Suicide and History of Childhood Trauma Among Street YouthScott E. Hadland, MD, MPH1,2, Brandon D. L. Marshall, PhD3,4, Thomas Kerr, PhD3,5, JiezhiQi, MSc3, Julio S. Montaner, MD3,5, and Evan Wood, MD, PhD3,51Boston Medical Center, Department of Pediatrics, One Boston Medical Center Place, Boston,MA, USA, 021182Children’s Hospital Boston, Department of Medicine, 300 Longwood Avenue, Boston, MA, USA,021153British Columbia Centre for Excellence in HIV/AIDS, St. Paul's Hospital, 608-1081 BurrardStreet, Vancouver, BC, Canada, V6Z 1Y64University of British Columbia, School of Population and Public Health, Mather Building, 5804Fairview Avenue, Vancouver, BC, Canada, V6T 1Z35University of British Columbia, Faculty of Medicine, Vancouver, BC, Canada, V6T 1Z3AbstractBackground—Street youth represent a marginalized population marked by early mortality andelevated risk for suicide. It is not known to what extent childhood abuse and neglect predispose tosuicide in this difficult-to-study population. This study is among the first to examine therelationship between childhood trauma and subsequent attempted suicide during adolescence andyoung adulthood among street youth.Methods—From October 2005 to November 2007, data were collected for the At Risk YouthStudy (ARYS), a cohort of 495 street-recruited youth aged 14–26 in Vancouver, Canada. Self-reported attempted suicide in the preceding six months was examined in relation to childhoodabuse and neglect, as measured by the Childhood Trauma Questionnaire (CTQ), using logisticregression.Results—Overall, 46 (9.3%) youth reported a suicide attempt during the preceding six months.Childhood physical and sexual abuse were highly prevalent, with 201 (40.6%) and 131 (26.5%) ofyouth reporting history of each, respectively. Increasing CTQ score was related to risk for suicideattempt despite adjustment for confounders (adjusted odds ratio [AOR], 1.45 per standarddeviation increase in score; 95% confidence interval [CI], 1.08–1.91).Limitations—Use of snowball sampling may not have produced a truly random sample, andreliance on self-report may have resulted in underreporting of risk behaviors among participants.Moreover, use of cross-sectional data limits the degree to which temporality can be concludedfrom the results of this study alone.Conclusions—There exists a strong and graded association between childhood trauma andsubsequent attempted suicide among street youth, an otherwise ‘hidden’ population. There is need© 2011 Elsevier B.V. All rights reserved.Send correspondence to: Evan Wood, MD, PhD, BC Centre for Excellence in HIV/AIDS, St. Paul’s Hospital, 608 - 1081 BurrardStreet, Vancouver, BC, Canada V6Z 1Y6, Phone: (604) 806-9116, Fax: (604) 806-9044, uhri-ew@cfenet.ubc.ca.Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to ourcustomers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review ofthe resulting proof before it is published in its final citable form. Please note that during the production process errors may bediscovered which could affect the content, and all legal disclaimers that apply to the journal pertain.NIH Public AccessAuthor ManuscriptJ Affect Disord. Author manuscript; available in PMC 2013 February 1.Published in final edited form as:J Affect Disord. 2012 February ; 136(3): 377–380. doi:10.1016/j.jad.2011.11.019.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author Manuscriptfor effective interventions that not only prevent maltreatment of children but also aid youth atincreased risk for suicide given prior history of trauma.MeSH Keywordshomeless youth; suicide; child abuse; child neglect; depressionINTRODUCTIONAmong adolescents and adults in the general population, risk of suicide is elevated amongthose with history of childhood physical, sexual or emotional abuse (Dube et al., 2001; Dukeet al., 2010). Street youth, a term generally referring to young people who live on the streetfull- or part-time, demonstrate early mortality and excess risk of suicide compared to theirmainstream peers (Roy et al., 2004). Since street youth represent a ‘hidden’ population thatoften evades standard school- and population-based sampling (Public Health Agency ofCanada, 2006), it is unknown to what extent childhood trauma may be related to risk ofsuicide in this highly marginalized and at-risk group.The present study attempts to replicate the finding that childhood trauma is associated withattempted suicide in a novel population, street youth, who are at once highly vulnerable andgreatly understudied.METHODSThe At Risk Youth Study (ARYS) follows a cohort of youth with extensive streetinvolvement in Vancouver, Canada, and has been described previously (Wood et al., 2006).Inclusion criteria included (1) age 14 to 26 years at enrollment, and (2) use of an illicit drugother than or in addition to marijuana in the month prior to enrollment. Recruitmentemployed street-based outreach methods and snowball sampling. All participants completeda nurse-administered questionnaire, thus providing sociodemographic information and dataon suicidal ideation and attempts, as well as drug-related and sexual risk behaviors. ARYSwas approved by the University of British Columbia/Providence Health Care ResearchEthics Board.The ARYS cohort is a sample of young people, the majority of whom are homeless and havenot completed high school (Wood et al., 2008). A small proportion of participants have beeninvolved in the sex trade, but most report a history of prior incarceration for at least briefperiods of time. Approximately half have previously been a victim of violence, and amajority report having been the perpetrator of a violent act. Although some youth are onlytransiently on the street, a number report having been intensely street-involved for longperiods of time, often measured in years (Fast et al., 2009).Active drug use in the greater street youth population in Vancouver, British Columbia ishigh. Commonly used substances include crystal methamphetamine, crack, cocaine, heroinand marijuana. In the ARYS cohort, injection drug use is common among older youth(Hadland et al., 2010), and many participants self-describe their drug use habits asproblematic (Fast et al., 2009). Many report a prior history of accessing or attempting toaccess drug treatment (Hadland et al., 2009). In general, ARYS youth are heavily involvedin the drug scene, and report being consumed by surviving despite homelessness, chronicpoverty and often dangerous income generation activities (Fast et al., 2009).We hypothesized that among street youth, history of childhood trauma would be stronglyassociated with suicidal attempt later in adolescence and young adulthood. Self-reportedHadland et al. Page 2J Affect Disord. Author manuscript; available in PMC 2013 February 1.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author Manuscriptsuicidal ideation and attempts, respectively, were assessed in the ARYS questionnairethrough the questions, “In the last 6 months, have you ever seriously thought of taking yourown life?” and “In the last 6 months, have you actually attempted suicide?” At-riskparticipants were referred for further psychiatric services as indicated. Childhood traumawas assessed using the Childhood Trauma Questionnaire (CTQ), a 28-item survey assessingphysical, sexual and emotional abuse and physical and emotional neglect that is valid andreliable when applied to adolescent and substance-using samples (Bernstein et al., 2003).Logistic regression was employed with recent suicide attempt as the dependent variable andtotal CTQ score as an independent variable. To adjust for possible confounders, bivariateassociations between recent suicide attempt and a range of sociodemographic, drug use andbehavioral variables were explored; variables associated with suicide attempt with p < 0.10were included as covariates in the final logistic regression model. Gender was additionallyforced into the model to specifically examine the effect of this variable. Subsequently, therole of depressive symptoms as a mediator of the relationship between childhood trauma andrisk of attempted suicide, which has been demonstrated previously (Dube et al., 2001), wasexplored using the Sobel test (Sobel, 1982). Depressive symptoms were measured using theCenter for Epidemiological Studies Depression (CES-D) scale (Radloff, 1991) and scoreswere dichotomized to <22 vs. ≥22.All statistical analyses were performed using SAS version 9.1 (SAS Institute, Inc, Cary,North Carolina). All reported p values are two-sided and considered significant at p < 0.05.RESULTSBetween October 2005 and November 2007, 495 youth were recruited into the ARYS cohortand provided data on history of attempted suicide. Of these, 182 (36.8%) reported lifetimehistory of suicidal ideation and 46 (9.3%) reported an actual attempt during the precedingsix months. Table 1 demonstrates characteristics of youth surveyed. Figure 1 demonstratesthe prevalence of childhood trauma among study participants.Three variables met criteria for inclusion in the final logistic regression model: total CTQscore, age and recent sex trade involvement. A fourth, gender, was forced into the model. Inthe final model, total CTQ score strongly retained its significance (adjusted odds ratio[AOR], 1.45 per 20 point increase [i.e., one standard deviation increase in score]; 95%confidence interval [CI], 1.08 – 1.91). Age additionally retained its significance (AOR, 1.13per year older; 95% CI, 1.01 – 1.27). Notably, when depressive symptoms (CES-D score<22 vs. ≥22) were added to the logistic regression model, they were strongly associated withrecent suicide attempt (AOR, 3.88; 95% CI, 1.87 – 8.05) and significantly reduced the effectof CTQ score on odds of suicide attempt (AOR, 1.01; 95% CI, 0.99 – 1.03), suggesting thatdepressive symptoms may mediate the effects of childhood trauma on attempted suicide. Aformal Sobel test evaluating the role of depressive symptoms as a mediator of therelationship between childhood trauma and odds of attempted suicide was highly significant(p = 0.002) (Sobel, 1982).DISCUSSIONOur results demonstrate that among street youth, a population marked by high rates ofcompleted suicide (Roy et al., 2004), childhood trauma may predispose youth to attemptedsuicide during adolescence and early adulthood. Previous studies have been limited by aninability to establish temporality between childhood/adolescent trauma and subsequentsuicidality (Dube et al., 2001). Our findings also support the role of depressive symptoms asa mediator of this relationship, although it remains unclear based on our results alone howHadland et al. Page 3J Affect Disord. Author manuscript; available in PMC 2013 February 1.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author Manuscriptdepressive symptoms and attempted suicide are temporally related. Other limitations of ourwork include a reliance on self-report as well as on snowball sampling, a method that doesnot produce a truly random sample.Genetic and environmental mediators of the relationship between childhood trauma andsuicide are not yet fully elucidated (Duke et al., 2010). Evidence from neurodevelopmentalstudies suggest that abuse and neglect may lead to ‘hardwired’ cognitive and emotionalchanges that persist into adulthood (Lee and Hoaken, 2007). For example, autopsy studies ofsuicide victims with history of childhood maltreatment demonstrate altered hypothalamic-pituitary-adrenal function relative to non-maltreated suicide victims, suggesting earlyneurodevelopmental changes in response to childhood trauma (McGowan et al., 2009). Suchfindings highlight the need for effective interventions that not only prevent maltreatment ofchildren, but also provide appropriate services for youth who may be at increased risk forsuicide given prior history of trauma.AcknowledgmentsThe authors thank the study participants for their contribution to the research, as well as current and pastresearchers and staff. We would specifically like to thank Deborah Graham, Peter Vann, Caitlin Johnston, SteveKain, and Calvin Lai for their research and administrative assistance.ReferencesBernstein DP, Stein JA, Newcomb MD, Walker E, Pogge D, Ahluvalia T, Stokes J, Handelsman L,Medrano M, Desmond D, Zule W. Development and validation of a brief screening version of theChildhood Trauma Questionnaire. Child Abuse Negl. 2003; 27:169–190. [PubMed: 12615092]Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, householddysfunction, and the risk of attempted suicide throughout the life span: findings from the AdverseChildhood Experiences Study. JAMA. 2001; 286:3089–3096. [PubMed: 11754674]Duke NN, Pettingell SL, McMorris BJ, Borowsky IW. Adolescent violence perpetration: associationswith multiple types of adverse childhood experiences. Pediatrics. 2010; 125:e778–786. [PubMed:20231180]Fast D, Small W, Wood E, Kerr T. Coming 'down here': Young people's reflections on becomingentrenched in a local drug scene. Soc Sci Med. 2009; 21:21.Hadland SE, Kerr T, Li K, Montaner JS, Wood E. Access to drug and alcohol treatment among acohort of street-involved youth. Drug Alcohol Depend. 2009; 101:1–7. [PubMed: 19081203]Hadland SE, Kerr T, Marshall BD, Small W, Lai C, Montaner JS, Wood E. Non-injection drug usepatterns and history of injection among street youth. Eur Addict Res. 2010; 16:91–98. [PubMed:20130409]Lee V, Hoaken PN. Cognition, emotion, and neurobiological development: mediating the relationbetween maltreatment and aggression. Child Maltreat. 2007; 12:281–298. [PubMed: 17631627]McGowan PO, Sasaki A, D'Alessio AC, Dymov S, Labonte B, Szyf M, Turecki G, Meaney MJ.Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhoodabuse. Nat Neurosci. 2009; 12:342–348. [PubMed: 19234457]Radloff LS. The use of the Center for Epidemiologic Studies Depression Scale in adolescents andyoung adults. J Youth Adolesc. 1991; 20:149–166.Roy E, Haley N, Leclerc P, Sochanski B, Boudreau JF, Boivin JF. Mortality in a cohort of street youthin Montreal. JAMA. 2004; 292:569–574. [PubMed: 15292082]Sobel, ME. Asymptotic intervals for indirect effects in structural equation models. In: Leinhardt, S.,editor. Sociological methodology. Jossey-Bass Publishers; San Francisco, CA: 1982.Street youth in Canada: findings from the enhanced surveillance of Canadian street youth, 1999–2003.Public Health Agency of Canada; Ottawa, ON: 2006.Hadland et al. Page 4J Affect Disord. Author manuscript; available in PMC 2013 February 1.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptWood E, Stoltz JA, Montaner JSG, Kerr T. Evaluating methamphetamine use and risks of injectioninitiation among street youth: the ARYS study. Harm Reduction Journal. 2006; 3:18. [PubMed:16723029]Wood E, Stoltz JA, Zhang R, Strathdee SA, Montaner JS, Kerr T. Circumstances of first crystalmethamphetamine use and initiation of injection drug use among high-risk youth. Drug AlcoholRev. 2008; 27:270–276. [PubMed: 18368608]Hadland et al. Page 5J Affect Disord. Author manuscript; available in PMC 2013 February 1.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptFigure 1.Proportion of youth reporting types of abuse/neglect (n = 495). Error bars represent +/− 95%confidence intervals for the estimate.Hadland et al. Page 6J Affect Disord. Author manuscript; available in PMC 2013 February 1.NIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptNIH-PA Author ManuscriptHadland et al. Page 7Table 1Sociodemographic factors, risk behaviors and childhood trauma and their associations with suicide attempt in the preceding six months (n = 495)CharacteristicTotal (%) n = 495Recent suicide attemptaUnadjusted Odds Ratio (OR) (95% CI)p valueYes (%) n = 46No (%) n = 449Demographic/Behavioral Male gender340 (68.7)27 (58.7)313 (69.7)3.08 (2.13 – 4.45)<0.125 Age (mean, SD)21.8 (2.8)22.5 (2.8)21.7 (2.8)1.11 (0.93 – 1.24) b0.067 High school educationc202 (40.8)23 (50.0)179 (39.9)1.51 (0.82 – 2.77)0.183 Recently homelessa375 (75.8)32 (69.6)343 (76.4)0.71 (0.36 – 1.37)0.304 Daily alcohol consumptiona68 (13.7)7 (15.2)61 (13.4)1.14 (0.49 – 2.67)0.760 Sex trade involvement56 (11.3)9 (19.6)47 (10.5)2.08 (0.95 – 4.58)0.064 CES-D score ≥22d197 (39.8)34 (73.9)163 (36.3)4.97 (2.50 – 9.87)<0.001Abuse/Neglect Physical abuse201 (40.6)25 (54.4)176 (39.2)1.85 (1.00 – 3.40)0.046 Sexual abuse131 (26.5)16 (34.8)115 (25.6)1.54 (0.81 – 2.95)0.180 Emotional abuse249 (50.3)33 (71.7)216 (48.1)2.74 (1.40 – 5.34)0.002 Physical neglect229 (46.3)28 (60.9)201 (44.8)1.91 (1.03 – 3.57)0.037 Emotional neglect176 (35.6)24 (52.2)152 (33.9)2.13 (1.16 – 3.93)0.013 Total CTQ score (mean, SD)e51 (20)60 (22)49 (20)1.54 (1.17 – 2.03)0.003a Refers to the preceding six monthsb OR reported is per year increase in agec Has already completed or is currently enrolled in high schoold Center for Epidemiological Studies Depression scale [8]e Childhood Trauma Questionnaire [6]J Affect Disord. Author manuscript; available in PMC 2013 February 1.