Abstract
Objective:
By analysing the suicide notes from suicide victims 15 years and younger, we attempt to gain insight into the process leading to youth suicide and explore the characteristics of the note writers.
Method:
During a psychological autopsy on 42 youths, we were entrusted with 23 suicide notes in which we explored the themes and compared the note writers to the non-note writers.
Results:
The key messages were as follows: explain the reason for suicide, declare love and give instructions. The note writers had showed suicidal behaviour more frequently than non-note writers.
Conclusion:
In the notes from youth suicide victims they present themselves as fully responsible and without confusion or overwhelming despair. The notes are likely equally informative as the notes of older victims.
Introduction
There is no single answer to why someone takes his life, regardless of his age. This was Leenaars’ central argument in his paper in which he called suicide “an intrapsychic drama on an interpersonal stage” (Leenaars, 1996). He stated that, despite great similarities, there are differences in suicide across the life span, and that young adults appear to have more interpersonal difficulties compared to older adults. However, he did not address suicides in children and adolescents. In 2004, suicide represented the third leading cause of death in children 14 years and younger in the United States (Dervic, Brent, & Oquendo, 2008; Kloos, Collins, Weller, & Weller, 2007), and in Norway in 2011, suicide was the second leading cause of death in this age group, equal to accidents and behind tumours (www.ssb.no). To understand the act of suicide, information is normally collected through interviews with one or more proxy respondents of the deceased, by the method called psychological autopsy (Brent, Perper, Kolko, & Zelenak, 1988). To obtain a first-hand understanding of the motives behind the act of suicide and the thoughts of the suicidal mind, the unsolicited written or recorded communication by the victim himself is the best material available. In the majority of cases, all other information will be second-hand. According to the Random House Dictionary (www.randomhouse.com/), a suicide note is defined as “a message that states the author has died by (or plans to die by) suicide, and left to be discovered and read in anticipation of suicide”. The incidence of leaving notes is reported to be 3–38% among adult suicide victims (Foster, 2003; Goren, Gurkan, Tirasci, & Ozen, 2003; Ho, Yip, Chiu, & Halliday, 1998) and approximately 10–45% among child and adolescent suicide victims (Posener, LaHaye, & Cheifetz, 1989; Shaffer, 1974). Groholt et al. found that 14% of suicide victims under 15 years of age and 39% of those between 15 and 19 years of age left a note (Groholt, Ekeberg, Wichstrom, & Haldorsen, 1998). We found surprisingly few studies that dealt with the contents of suicide notes from children and adolescents (Hokans & Lester, 2007; Leenaars, Wilde, Wenckstern, & Kral, 2001; Posener et al., 1989). Although some studies included a small number of suicide notes written by younger persons, these were not analysed separately (Bhatia, Verma, & Murty, 2006; Olson, Wahab, Thompson, & Durrant, 2011). This paper will add to our understanding of the pathology of suicide through an examination of suicide notes from victims 15 years and younger.
Using the data from the psychological autopsy study of suicide victims 15 years and younger (n = 42), conducted in 2007–2009 in which 18 (43%) had left a note directly connected to their suicide, we wanted to (1) assess if there were differences between the note writers and the non-note writers and what the differences consisted of, and (2) describe and examine the content of the notes and comment on their messages. Our hypothesis is that suicide notes written by children and young adolescents have many similarities to the notes of older adolescents and young adults, but, due to the younger age and immaturity of their authors, present more confusion. The prevalence of mental disorders is lower among the youngest suicide victims (Beautrais, 2001; Freuchen, Kjelsberg, Lundervold, & Groholt, 2012; Groholt et al., 1998; Shaffer, 1974); hence, their notes will most likely present fewer signs of inner pain.
Method
Sample
Of the 91 residents 15 years and younger whose death was classified as suicide in Norway from 1993 to 2004, 42 (46%) had parents that agreed to be interviewed. The interview procedure is described in a previous paper (Freuchen et al., 2012). We were made aware of 18 youths who had left 23 notes, all of which were perceived as suicide notes. We defined suicide message as any written note, essay or voice message directly connected to the suicide in time and/or content. For those who had written more than one note, we used the word count of the longest note.
The only information we received on the non-participants was their age (mean age was 13.8) and the method of suicide, which were not significantly different from age and method of the participants. For the age group 10–14 year-olds, there was no significant gender difference between the participants and the non-participants comparing with official statistics.
The data collecting period was 2007–2009.
Measurements
The assessment of psychiatric diagnoses, in accordance with the Diagnostic and Statistical Manual of Mental Disorders 4th edition (DSM-IV, American Psychiatric Association, 1994), was made by the first author using all available information, including interviews with parents with The Schedule for Affective Disorders and Schizophrenia for School Aged Children (6–18 years): Present and Lifetime Version (The Kiddie-SADS-PL) (Kaufman et al., 1997). Depressive symptoms that failed to meet the full criteria for depression were noted as sub-threshold depression, in accordance with the definition given by Fergusson (Fergusson, Horwood, Ridder, & Beautrais, 2005). In the original autopsy study, which included the 42 suicide victims presented in this paper (Freuchen et al., 2012), another psychiatrist made the diagnoses independently for 14 randomly selected Kiddie-SADS-PL interviews, by blinded assessment, using the Statistical Package of Social Sciences version 16.0 (SPSS) for the selection. An acceptable reliability was obtained, with a kappa of 0.82.
Furthermore, the parents reported whether their child had had mental problems or not and stated if the child had been in contact with Child and Adolescent Mental Health Services (CAMHS). The parents described the child’s personality traits, and by doing this, they used adjectives that we could put into four categories (vulnerable-touchy, worried-anxious, self and socially confident, impulsive-temperamental). Each category was rated as being present or not, and the categories were not mutually exclusive. We asked if the child was openhearted towards his/her parents and whether there was open discussion about his/her difficulties, joys and sorrows (yes/no). We noted suicides in the family and in the local community in the past 2–3 years that had affected the youth emotionally. We also noted any loss experienced throughout her/his life that had an apparent emotional impact, such as the loss of a family member, friend, other significant person or a pet. In addition, we asked if there had been any recent conflict (with parents, school, peers, police or others), and we noted if the child had been bullied. To get an idea of the child’s suicidal ideations, we surveyed whether she/he had been interested in the topic of suicide (i.e., talked about it in different contexts, wanted to discuss it, read poems or books about it, etc.). Suicide threats and attempts are covered in Kiddie-SADS-PL. We further noted if the child had shown altered behaviour shortly preceding the suicide. Finally, we asked if a suicide note was found, defined as any written or recorded verbal communication with content directly connected to the suicide.
We obtained the police reports in 38 (91%) cases, which provided additional information on the external circumstances of the death scene and the presence of a suicide note.
On the basis of all available information, the Suicide Intent Scale (SIS) was scored (Beck, Beck, & Kovacs, 1975). The SIS consists of two parts: part I (items 1–8) covers the objective circumstances of the suicide. It describes three main components: precautions, planning and communications (Freedenthal, 2008). All eight items were rated 0, 1 or 2, yielding a total range of 0–16. Higher scores are indicative of greater suicide intent. Part II is based on reports from individuals who have attempted suicide and is therefore not relevant for this study.
The suicide notes were read to obtain an understanding of the general content; meaningful themes were noted, as well as the sequence of these themes. We also noted the word count, what type of paper was used and how they were written (by hand or typed).
Analyses
The statistical analyses were performed using the Statistical Package of Social Sciences version 16.0 (SPSS Inc., Chicago, IL USA). Pearson chi-square or Fisher’s exact test were used for cross tabulations of categorical variables, and a t-test was used to conduct independent group comparisons for continuous variables. The significance level was set at p ≤ 0.05; not-significant is abbreviated n.s. and 95% confidence interval 95%CI. The note writers were compared with the non-note writers on selected variables. To explore a possible relation between openness and personality traits, we performed a binary logistic regression analysis, using the enter method, with openness as the dependent variable.
The content examinations were done in three steps: the notes were read through by both authors of this paper, meaningful themes were noted, and we recorded in how many notes these themes were present. The themes that occurred most frequently constituted the basis for saying something about the youths who wrote these notes and what the notes communicated.
Ethics
Required permission was obtained from the Norwegian Directorate for Health and Social Affairs, the Norwegian Social Science Data Services, the Director General of Public Prosecution, the Directory of Residents, and Statistics Norway. The Regional Committees for Medical and Health Research Ethics approved the study. We obtained parental approval to use the presented note.
Results
Suicide notes
Among the 18 (43%) children and young adolescents who left a suicide message, 11 (61%) were 14 years or younger, and 7 (39%) were 15 years. Twelve youths left one note each, one left two notes, one left five notes, and one left a voice mail. In the analyses, these notes are treated as one note per individual, as there was considerable overlap in the contents of the different notes written by the same person. In addition, there were three school essays, written by three different youths; one of these essays had been destroyed by the parents, but we were retold its contents. Essays excluded, the mean word count of the notes was 47, with a range of 2–144 (with no significant difference between the notes from those 14 years and younger compared to those from the 15-year-olds). One note was typed on a computer, the rest, and all three school essays, were hand-written. The paper that was used varied from a yellow post-it note to plain white A4 paper, with or without lines. There was one voice mail on a cell phone, of which the content was reported to us.
Each of the three essays was written 1–2 months before the suicide; two told in detail how and where the suicide would take place. The third essay, written by a 15-year-old boy, was more philosophical than the other two. In this he revealed his ambivalence, debating whether to mobilize enough strength to continue his difficult life or to choose the relief of death. The two school essays we had in print were corrected by the teacher in the usual way, concerning grammar and phrasing. None of the essays prompted the school to contact the parents or to take any other precautions. The parents were made aware of the essays only in retrospect.
Note writers versus non-note writers
We found no differences between note writers (n = 18) and non-note writers (n = 24) regarding variables related to demographic factors, circumstances of the suicide, mental health, treatment received and stressors (Table 1). However, the note writers had previously attempted suicide more often than the non-note writers, but had not threatened with suicide more often. The note writers also had shown an altered behaviour prior to suicide more often (being helpful, cuddly, silent, joyful, tense or relaxed) (Table 1).
Note leavers compared to non-note leavers by cross tabulations.
Non-note leavers n = 22.
CAMHS: Child and Adolescent Mental Health Services.
There were no differences between the groups regarding experiences of suicide in the family and local community. The note writers were non-significantly less openhearted towards their parents (p = .07) and did not differ from the non-note writers in regard to personality traits. We found no relationship between openheartedness and personality traits in either group.
We had insufficient information to score the SIS on one non-note writer (thus, n = 23). In the SIS scores presented in Table 2, item seven (suicide note) is excluded.
The note writers compared to the non-note writers on Suicide Intent Scale a by cross tabulations.
Item 7, the presence of suicide note is excluded.
Timing, note writers: n=17.
Total score 1 = item 7 suicide note, excluded; total score 2 = item 7 suicide note, included.
The SIS showed that the note writers had performed some final acts more often (given away belongings, written a will, etc.), they had done more preparations (put on certain music in the CD-player, cleared their room, one boy put on a red sweater he otherwise rarely wore but that would make it easier to find him, etc.), they had communicated their suicidal thoughts and/or plans more often than the non-note writers (mostly to friends), and they had more often contacted somebody shortly before suicide (friends, mother). Thus, the total mean score showed a significant difference between the note writers and the non-note writers (8.4 versus 6.2, p = .005). When item 7 (suicide note) was included, the total score differed even more (10.3 versus 6.2, p ≤ .001).
Notes’ contents
Whereas all written notes, essays and the voice mail were included in the analyses presented above, the voice mail was excluded in the examination of content given below.
Reading the notes thoroughly and repeatedly, they left an overall impression of being coherent, easy to read, homogenous, logical and without signs of confusion or major distress. We registered 10 meaningful themes. The frequencies in which each theme was found in the notes are given in Table 3.
The content of the suicide notes of girls compared to those of boys by cross tabulations.
No significant gender differences were found (p = .29–1.00).
The sequence in which they appeared did to a large extent follow the same order as presented in Table 3. In eight cases, the notes were addressed to someone: five were addressed to the parents, two to a named friend, and one to “dear all”. They gave reasons for the suicide: “I hit a boy at school today, and he claimed he had lost his hearing because of it”, “I shall never steal any more, that’s for sure”. They declared their love: “You mean all to me”, “I will miss you”, “I am so very fond of you”. Those who confronted their own life, that is, had a settlement with themselves, did it in this way: “I have a fucking life”, “it’s better for me to be dead”, “I’m sorry that I was mean to you”. The ones who had a settlement with others wrote: “I do this for you, dad”, “I felt like being someone everybody could pick on”. Some gave consolations and good wishes: “this hurts me, as it will hurt you”, “good luck in the future”, “you have always been kind to me”, “take care of yourselves”. A few expressed aggression: “I guess I’m too stupid for this family”, “you bastards”. The instructions in the notes were most often detailed and testament-like: “give Peter Playstation 2”, “1000 NOK for cat food to Bessie”, “send Warhammer to Jack”. One wrote the text he desired for his tomb-stone, and two told where to be found. The notes left an overall feeling of inner pain after the child had made his final decision. Six signatures were in full name, both first and family name, which left a certain formal impression, very much in contrast to the rest of the design, which was hand-written and undated. Only three (21%) of the 14 notes (essays excluded) were dated.
In terms of what the notes tell about the internal reality they represented, the suicidal mind, the motives and the emotions, we present parts of a note that is representative, both according to meaningful themes and the sequence in which they appear, well aware of the fact that a suicide note is an expression of how the deceased wants to be perceived (translation by first author): We shall call him John, 15.5 years old. His parents described him as a quiet boy, a little stubborn, and vulnerable. His uncle, with whom he had a close relation, had died by suicide 1.5 years earlier, a death he mourned. John had shown interest in the item of suicide, but he had never made a suicide attempt, as far as his parents knew. His suicide note initiated with, “I don’t know where to begin, but I start by saying I’m terribly sorry for what I have done/am about to do. Hope you can forgive me. I just couldn’t live the way I do.” The note leaves a sense of inner pain, the pain of a young boy who could not figure out how to live his life; “Well, I’m tired of being a coward and now it’s over. This life, it doesn’t work.” He declares his love, blaming no one; “I just want to say that I am incredibly fond of you all equally. You mean everything to me.” John had many friends, both boys and girls, and in his note, he favours some of them, listing to whom his money and major assets shall be given, including a sentence relating to a girlfriend who needs a computer, for which he wants to pay.
Discussion
The note writers present signs that may indicate higher suicide intent than do the non-note writers, beyond the fact that they left a note. The notes are coherent and do not reveal confusion or overwhelming emotions. The children and young adolescents emphasize their consciousness of what they are about to do and they take full responsibility. When still alive, they did not communicate to the parents the factors leading up to the suicide, and they did not seek advice or help from them to find alternative solutions. The notes show several similarities to those written by older adolescents (Leenaars t al., 2001; Lester, Haines, & Williams, 2010; Posener et al., 1989) and do not confirm our hypothesis of the young note writers being confused. In addition, our assumption about less inner pain was disproved, as inner pain was frequently present.
The number who left a suicide note was high in the present study (43%). If we omit the school essays and the voice mail, 36% still left a note. Posener, who has done one of the few studies on suicide notes by adolescents, found that only 10% of 10–19-year-old suicide victims left notes, Shaffer found 45% and Groholt 14% of those younger than 15 years (Groholt et al., 1998; Posener et al., 1989; Shaffer, 1974), compared to our findings of 42% in the same age group. Posener also found that the note writers were more likely to choose shooting as the suicide method than non-note writers, whereas we found no difference in suicide method between the two groups and that strangulation was the chosen method, both among the note writers and non-note writers. This finding may be explained not only by a lower mean age in our sample (14.0 compared to 17.9), as strangulation is the most prevalent method in the youngest age groups (Freuchen et al., 2012; Groholt et al., 1998), but also by the fact that Posener’s paper was written in 1989, and there has been a tendency towards strangulation relative to firearms as the preferred suicide method since the 1990s (Lubell, Swahn, & Crosbsy, 2004). Ho found that suicide notes tend to be left by those without previous suicidal attempts or suicidal behaviour (Ho et al., 1998), which is in contrast to our findings of significantly more suicidal behaviours among the note writers. However, this comparison is complicated, as Ho did not analyse the younger age group (≤24 years) separately on this item.
Unlike Posener et al. (1989), we found the notes to be logical, without signs of confusion or desperation. The note writers leave an impression of control and of having thought things through; this is underlined by the fact that they had also given away assets, done active preparations and, in addition, they had exhibited an altered behaviour prior to suicide. All of these actions may be indications of a higher degree of suicide intent among the note writers; a decision had been made, beyond the fact that they actually left a suicide note. Surprisingly, they had also acted to get help and communicated their suicidal ideations more often than the non-note writers, although in an ambiguous manner. It is tempting to interpret this as a sign of ambivalence: “see me, stop me!” Unfortunately, we did not succeed in finding studies of young suicide note writers supporting or contradicting these findings and interpretations.
The support from friends and parents is critical in moderating the feelings of entrapment in an unbearable situation and in preventing suicidal behaviour (Sanger & Veach, 2008), but the children and young adolescents in our study did not seek such counselling, at least not from their parents (we had no information about advice-seeking from friends). In a case-control study of suicide attempters by Groholt et al. (Groholt, Ekeberg, Wichstrom, & Haldorsen, 2000), reanalysed for children and adolescents under 16 years of age, the adolescents were asked who they would ask for support if they felt low and depressed: among the suicide attempters under 16 years, 36% would ask the parents, while 61% among the average controls would do the same. In both groups 72% would turn to friends for support. This is in contrast to the fact that they did leave a note, which per se is an act of communication (Sanger & Veach, 2008), and the notes often contained repeated declarations of their love for the parents or other persons. In accordance with Posener et al. (1989), we found that the notes were likely to be addressed to the parents. This finding may indicate a sense of belonging in a relation; they clearly see and care about their parents, knowing the suicide will hurt them. Thus, the suicide notes support Leenaars’ statement about suicide: as an intrapsychic drama, on an interpersonal stage (Leenaars, 1996). We did not have sufficient information to evaluate the parent–child relation in terms of attachment, which could have enlightened the question about why they did not ask the parents for help.
The notes presented declarations of love, often repeatedly, and mostly to the parents, and thereby the children and young adolescents both deprive the bereaved of guilt and, like Posener found, they communicate their responsibility for the act, and they do this to a greater extent than do older adolescents (Posener et al., 1989). This reassertion, declaring of love and giving of explanations, seem to be of major importance, which is also stated in the study of young adults (Bhatia et al., 2006), even though they are aware of the pain the suicide will inflict on the beloved ones. This finding leaves an impression of the children and young adolescents, well aware of what they are about to do, as responsible and attending persons, both in relation to parents and friends. This is in accordance with the findings of Darbonne, who analysed the notes of young adults (20–39 years; Darbonne, 1969). The note writers take full responsibility, and thereby indicate maturity and having thought things over. However, they write nothing about possible alternative solutions, nor do they ask for help or advice in the process, and they reveal an inability to adjust (Leenaars et al., 2001), which remind us of their young age and intellectual immaturity. Suicidal children seem to have poor reality testing, to be impulsive and to have problems with emotional and social problem solving (Pfeffer, Hurt, Peskin, & Siefker, 1995). With development of the brain during adolescence, changes occur in cognitive ability, especially in the context of overriding competing thoughts and actions (Durston & Casey, 2006). Despite the suicidal process, in which the child may tend to think that the way things are today they will stay forever, that things will never change, they do not display overwhelming despair in the notes nor in their behaviour that was visible to the parents. These children and young adolescents somehow retain their dignity; they act like decent people do, they bear their pain alone, and even manage to take care of others by leaving detailed instructions with respect to giving away their assets. This is in agreement with Ho et al. (1998) but opposed to Posener et al. (1989) who found that older adolescents left specific instructions, whereas younger adolescents were less likely to do so. Like Ho et al. (1998), we found that a minority of the notes were dated, suggesting that they were written close to the time of suicide.
One may question the representativeness of the notes and ask if the suicidal child or young adolescent, in that given state of mind, is capable of writing a full and explanatory psychological note (Leenaars et al., 2001). In the notes presented in this paper, there is little that indicates otherwise. One may also wonder if the children’s immaturity makes it difficult for them to see the suicide as the tragedy it is, when they also see it as a solution. The fact that the children and young adolescents in our study did not communicate their despair prior to suicide, at least not in a way that evoked the parents’ worry, may be more of a general attitude, as we, in a previous study, found that the majority of suicide victims 15 years and younger appeared not to differ much from their peers (Freuchen et al., 2012). Thus, further research on the quality of the relationship between children and young adolescents and their caregivers should be performed, with an attempt to reveal attitudes and qualities important for both youths and parents that could be applied in preventive work. The suicide notes of children and young adolescents have more similarities than differences with the notes written by older age groups and should thus be considered equally important.
We have little information about earlier trauma of the children in this study. Children who have experienced interpersonal trauma may develop Post Traumatic Stress Disorder (PTSD), but they may also present a vast range of symptoms (D’Andrea, Ford, Stolbach, Spinazzola, & van der Kolk, 2012). Interpersonal trauma includes problems and abuse in important interpersonal, including parents, peers and other important relationships. Generally, traumatized children have problems with affect and impulse regulation. They often show reduced self-esteem and may have false-negative recalls about themselves. Studies have shown that a secure parental attachment gives the child a kind of emotional buffer that allows him to cope with stressors appearing in life (Bowlby, 1988).
Thus, future research should look more closely at earlier trauma and at parental attachment as contributions to childhood suicide.
Limitations and strengths
The results presented in this paper are based on a small number and can only give tentative interpretations of what the contents of suicide notes tell us about suicide victims 15 years and younger. We are reluctant to generalize and draw firm conclusions. Most studies of suicide notes contain quite a large number of notes, collected from official records, from a broad age range (Chavez-Hernandez, Paramo, Leenaars, & Leenaars, 2006; Ho et al., 1998; Sanger & Veach, 2008). The notes presented here were collected during an autopsy study from a restricted age group. This can make comparisons difficult. The time span of 11 years in which our participants took their lives, hence the technological development during these years (1993–2004), may bias the way the notes are written or communicated (on computer, email, voice mail, cell phone message, etc.), but should not necessarily alter the contents. Since the period for this study (1993–2004), communications on social media (Facebook, Twitter, etc.) have developed rapidly, and items like suicidal ideations and suicide plans are often discussed among the youths. This should be noted for further research. The mean time interval of 10 years between the child’s death and the data collection represents a considerable source of error (Freuchen et al., 2012). However, this was not an issue concerning the suicide notes. As we had very little available information on the non-participants, we could not evaluate the participants’ representativeness.
We did not succeed in finding comparable studies using SIS scores on note writers/non-note writers (Table 2) or on the variables representing risk factors for suicide (Table 1).The examination of the contents was not conducted by a qualitative research method, for example, text condensation (Malterud, 2002, 2011), as this was outside the realm of this paper.
Conclusions
This is one of very few studies on suicide notes written by children and young adolescents 15 years and younger. In the notes, they communicate, mostly to their parents, some of the difficulties that made them commit suicide. They present themselves as fully responsible, without confusion or overwhelming despair, and they do not seek advice or help from their parents in the suicidal process, which could have helped them to find alternative solutions. The suicide notes of children and young adolescents have more similarities than differences with the notes written by older age groups and are thus most likely equally informative.
Footnotes
Acknowledgements
We want to thank the parents who participated and had the courage to present the notes to us.
Author contributions
B Grøholt and A Freuchen contributed to the conceptualization of the study, were involved in the data analysis, and contributed to the writing of the manuscript. A Freuchen drafted the manuscript. Both authors read and approved the final manuscript.
Funding
The Research Council of Norway and Sørlandet Sykehus HF funded the study, and the Norwegian Directorate for Health and Social Affairs gave financial support.