SciELO - Scientific Electronic Library Online

vol.90 issue4Arterial oxygen saturation and heart rate in term newborn in the first hour after birthValidation of “The Parental Stressor Scale Infant Hospitalization modified, to Spanish” in a pediatric intensive care unit author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Revista chilena de pediatría

Print version ISSN 0370-4106

Rev. chil. pediatr. vol.90 no.4 Santiago Aug. 2019 


Association between attachment and suicidal risk in Colombian adolescent students

Yuly Suárez Colorado1 

Adalberto Campo-Arias2 

1 Universidad Cooperativa de Colombia, Colombia.

2 Universidad del Magdalena. Colombia, Colombia.



Attachment is considered an important element in mental health, however, the relationship between attachment dimensions and suicidal risk has been little studied.


To establish the association between trust, communication, and alienation and suicide risk in Colombian school adolescents.

Subjects and Method:

A cross-sectional observational study with an analytical component was de signed in a randomized sample of 399 school adolescents (n = 1,901), according to a 30% prevalence of high suicide risk, members of the official educational institutions of Santa Marta (Colombia), the commune with the highest index of unsatisfied basic needs. Participants completed the Armsden &

Greenberg attachment inventory and the Plutchik suicide risk scale, both validated locally. Reliability and association tests were calculated.


Adolescents were 339 adolescents between 13 and 19 years of age (M = 15.7, SD = 1.1), 57.8% were women. 59.9% of adolescents showed low trust in the mother, 57.2% low communication with the mother, 54.9% low alienation with the mother, 46.3%, low trust in the father, 49% low communication with the father, 48.7% low alienation with the father, and 28.6% high suicide risk. There was an association (adjusted) between high suicide risk and trust in the mother (OR = 2.00, 95%CI 1.12-3.57), communication with the mother (OR = 3.80, 95%CI 2.13-6.75), trust in the father (OR = 2.39, 95%CI 1.41-4.03), and communication with the father (OR = 2.01, 95%CI 1.19-3.37).


Low trust and low communication with mother and father are risk factors for high suicide risk in Colombian school adolescents. Further research on this association in other populations is needed, as well as to consider other mediating factors.

Keywords: Suicide; attachment; adolescents; cross-sectional studies


Suicide is a global public health problem, especially for young people1. In Colombia, 19,977 suicide cases were recorded between 2008 and 2017 at all ages. For example, 2,571 cases were reported in 2017; out of the se, 415 cases were persons aged 15-19, representing a 5.0% increase compared to 20162. Although there are global protocols for school, community and primary care intervention, a better understanding of the factors associated with the prevention of suicidal behavior in adolescents is needed3.

Suicide in adolescents is associated with factors related to psychological, cognitive, personality, social, and family resources4. Suicidal risk is recognized as the set of factors that multiply the risk of committing suicide5. Empirical studies are limited in establishing the association between suicidal risk and the originally theorized dimensions of attachment, trust, commu nication, and alienation6,7,8,9. However, several studies explored the association between different dimensions theorized for attachment with specific mental disor ders such as personality disorders, eating disorders, obsessive-compulsive disorder, post-traumatic stress disorders, and self-injurious behaviors, non-suicidal self-injuries, and suicide attempts10.

In current psychology, theories of development position attachment as an innate pattern, fundamental for the survival of human beings and for the organi zation of secure basic behaviors that are consolidated in the interaction with significant persons, such as parents11. Children with safe behaviors can explore, learn to regulate stress levels, and adults will be more likely to establish safer relationships12,13. According to this model, attachment is positively established in a secure style and negatively in insecure-ambivalent and insecure-avoidant styles12.

Attachment bonds can predict in high propor tion the positive evaluation of stress experiences, the implementation of better strategies to face stressful events, well-being, psychological adjustment, and mental health in general13,14,15. Empirical studies show that insecure attachment styles -ambivalent and avoi dant- are associated with suicidal ideation5 and suicide attempts8,9.

The attachment styles assessment depends on com binations in dimensions such as trust, communication, and alienation from the parent or caregiver who repre sents these figures and even peers16. Trust appears in the relationship of understanding and respect; com munication appears related to the availability and qua lity of verbal communication; and alienation related to isolation, resentment, or hatred towards one of the figures and preferential inclination towards the other17.

Studies show divergent associations between trust, communication and alienation, and suicidal behaviors in adolescents. Some authors reported a significant association between trust and communication with parents and suicidal ideation or self-injurious suicidal or non-suicidal behaviors18,19,20, however, others found independence between these variables(20, 21).

Nature is different in forms of suicidal behavior22, therefore, the study of the attachment dimensions and the suicide risk is relevant in the face of scarce empiri cal evidence. It is also necessary to recognize this asso ciation in the Colombian context, a non-clinical ado lescent population with particular social, economic, cultural, and historical characteristics.

The objective of this research was to establish the association between attachment dimensions and high suicidal risk among school adolescents in a city in Co lombia. Trust and communication with parents were expected to be associated with low suicidal risk and alienation with high suicidal risk.

Subjects and Method

Observational cross-sectional study with analytical component23.

The population was made up of 1901 school ado lescents who attend to the official educational institu tions of the Santa Marta commune, Colombia. This commune is among the three with the highest unsatis fied basic needs index. The calculated sample was 399 participants, for a 30% expected prevalence of high suicide risk, a 5% error margin, a 5% alpha error, and 80% beta error. This sample size included a 20% repla cement for losses. Participants were randomly selected from the ninth, tenth, and eleventh grades of each ins titution. In the same way, this sample size would allow for associations and adjustments if necessary. Students who reported having a paternal or maternal figure, or someone else playing the role of caregiver were inclu ded.

The Plutchik5 Suicide Risk Scale adapted for Spain by Rubio, Montero, Jáuregui, Villanueva, Casado, Ma rín, & Santo Domingo was used24. This instrument has 15 items with dichotomous questions (yes or no). The scale investigates factors related to the suicide risk such as previous self-injury attempts, intensity of the current ideation, depression, hopelessness, and others. The suicidal risk would indicate the presence of psycho social factors that can precipitate suicidal behavior in adolescents. The scale has internal consistency between 0.75 and 0.89 in different studies25. Plutchik suicide risk scale in this study presented Kuder Richardson’s reliability of 0.77.

For the attachment dimensions measurement, we used the Inventory of Parents and Peers Attachment (IPPA) by Armsden & Greenberg15 adapted for Co lombia by Pardo, Pineda, Carrillo & Castro14. This ins trument was designed to assess the quality of percei ved attachment with father-mother-peers in 25 items for each attachment figure, with Likert-scale respon se options graded from one (never or seldom) to five (usually or always). Subscales for father and mother were used in this study. Each subscale is composed of three dimensions: trust, communication, and aliena tion. For attachment to the mother, there was internal consistency in trust between 0.80 and 0.87, communication between 0.81 and 0.84, and alienation between 0.54 and 0.60. For attachment to the father, there was internal consistency in trust between 0.91 and 0.93, communication between 0.88 and 0.99, and alienation between 0.68 and 0.7914,15,26. In this work, Cronbach’s alpha scale showed for the mother subscale trust 0.84, communication 0.77, and alienation 0.44; and for the father subscale trust 0.89, communication 0.82, and alienation 0.59.

The research project was approved by an ethics committee and qualified as a minimum risk to the health of the participants. Subsequently, authorization was requested from the Santa Marta school board (Co lombia) and informed consent was sent to the parents. The adolescents who submitted the signed consent form and confirmed their willingness to participate in the assent filled out the booklet in the classrooms. The research adopted the ethical considerations set forth in the Declaration of Helsinki of the World Medical Association27 and Resolution 8430 of the Ministry of Health of Colombia28.

The information was collected between October and November 2016 in the classrooms of the educa tional institutions of Santa Marta, Colombia, by a pro fessional in psychology and with the help of teachers. Each collection process lasted between 45 and 60 mi nutes.

Variables were dichotomized for descriptive analy sis. Odds Ratio (OR) with 95% confidence intervals (95%CI) were then observed, and an adjustment analy sis of associations was subsequently performed using a logistic regression model, as recommended by Greenland29. At the final adjustment, the Hosmer-Lemeshow test was calculated in order to know the goodness of fit of the model30. Adjusted OR probabilities showing 95%CI that did not include the unit were accepted as significant associations. The analysis was performed using the IBM-SPSS version 22 statistical software.


Final sample of 339 adolescents, 42.2% was male and 57.8% female (M = 16.3; SD = 1.4). 15% loss was obtained due to non-attending classes on the applica tion day, desire for non-participation, withdrawal at any time from the study, and lack of informed consent signed by parent or guardian. (Table 1) shows the des criptive analysis of sociodemographic variables, atta chment dimensions, and suicidal risk.

Table 1 Sociodemographic characteristics. 

(Table 2) shows suicidal risk scores for school adoles cents in Santa Marta, Colombia.

Table 2 Suicidal Risk Score in School Adolescents. 

The bivariate analysis between the dimensions of attachment to the mother and suicidal risk indicated that in adolescents with high suicidal risk is often low trust, low communication with mother, as well as high alienation with mother. These associations were main tained in a significant range by adjusting for gender and victimization. Regarding the dimensions of atta chment to the father, it was recorded in the bivariate analysis that adolescents with high suicidal risk presen ted low trust, communication, and alienation with the father. (Table 3) shows that when adjusting for gender and victimization, only the association with trust and communication remained significant.

Table 3 Crude and adjusted association between attachment to mother and father and suicidal risk in Colombian adolescents. 


In this study with school adolescents in Santa Mar ta, Colombia, it is observed that low trust and com munication with the mother and father, adjusted for gender and victimization experiences, are associated with high suicidal risk.

In this investigation, a statistically significant asso ciation was observed between trust and communica tion with parents and high suicidal risk. These findings are consistent with previous research. In Mexico, Pérez-Amezcua et al18, in adolescents aged 14 to 19 years, found a significant relationship between poor trust/ communication with parents and suicidal ideation (OR = 1.30, 95%CI 1.13-1.50) and suicide attempts (OR = 1.54, 95%CI 1.19-1.99). In another study in Bel gium, Gandhi et al19, in school adolescents between 11 and 19 years of age, showed that trust in the mother is associated with minor self-injuries mediated by iden tity synthesis (P = -0.22; 95%CI 0.42-0.08), and iden tity confusion (P = -0.17; 95%CI 0.34-0.03). Likewise, maternal alienation was associated with self-injuries mediated by identity synthesis (P = 0.28; 95%CI 0.10 0.55) and identity confusion (P = 0.33; 95%CI 0.08 0.63).

However, Nrugham et al20, in Norwegian school adolescents with several depressive symptoms reported that trust with mother (OR = 0.99, 95%CI 0.95-1.04), communication with mother (OR = 1.00, 95%CI 0.90 1.05), trust with father (OR = 1.01, 95%CI 0.90-1.06), and communication with father (OR = 0.98, 95%CI 0.94-1.03) were not associated with suicidal beha vior. On the other hand, Turner et al21 in a sample of university students between 17 and 54 years of Cau casian and Asian origin, showed a lack of association between self-injurious behaviors and trust with pa rents (OR=0.99; 95%CI 0.92-1.06), communication with parents (OR = 1.00, 95%CI 0.94-1.07) in Asians, and trust with parents (OR = 0.10, 95%CI 0.99-0.92), communication with parents (OR = 0.88, 95%CI 1.04 0.96) in Caucasians. The study did not discriminate the association for attachment to mother and father dimensions.

In this study, the alienation dimension in both pa rents was independent of suicidal risk. This observa tion is similar to that reported by Nrugham et al20 who documented that alienation with mother (OR = 0.98, 95%CI 0.92-1.06) and father (OR = 1.00, 95%CI 0.90 1.07) were not related to suicidal behaviors. In the same direction, in Canada, Turner et al21 showed that alienation with parents (OR = 1.07; 95%CI 0.98-1.16) was not associated with self-injuries in Asians or Cau casians (OR = 1.07; 95%CI 0.98-1.17).

As noted, some attachment dimensions are asso ciated with suicidal behavior, therefore, this factor is relevant in the study of suicide risk in adolescents. The complex configuration of factors predicting suicidal behavior requires consideration of mediating variables in the absence of association4. In the same way, it is important to recognize the cultural characteristics in the attachment formation, due to the different paren ting patterns that modify trust, communication, and alienation between parents and children21.

The empirical evidence is clear in pointing out the importance of attachment in the different suicidal behaviors6,7,8,9,16. However, studies are limited from the di mensional perspective of attachment and association with high suicidal risk. To date, studies have focused on self-injurious thoughts and behaviors rather than on suicide risk assessment(31, 32). It may be considered that the needs of emotional attachment should be ad dressed for suicide risk prevention, given that attach ment builds cognitive schemes based on repeated ex periences. Life history with caregivers relates to future interpersonal relationships, identification of one’s ne eds, seeking support, and representation about oneself and others33.

The attachment clinical implications show the rele vance of dimensionality in suicide risk. Insecure atta chment is a vulnerability factor for manifestations such as somatization, obsessions, interpersonal sensitivity, depression, anxiety, hostility, phobias, referential idea tion, and psychoticism16. These symptoms are present in some mental disorders and are known to be an im portant risk factor for suicidal behaviour4.

In this way, if adolescents go through situations that generate a lack of trust and communication with the mother or father, the risk of suicide could be in creased due to distance responses, lack of recognition of needs, inability to seek help, negative self-unders tanding, negative perception of others, and experience of intense and intolerable negative emotions34.

This study allowed identifying the attachment di mensions associated with suicidal risk, which may su ggest with greater precision promotion and prevention actions in families and adolescents. Attachment theory is positioned as a fundamental model in the health field and represents a protective or risk factor in a pu blic health problem such as suicide6,7,8,9. The approach to suicidal risk from an attachment dimensional pers pective responds in part to challenges in the public mental health field related to the need to integrate so cial-cultural theoretical perspectives, and professional perspectives in health among physicians, psychiatrists, psychologists, and nurses to understand and imple ment actions that save lives(34, 35).

Although the study presents limitations such as the number of scales used to measure each variable and does not consider other covariates that may in tervene in the model such as depression, it allows us to approximate which aspect of attachment intervenes in the high suicidal risk in school adolescents in Co lombia. The evidence is still limited to strictly compare the results between the dimensions of attachment and suicidal risk in Colombia. This gap gives relevance to these results by generating knowledge about the con tribution of the attachment dimensions with mother and father in suicidal risk, especially in a non-clinical population belonging to households with unsatisfied basic needs.

In future research, it is recommended to consider variables with clinical population, attachment with peers, and other mediating mental health variables, and also analyze the psychometric behavior of the alie nation dimension in Colombian adolescents.

In conclusion, low trust and low communication with mother and father are risk factors for high suicidal risk in school adolescents in Santa Marta, Colombia, with unsatisfied basic needs. It is necessary to broaden the research of this association in other populations, as well as to consider other mediating factors.

Ethical Responsibilities

Human Beings and animals protection: Disclosure the authors state that the procedures were followed ac cording to the Declaration of Helsinki and the World Medical Association regarding human experimenta tion developed for the medical community.

Data confidentiality: The authors state that they have followed the protocols of their Center and Local regu lations on the publication of patient data.

Rights to privacy and informed consent: The authors have obtained the informed consent of the patients and/or subjects referred to in the article. This docu ment is in the possession of the correspondence author.

Financial Disclosure: Authors state that no economic support has been asso ciated with the present study.

Conflicts of Interest: Authors declare no conflict of interest regarding the present study.

Aknowledgments: To Colciencias and the Magdalena Government for funding - Call 672/ Magdalena Royalties.


1. Varnik P. Suicide in the world. Int J Environ Res Public Health. 2012;9(3):760-71. [ Links ]

2. Montoya B. Comportamiento del suicidio en Colombia, 2017. Forensis. 2018;19(1):349-83. [ Links ]

3. Calear A, Christensen H, Freeman A, et al. systematic review of psychosocial suicide prevention interventions for youth. Eur Child Adolesc Psychiatry. 2016;25(5):467-82. [ Links ]

4. Franklin J, Ribeiro J, Fox K, et al. Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychol Bull. 2017;143(2):187-232. [ Links ]

5. Plutchik R, Van Praag H, Conte H, Picard S. Correlates of suicide and violent risk: The suicide risk measure. Compr Psychiatry. 1989;30(4):296-302. [ Links ]

6. Ozouni R, Valizadehb S, Nikam M. The relationship between attachment styles and suicide ideation: the study of Turkmen students, Iran. Procedia. 2010;5:1190-4. [ Links ]

7. Palitsky D, Mota N, Afifi T, Downs A, Sareen J. The association between adult attachment style, mental disorders, and suicidality: findings from a population-based study. J Nerv Ment Dis. 2013;201(7):579-86. [ Links ]

8. Sheftall A, Mathias C, Furr R., Dougherty D. Adolescent attachment security, family functioning, and suicide attempts. Attach Human Dev. 2013;15(4):368-83. [ Links ]

9. Glazebrook K, Townsend E, Sayal K. The role of attachment style in predicting repetition of adolescent self-harm: A longitudinal study. Suicide Life Threat Behav. 2015;45(6):664-78. [ Links ]

10. Mikulincer M, Shaver PR. An attachment perspective on psychopathology. World Psychiatry. 2012;11(1):11-5. [ Links ]

11. Bowlby J. Attachment and loss. New York: Basic books; 1980. [ Links ]

12. Sroufe L. Attachment and development: A prospective, longitudinal study from birth to adulthood. Attach Human Dev. 2003;7(4):349-67. [ Links ]

13. Pietromonaco P, Powers S. Attachment and health-related physiological stress processes. Cur Op Psychol. 2015;1(1):34-9. [ Links ]

14. González R, Ysern L, Pallás M, Marqués C, Martín M. Relaciones entre psicopatología y apego en la adolescencia. Rev Iberoam Diagn Eval. 2010;29(1):9-26. [ Links ]

15. Moral M, Quintana S. Ideaciones suicidas en adolescentes, relaciones paternofiliales y apego a los iguales. Rev Int Psicol Ter Psicol. 2018;18(2):163-77. [ Links ]

16. Pardo M, Pineda S, Carrillo S, Castro J. Análisis psicométrico del inventario de apego con padres y pares en una muestra de adolescentes colombianos. Interam J Psychol. 2006;40(3):289-302. [ Links ]

17. Armsden G, Greenberg M. The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in adolescence. J Youth Adolesc. 1987;16(5):427-54. [ Links ]

18. Pérez-Amezcua B, Rivera-Rivera L, Atienzo E, de Castro F, Leyva-López A, Chávez-Ayala R. Prevalencia y factores asociados a la ideación e intento suicida en adolescentes de educación media superior de la República Mexicana. Salud Publica Mex. 2010;52(4):324-33. [ Links ]

19. Gandhi A, Claes L, Bosmans G, et al. Non-suicidal self-injury and adolescent’s attachment with peers and mother: The mediating role of identity synthesis and confusion. J Child Fam Stud. 2016;25(6):1735-45. [ Links ]

20. Nrugham L, Larson B, Sund M. Predictors of suicidal acts across adolescence: influences of familial, peer and individual factors. J Affect Disord. 2008;109(1-2):35-45. [ Links ]

21. Turner B, Arya S, Chapman A. Nonsuicidal self-injury in Asian versus Caucasian university students: Who, how, and why?. Suicide Life Threat Behav. 2015;45(2):199-216. [ Links ]

22. Klonsky D, May A. Differentiating Suicide Attempters from Suicide Ideators: A Critical Frontier for suicidology research. Suicide Life Threat Behav. 2013;44(1):1-5. [ Links ]

23. Londoño JL. Metodología de la investigación epidemiológica. Bogotá: Manual Moderno; 2004. [ Links ]

24. Rubio G, Montero I, Jáuregui J, Villanueva R, Casado M, Marín J, Santo Domingo J. Validación de la escala de riesgo suicida de Plutchik en población española. Arch Neurobiol. 1998;61(2):143-52. [ Links ]

25. Aristizábal C, Cañón S, Castaño J, et al. Riesgo suicida y factores asociados en instituciones de rehabilitación para adictos a las drogas en la ciudad de Manizales (Colombia), 2012. Arch Med. 2013;13(1):11-23. [ Links ]

26. Delgado L, Penelo E, Fornieles A, Brun- Gasca C, Ollé M. Estructura factorial y consistencia interna de la versión española del Inventario de Apego a Padres y Pares para Adolescentes (IPPA). Univ Psychol. 2016;15(1):327-38. [ Links ]

27. Declaración de Helsinki. Disponible en: Disponible en: . Fecha de acceso: marzo 29 de 2005. [ Links ]

28. Ministerio de Salud. Resolución 008430 por la cual se establecen las normas científicas, técnicas y administrativas para la investigación en salud. Santafé de Bogotá; 1993. [ Links ]

29. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health. 1989;79(3):340-9. [ Links ]

30. Hosmer D, Taber S, Lemeshow S. The importance of assessing the fit of logistic regression models: a case study. Am J Public Health. 1991;81(12):1630-5. [ Links ]

31. Bureau JF, Martin J, Freynet N, Poirier A, Lafontaine M, Cloutier P. Perceived dimensions of parenting and non-suicidal self-injury in young adults. J Youth Adolesc. 2010;39(5):484-94. [ Links ]

32. West M, Spreng S, Rose S, Ada K. Relationship between attachment-felt security and history of suicidal behaviours in clinical adolescents. Can J Psychiatry. 1999; 44(6):578-82. [ Links ]

33. Grimalt L, Heresi E. Estilos de apego y representaciones maternas durante el embarazo. Rev Chil Pediatr. 2012;83(3):239-46. [ Links ]

34. Nanjappa S, Chambers S, Marcenes W, Richards D, Freeman R. A theory led narrative review of one-to-one health interventions: the influence of attachment style and client-provider relationship on client adherence. Health Educ Res. 2014;29(5):740-54. [ Links ]

35. Wahlbeck K. Public mental health: the time is ripe for translation of evidence into practice. World Psychiatry. 2015;14(1):36-42. [ Links ]

Received: November 19, 2018; Accepted: March 11, 2019

Correspondence: Yuly Suárez Colorado. E-mail:

Creative Commons License Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons