correlation and robust standard errors were used in the analyses.

The regression models were adjusted for probable confounders determined a priori: age, gender, ethnicity, history of reported mental illness in first-degree family members, and maternal and paternal education. Due to a lack of power, we did not include the “other” gender group (0.7% of the sample) in the analyses. A conceptual model of the relationship between exposure, outcome, and potential mediators and confounders was formulated using a DAG (Supplementary Figure 1).¹⁹

Before adjustment, effect modification by the variables mentioned above was assessed using their interaction terms with childhood adversities in the models (P-value <0.05 as evidence of modification). If effect modification was ruled out, these covariates were adjusted for in the subsequent models. Then, potential mediators were adjusted, one at a time, to examine whether those adjusted estimates for childhood adversity and mental health weakened (or lost statistical significance), which would suggest mediation of the association. Instead of creating a cumulative exposure variable, these covariates at each time point were allowed to be described independently in the GEE models.³⁶ Descriptive results are presented as percentages, means, and medians along with corresponding 95% confidence intervals (CIs), SDs, and interquartile range (IQR), respectively. Results from the regression models are presented as odds ratios (ORs) and 95% CIs. All analyses were conducted using STATA software version 16.0.³⁷ The study was reviewed for ethical compliance and approved by the Queen’s University and Affiliated Teaching Hospitals Research Ethics Board (HSREB PSIY-609-18) and the University of Calgary Health Research Ethics Board (REB20-0436).

Results

Table 1 presents the sample characteristics in the combined sample (n = 5,943). The mean age of the sample at baseline was 18.1 years (SD 1.6), and the majority were females (67.1%). Around 13% of the sample reported past-month tobacco or vape use, 17% reported weekly or more frequent cannabis use, and 24% reported poor sleep quality. Approximately one-third of students reported exposure to at least one form of childhood abuse. The most common childhood adversity was peer bullying (20.9%) and the least reported was physical abuse (7.4%).

At baseline, 53% (95% CI, 51.8 to 54.4) of the respondents screened positive for at least one of the mental health outcomes, which remained roughly similar throughout the follow-up, even though suicidality at baseline was reported for a lifetime and subsequently for over the past 6 months (since the last survey). The prevalence of mental health outcomes was consistently higher among those exposed to any childhood adversity at each time point (Supplementary Table 4). The prevalence of each mental health outcome at all time points in the combined sample is reported in the Supplementary Table 3. At baseline, 42.8% (95% CI, 40.4 to 45.3) and 44.0% (95% CI, 41.7 to 46.4) of the respondents who had clinically significant levels of depression and anxiety symptoms reported functional impairment.

Table 3.

Association Between Potential Mediators and Screening Positive for Mental Health Outcome in the Combined Sample.

Characteristics	Crude OR (95% CI)
Past-month tobacco or vaping use:1–10 times or moreNever or rarely	1.55 (1.34 to 1.80)Ref
Past-month binge drinking:Weekly or more frequentlyLess than weekly or never	1.06 (0.96 to 1.17)Ref
Past-month cannabis use:Weekly or more frequentlyLess than weekly or never	1.32 (1.19 to 1.45)Ref
Past-month caffeine use:4 or more times a day2–3 times a day or less	2.24 (1.57 to 3.19)Ref
Past-month exercise:Once a week or moreLess than once a week or never	1.44 (1.32 to 1.56)Ref
Clinically significant insomnia/poor sleep quality:YesNo	5.35 (4.78 to 5.99)Ref
BMI groups:ObeseOverweightUnderweightNormal weight	1.18 (0.96 to 1.45)1.20 (1.04 to 1.37)1.20 (1.01 to 1.43)Ref
Self-esteem score (range = 0–30)^a	0.79 (0.78 to 0.81)
Perceived stress score (range = 0–16)^a	1.58 (1.54 to 1.61)
Social competence score (range = 0–20)^a	0.91 (0.90 to 0.92)
Social support score (range = 0–20)^a	0.92 (0.91 to0.93)

Notes. BMI = body mass index; CI = confidence interval; OR = odds ratio.

Continuous variable: Higher score indicates higher construct under study.

Table 4.

Models Examining the Association Between Childhood Adversity and Screening Positive for any Mental Health Outcome^a Adjusted for Each of the Potential Mediators Separately.

	Childhood abuse and any mental health outcome	Parental separation/divorce and any mental health outcome
Potential mediator	OR (95% CI)^b	OR (95% CI)^c
Overall (no mediator)	2.89 (2.58 to 3.23)	1.29 (1.12 to 1.50)
Past-month tobacco/vaping use (≥1–10 times)	3.08 (2.73 to 3.49)	1.30 (1.12 to 1.52)
Past-month binge drinking (at least once/week)	3.20 (2.82 to 3.62)	1.29 (1.10 to 1.51)
Past-month cannabis use (≥weekly)	3.01 (2.68 to 3.39)	1.27 (1.09 to 1.47)
Past-month caffeine use (≥4 times a day)	3.15 (2.79 to 3.56)	1.31 (1.12 to 1.53)
Past-month exercise (<once a week)	2.89 (2.58 to 3.24)	1.29 (1.11 to 1.49)
Adiposity subgroups (normal weight as reference)	2.87 (2.56 to 3.22)	1.29 (1.11 to 1.49)
Clinically significant insomnia	2.51 (2.23 to 2.82)	1.25 (1.08 to 1.46)
Self-esteem (range = 0–30)	2.38 (2.11 to 2.69)	1.23 (1.05 to 1.45)
Perceived stress (range = 0–16)	2.29 (2.03 to 2.59)	1.17 (1.00 to 1.36)
Social competence (range = 0–20)	3.05 (2.71 to 3.44)	1.33 (1.14 to 1.55)
Social support (range = 0–20)	2.75 (2.46 to 3.08)	1.27 (1.10 to 1.47)

Notes. CI = confidence interval; OR = odds ratio.

Any mental outcome includes clinically significant symptoms of anxiety and depression, self-harm, and suicidality.

Association between any childhood abuse and any mental health outcome adjusted for known confounders; further adjusted for each of the potential mediators separately.

Association between parental separation/divorce and any mental health outcome adjusted for known confounders; further adjusted for each of the potential mediators separately.

As decided a priori, the associations were adjusted for age, gender, ethnicity, family history of mental illness, and parental education. There was no evidence of meaningful effect modification by these variables, as the interaction term P-values were >0.05. Exposure to at least one form of childhood abuse and parental separation or divorce were significantly associated with the mental health outcomes, adjusted ORs being 2.89 (95% CI, 2.58 to 3.23) and 1.29 (95% CI, 1.12 to 1.50), respectively (Table 2). Both categories of adversity were independent predictors of the mental health outcomes, with a stronger association observed for childhood abuse (OR 2.87: 95% CI, 2.56 to 3.21) than parental separation or divorce prior to the age of 16 years (OR 1.19: 95% CI, 1.03 to 1.38), each adjusted for the other adversity and the confounders.

Table 3 presents the association between each potential mediator and the composite mental health outcome. Most of these psychosocial and behavioral variables were significantly associated with the mental health outcome.

Table 4 presents the association between childhood adversities and mental health outcomes, adjusting for each potential mediator separately. The association between any abuse and the composite indicator of mental health outcome remained stable when adjusted for most covariates (such as cannabis use, binge drinking, and social support). However, it weakened with adjustment for perceived stress, self-esteem, and clinically significant insomnia separately (Table 5). When these three variables were adjusted simultaneously in the full mediation model, the adjusted OR became 2.05 (95% CI, 1.80 to 2.34). Similarly, only adjustment for self-esteem weakened the association between parental divorce and the mental health outcomes (adjusted OR 1.17 (95% CI, 1.00 to 1.36)).

Discussion

This analysis identified that childhood abuse and parental separation or divorce are associated with common mental health concerns in undergraduate students. Potentially modifiable psychological factors, namely perceived stress, self-esteem, and insomnia appeared to mediate the association between childhood abuse and adverse mental health. Self-esteem also mediated the association of parental divorce with poor mental health.

In this study, first-year university students reported high levels of childhood adversity in keeping with epidemiological data.^2,15,38,39 Further, the high prevalence of mental health concerns among university students found in the study has also been reported consistently in the literature.^5,40,41 The prevalence estimates reported for specific mental health outcomes in this study were slightly higher than those reported in a study of Canadian postgraduate students,⁵ potentially due to the difference in the sample included (household residents only vs. living on- or off-campus) and the measurement instrument used (diagnostic vs. screening tool). Also, there was an overrepresentation of females in our study, and females report a higher prevalence of mental health problems than males.⁵

The finding that childhood adversity is a strong predictor of psychopathology in university students aligns with the extant literature.^38,42 For instance, a study of South African university students reported that at least one adverse exposure (such as bullying, physical and sexual abuse, and parental psychopathology) was significantly associated with depression.¹⁵ A recent systematic review reported a significant association between physical and sexual abuse and suicidal attempts among youth.⁴³ The importance of parental divorce in childhood as a risk factor for adverse mental health, including depression, anxiety, suicidal thoughts, and attempts, has also been reported among college students.^44,45 These findings suggest that early recognition of risk factors, including childhood adversity in help-seeking students, might help map intervention to specific student mental health needs. However, further intervention studies are required to generate evidence-based interventions.

The specific adversities were independent predictors of adverse mental health in adulthood, independent of the other adversity and known confounders. However, our analysis indicated two different dimensions of adversity; the ORs were statistically significant but consistently lower for parental separation/divorce compared to childhood abuse. This finding supports the emerging argument that childhood adversities have various dimensions, such as threat-based (e.g., physical abuse, emotional abuse) and deprivation-based (e.g., parental divorce and substance abuse),⁴⁶ and may differentially contribute to reduced health.⁴⁷

Another important finding was the potential mediating effect of perceived stress, self-esteem, and clinically significant insomnia in the relationship between childhood adversities and mental health outcomes. These findings are consistent with the existing literature that reported childhood abuse is significantly associated with increased perceived stress and maladaptive coping strategies which may be due to the adverse effect of abuse and deprivation on the maturation of self-regulatory systems and abnormally heightened stress response.⁴⁸ Stress has been reported to be a significant risk factor for adverse mental health,⁴⁹ and persons with high stress are also reported to use mental health services frequently.⁵⁰ Also, it has been reported earlier that adolescents with childhood adversity are more likely to experience insomnia,⁵¹ which may be a precipitating factor for mental illness.⁵² Similarly, students with childhood abuse and parental divorce or separation may have low self-esteem, leading to maladjustment and social withdrawal, which may increase vulnerability to mood-related problems.^53–55 This finding suggests that healthy stress coping interventions,^48,56–58 promotion of a healthy lifestyle (especially targeting insomnia symptoms),⁵¹ and strategies for recovery of self-esteem (e.g., encouraging positive self-evaluations)^1,54 need to be examined as potential approaches in prevention and management of mental health concerns in university students. The covariates were measured longitudinally from the start of university, and adversities reportedly occurred in childhood, suggesting mediation; however, due to the nature of data collection (data on adversities and covariates collected simultaneously), the mediation hypothesis could not be established deterministically.

The study has several strengths and limitations to consider. The longitudinal nature of the study, large sample size, and the use of validated measures to assess the outcomes are the strengths. Females were overrepresented throughout the survey, and the estimates were adjusted for gender to make them comparable across gender categories. Although 57% is a reasonable response rate for this type of survey, there is still a potential for selection bias due to nonparticipation and attrition during follow-up. The magnitude of any such bias cannot be quantified using existing data. Nonetheless, the samples were broadly representative of the eligible first-year student population according to the program of study, international student status, and ethnicity.²⁵ The retrospective reports of childhood adversities may have introduced recall bias and possibly strengthened the associations if the misclassification may have been differential (i.e., students with poor mental health may be more likely to recall adverse experiences than their counterparts).⁵⁹ The analysis was not limited to only those students indicating functional impairment associated with symptoms of depression and anxiety. Future studies should seek to quantify the impact on functional impairment, given its importance for informing targeted intervention strategies. Finally, the generalization of findings to other university student populations requires caution, especially when personal and attitudinal variables are analyzed.⁶⁰

Conclusion

The findings suggest that childhood adversities are common and associated with adverse mental health among undergraduate university students. Characterization of this association provides supportive evidence for incorporating risk factors such as early adversity into a more individualized risk profile, which may help with clinical triage and mapping to the appropriate supports and level of care for students seeking mental health support.^1,61 Psychosocial factors (i.e., perceived stress, low self-esteem, and low sleep quality) identified as mediators in the study should be further explored through intervention studies as targets to mitigate the harms to mental health associated with childhood adversities. Student-tailored mental health literacy may support the development of healthy socioemotional coping resources and facilitate timely help-seeking through reduced stigma⁶² and increased emotional self-awareness.⁶³

Supplemental Material

sj-docx-1-cpa-10.1177_07067437221111368 - Supplemental material for Childhood Adversity and Mental Health Outcomes Among University Students: A Longitudinal Study

Supplemental material, sj-docx-1-cpa-10.1177_07067437221111368 for Childhood Adversity and Mental Health Outcomes Among University Students: A Longitudinal Study by Asmita Bhattarai, Nathan King, Kamala Adhikari, Gina Dimitropoulos, Dan Devoe, Jin Byun, Melanie Li, Daniel Rivera, Simone Cunningham, Andrew G.M. Bulloch, Scott B. Patten and Anne Duffy in The Canadian Journal of Psychiatry

Footnotes

Author’s Contribution

All co-authors meet the authorship criteria. A.D. is the nominated principal investigator of the U-Flourish Student Well-Being and Academic Success study. A.B. and S.B.P. conceived the study, guided also by discussions with A.D., A.G.M.B., G.D., and K.A. A.B. analyzed the data and prepared the first draft. A.D., S.B.P., K.A., and N.K. were involved in the discussion of data analysis and interpretation of the results. G.D., D.D, M.L., J.B., D.R., and S.C. also provided intellectual content to the study at all stages. All authors assisted with the preparation of this manuscript and read and approved the final draft.

Data Availability

Access to the data used in the study may be provided upon request to the principal investigator of the U-Flourish Student Well-Being and Academic Success study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. This work was supported by the Alberta Innovates Health Solutions (AIHS) (grant number 201810552) Graduate Studentship, Canadian Institute of Health Research (CIHR), and Cuthbertson and Fischer Chair in Pediatric Mental Health, Rossy Family Foundation.

ORCID iDs

Asmita Bhattarai

Nathan King

Kamala Adhikari

Andrew G.M. Bulloch

Scott B. Patten

Anne Duffy

Supplemental Material

Supplemental material for this article is available online.

References

Duffy

Keown-Stoneman

Goodday

, et al. Predictors of mental health and academic outcomes in first-year university students: identifying prevention and early-intervention targets. BJPsych Open. 2020;6(3):1-8.

Myers

Bantjes

Lochner

Mortier

Kessler

Stein

. Maltreatment during childhood and risk for common mental disorders among first year university students in South Africa. Soc Psychiatry Psychiatr Epidemiol. 2021:1‐13.

Robinson

Jubenville

Renny

Cairns

. Academic and mental health needs of students on a Canadian campus. Canadian Journal of Counselling and Psychotherapy. 2016;50(2):108-123.

Linden

Grey

Stuart

. National standard for the psychological health and safety of post-secondary students – phase I: scoping literature review. Kingston. ON: Queen’s University. 2018, pp. 1-52.

Wiens

Bhattarai

Dores

, et al.

Mental health among Canadian postsecondary students: a mental health crisis?

The Canadian Journal of Psychiatry. 2020;65(1):30‐35.

Anda

Dong

Brown

, et al. The relationship of adverse childhood experiences to a history of premature death of family members. BMC Public Health. 2009;9(1):1‐10.

Felitti

Anda

Nordenberg

, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the adverse childhood experiences (ACE) study. Am J Prev Med. 1998;14(4):245‐258.

Bellis

Hughes

Leckenby

, et al. Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bull W H O. 2014;92:641‐655.

Chartier

Walker

Naimark

. Separate and cumulative effects of adverse childhood experiences in predicting adult health and health care utilization. Child Abuse Negl. 2010;34(6):454‐464.

10.

Hughes

Bellis

Hardcastle

, et al. The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. The Lancet Public Health. 2017;2(8):e356‐ee66.

11.

McLaughlin

. Science brief: the long shadow of adverse childhood experiences. Washington, DC, USA: American Psychological Association (APA); 2017.

12.

Chapman

Whitfield

Felitti

Dube

Edwards

Anda

. Adverse childhood experiences and the risk of depressive disorders in adulthood. J Affect Disord. 2004;82(2):217‐225.

13.

Windle

Haardörfer

Getachew

, et al. A multivariate analysis of adverse childhood experiences and health behaviors and outcomes among college students. J Am Coll Health. 2018;66(4):246‐251.

14.

Bruffaerts

Mortier

Kiekens

, et al. Mental health problems in college freshmen: prevalence and academic functioning. J Affect Disord. 2018;225:97‐103.

15.

Mall

Mortier

Taljaard

Roos

Stein

Lochner

. The relationship between childhood adversity, recent stressors, and depression in college students attending a South African university. BMC Psychiatry. 2018;18(1):1‐10.

16.

Nurius

Green

Logan-Greene

Borja

. Life course pathways of adverse childhood experiences toward adult psychological well-being: a stress process analysis. Child Abuse Negl. 2015;45:143‐153.

17.

Kuh

Ben-Shlomo

Lynch

Hallqvist

Power

. Life course epidemiology. J Epidemiol Community Health. 2003;57(10):778.

18.

Myers

McLaughlin

Wang

Blanco

Stein

. Associations between childhood adversity, adult stressful life events, and past-year drug use disorders in the National Epidemiological Study of Alcohol and Related Conditions (NESARC). Psychol Addict Behav. 2014;28(4):1117‐1126.

19.

Greenland

Pearl

Robins

. Causal diagrams for epidemiologic research. Epidemiology. 1999;10(1):37‐48.

20.

Goodday

Rivera

Foran

, et al. U-Flourish university students well-being and academic success longitudinal study: a study protocol. BMJ open. 2019;9(8):e029854.

21.

Bifulco

Brown

Harris

. Childhood Experience of Care and Abuse (CECA): a retrospective interview measure. Journal of Child Psychology and Psychiatry. 1994;35(8):1419‐1435.

22.

Kroenke

Spitzer

Williams

. The PHQ–9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606‐613.

23.

Spitzer

Kroenke

Williams

Löwe

. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092‐1097.

24.

Posner

Brown

Stanley

, et al. The Columbia–suicide severity rating scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry. 2011;168(12):1266‐1277.

25.

King

Pickett

McNevin

, et al. Mental health need of students at entry to university: baseline findings from the U–flourish student well–being and academic success study. Early Interv Psychiatry. 2021;15(2):286‐295.

26.

Brown

Cohen

Johnson

Salzinger

. A longitudinal analysis of risk factors for child maltreatment: findings of a 17-year prospective study of officially recorded and self-reported child abuse and neglect. Child Abuse Negl. 1998;22(11):1065‐1078.

27.

Park

Fuhrer

Quesnel-Vallée

. Parents’ education and the risk of major depression in early adulthood. Soc Psychiatry Psychiatr Epidemiol. 2013;48(11):1829‐1839.

28.

Pirkola

Isometsä

Aro

, et al. Childhood adversities as risk factors for adult mental disorders. Soc Psychiatry Psychiatr Epidemiol. 2005;40(10):769‐777.

29.

Glasser

. Bias, confounding, and effect modification (interaction). Essentials of Clinical Research: Springer. 2014:363‐372.

30.

Doepker

Franke

Myers

, et al. Key findings and implications of a recent systematic review of the potential adverse effects of caffeine consumption in healthy adults, pregnant women, adolescents, and children. Nutrients. 2018;10(10):1536.

31.

Espie

Kyle

Hames

Gardani

Fleming

Cape

. The sleep condition indicator: a clinical screening tool to evaluate insomnia disorder. BMJ open. 2014;4(3):e004183.

32.

Cohen

Kamarck

Mermelstein

. A global measure of perceived stress. J Health Soc Behav. 1983;24(4):385‐396.

33.

Rosenberg

. Society and the adolescent self-image. Princeton, NJ: Princeton University press; 2015.

34.

Hjemdal

Friborg

Stiles

Martinussen

Rosenvinge

. A new scale for adolescent resilience: grasping the central protective resources behind healthy development. Meas Eval Couns Dev. 2006;39(2):84‐96.

35.

Wang

. Generalized estimating equations in longitudinal data analysis: a review and recent developments. Advances in Statistics. 2014;2014:11, Article ID 303728.

36.

Cook R

Kelso N

Brumback B

Chen

. Analytic strategies to evaluate the association of time-varying exposures to HIV-related outcomes: alcohol consumption as an example. Curr HIV Res. 2016;14(2):85‐92.

37.

StataCorp

. Stata statistical software: release 16. Stata Press: College Station, TX; 2019.

38.

Mortier

Alonso

Auerbach

, et al. Childhood adversities and suicidal thoughts and behaviors among first-year college students: results from the WMH-ICS initiative. Soc Psychiatry Psychiatr Epidemiol. 2021:1‐11. doi:10.1007/s00127-021-02151-4.

39.

Knox

Zusman

DeCuzzi

. The effect of parental divorce on relationships with parents and romantic partners of college students. Coll Stud J. 2004;38(4):597‐602.

40.

Pedrelli

Nyer

Yeung

Zulauf

Wilens

. College students: mental health problems and treatment considerations. Acad Psychiatry. 2015;39(5):503‐511.

41.

Shangraw

Silvers

Warholak

Vadiei

. Prevalence of anxiety and depressive symptoms among pharmacy students. Am J Pharm Educ. 2021;85(2):8166.

42.

Wang

Kelifa

Yang

. Classes of childhood adversities and their associations to the mental health of college undergraduates: a nationwide cross-sectional study. Environ Health Prev Med. 2021;26(1):1‐16.

43.

Angelakis

Austin

Gooding

. Association of childhood maltreatment with suicide behaviors among young people: a systematic review and meta-analysis. JAMA Network Open. 2020;3(8):e2012563.

44.

Short

. The effects of parental divorce during childhood on college students. J Divorce Remarriage. 2002;38(1–2):143‐155.

45.

Størksen

Røysamb

Moum

Tambs

. Adolescents with a childhood experience of parental divorce: a longitudinal study of mental health and adjustment. J Adolesc. 2005;28(6):725‐739.

46.

McLaughlin

DeCross

Jovanovic

Tottenham

. Mechanisms linking childhood adversity with psychopathology: learning as an intervention target. Behav Res Ther. 2019;118:101‐109.

47.

Chen

Turiano

Mroczek

Miller

. Association of reports of childhood abuse and all-cause mortality rates in women. JAMA psychiatry. 2016;73(9):920‐927.

48.

Hyman

Paliwal

Sinha

. Childhood maltreatment, perceived stress, and stress-related coping in recently abstinent cocaine dependent adults. Psychol Addict Behav. 2007;21(2):233‐238.

49.

Karatekin

Ahluwalia

. Effects of adverse childhood experiences, stress, and social support on the health of college students. J Interpers Violence. 2020;35(1–2):150‐172.

50.

Prior

Vestergaard

Larsen

Fenger-Grøn

. Association between perceived stress, multimorbidity and primary care health services: a Danish population-based cohort study. BMJ open. 2018;8(2):e018323.

51.

Wang

Raffeld

Slopen

Hale

Dunn

. Childhood adversity and insomnia in adolescence. Sleep Med. 2016;21:12‐18.

52.

Pigeon

Bishop

Krueger

. Insomnia as a precipitating factor in new onset mental illness: a systematic review of recent findings. Curr Psychiatry Rep. 2017;19(8):1‐11.

53.

Lee

Bae

Rim

H-D

, et al. Mediating effect of resilience on the association between emotional neglect and depressive symptoms. Psychiatry Investig. 2018;15(1):62‐69.

54.

Mann

Hosman

Schaalma

De Vries

. Self-esteem in a broad-spectrum approach for mental health promotion. Health Educ Res. 2004;19(4):357‐372.

55.

Block

Spiegel

. The impact of parental divorce on emerging adults’ self-esteem. New York University’s Applied Psychology OPUS. 2013.

56.

Futa

Nash

Hansen

Garbin

. Adult survivors of childhood abuse: an analysis of coping mechanisms used for stressful childhood memories and current stressors. J Fam Violence. 2003;18(4):227‐239.

57.

Xiang

Wang

Guan

. The relationship between child maltreatment and dispositional envy and the mediating effect of self-esteem and social support in young adults. Front Psychol. 2018;9:1054.

58.

Karatekin

. Adverse childhood experiences (ACEs), stress and mental health in college students. Stress Health. 2018;34(1):36‐45.

59.

Patten

Wilkes

Williams

, et al. Retrospective and prospectively assessed childhood adversity in association with major depression, alcohol consumption and painful conditions. Epidemiol Psychiatr Sci. 2015;24(2):158‐165.

60.

Hanel

Vione

. Do student samples provide an accurate estimate of the general public? PloS one. 2016;11(12):e0168354.

61.

Bhattarai

Dimitropoulos

Marriott

, et al.

Can the adverse childhood experiences (ACEs) checklist be utilized to predict emergency department visits among children and adolescents?

BMC Med Res Methodol. 2021;21(1):1‐16.

62.

Chisholm

Patterson

Torgerson

Turner

Jenkinson

Birchwood

. Impact of contact on adolescents’ mental health literacy and stigma: the SchoolSpace cluster randomised controlled trial. BMJ open. 2016;6(2):e009435.

63.

Kelly

Jorm

Wright

. Improving mental health literacy as a strategy to facilitate early intervention for mental disorders. Med J Aust. 2007;187(S7):S26‐S30.

64.

Bulloch

Williams

Lavorato

Patten

. Trends in binge drinking in Canada from 1996 to 2013: a repeated cross-sectional analysis. Canadian Medical Association Open Access Journal. 2016;4(4):E599‐E604.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.41 MB