Abstract
Self-harm is an important public health problem in New Zealand [1]. In 1998, there were 2596 hospital discharges coded as non-fatal self-inflicted injury in New Zealand, which accounted for 6% of all injuryrelated hospitalizations in that year [2]. Studies of helpseeking following self-harm are generally based on people who have presented to medical services [3]. Little is known about help-seeking by people who self-harm and are not known to medical services.
Appleby et al. highlighted the potential value of early treatment for those who repeatedly self-harm [4]. Among those not presenting to health services, there may be a group who could be ‘caught early’, so it is important to determine their patterns of help-seeking and barriers to seeking help. A further gap in current knowledge is related to help-seeking for self-harm that falls short of the traditional definition for self-harm. Behaviours such as deliberately hitting oneself (self-battery), although often regarded as medically trivial, might repay early intervention in the hope of averting more life-threatening self-harm. Thus, it is desirable to assess help-seeking for these behaviours in representative population-based samples, not just in selected groups presenting to medical services.
One reason it is desirable to know about help-seeking among people who self-harm is because of the link between self-harm and treatable psychiatric disorder. This has been found in both hospital [5, 6] and communitybased studies [7, 8]. Many with mental health problems do not seek or receive help [9, 10]. One route to assessing and treating them is if they seek help in relation to selfharm, which may be the event that finally brings a person to the attention of medical services.
The role of family and friends in providing help has received some attention for mental health problems, in general [11, 12], but little is known about this aspect of help-seeking among those who self-harm.
Given the high risk for repetition of self-harm and suicide, by those who have made a prior suicide attempt [13], it is important to identify barriers to seeking help among those who engage in self-harm. Research into barriers to help-seeking in people with lifetime mental health problems has shown that they were often attitudinal (e.g. a feeling that the problem would get better by itself) rather than practical. Family or environmental characteristics such as receiving household public assistance have been found to predict service use particularly in urban users [10], and low socioeconomic status was found to predict help-seeking for emotional problems [14]. Help-seeking for mental health problems was found to be less likely among those with lower educational qualifications [15], but there have been conflicting findings with regards to low income [15].
The aim of the present study was to examine helpseeking behaviours not only for self-harm involving traditional methods of suicide, but also for other selfharm such as self-battery and self-biting in a representative population-based sample of young adults. The specific aims were to: (i) determine the prevalence of help-seeking for self-harm in the previous year; (ii) identify sources of help, levels of satisfaction associated with help received from each source, and correlates of help-seeking; and (iii) identify barriers and correlates of barriers to help-seeking.
Methods
Sample
The participants were aged 26 years and belong to a cohort of 1037 children born in Dunedin between 1 April 1972 and 31 March 1973 [16]. Since their birth they have completed a variety of health, behaviour, and background assessments. They were first followed-up at age 3 years, thereafter biennially until age 15 years, and later at ages 18, 21 and 26 years. The present study is based on 965 participants (471 women and 494 men, representing 95% of the surviving cohort) with complete data for self-harmful behaviours and help-seeking. The sample adequately represents all socioeconomic levels in New Zealand [17]. Seven percent of the sample self-identified as Mãori.
Measures
Participants completed a 20 minute semistructured interview on self-harm thoughts and behaviours [18]. In face-to-face interviews, the self-harm questions were asked in the context of dealing with mental or psychological pain, emotions, or stress rather than using the term ‘suicide’ initially. At the end of their interviews, all participants received a list of support services for self-harm and its related aspects. Most self-harm interviews were conducted by the self-harm research team members (including the first two authors), who have postgraduate qualifications in psychology and counselling or clinical skills. The interviewers were specially trained for the study.
Two categories of self-harmful behaviour reported in the previous year were examined. The first category, ‘ICD-self-harm’, was defined according to methods specified in International Classification of Diseases (ICD-9) E-codes 950–958 for suicide and self-inflicted injury. The second category, ‘other self-harmful behaviours’ or lesser forms of self-harm, comprised deliberately hitting oneself or putting one's fist through a wall (i.e. self-battery), denying oneself a necessity such as food as a means of punishment, exercising excessively to deliberately hurt oneself, and self-biting or wounding. We also examined intoxication by alcohol or drugs to deal with emotional pain.
For any self-harm behaviours listed above, occurring in the year prior to the interview, participants were asked if they received help for either physical consequences or other aspects (generally psychological) related to those behaviours. Affirmative responses led to further questions about formal (professional) and informal (family and/ or friends) sources of help consulted or received, satisfaction with help received from each source, and main reasons for seeking help. Those who had not sought help, but reported self-harm behaviour in the past year were asked to indicate barriers to help-seeking from a list of potential barriers [9].
Demographic measures from a separate assessment included level of educational attainment, employment, and socioeconomic status in the previous year.
Analyses
Chi-square tests or Fisher's exact tests as appropriate were used to examine associations between help-seeking and self-harm behaviours, and gender differences.
Results
Of the 965 in the sample with data on self-harm behaviours and helpseeking for the previous year, 25 (3%) reported one or more episodes of ICD-self-harm. More than half in this group (n = 15) reported an overdose and nine had deliberately cut themselves. A total of 119 (12%) reported one or more episodes of other (lesser forms of) selfharmful behaviours in the past year. These comprised deliberately hitting oneself or putting one's fist through a wall (i.e. self-battery, n = 92), denying oneself a necessity such as food as a means of punishment (n = 27), exercising excessively to deliberately hurt oneself (n = 8), and self-biting or wounding (n = 4). Of the remainder in the 602 HELP-SEEKING FOR SELF-HARMFUL BEHAVIOURS sample, 138 (14%) only reported one or more episodes of intoxication by alcohol or other drugs, specifically to deal with emotional pain. A total of 683 (71%) reported no episodes of ICD-self-harm, other selfharmful behaviours, or intoxication as a way of dealing with emotional pain in the previous year. Of the total number of episodes reported for each behaviour, suicidal intent was involved in 9% of the 213 ICD-selfharm episodes, none of the 1573 other self-harmful behaviour episodes, and 0.2% of the 2296 intoxication episodes. Similarly, an injury was a consequence of 11% of the ICD-self-harm episodes, 15% of the other self-harmful behaviour episodes, and 0.8% of the intoxication episodes.
Help-seeking for self-harm
A total of 80% of the ICD-self-harm group (7 of 10 women and 13 of 15 men) had consulted and/or received formal (professional) or informal (from family or friends) help for either physical consequences of self-harm or related aspects. This compared with 40% of the other self-harmful behaviour group (23 of 46 women and 24 of 73 men) who had sought help. The proportions of the two groups who had sought help from any source on more than one occasion were similar (75% of the ICD-self-harm group compared with 70% of the other self-harmful behaviour group). The main reasons given for seeking help were similar for both groups; emotional or psychological aspects related to self-harm (e.g. depressed mood) was a reason given by 75% (n = 15) of the ICD self-harm group compared with 62% (n = 29) of the other self-harmful behaviour group; specifically for self-harm reported by 30% (n = 6) compared with 9% (n = 4); or for an injury following self-harm reported by 25% (n = 5) compared with 6% (n = 3).
Formal help-seeking for self-harm
Formal help was defined as help received from health or other professional services. Just over half the ICD-self-harm group (n = 14, 56%) and 20% (n = 24) of the other self-harmful behaviour group had received formal help in the previous year. Overall, this represented 26% of the combined self-harm groups.
Table 1 summarizes the types of health professionals consulted by each of the two self-harm groups in the previous year. For both groups, general practitioners were the most common health professional consulted, followed by counsellors or psychologists, and emergency services. Significantly more women (59%) than men (21%) had consulted a counsellor or psychologist (χ 2 = 5.39, 1 df, p < 0.05), and significantly more men (53%) than women (18%) had consulted an emergency service (χ 2 = 4.76, 1 df, p < 0.05). A total of 7 of the 25 in the ICD-self-harm group and 6 of the 119 in the other self-harmful behaviour group (9% overall) had consulted other professional services such as social or educational services, government departments (e.g. welfare services), telephone counselling, self-help or support groups, and religious services.
Type of health services consulted in the past year by women and men who reported ICD self-harm and other self–harmful behaviours at age 26 years
Informal help-seeking for self-harm
Informal help (i.e. help received from family or friends) was reported by 68% (n = 17) of the ICD-self-harm group and 30% (n = 36) of the other self-harmful behaviour group. Combining data for both self-harm groups, similar proportions had consulted family (22%) or friends (29%). Significantly more women (32%) than men (16%) consulted family (χ 2 = 5.22, 1 df, p < 0.01). Those who reported informal helpseeking (47%) were significantly more likely than those who did not (14%) to also report formal help-seeking (i.e, from professional services) (χ 2 = 18.65, 1 df, p < 0.01).
Satisfaction with help received
Participants who sought help for self-harm in the previous year rated their satisfaction with the help received from each source on a 4-point scale of ‘very unhappy’ to ‘very happy’. Results are reported for the combined two groups of self-harm because few in each group reported seeking help from each source. General practitioners (84% or 16 of 19 who had sought help from a GP), psychiatrists (83% or 5 of 6), and psychologists or counsellors (79% or 11 of 14) were rated favourably by most who had sought help from each source. However, a lower proportion of those attending emergency services (62% or 8 of 13) rated the help they received favourably. Family were rated favourably by 91% (10 of 11 who consulted family) in the ICD-self-harm group and 95% (20 of 21) in the other self-harmful behaviour group. Similarly, friends were rated favourably by 92% (12 out of 13) and 100% (all 28) in the two self-harm groups, respectively.
Demographic characteristics of formal help-seekers for self-harm
Numbers in each group were too few to enable separate analyses to be conducted for each self-harm group. Formal help-seekers for selfharm (i.e ICD-self-harm or other self-harmful behaviours, 17 women and 21 men) were compared on demographic characteristics with those who did not seek formal help (39 women and 67 men). A higher proportion of women formal help-seekers (65%) compared to 36% of those who did not seek formal help were unemployed at the time of their interview, but this difference was not statistically significant. The groups did not differ significantly in educational achievement or socioeconomic status. Significantly more men who reported formal help-seeking (57%) were from a low socioeconomic status compared to those who had not sought formal help (28%) (χ 2 = 5.81, df = 1, p < 0.05).
Help-seeking for intoxication
A total of 138 reported intoxication by alcohol or drugs to deal with emotional pain in the previous year, but none of any of the other forms of self-harmful behaviours enquired about in the self-harm interview. About one-third (34%) of this group reported formal or informal helpseeking. Eleven percent reported formal help-seeking. Almost 1 in 10 had consulted health services; the commonest sources of help were a psychologist or counsellor (9 of 13) or a general practitioner (6 of 13).
Family was consulted by 17% and friends by 31%. Most (81%) reported seeking help for emotional or psychological reasons rather than intoxication per se or associated injury.
All who consulted a counsellor, psychologist, or general practitioner rated them favourably, as did two of the three participants who had consulted a psychiatrist. The only person who attended an emergency service rated the service unfavourably. More than 80% of the 46 who consulted family or friends rated the help they received from them favourably.
Barriers to help-seeking
Five of the 25 in the ICD-self-harm group and 72 of the 119 in the other self-harmful behaviour group reported not seeking any formal or informal help in the previous year. Given the small numbers in the ICD self-harm group, results from the analyses examining potential barriers to help-seeking are reported for the two self-harm groups combined for women and men separately. Of the 144 in the combined self-harm group, 77 (54%) had not sought any help for self-harm or related behaviours. Most (83%) considered it unnecessary to consult anyone. Attitudinal barriers (e.g. thought should be strong enough to handle problem on their own; thought problem would resolve itself; did not think anyone could help; or too embarrassed to discuss it with anyone) were reported by 39% each of the women and men (i.e. 10 and 20, respectively). Practical barriers (e.g. lack of time; unable to get to health service; could not afford to pay bill; did not know of any place to go for help; inconvenient hours, or appropriate service did not exist) were reported by 12% of the women (n = 3) and 10% (n = 5) of the men. Barriers indicative of fear or stigma (e.g. afraid of what others may think; afraid of being hospitalized; afraid of treatment; hated answering personal questions; or a member of family objected) were reported by 8% (n = 2) of the women and 6% (n = 3) of the men. About 5% in the group reported that although others thought it necessary for them to seek help, they themselves considered it unnecessary.
In the intoxication group of 138, 66% reported no formal or informal help-seeking in the previous year, mostly because they considered it unnecessary. The proportions of women (53%, n = 19) and men (38%, n = 21) who reported attitudinal barriers did not differ significantly. Practical barriers were reported by 14% (n = 5) of the women and 2% (n = 1) of the men. Fear or stigma-related barriers were reported by none of the women and 7% (n = 4) of the men.
Discussion
In this population sample of young adults, nearly half of those who reported self-harm (i.e. ICD-self-harm or other forms of self-harmful behaviours) had sought help either from professionals or from family or friends in the past year. Formal help-seeking was quite common, being reported by one in four in the self-harm group as a whole, and half of the ICD self-harm group. It was encouraging to find that many young adults who selfharmed had actively sought assistance from a variety of professional and other sources. This may have provided an opportunity for receiving advice and treatment and might even have prevented progression to more serious self-harm.
Young adults who reported self-harm, which was most often minor, were more likely to have consulted general practitioners, counsellors, and psychologists than psychiatrists or emergency services. The finding that this group of professionals was generally rated favourably supports the importance of this group in the assessment and prevention of self-harm by young adults (e.g. guidelines recently developed by the Royal New Zealand College of General Practitioners for the detection and management of young people at risk of suicide).
Medically serious self-harm is more likely to come to the attention of emergency services than general practitioners or counsellors. Thirteen young adults in our study, mostly men, reported having attended an emergency service in the past year after self-harm. It was of concern to note that nearly one-third of them, all men, rated the help they received unfavourably. No data were collected on the reasons for unfavourable ratings, however, it may be reasonable to speculate that resource and staff constraints of many emergency services and the need for triage may result in more dissatisfaction with the service received from emergency services than from general practitioners or counsellors. Attendees may be more unwell than self-harmers who present to other health services, another factor that may influence the perception of emergency service attendees’ ratings of help or support received. The small number reporting 604 HELP-SEEKING FOR SELF-HARMFUL BEHAVIOURS emergency service use suggests that the above findings should be interpreted with caution.
Studies that focus only on formal help-seeking cannot consider the contributions that informal help-seeking can make to support those who self-harm. In the present study, just over one-third in the combined self-harm group had consulted family or friends, with friends being the preferred source. Furthermore, those who reported informal help-seeking were significantly more likely to also report formal help-seeking for self-harm. These findings support those from a study of French adolescents’ help-seeking behaviours for psychiatric problems which found that the choice of a confidante was important in determining whether a mental health professional was consulted [11]. Confiding in a family member or friend may provide support and encouragement for selfharming young adults to seek formal help, if required; alternatively, young adults may have a general tendency towards help-seeking [12].
Very few in the present study who had self-harmed had sought help from other professional services such as telephone counselling or crisis centres. This seemed somewhat surprising given that services such as Lifeline and Youthline are free and their availability brought to the attention of potential consumers by a variety of media. Hawton [19] commented that research on the Samaritans in the UK had not shown a significant impact on suicide rates, possibly because clients who contact the Samaritans may be different from patients who attempt suicide.
Treating substance use problems, which are known to be strongly associated with self-harm including suicidal behaviours [20] may be a way of preventing self-harm. Our finding that about 10% in the intoxication group had sought help from general practitioners, counsellors, or psychologists, suggests that those health professionals do get opportunities to address young adults’ substance use problems. An Australian prospective study found that men diagnosed with substance misuse were at high risk for suicide, but concluded that clinicians found this particular group difficult to engage in treatment [13].
A limitation of the present study was that even with 965 participants, there were too few reporting ICD-selfharm in the previous year for barriers to help-seeking to be analyzed separately for them. For the combined self-harm group (ICD-self-harm or other self-harmful behaviours), just over half had not sought help, with the commonest reason being that it was not considered necessary. This may have been appropriate, particularly for those engaging in lesser forms of self-harm such as self-battery. Recent findings from the National Comorbidity Survey [21] indicated that most people with less severe or impairing psychopathology did not perceive a need to seek help, perhaps because they recognized the self-limiting nature of their problems. This could well have been true for some who reported self-harm in the present study. Nevertheless, one-third reported attitudinal barriers to help-seeking, and it is likely that some who were in need of help did not seek help, largely because of these attitudinal barriers rather than practical ones.
Conclusion
We agree with Pirkis et al. [22] who commented that more research is required on the patterns of help-seeking by those who engage in self-harm (including suicidal behaviour) to determine if their patterns are unusual and whether they have particular risk factors. Many in this representative sample of young adults who self-harmed did not seek help from professional or other sources. To encourage help-seeking, it may be necessary to address attitudinal barriers, raise awareness and recognition of mental health problems and promote appropriate health service use early in the life course.
Footnotes
Acknowledgements
We are grateful to the study members and their families for their support. The self-harm study was funded by Project Grant 98/148 from the Health Research Council of New Zealand (HRC) to the Injury Prevention Research Unit (IPRU) and a Community Trust of Otago Research Fellowship. We thank Paula Sowerby, John Langley, Richie Poulton, Phil Silva and DMHDS staff. The IPRU is supported by the HRC and the Accident Compensation Corporation.