Introduction

Rates of adolescent deliberate self-harm have increased throughout the world in recent decades. The majority of studies have focused on adolescents referred to the Accident and Emergency departments after an act of deliberate self-harm, although large community studies show that the majority of self-harmers do not receive medical attention for their behaviour [4, 7, 9, 21, 26]. There is evidence of a difference between the method of self-harm used by young people admitted to hospital following deliberate self-harm and self-harmers who do not come to the attention of health care services: most hospital-admitted self-harmers use self-poisoning while self-cutting infrequently leads to hospital presentation [19, 26]. Rodham and colleagues showed that self-cutters think about self-harming for a shorter period than self-poisoners before acting on the thoughts. The authors suggested that taking an overdose requires more time and planning, may indicate more serious intent, and as a consequence is more likely to need medical attention [21]. Despite the fact that female adolescents report more episodes of deliberate self-harm than males, there is no gender difference in the proportion of self-harmers who present to hospital [9].

In addition to premeditation and seriousness of intent, several authors have addressed the reasons or motives for adolescent deliberate self-harm. When adolescents are provided with a list of potential motives for self-harm, the most frequently reported motives are ‘to die’, ‘to escape’ and ‘to obtain relief’, while more externally directed motives (e.g. ‘to change someone else’s behaviour’) are less frequently reported [3, 8, 16]. Several studies have examined intentions, motives and/or reasons for engaging in suicidal behaviour and found underlying dimensions [10–13, 18, 22, 23]. Using the reasons for attempting suicide questionnaire (RASQ), which includes 14 items, Holden and colleagues identified a two-dimensional motivational model (internal perturbation-based reasons and extrapunitive/manipulative motivations) [11–13]. Skögman and öjehagen [23], Hjelmeland et al. [10] Schnyder et al. [22] and Michel et al. [18] used the 14-items of the motives for parasuicide questionnaire (MPQ)—a scale from the European Parasuicide Study Interview Schedule [15]—and concluded that at least two different dimensions can be found. These studies all found evidence for one dimension of intrapersonal reasons (represented by ‘final and temporarily exit’, ‘personal motives’, or ‘escape reasons’), but disagreed on other dimensions. Some studies suggested interpersonal reasons, while others pointed at more manipulative, communicative, attention or care-seeking dimensions. Thus, most studies agree on a dimension of intrapersonal reasons, which include: ‘I wanted to die’, ‘I wanted to get relief from a terrible state of mind’, ‘I wanted to punish myself’, ‘I wanted to escape for a while from an impossible situation’, ‘It seemed that I lost control over myself and I do not know why I did it’, ‘I wanted to escape from a situation that was so unbearable, that I could not think of any other alternative’, ‘I am a failure’, and ‘I wanted to sleep for a while’.

In this study we examined the self-reported reasons for deliberate self-harm using data from the child and adolescent self-harm in Europe (CASE) study [17]. Among the 30,477 adolescents, 11.1% (n = 3,293) reported at least one episode of self-harm in their lifetime of which 8.8% (n = 2,548) had carried out acts meeting the study criteria for deliberate self-harm (abbreviation not used earlier). Females (13.4%) were more likely than males (4.3%) to report episodes of deliberate self-harm meeting the study criteria.

In accordance with the cry of pain hypothesis [20, 25], suicidal behaviour is conceptualised as the response (‘cry’) to a situation that has three components: defeat, no escape and no rescue. In this study the dimensions underlying the reasons adolescents give for deliberate self-harm were also analysed. The hypothesis that one of these dimensions can be associated with the cry of pain model was also tested. Another dimension is expected to be more interpersonal, communicative and help seeking.

Method

30,477 school pupils (51.3% males and 48.7% females) aged 14–17 years within the framework of the child and adolescent self-harm in Europe (CASE) Study in 2000–2001 completed the Lifestyle and Coping Questionnaire. The anonymous questionnaire was administered in a classroom setting during a single lesson lasting 30 min followed by discussion/debriefing. Seven countries took part in the study, including six European countries (Belgium, England, Hungary, Ireland, the Netherlands and Norway) and Australia. The study was approved by relevant research ethics committees and school authorities in the participating countries. At international level, the study was coordinated by the National Children’s Bureau in London, United Kingdom.

One part of the questionnaire focused on self-harm behaviour and included the following questions: ‘Have you ever deliberately taken an overdose (e.g. of pills or other medication) or tried to harm yourself in some other way (such as cut yourself)?’ Related to the last episode of deliberate self-harm an open-ended question was presented: ‘describe what you did to yourself on that occasion. Please give as much details as you can—for example, the name of the drug taken in an overdose’. Strict criteria for classification of deliberate self-harm were developed for this study, and responses to these two questions were taken into account in determining whether or not deliberate self-harm had occurred.

The criteria for self-harm were [9]:

An act with a non-fatal outcome in which an individual deliberately did one or more of the following:

  • Initiated behaviour (for example, self-cutting, jumping from a height), which they intended to cause self-harm.

  • Ingested a substance in excess of the prescribed or generally recognised therapeutic dose.

  • Ingested a recreational or illicit drug that was an act that the person regarded as self-harm.

  • Ingested a non-ingestible substance or object.

A manual was developed to aid decision making about whether a description of self-harm met these criteria [6 or available from authors on request].

Each team’s decisions were also reviewed by colleagues at the National Children’s Bureau to ensure consistency across centres.

Eight possible reasons for self-harm were offered: ‘I wanted to show how desperate I was feeling’, ‘I wanted to die’, ‘I wanted to punish myself’, ‘I wanted to frighten someone’, ‘I wanted to get my own back on someone’, ‘I wanted to get relief from a terrible state of mind’, ‘I wanted to find out whether someone really loved me’, and ‘I wanted to get some attention’. The youngsters were asked to indicate the reasons that applied to their self-harm behaviour by the response categories ‘yes’ or ‘no’. They could choose as many as they wished.

The statistical analyses were conducted using SPSS for Windows (version 12.0). Chi-square analyses and a principal component analysis were applied. The dataset was weighted by age for 14 and 15 year olds, and for 16 and 17 year olds, to prevent confounding by differing age profiles in the national samples. Within each age group and for each country, gender was equally distributed and therefore gender was not taken into account in age standardisation. The results on deliberate self-harm are based on the pupils that had carried out self-harm acts that met the study criteria.

Results

Analyses regarding the motives for deliberate self-harm at any time (i.e. past month/past year/lifetime) show that ‘I wanted to get relief from a terrible state of mind’ was most commonly reported (71.5, 70.9 and 66.5%, respectively) followed by ‘I wanted to die’ (58.4, 59.0 and 56.7%, respectively). The three most commonly reported reasons for self-harm acts in the past month/past year/lifetime were similar for males and females. The least commonly reported motives were: ‘I wanted to frighten someone’, ‘I wanted to get my own back on someone’, and ‘I wanted to get some attention’. Separate analyses regarding the reasons for the last episode of self-harm when it occurred in the past month showed that male youngsters more often reported that they wanted to ‘frighten someone’ with their act while females more often wanted to ‘get relief from a terrible state of mind’. The majority of adolescent respondents (n = 349; 81.0%) reported more than one reason for carrying out their episode of deliberate self-harm in the past month (Table 1).

Table 1 Reasons given for last episode of self-harm meeting the study criteria in the past month, by gender
Full size table

Significant gender differences were also found for the reasons which adolescents reported for their last episode of deliberate self-harm during the past year (Table 2). Females more often reported they wanted to ‘show how desperate they were feeling’, ‘to die’, ‘to punish themselves’, and ‘to get relief from a terrible state of mind’ than males.

Table 2 Reasons given for last episode of self-harm meeting the study criteria in the past year, by gender
Full size table

With regard to lifetime prevalence, females significantly more often reported the following reasons for their last episode of self-harm: ‘to show how desperate they were feeling’, ‘to die’, and ‘get relief from a terrible state of mind’ (Table 3).

Table 3 Reasons given for last episode of self-harm meeting the study criteria for lifetime history of self-harm, by gender
Full size table

Principal component analysis with varimax rotation showed that there were two underlying dimensions in the selection of the eight reasons (Table 4). Together these two factors explained 41.7% of the variance. The first factor consisted of five reasons for carrying out the last episode of self-harm (25.85% of variance explained; reliability statistic Cronbach’s alpha α = 0.641): ‘I wanted to show how desperate I was feeling’, ‘I wanted to frighten someone’, ‘I wanted to get my own back on someone’, ‘I wanted to find out whether someone really loved me’, and ‘I wanted to get some attention’. These reasons are externally directed and can be labelled as the ‘cry for help factor’. Overall, 67.5% of the self-harmers reported at least one cry for help motive. The other factor consisted of the reasons ‘I wanted to die’, ‘I wanted to punish myself’, and ‘I wanted to get relief from a terrible state of mind’. These reasons are inward directed (15.85% of variance explained; reliability statistic Cronbach’s alpha α = 0.282), which can be labelled as the ‘cry of pain factor’. At least one cry of pain reason was mentioned by 86.9% of self-harmers. Females reported more reasons in general for why they had carried out their last episode of self-harm than males (χ 2 = 89.57; df = 8; P < 0.001). The same pattern was found for the two dimensions separately, with females more often reporting at least one cry of pain motive (89.0 vs. 80.4%, χ 2 = 27.502; df = 1; P < 0.001) and at least one cry for help motive than males (70.1 vs. 59.1%, χ 2 = 21.902; df = 1; P < 0.001). For the total group there were no age differences, but examining the two factors separately, the 16–17 year olds significantly more often reported a cry for help reason for their lifetime prevalence of deliberate self-harm (χ 2 = 7.422; df = 1; P < 0.01) than the younger age group of 14–15 year. When examined more in detail, the age difference was only prevalent for the female sample (χ 2 = 12.335; df = 1; P < 0.001).

Table 4 Principal component analysis with varimax rotation (rotated component matrix) of the reasons for carrying out the last episode of self-harm (meeting the study criteria for lifetime history of self-harm)
Full size table

There was a broad level of consistency across gender and country for reasons adolescents give for their last episode of self-harm during the past year. There were however some significant differences, with ‘I wanted to show how desperate I was feeling’ and ‘I wanted to die’ being selection most frequently by the Irish males, who were equal with Norwegian males in their choice of the highest reason ‘I wanted to punish myself’. The Norwegian males also had the highest percentages reporting ‘I wanted to get relief from a terrible state of mind’ and ‘I wanted to find out whether someone really loved me’. Hungarian males had the highest percentage reporting ‘I wanted to frighten someone’ and ‘I wanted to get some attention’, while Belgian males most commonly reported ‘I wanted to get my own back on someone’. Australian males had the lowest percentages for all reasons, except ‘I wanted to find out whether someone really loved me’ and ‘I wanted to get some attention’, which was selected by even fewer of the Dutch males (Table 5).

Table 5 Reasons given for last episode of self-harm meeting the study criteria in the past year, by country, for males (percentages)
Full size table

The female sample showed a similar picture, although they generally indicated more reasons. Norwegian girls had the highest percentage reporting ‘I wanted to punish myself’, ‘I wanted to frighten someone’ and ‘I wanted to get my own back on someone’. Hungarian girls more commonly reported: ‘I wanted to die’, ‘I wanted to get relief from a terrible state of mind’ and ‘I wanted to get some attention’, but they least often chose ‘I wanted to punish myself’. ‘I wanted to show how desperate I was feeling’ was chosen most frequently by the Irish girls, while they least often indicated ‘I wanted to find out whether someone really loved me’. Belgian females most commonly reported the latter motive. Dutch girls least often chose ‘I wanted to frighten someone’, and Australian females indicated the other five reasons the least (Table 6).

Table 6 Reasons given for last episode of self-harm meeting the study criteria in the past year, by country, for females (percentages)
Full size table

Discussion

This school-based survey in seven countries aimed at investigating the reasons adolescents report for deliberate self-harm. The majority of adolescents reported more than one reason for carrying out their episode of self-harm, which is consistent with the finding from earlier studies, possibly reflecting ambivalence [8, 22]. In keeping with previous studies [3, 8, 16], the reasons most frequently reported in this survey by adolescents who deliberate self-harmed included: ‘I wanted to get relief from a terrible state of mind’ and ‘I wanted to die’. Schnyder et al. [22], Michel et al. [18] and Bancroft et al. [1] reported similar findings for deliberate self harm patients of all ages. The most infrequent reason reported by male adolescents in our survey was ‘I wanted to get some attention’, while female adolescents least often reported the reason ‘I wanted to frighten someone’. The more infrequent endorsement of the externally directed motives is in keeping with the findings of Michel et al. [18] and Hawton et al. [8]. The latter authors showed that such motives were more frequently attributed to adolescent self-harmers by clinicians.

Our results confirm the hypothesis that two underlying dimensions co-exist for the eight reasons for adolescent self-harm that were used in this study. The first factor represents the following five externally directed reasons, i.e. ‘I wanted to show how desperate I was feeling’, ‘I wanted to frighten someone’, ‘I wanted to get my own back on someone’, ‘I wanted to find out whether someone really loved me’, and ‘I wanted to get some attention’. These reasons are more interpersonal and communicative and comprise what we have referred to as a cry for help factor. The second dimension is in line with the cry of pain model, and represents the three inward directed reasons ‘I wanted to die’, ‘I wanted to punish myself’, and ‘I wanted to get relief from a terrible state of mind’. These findings are in keeping with those of previous studies in adults [11, 12, 17, 18, 22, 23] and also a similar European cross-cultural study in adults [10].

Female adolescents reported more reasons in general and on the two dimensions separately for their deliberate self-harm, which could imply that they have a higher need to explain their behaviour and communicate their distress or have a greater understanding of the complexity of motives. They more often reported both cry of pain and cry for help motives than male adolescents. Girls aged 16–17 year more frequently reported a cry for help motive than younger girls, which could be associated with a higher awareness of the interpersonal connotation of their self-destructive behaviour. However, these results are in contrast with the results from the studies of Hjelmeland et al. [10] and Skögman and öjehagen [23], in which the reasons for carrying out an act of deliberate self-harm did not vary greatly by gender or age, although these samples mostly comprised adults. Michel et al. [18] found—more in keeping with the results of our study—that younger persons more often reported the cry of pain motive ‘wanted to find relief from an unbearable state of mind’.

There was a reasonable level of consistency across countries and gender in the reasons adolescents reported for their last episode of self-harm during the past year. Apart from the top ranking of reasons that was reversed in Australian and Dutch adolescents, the order of the reasons was similar in the different countries. Only Australian youngsters had consistently lower percentages on all reasons given for their last episode of self-harm, which might be explained by a cultural difference in the extent to which they felt the necessity to provide detailed explanations for deliberate self-harm. Regarding the European countries, adolescents in the northern countries Norway, Ireland and Hungary, chose more reasons for deliberate self-harm than adolescents in the Netherlands, England and, to a lesser extent, Belgium.

Methodological limitations

This school-based survey investigated reasons for adolescent deliberate self-harm. The use of a standardised list of reasons for deliberate self-harm has both advantages and disadvantages. The use of a anonymous self-report questionnaire might be advantageous in that the individual is less driven by social desirability associated with the need to justify the behaviour and/or evoke concern and sympathy. Other advantages include that a standard is offered to compare between countries [10] and other subgroups, and that interpretation is less prone to notions of the researchers [2]. Disadvantages include the possible suggestive effect of the list as such, and as a consequence the loss of spontaneous answers, and the conception that the list is exhaustive. Nevertheless, one could argue that the items in the questionnaire have been chosen from items commonly reported spontaneously by self-harm patients [2]. A further limitation was that there was considerable variation in how much time had elapsed between the time of the survey and of the deliberate self-harm episodes. Respondents may have rationalized their behaviour over time, may have interpreted their behaviour differently after treatment, or may have had distorted memory recollections about the cognitions preceding the act [21]. Self-report biases were minimised by the anonymity offered by our questionnaire. Strengths of this study include a comparable methodology, adopted in each of the countries, along with rigorous criteria for the assessment of self-harm.

Clinical implications

The findings of this study show that adolescents who deliberately self-harm often report both cry of pain and cry for help motives. The majority of youngsters wanted to get relief from a terrible state of mind and/or wanted to die with their act of self-harm. Although the study shows that there is also a cry for help, this type of motive seems to be less prominent than the cry of pain, which is inconsistent with the popular notion that adolescents deliberate self-harm is ‘only’ a cry for help.

Rodham et al. [21], Boergers et al. [3], and Hawton et al. [8] state that many adolescents seem to engage in deliberate self-harm in order to cope with distress. Bancroft et al. [2] suggested that the ‘wish to die’ reason is used by a proportion of (adult) patients as a means of gaining social acceptability after the event, and serves to obtain sympathy and understanding of the pain related to the performing of the act. In accordance with the results of our study, Hawton and James [5] state that although adolescents who self-harm may claim they want to die, the motivation is often associated with an expression of distress and desire for escape from troubling situations. Michel et al. [18] found in their study that even the concept of a cry for help might be misleading, as the majority of (adult) patients reported that nobody could have prevented the event, and half of the sample reported they would not have accepted help before the suicide attempt. Many deliberate self-harm patients are not actually looking for professional help, but for a more general helping response from people in their close environment (e.g. practical/financial problems).

Prevention of deliberate self-harm in young people can be difficult due to the impulsive nature of the behaviour. Acquiring more constructive ways of dealing with distress by educational programs, creating a more communicative relationship with significant others, and active listening might be helpful in reducing the high prevalence of the self-harming behaviour. Educational programs could address the opportunity to communicate with teachers and/or peers about self-harming thoughts if such programs are embedded in a more global mental health framework. As Hawton and Rodham [6] indicate, a successful mental health program should include a focus on addressing and managing the difficulties faced by young people and equipping them with the skills to cope. Peers need help in recognising problems in their friends and school staff should become more aware of the difficulties and pressures youngsters face as well as the resources that are available to them. Peer helping training, skills training, and gatekeeper training are all programs focused on increasing awareness of suicidal behaviour in order to facilitate self-disclosure and prepare teenagers to identify at-risk peers and take responsible action [14].

Understanding the reason for why the deliberate self-harm occurred is important to provide the most appropriate treatment, and could prevent a future episode of deliberate self-harm. Holden et al. [13] and Boergers et al. [3] showed that patients who attribute internal perturbation-based reasons for their non-fatal suicidal behaviour (closely related to our cry of pain factor) have an increased suicide risk. In addition, Varadaraj et al. [24] have shown that one goal of therapy may be to increase family awareness of the patient’s internal pain, because significant others are often unaware of the intropunitive reasons a person may have for deliberate self-harm.

Research implications

The findings from this study add substantially to our knowledge of reasons for self-harming behaviours among adolescents. However, this study also raises questions about motives and intentions, which should be addressed in future research. It is unclear to what extent adolescents who stated ‘wanting to die’ as a reason for self-harm actually believed that death might have resulted from their actions. A measure of lethality, which would have been helpful in assessing the relative physical danger of the acts of deliberate self-harm, was not included in this study. While the questions about deliberate self-harm in the questionnaire used for this study indicated fairly clearly the type of behaviour that was being investigated, adolescents who report such behaviour might be interviewed to determine whether they definitely regarded this behaviour as self-harm. For example, intoxication may have been the reason for deliberate self-harm in some of the adolescents. Future studies should also include a perspective of time, including information about the time between the last act of self-harm and completion of the questionnaire, as reasons for acts of self-harm may change over time.