Review Paper

Suicide Attempts: Prevention of Repetition

Marc S Daigle, PhD1; Louise Pouliot, PhD2; Franois Chagnon, PhD3; Brian Greenfield, MD4; Brian Mishara, PhD3
Objective: To present an overview of promising strategies to prevent repetition of suicidal behaviours. Method: This literature review on tertiary preventive interventions of suicide attempts was produced using the computerized databases PubMed and PsycINFO from January 1966 to September 2010, using French- and English-language limits and the key words: suicid* or deliberate self-harm and treatment* or therapy or intervention* or management. Results: Thirteen of the 35 included studies showed statistically significant effects of fewer repeated attempts or suicides in the experimental condition. Overall, 22 studies focused on more traditional approaches, that is, pharmacological or psychological approaches. Only 2 of the 6 pharmacological treatments proved significantly superior to a placebo a study of lithium with depression and flupenthixol with personality disorders. Eight out of 16 psychological treatments proved superior to treatment as usual or another approach: cognitive-behavioural therapy (CBT) (n = 4), (including dialectical behaviour therapy [n = 2]); psychodynamic therapy (n = 2); mixed (CBT plus psychodynamic therapy [n = 1]); and motivational approach and change in therapist (n = 1). Among the 8 studies using visit, postal, or telephone contact or green-token emergency card provision, 2 were significant: one involving telephone follow-up and the other telephone follow-up or visits. Hospitalization was not related to fewer attempts, and 1 of the 4 outreach approaches had significant results: a program involving individualized biweekly treatment. The rationale behind these single or multiple approaches still needs to be clarified. There were methodological flaws in many studies and some had very specific limited samples. Conclusions: There is a need for more research addressing the problem in definitions of outcomes and measurement of the dependent variables, gender-specific effects, and inclusion of high-risk groups. There is a need for the development and evaluation of new approaches that support collaboration with community resources and more careful assessment and comparisons of existing treatments with different populations. Can J Psychiatry. 2011;56(10):621629.

Highlights We reviewed treatments to prevent relapse of suicidal behaviours. Only 2 pharmacological treatments proved significantly superior to a placebo. Eight out of 16 psychological treatments proved superior to treatment as usual. CBT and psychoanalytically oriented therapy are promising interventions. Two visit or phone contact approaches and 1 intensive outreach program proved effective over treatment as usual.

Key Words: suicide, attempt, self-harm, relapse, prevention, treatment, aftercare, repetition

The Canadian Journal of Psychiatry, Vol 56, No 10, October 2011 W

621

Review Paper

n the general population, the lifetime prevalence of nonfatal suicidal behaviour ranges from 2% to 8%.1,2 However, intentionality, lethality, and outcomes of those suicidal behaviours vary largely. In specific populations, including Aboriginal people, young people living in youth centres, inmates, alcoholics, drug abusers, and people with mental disorders, the prevalence is much higher and represents a substantial problem at individual high-risk and population levels.3 Our paper presents current definitions of suicide attempts and deliberate self-harm and data on the risk of repetition in suicide attempters. We then critically review the differential effectiveness of various approaches to prevent repetition.

than clear ideological differences. Complications arise between study comparisons, as different researchers often use different criteria to determine when there is a relapse or re-occurrence of suicidal behaviours. Self-harm and suicide attempts can also be distinguished by criteria other than the intent to die.9 One may instead use the criteria of the lethality of the means employed and overall potential lethality of the incident. For clinicians, intent to die may appear to be the best criterion for categorizing suicidal behaviours.10 However, as it is impossible to obtain direct knowledge about a persons intentions at the time an attempt is initiated, researchers must rely on retrospective assessments involving asking attempters to identify their intentions at the time of the past event. We have no way of verifying the validity of these self-reports, and it is likely that many people may have trouble remembering what they truly intended, particularly when they were under the influence of alcohol or drugs at the time. People may also deny or try to hide their true intentions to try to avoid psychiatric hospitalization or minimize the importance of their past behaviours once they are feeling less suicidal.1115 Further, many people have been found to be ambivalent or confused when questioned about their past suicide attempts.10,16 They may reconstruct their interpretation of the experience because of pressure to respond in a socially desirable manner, leaving them to perhaps deny, minimize, or exaggerate matters. To resolve the problems involved in retrospective assessments of intent, some researchers have sought to specify the intent to die exclusively on the basis of observable behaviours surrounding the event.17 These observable behaviours correspond, incidentally, to the first section of the Beck Suicide Intent Scale, which is based only on the objective circumstances of the event.4 One way of trying to avoid the problems associated with assessing the intent to die is to categorize the lethality of the means used in the incident. The premise is that the choice of a more lethal method indicates a more genuine intent to die. Although this may appear to be logical, the use of the lethality of the means as an indication of intent to die does not stand up to careful scrutiny. Research has shown that a wide variety of other factors are associated with the choice of a method, including: gender, education, family background, ruralurban residence, knowledge of specific means, prior experiences, knowledge of examples of people using specific means, and, in particular, availability and ease of access.1822 In fact, the correlation between intent to die (measured with the Beck Suicide Intent Scale) and the medical lethality of a gesture has been shown to be negative for people with inaccurate expectations of this lethality (0.34). For the attempters with accurate knowledge and expectations regarding the likelihood of dying by these specific means, the correlation is positive but moderate (0.45), accounting for only 20% of the variance.23 A third possibility is to assess the lethality of the incident. This takes into account both the lethality of the means and
W La Revue canadienne de psychiatrie, vol 56, no 10, octobre 2011

Definitions of Suicide Attempts and Deliberate Self-Harm

Various nomenclature is used in scientific reports to refer to suicidal behaviours, including: intentional selfharm, attempted suicide, self-injury, self-mutilation, selfpoisoning, suicidal gesture, abortive suicide, simulated suicide, pseudo- suicide, subintentional suicide, lifethreatening behaviours, and parasuicide.4,5 Silverman et al6,7 proposed a nomenclature for the study of suicide and suicidal behaviours that revises the version by OCarroll et al5 adopted by the American Psychiatric Association8 in its practice guidelines for the assessment and treatment of patients with suicidal behaviours. This nomenclature establishes distinctions based on 2 elements: outcomes of the gesture (no injury, nonfatal injury, or death) and the intent to die by suicide. According to this new nomenclature,7 self-harm is defined as a self-inflicted, potentially injurious behaviour for which there is evidence (either implicit or explicit) that the person did not intend to kill himself/herself (i.e., had no intent to die).p 272 A suicide attempt is defined as a self-inflicted, potentially injurious behaviour with a nonfatal outcome for which there is evidence (either explicit or implicit) of intent to die.p 273 The goal of these authors7 is to standardize the common use of the terms. However, as indicated in our paper, currently many authors do not adequately define the terms they use. It is often impossible to understand which behaviours are included and which are excluded from their investigations. Namely, many authors cited in our review use the term deliberate self-harm, adding sometimes the idea of intentionality which is not included in the Silverman et al6,7 definition. Researchers use various different terminologies, sometimes reflecting regional habits rather Abbreviations
CBT DBT LSARS RCT cognitive-behavioural therapy dialectical behaviour therapy Lethality of Suicide Attempt Rating Scale randomized controlled trial

TAU

treatment as usual

622

Suicide Attempts: Prevention of Repetition

the circumstances surrounding the event. The lethality of the incident has been assessed using the LSARS, which ranges from 0 to 10 in even increments.14,24 There is a moderate correlation (0.67) between LSARS scores and intent to die as rated on a 6-point scale.25 The LSARS is a relatively sophisticated instrument used to quantitatively evaluate not only suicide attempts but also other suiciderelated behaviours.9,26 In the LSARS, lethality refers to the potential fatal impact of the specific act and of the means used. The evaluation of the lethality of the method used is modified by an analysis of the circumstances surrounding the gesture: how much medical intervention was required which is only one aspect of lethality27but also was the person alone, did he or she tell anyone, could he or she have expected someone to disrupt the attempt, and so on.

5. the primary targets of the intervention were patients; and 6. the design of the study was prospective.

Results

Need for Prevention of Repetition

Considering only suicide attempts, which is the main focus of our paper, repetition is common, even after hospital presentation, with most studies reporting between 12% and 25% of patients repeating and again presenting to hospital within a year.28 About one-half of attempters seen for the first time by a clinician previously attempted suicide at least once. Further, the risk of a fatal repetition of a suicide attempt is highest in the period immediately following the attempt.29 In prospective studies of suicide attempters, up to 3% will die by suicide within 1 year, 9% within 5 years, and in studies of longer duration, mortality rates are close to 11%.29 The risk of dying by suicide is 100 times higher in people who have made prior attempts, but most will not die by suicide.30 In the next section, we present a review on the efficacy of different approaches addressing the issue of repetition in suicide attempters.

eTable 1 lists the 35 studies or series of studies4382 meeting the inclusion criteria, by alphabetical order of the first authors and classified within 6 types of programs or interventions offered to the experimental group (as compared with a control group): pharmacological treatments (n = 6); psychological treatments (n = 16); visit, postal, or telephone contact (n = 5); green-token emergency card provision (n = 3); hospitalization (n = 1); and outreach programs (n = 4). Note that the van der Sande et al study65 is not included in the hospitalization section although it also tests a brief admission to a special crisis intervention unit of a university hospital. The authors65 focus mainly on the psychosocial (problem-solving) skills of the patient and, consequently, are included in the psychological treatments section. The table also indicates the duration of the programs, the longest reported follow-up periods, the characteristics of the samples, principal dependent variables (using the authors specific terminology), the source of the data (collection), and the percentages of repeaters for each condition. Note that reported rates are always for the longest follow-up period; this is a more conclusive period but, at the same time, it may hide some better results for shorter periods and it may be unfair when comparing with other shorter studies. Thirteen of these 35 studies had statistically significant results, indicating fewer repeaters or suicides in the experimental condition. Overall, 22 studies focused on more traditional approaches; that is, pharmacological or psychological.4366 Only 2 of the 6 pharmacological treatments proved significantly superior to a placebo, a study of lithium with depression and flupenthixol with personality disorders.45,46 Eight out of 16 psychological treatments proved superior to TAU or another approach. The effective psychological treatments are: CBT (n = 4), including Linehans DBT (n = 2); psychodynamic therapy (n = 2); mixed (CBT + psychodynamic) (n = 1); and, motivational approach and change in therapist (n = 1). Within the 8 studies using visit, postal, or telephone contact6773 or green-token emergency card provision (after initial regular treatment)7477 only 2 had statistically significant effects: a program involving telephone follow-up and a program with telephone contacts or visits.67,68,73 Finally, hospitalization78 was not related to fewer attempts and 182 of the 4 outreach approaches7982 had significant results, in favour of the experimental group, a program involving biweekly individualized treatment. Many of the RCTs have insufficient sample sizes, which may explain the lack of significant differences between the intervention treatment and the comparison intervention in many cases. In the same vein, the TAU received by control groups may have included some aspects of the experimental
623

Method

Although the objective was to focus on suicide attempts, as defined by Silverman et al,7 the research strategy was larger to cover the variations in the nomenclature and to take into account that many authors do not specify their definitions of a suicide attempt or use other poorly defined terminology. Our literature review on tertiary preventive interventions of suicidal attempts was produced using the computerized databases PubMed and PsycINFO from January 1966 to September 2010, using French- and English-language limits and the key words: suicid* or deliberate self-harm and treatment* or therapy or intervention* or management. In addition, other articles were identified using the reference lists of previous literature reviews on the same topic.27,3142 Studies were included if they met the 6 following criteria: 1. the intervention occurred following a suicide attempt; 2. recurrence of suicide attempt or completed suicide was among the measured dependent variables; 3. participants were included in the study shortly after their suicide attempt; 4. participants were assigned randomly to either intervention or control intervention group (RCT);
The Canadian Journal of Psychiatry, Vol 56, No 10, October 2011 W

Review Paper

treatment, with a confounding effect on the results of the trials. The authors often failed to specify what the TAU consisted of. An important limitation of these studies is that, as is general practice in RCTs, participants considered from a clinical viewpoint to be at high risk of suicide were excluded for ethical considerations. Although we do not have access to the basis of exclusion of high-risk patients from the studies, 18 studies appear to have excluded patients suffering from schizophrenia, major mood disorder, or having problems of substance abuse or dependence.4551,53,54,56,5865,73,7576,80 Owing to spatial constraints, all of these specific exclusion criteria are not included in eTable 1, but they vary from study to study and have certainly biased the final results. For example, in the study by Lauterbach et al,45 all the exclusion criteria led to retaining only 18% of the prospective sample. In such a case, it is obvious that the conclusions only apply to a very specific sample. It is also important to note that in 28 out of 35 studies that specified the gender of participants, women were overrepresented in all the significant studies (up to 100% in the Linehan et al studies59,60) and in most of the nonsignificant ones. Gender differences in treatment efficacy were almost never reported, possibly owing, in part, to the low statistical power of small sample sizes. Overall, there is limited documentation on the reliability and validity of data on the recurrence of suicide attempts. Regarding psychosocial interventions, there is little information on how therapists were selected, trained, and coached to deliver effective treatments. None of the reports clearly describe the process used to ensure implementation reliability for those interested in reproducing the treatment, and there is little focus on how others can implement the experimental interventions. The heterogeneity in the labelling of the target outcomes (dependent variables listed in eTable 1) is another important aspect that has been reviewed above. These labels may refer to different nonfatal behaviours, with different lethality and intentionality, but this was not generally taken into account in the 35 studies, except for 3: Slee et al,63 Wood et al,66 and Vaiva et al.73 This, again, makes comparison between treatments difficult. However, they were all considered suicide attempts for the purposes of our review. Only Motto and Bostrom,72 Lauterbach et al45 and Fleischmann et al68 included suicide (fatal outcome) as a dependent variable, although the samples were neither always large (167 to 1867 participants) nor the follow-up periods very long (12 to 120 months). Although our classification of 6 categories of intervention is heuristic, there are also large differences within them. The duration of the treatments presented in eTable 1 ranged from 1 day to 18 months. Further, the follow-up period was generally from 3 to 36 months. In Motto and Bostroms study,72 the intervention of sending postcards and the follow-up extend to 180 months. Outcomes were generally assessed by 2 different data collection strategies: self-reports, in which patients are asked at some point
624

during the follow-up period if they had attempted suicide, and by reviews of patients clinical records. Participants clinical and sociodemographic profiles vary greatly. For example, eTable 1 shows that some programs are for children and adolescents and others for adults; thus the mean age of participants varies between 14 and 41 years. Also, some interventions were for people with very specific problems; for example, drug abuse or personality disorders. The time of entry in the study also differs substantially between studies. In some instances, participation began in the hospital ward to which the patient was admitted following the suicide attempt, while in others, participation began days or months after discharge.

Discussion

Before presenting our general conclusions, we must discuss some general limitations to the current state of knowledge on the efficacy of treatments to prevent repetition of suicide attempts. First there are important limitations in categorizing interventions. Although interventions may share similar therapeutic approaches, the specific activities of the intervention, their intensity, and the reliability of the implementation may vary considerably between studies. Second, most studies have samples that overrepresent women and none compare the effectiveness of interventions by gender, except the study by Hawton et al.56 Several recent studies suggest that men may have the same prevalence of nonlethal suicidal behaviours in nonclinical populations.12,8385 It is not clear that promising interventions are equally effective for men, who are generally at higher risk of dying by suicide than women.86 Only one of the studies, the postcard follow-up by Motto and Bostrom,72 who used completed suicide as the outcome, addressed possible gender difference, and with sufficient power. In this case, the intervention was proactive (postal contact), as is sometimes recommended with men, and the gender ratio was almost equal. However, the results were not statistically significant, although differences were in the predicted direction. Third, most samples excluded people at high risk of attempting suicide, thus effectively eliminating from the studies people who, one would hope, could most benefit from the interventions. Thus our ability to generalize about the usefulness of many interventions to prevent recurrence in high-risk groups is limited. This reservation about the efficacy of the interventions with high-risk groups is supported by the fact that, in most instances, people who have attempted suicide present an Axis I major psychiatric disorder (that is, depression, schizophrenia, or substance disorder), often in comorbidity with a personality disorder. People with precisely those diagnoses were excluded from some of the clinical trials. Fourth, in most studies outcome assessments were based on patients self-reports. Whenever self-reports are used, social desirability can be a major issue. Suicide is a sensitive matter that participants may be reluctant to report, and none of the trials under review indicate that there was an attempt to validate the
W La Revue canadienne de psychiatrie, vol 56, no 10, octobre 2011

Suicide Attempts: Prevention of Repetition

self-report data. It is possible that the lack of findings or some of the positive findings indicating the effectiveness of some treatments in preventing recurrence, based on self-reports, may be related to this methodological pitfall. The low to moderate effectiveness of some treatments in preventing recurrence should be considered in light of the fact that suicide is a multifaceted phenomenon.As suicidal behaviour is related to multiple etiological factors, a monolithic approach to prevention may have severe limitations.87 Only a handful of trials appear to have recognized this fact by combining 2 or more intervention approaches. However, none of these studies offer an explicit statement of the rationale guiding the decision to combine, or not to combine, intervention approaches in their experimental conditions. Adding interventions without a clear rationale is no assurance that the combination has advantages over any specific components. Even with a clear rationale, it is also more difficult to ascertain if the implementation of all the components was well done. When programs that have multiple components are compared with control conditions, if there are significant effects, it is impossible to determine if the effects are due to one of the components or the combination that was studied. Also, as few studies assess less traditional intervention methods, we have less data to determine the effectiveness of new and innovative practices.

Psychological Treatments

There is a growing consensus in the literature that, in the management of suicidal behaviour, treatment interventions should address coping deficiencies and symptoms of psychological distress in patients who have attempted suicide.8 In the last 3 decades, several psychotherapies have proven to be effective in preventing the recurrence of suicide attempts. In fact, 9 of the 16 psychotherapy studies proved effective, if one includes the Davidson et al study.53 eTable 1 shows nonsignificant results for this study53 in its effect on the percentage of repeaters. However, they found that the number of repetitions (mean number of suicidal acts per participant of CBT) was significantly less than in the TAU. Several studies have found that suicide attempters have deficits in problem solving abilities.9097 Nevertheless, none of the studies whose interventions were specifically related to problem solving approaches had significant effects in decreasing repetition. The principal components of problem solving therapies involve clarification of the problem behind the attempt and exploration of alternative solutions. Several CBTs had significant effects in decreasing repetition of attempts. Four (5 including the Davidson et al study53) of the RCTs had significant results. The 2 DBT forms of CBT proved significant, but only with patients having a borderline personality disorder.59,60 These 2 last studies give no indication about the effectiveness of DBT with other diagnoses. There was also some form of DBT in the mixed approach of Wood et al66 with children and adolescents,66 but these findings were later contradicted in the studys replication by Hazell et al.57 There are fewer studies of the effects of psychodynamic approaches. However, 2 RCTs indicated significant effects.4951,55 In one study, by Bateman and Fonagy,51 the experimental group was exposed not only to an 18 months treatment but also to another 18 months of a maintenance program. Also, the study by Wood et al,66 which had significant effects, included both CBT and a psychodynamic approach. Nevertheless, the findings of our review indicate that the psychological approaches of CBT and psychodynamic therapies are promising and may prevent repetition of suicidal behaviour. As for the Mller study,61 it shows the effectiveness of a short motivational interview, but only in the context of a reference to a suicide prevention centre where there is a change of therapist. There is not much information in the report61 to help us understand which conditions are associated with the success, the change of therapist or the referral to a suicide prevention centre.

Pharmacological Treatments

In the past 30 years, only a handful of studies have tested the effectiveness of pharmacological treatments in preventing relapse of suicide attempts. In our review, only the studies by Montgomery and Montgomery46 (on flupenthixol with personality disorders) and by Lauterbach et al45 (on lithium with depression) had significant findings. In the first study,46 the sample was not large but the results were very clear: 21% repetition in the experimental group compared with 75% in the control group. This study focused on a single behavioural dependent variable, namely, a suicidal act (not defined at all by the authors), and included only patients with mainly borderline or histrionic personality disorders and not suffering from schizophrenia or depression. The study by Lauterbach et al45 has a larger sample and a longer time of treatment and follow-up. Nevertheless, the treatment was only effective in preventing suicides and not suicide attempts. Also the repetition rates were very low, both in experimental and in control conditions, as compared with the findings in the Montgomery and Montgomery study,46 which clearly reflects the profile of the participants: personality disorders compared with depressive disorders. The study with paroxetine by Verkes et al,48 which was not significant overall with patients without major depression, did have significant findings in 2 subgroups of the sample: people with fewer than 5 previous attempts and people meeting fewer than 15 criteria for cluster B personality disorder. However, one must not forget that there is abundant research showing the effectiveness of pharmacological treatments for the condition most associated with suicide attempts and completed suicides, namely, depression.88,89
The Canadian Journal of Psychiatry, Vol 56, No 10, October 2011 W

Visit, Postal, or Telephone Contact and Green-Token Card Provision

Limited financial and human resources is an important issue in many hospital settings. Therefore, brief, low-cost visits, telephone, or postal contact initiatives have been developed to address this widespread problem. Five minimal interventions have been empirically tested in various countries using RCTs. When we examine the percentage of repeaters, 2 studies produced significant results. In the
625

Review Paper

large study reported by Bertolote et al67 and Fleischmann et al,68 participants had a 1-hour information session but the intervention relied mainly on follow-up contacts by visits or phone calls. The program was proven effective, but only in reducing the number of suicides. In the Vaiva et al study,73 contacting patients by phone 1 month after discharge was associated with significant decreases in attempts; but contact after 3 months was not conclusive (although the results were in a positive direction). Nevertheless, considering only the number of readmissions for self-poisoning per person, the Carter et al study69,70 shows a significant reduction in the experimental group. The study by Motto and Bostrom72 also seems promising. Designed for a 10-year follow-up (total = 15 years), this study had significant results in the first 2 years. However, although there were still differences in suicide attempt rates in the predicted direction 5 years later, the difference between the experimental and control groups were not statistically significant at that time. At 15 years, the predicted direction of the results was the opposite, although there were nonsignificant differences between experimental and control groups. In this study,72 patients in the contact group were sent a personalized letter on a monthly basis for the first 4 months, then every second month for 8 months, and every 3 months for 4 years. Altogether, these postal or telephone approaches are attractive, especially with populations who may be reluctant to use available services; for example, very high-risk groups or men.86 Here again, a gender-specific analysis could provide further clarification of the potential effects of these methods. It should be noted also that, overall in our review, none of the other studies followed their patients for more than 2 years as in the Motto and Bostrom study, except the 5 years of the Bateman and Fonagy study,4951 and we do not know about their long-term effectiveness. Three RCTs7477 also tested the effectiveness of providing suicide attempters with a green card, or token, giving access to telephone consultation with a psychiatrist and on-demand admission to their local hospital. None of these studies found significant differences between the experimental and control groups in the proportion of those repeating suicide attempts at 6- and 12-month follow-up intervals.

but a number of intensive outreach programs have also been proposed to circumvent these problems and prevent the recurrence of suicide attempts. Our review identified 4 RCTs7982 in this area and only 1 was not intensive.81 These programs involved, to differing extents, a combination of home visits, individualized treatment plans, referrals to resources, a schedule of appointments, and compliancerelated measures. The findings from these trials failed, in most cases, to demonstrate a decrease in the repetition of suicidal behaviour. However, Welus treatment program82 appeared to be effective in reducing suicide re-attempts. It should be noted that the follow-up period (4 months) was shorter than in many other programs. Therefore, results from this study82 do not provide evidence of the programs long-term effectiveness.

Conclusion

On the whole, more traditional approaches, especially psychotherapies based on CBT, including DBT, as well as psychoanalytically oriented therapies, appear to significantly reduce repetition of suicide attempts. The minimal approach of using regular visits, postal, or telephone contacts appears to be a promising low-cost intervention that is worthy of further investigation. The green or token emergency card provision fail to have statistically significant effects. Intensive outreach follow-up approaches often use home visits, which are both intensive and personalized. One such experience82 proved efficient. One may argue that it would be best to combine all the approaches, pharmacological, psychotherapeutic, and follow-up, to have a maximum effect. However, adding active ingredients does not necessarily yield better results according to empirical findings. If interventions are combined, a clear rationale should be elaborated. Another concern is that each approach is costly, and the sum of several is even more costly, with no empirical evidence, to date, indicating greater benefits Given that there were methodological flaws in many studies and that some had very limited samples, there is still a need for more sound research. Future studies should address the problem of the definition of the dependent variables, and the determination of the reliability and validity of measures. There is a need to have sufficiently large samples to examine gender-specific effects. Further, we must obtain data on high-risk groups rather than systematically excluding them from investigations. More thought should also be put into approaches that support collaboration with community resources and new technologies, such as the Internet and text messaging,100 to explore new initiatives that may potentially prevent repeated attempts and save lives.

Hospitalization

Waterhouse and Platt78 compared the effects of general hospital admission with home discharge in preventing repetition, but both conditions had a comparable number of what they called parasuicides. However, if one only considers the rate of rehospitalization as the outcome variable, we know that outpatient services may be better than hospitalization, at least with adolescents.98

Outreach Program

A large proportion of patients who have attempted suicide discontinue their treatment and fail to attend a follow-up appointment after their discharge from hospital facilities.99 Visit, postal, or telephone contacts and greentoken emergency card provision may help in this regard,
626

Acknowledgements

Preparation of this article was supported by the team of the Centre for Research and Intervention on Suicide and Euthanasia. Thanks, especially, to Louise Sansoucy for her help.
W La Revue canadienne de psychiatrie, vol 56, no 10, octobre 2011

Suicide Attempts: Prevention of Repetition

References

1. Kerkhof A. Attempted suicide: patterns and trends. In: Hawton K, van Heeringen K, editors. The international handbook of suicide and attempted suicide. Chichester (GB): John Wiley & Sons Ltd; 2000. p 4964. 2. Nada-Raja S, Skegg K, Langley J, et al. Self-harmful behaviors in a population-based sample of young adults. Suicide Life Threat Behav. 2004;34:177186. 3. Hawton K, van Heeringen K, editors. The international handbook of suicide and attempted suicide. Chichester (GB): John Wiley & Sons Ltd; 2000. 4. Beck AT, Schuyler D, Herman I. Development of suicide intent scales. In: Beck AT, Resnick HCP, Lettieri D, editors. The prediction of suicide. Bowie (MD): Charles Press; 1974. p 4569. 5. OCarroll PW, Berman A, Maris R, et al. Beyond the Tower of Babel: a nomenclature for suicidology. Suicide Life Threat Behav. 1996;26(3):237252. 6. Silverman S, Berman AA, Sanddal N, et al. Rebuilding the Tower of Babel: a revised nomenclature for the study of suicide and suicidal behaviors. Part I: background, rationale, and methodology. Suicide Life Threat Behav. 2007;37(3):248263. 7. Silverman S, Berman AA, Sanddal N, et al. Rebuilding the Tower of Babel: a revised nomenclature for the study of suicide and suicidal behaviors. Part II: suicide-related ideations, communications and behaviors. Suicide Life Threat Behav. 2007;37(3):264277. 8. American Psychiatric Association. Practice guidelines for the assessment and treatment of patients with suicidal behaviours. Am J Psychiatry. 2003;160(Suppl):160. 9. Daigle M, Ct G. Nonfatal suicide-related behavior among inmates: testing for gender and type differences. Suicide Life Threat Behav. 2006;36(6):670681. 10. Wagner BM, Wong SA, Jobes DA. Mental health professionals determinations of adolescent suicide attempts. Suicide Life Threat Behav. 2002;32:284300. 11. Farberow N. Indirect self-destructive behavior: classification and characteristics. In: Farberow N, editor. The many faces of suicide: indirect self-destructive behavior. New York (NY): McGraw-Hill; 1980. p 1527. 12. Kjller M, Nrlev J, Davidsen M. Recall bias and suicidal behaviour in a six year follow-up study. Paper presented at the 10th European Symposium on Suicide and Suicidal Behaviour; 2004 Aug; Copenhagen, Denmark. 13. Menninger K. Man against himself. New York (NY): Harcourt, Brace, World; 1938. 14. Smith K, Conroy RW, Ehler BD. Lethality of suicide attempt rating scale. Suicide Life Threat Behav. 1984;14(4):215243. 15. Winterbotham RG. Self-reports of parasuicidal behaviours and social desirability effects in a correctional setting (thesis). [Vancouver (BC)]: Simon Fraser University; 1989. 16. Goldston DB. Conceptual issues in understanding the relationship between suicidal behavior and substance use during adolescence. Drug Alcohol Depend. 2004;76 Suppl:S79S91. 17. Denning DG, Conwell Y, King D, et al. Method choice, intent and gender in completed suicide. Suicide Life Threat Behav. 2000;30:282288. 18. Cantor CH. Suicide in the Western World. In: Hawton K, van Heeringen K, editors. The international handbook of suicide and attempted suicide. Chichester (GB): John Wiley & Sons Ltd: 2000. p 928. 19. Clarke RVG, Lester D. Suicide: closing the exits. New York (NY): Springer Verlag; 1989. 20. De Moore GM, Robertson AR. Suicide attempts by firearms and by leaping from heights: a comparative study of survivors. Am J Psychiatry. 1999;9:14251431. 21. Lester D. Why do people choose particular methods for suicide? Act Nerv Super (Praha). 1988;30(4):312314. 22. Marzuk PM, Leon AG, Tardiff K, et al. The effect of access to lethal methods of injury on suicide rates. Arch Gen Psychiatry. 1992;49(6):451458.

23. Brown GK, Henriques GR, Sosdjan D, et al. Suicide intent and accurate expectations of lethality: predictors of medical lethality of suicide attempts. J Consult Clin Psychol. 2004;72:11701174. 24. Berman AL, Shepherd G, Silverman MM. The LSARS-II: Lethality of Suicide Attempt Rating Scale-Updated. Suicide Life Threat Behav. 2003;33(3):261276. 25. Andrews JA, Lewinsohn PM. Suicidal attempts among older adolescents: prevalence and co-occurrence with psychiatric disorders. J Am Acad Child Adolesc Psychiatry. 1992;31:655662. 26. Lohner J, Konrad N. Deliberate self-harm and suicide attempt in custody: distinguishing features in male inmates self-injurious behavior. Int J Law Psychiatry. 2006;29(5):370385. 27. Linehan MM. Behavioral treatments of suicidal behaviors: definitional obfuscation and treatment outcomes. In: Maris RW, Canetto SS, McIntosh JL, et al, editors. Review of suicidology. New York (NY): The Guilford Press; 2000. p 84112. 28. Hawton K, Sinclair J. The challenge of evaluating the effectiveness of treatments for deliberate self-harm. Psychol Med. 2003;33(6):955958. 29. Sakinofsky I. Repetition of suicidal behaviour. In: Hawton K, van Heeringen K, editors. The international handbook of suicide and attempted suicide. Chichester (GB): John Wiley & Sons Ltd; 2000. p 385404. 30. Owens D, Horrocks J, House A. Fatal and non-fatal repetition of selfharm. Br J Psychiatry. 2002;181:193199. 31. Arensman E, Towsend E, Hawton K, et al. Psychosocial and pharmacological treatment of patients following deliberate self-harm: the methodological issues in evaluating effectiveness. Suicide Life Threat Behav. 2001;31(2):169180. 32. Burns J, Dudley M, Hazell P, et al. Clinical management of deliberate self-harm in young people: the need for evidencebased approaches to reduce repetition. Aust N Z J Psychiatry. 2005;39:121128. 33. Crawford MJ, Thomas O, Khan N, et al. Psychosocial interventions following self-harm: systematic review of their efficacy in preventing suicide. Br J Psychiatry. 2007;190:1117. 34. Comtois KA, Linehan MM. Psychosocial treatments of suicidal behaviours: a practice-friendly review. J Clin Psychol. 2006;62(2):161170. 35. Gould MS, Greenberg T, Velting DM, et al. Youth suicide risk and preventive interventions: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 2003;42(4):386405. 36. Hawton K, Townsend E, Arensman E, et al. Psychosocial and pharmacological treatments for deliberate self-harm. Cochrane Database Syst Rev. 1999;4:133. 37. Hawton K. Psychosocial treatments following attempted suicide: evidence to inform clinical practice. In: Hawton K, editor. Prevention and treatment of suicidal behaviour: from science to practice. Oxford (GB): Oxford University Press; 2005. p 197219. 38. Hepp U, Wittmann L, Schnyder U, et al. Psychological and psychosocial interventions after attempted suicide: an overview of treatment studies. Crisis. 2004;25(3):108117. 39. Links PS, Bergmans Y, Cook M. Psychotherapeutic interventions to prevent repeated suicidal behavior. Brief Treat Crisis Interv. 2003;3:445464. 40. Mann J, Apter A, Bertolote J, et al. Suicide prevention strategies: a systematic review. JAMA. 2005;294(16):20642074. 41. Rudd D, Joiner TE Jr, Trotter D, et al. The psychosocial treatment of suicidal behavior: a critique of what we know (and dont know). In: Kleespies PM, ediror. Behavioral emergencies: an evidencebased resource for evaluating and managing suicidal behavior, violence, and victimization. Washington (DC): American Psychological Association; 2009. p 339350. 42. Tarrier N, Taylor K, Gooding P. Cognitive-behavioral interventions to reduce suicide behavior: a systematic review and meta-analysis. Behav Modif. 2008;32(1):77108. 627

The Canadian Journal of Psychiatry, Vol 56, No 10, October 2011 W

Review Paper
43. Battaglia J, Wolff TK, Wagner-Johnson DS, et al. Structured diagnostic assessment and depot fluphenazine treatment of multiple suicide attempters in the emergency department. Int Clin Psychopharmacol. 1999;14:361372. 44. Hirsch SR, Walsh C, Draper R. Parasuiside: a review of treatment interventions. J Affect Disord. 1982;4:299311. 45. Lauterbach E, Felber W, Mller-Oerlinghausen B, et al. Adjunctive lithium treatment in the prevention of suicidal behaviour in depressive disorders: a randomized, placebo-controlled, 1-year trial. Acta Psychiatr Scand. 2008;118:469479. 46. Montgomery SA, Montgomery D. Pharmacological prevention of suicidal behaviour. J Affect Disord. 1982;4(4):291298. 47. Montgomery SA, Roy D, Montgomery DB. The prevention of recurrent suicidal acts. Br J Clin Pharmacol. 1983;15:183S188S. 48. Verkes RJ, Van der Mast RC, Hengeveld MW, et al. Reduction by paroxetine of suicidal behaviour in patients with repeated suicide attempts but not major depression. Am J Psychiatry. 1998;155(4):543547. 49. Bateman A, Fonagy P. Effectiveness of partial hospitalization in the treatment of borderline personality disorder: a randomized controlled trial. Am J Psychiatry. 1999;156:15631569. 50. Bateman A, Fonagy P. Treatment of borderline personality disorder with psychoanalytically oriented partial hospitalization: an 18-month follow-up. Am J Psychiatry. 2001;158(1):3642. 51. Bateman A, Fonagy P. 8-year follow-up of patients treated for borderline personality disorder: mentalization-based treatment versus treatment as usual. Am J Psychiatry. 2008;165:631638. 52. Brown GK, Ten Have T, Henriques GR, et al. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. 2005;294(5):563570. 53. Davidson K, Norrie J, Tyrer P, et al. The effectiveness of cognitive behaviour therapy for borderline personality disorder: results from the borderline personality disorder study of cognitive therapy (BOSCOT) trial. J Pers Disord. 2006;20(5):450465. 54. Evans K, Tyrer P, Catalan J, et al. Manual-assisted cognitivebehaviour therapy (MACT): a randomized controlled trial of a brief intervention with bibliotherapy in the treatment of recurrent deliberate self-harm. Psychol Med. 1999;29:1925. 55. Guthrie E, Kapur N, Mackway-Jones, et al. Randomised controlled trial of brief psychological intervention after deliberate selfpoisoning. Br Med J. 2001;323:15. 56. Hawton K, McKeown S, Day A, et al. Evaluation of out-patient counselling compared with general practitioner care following overdoses. Psychol Med. 1987;17:751761. 57. Hazell PL, Martin G, McGill K, et al. Group therapy for repeated deliberate self-harm in adolescents: failure of replication of a randomized trial. J Am Acad Child Adolesc Psychiatry. 2009;48(6):662670. 58. Liberman RP, Eckman T. Behavior therapy vs insight-oriented therapy for repeated suicide attempters. Arch Gen Psychiatry. 1981;38:11261130. 59. Linehan MM, Armstrong HE, Suarez A, et al. Cognitive-behavioral treatment of chronically parasuicidal borderline patients. Arch Gen Psychiatry. 1991;48:10601064. 60. Linehan MM, Comtois KA, Murray AM, et al. Two year randomised controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Psychiatry. 2006;63:7571401. 61. Mller H-J. Attempted suicide: efficacy of different aftercare strategies. Int Clin Psychopharmacol. 1992;6(Suppl 6):5869. 62. Salkovskis PM, Atha C, Storer D. Cognitive-behavioural problem solving in the treatment of patients who repeatedly attempt suicide. A controlled trial. Br J Psychiatry. 1990;157:871876. 63. Slee N, Garnefski N, van der Leeden R, et al. Cognitivebehavioural intervention for self-harm: randomised controlled trial. Br J Psychiatry. 2008;192:202211. 628 64. Tyrer P, Thompson S, Schmidt U, et al. Randomised controlled trial of brief cognitive behavior therapy versus treatment as usual in recurrent deliberate self-harm: the POPMACT study. Psychol Med. 2003;33:969976. 65. van der Sande R, Van Rooijen L, Buskens E, et al. Intensive in-patient and community intervention versus routine care after attempted suicide. Br J Psychiatry. 1997;171:3541. 66. Wood A, Trainor G, Rothwell J, et al. Randomised trial of group therapy for repeated deliberate self-harm in adolescents. J Am Acad Child Adolesc Psychiatry. 2001;40(11):12461253. 67. Bertolote JM, Fleischmann A, De Leo D, et al. Repetition of suicide attempts: data from emergency care settings in five culturally different low- and middle-income countries participating in the WHO SUPRE-MISS study. Crisis. 2010;31(4):194201. 68. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bull World Health Org. 2008;86(9):703709. 69. Carter GL, Clover K, Whyte IM, et al. Postcards from the EDge project: randomised controlled trial of an intervention using postcards to reduce repetition of hospital treated deliberate self poisoning. BMJ. 2005;331(7520):805. 70. Carter GL, Clover K, Whyte IM, et al. Postcards from the EDge: 24-month outcomes of a randomised controlled trial for hospitaltreated self-poisoning. Br J Psychiatry. 2007;191:548553. 71. Cedereke M, Monti K, Ojehagen A. Telephone contact with patients in the year after a suicide attempt: does it affect treatment attendance and outcome? A randomised controlled study. Eur Psychiatry. 2002;17(2):8291. 72. Motto JA, Bostrom AG. A randomised controlled trial of postcrisis suicide prevention. Psychiatr Serv. 2001;52(6):828833. 73. Vaiva G, Ducrocq F, Meyer P, et al. Effect of telephone contact on further suicide attempts in patients discharged from an emergency department: randomised controlled study. BMJ. 2006;332(7552):12411244. 74. Cotgrove A, Zirinsky L, Black D, et al. Secondary prevention of attempted suicide in adolescence. J Adolesc. 1995;18:569577. 75. Evans MO, Morgan HG, Hayward A, et al. Crisis telephone consultation for deliberate self-harm patients: effects on repetition. Br J Psychiatry. 1999;175:2327. 76. Evans J, Evans M, Morgan HG, et al. Crisis card following selfharm: 12-month follow-up of a randomised controlled trial. Br J Psychiatry. 2005;187:186187. 77. Morgan HG, Jones EM, Owen JH. Secondary prevention of non-fatal deliberate self-harm. The green card study. Br J Psychiatry. 1993;163:111112. 78. Waterhouse J, Platt S. General hospital admission in the management of parasuicide: a randomised controlled trial. Br J Psychiatry. 1990;156:236242. 79. Allard R, Marshall M, Plante MC. Intensive follow-up does not decrease the risk of repeat suicide attempts. Suicide Life Threat Behav. 1992;22:303314. 80. Harrington R, Kerfoot M, Dyer E, et al. Randomised trial of a home-based family intervention for children who have deliberately poisoned themselves. J Am Acad Child Adolesc Psychiatry. 1998;37(5):512518. 81. Van Heeringen C, Jannes S, Buylaert W, et al. The management of non-compliance with referral to out-patient after-care among attempted suicide patients: a controlled intervention study. Psychol Med. 1995;25:963970. 82. Welu T. A follow-up program for suicide attempters: evaluation of effectiveness. Suicide Life Threat Behav. 1977;7(1):1730. 83. Boyer R, St-Laurent D, Prville M, et al. Ides suicidaires et parasuicides [Suicidal and parasuicidal ideation]. In: Daveluy C, Pica L, Audet N, et al, editors. Enqute sociale et de sant 1998. Qubec (QC): Institut de la statistique du Qubec; 2000. p 355367. 84. Langhinrichsen-Rohling J, Arata C, Bowers D, et al. Suicidal behavior, negative affect, gender, and self-reported delinquency in college students. Suicide Life Threat Behav. 2004;34(3):255266. W La Revue canadienne de psychiatrie, vol 56, no 10, octobre 2011

Suicide Attempts: Prevention of Repetition

85. Wichstrm L, Rossow I. Explaining the gender difference in self-reported suicide attempts: a nationally representative study of Norwegian adolescents. Suicide Life Threat Behav. 2002;32(2):101116. 86. Canetto SC, Lester D. Gender and the primary prevention of suicide mortality. Suicide Life Threat Behav. 1995;25(1):5869. 87. Pouliot L, De Leo D. Critical issues in psychological autopsy studies. Suicide Life Threat Behav. 2006;36(5):491510. 88. Adams SM, Miller KE, Zylstra RG. Pharmacologic management of adult depression. Am Fam Physician. 2008;77(6):785792. 89. Wolf NJ, Hopko DR. Psychosocial and pharmacological interventions for depressed adults in primary care: a critical review. Clin Psychol Rev. 2008;28(1):131161. 90. Arie N, Apter A, Orbach I, et al. Autobiographical memory, interpersonal problem solving, and suicidal behaviour in adolescent inpatients. Compr Psychiatry. 2008;49(1):2229. 91. Chagnon F. Coping mechanisms, stressful events and suicidal behaviour among youth admitted to juvenile justice and child welfare service. Suicide Life Threat Behav. 2007;37(4):439452. 92. Jeglic EL, Sharp IR, Chapman JE, et al. History of family suicide behaviours and negative problem solving in multiple suicide attempters. Arch Suicide Res. 2005;9(2):135146. 93. McAuliffe C, Corcoran P, Keeley HS, et al. Problem-solving ability and repetition of deliberate self-harm: a multicentre study. Psychol Med. 2006;36(1):4555. 94. Pollock LR, Williams JM. Problem solving in suicide attempters. Psychol Med. 2004;34(1):163167. 95. Roskar S, Zorko M, Bucik V, et al. Problem solving for depressed suicide attempters and depressed individuals without suicide attempt. Psychiatr Danub. 2007;19(4):296302. 96. Sakinofsky I, Roberts RS, Brown Y, et al. Problem resolution and repetition of parasuicide. A prospective study. Br J Psychiatry. 1990;156:395399. 97. Sakinofsky I, Roberts RS. Why parasuicides repeat despite problem resolution. Br J Psychiatry. 1990;156:399405. 98. Greenfield B, Larson C, Hechtman L, et al. A rapid-response outpatient model for reducing hospitalization rates among suicidal adolescents. Psychiatr Serv. 2002;53(12):15741579. 99. Van Heeringen C. The management of non-compliance with outpatient after-care in suicide attempters: a review. Int J Suicidology. 1992;2:7983. 100. Chen H, Mishara BL, Liu XX. A pilot study of mobile telephone message interventions with suicide attempters in China. Crisis. 2010;31(2):109112.

Manuscript received March 2011, revised, and accepted May 2011. 1 Professor, Universit du Qubec Trois-Rivires, Trois Rivieres, Quebec; Researcher, Centre for Research and Intervention on Suicide and Euthanasia (CRISE), Montreal and Philippe-Pinel Institute of Montreal, Montreal, Quebec. 2 Associate Professor, Universit du Qubec Montral, Montreal, Quebec; Researcher, Centre for Research and Intervention on Suicide and Euthanasia (CRISE), Montreal, Quebec. 3 Professor, Universit du Qubec Montral, Montreal, Quebec; Researcher, Centre for Research and Intervention on Suicide and Euthanasia (CRISE), Montreal, Quebec. 4 Associate Professor, McGill University, Montreal, Quebec; Doctor, Montreal Childrens Hospital, Montreal, Quebec; Researcher, Centre for Research and Intervention on Suicide and Euthanasia (CRISE), Montreal, Quebec. Address for correspondence: Dr M S Daigle, Universit du Qubec Trois-Rivires, PO Box 500, Trois-Rivires, QC G9A 5H7; marc.daigle@uqtr.ca

Rsum : Tentatives de suicide : prvention de la rptition

Objectif : Prsenter un aperu de stratgies prometteuses pour prvenir la rptition de comportements suicidaires. Mthode : Cette revue de la littrature sur les interventions prventives tertiaires des tentatives de suicide a t produite au moyen des bases de donnes informatiques PubMed et PsycINFO, de janvier 1966 septembre 2010, laide des restrictions de langue du franais et de langlais et des mots cls : suicid* ou autodestruction et traitement* ou thrapie ou intervention* ou prise en charge. Rsultats : Treize des 35 tudes incluses rvlaient des effets statistiquement significatifs des tentatives rptes ou des suicides en moins grand nombre dans la condition exprimentale. Globalement, 22 tudes portaient sur des approches plus traditionnelles, cest--dire, des approches pharmacologiques ou psychologiques. Seulement 2 des 6 traitements pharmacologiques se sont avrs significativement suprieurs un placebo une tude du lithium pour la dpression et du flupentixol pour les troubles de la personnalit. Huit des 16 traitements psychologiques se sont rvls suprieurs au traitement habituel ou une autre approche : la thrapie cognitivo-comportementale (TCC) (n = 4), (y compris la thrapie comportementale dialectique [n = 2]); la thrapie psychodynamique (n = 2); la thrapie mixte (TCC plus thrapie psychodynamique [n = 1]); et lapproche motivationnelle et le changement de thrapeute (n = 1). Parmi les 8 tudes o le contact se faisait par une visite, la poste ou le tlphone, ou la fourniture dune carte durgence verte, 2 taient significatives : lune comportait un suivi par tlphone et lautre, un suivi par tlphone ou des visites. Lhospitalisation ntait pas lie un moins grand nombre de tentatives, et 1 service dapproche sur 4 a eu des rsultats significatifs : un programme comportant un traitement individualis toutes les 2 semaines. La raison dtre de ces approches simples ou multiples demeure clarifier. Il y avait des dfauts mthodologiques dans bien des tudes et certaines avaient des chantillons limits trs spcifiques. Conclusions : Il faut plus de recherche pour aborder le problme des dfinitions des rsultats et de la mesure des variables dpendantes, les effets sexospcifiques, et linclusion de groupes risque lev. Il faut aussi laborer et valuer de nouvelles approches qui soutiennent la collaboration avec les ressources communautaires ainsi quune valuation et des comparaisons plus prcises des traitements existants pour diffrentes populations.

The Canadian Journal of Psychiatry, Vol 56, No 10, October 2011 W

629