Does asking about suicide and related behaviours induce suicidal ideation? What is the evidence?

Abstract

There is a commonly held perception in psychology that enquiring about suicidality, either in research or clinical settings, can increase suicidal tendencies. While the potential vulnerability of participants involved in psychological research must be addressed, apprehensions about conducting studies of suicidality create a Catch-22 situation for researchers. Ethics committees require evidence that proposed studies will not cause distress or suicidal ideation, yet a lack of published research can mean allaying these fears is difficult. Concerns also exist in psychiatric settings where risk assessments are important for ensuring patient safety. But are these concerns based on evidence? We conducted a review of the published literature examining whether enquiring about suicide induces suicidal ideation in adults and adolescents, and general and at-risk populations. None found a statistically significant increase in suicidal ideation among participants asked about suicidal thoughts. Our findings suggest acknowledging and talking about suicide may in fact reduce, rather than increase suicidal ideation, and may lead to improvements in mental health in treatment-seeking populations. Recurring ethical concerns about asking about suicidality could be relaxed to encourage and improve research into suicidal ideation and related behaviours without negatively affecting the well-being of participants.

Introduction

In their recent paper in Psychological Medicine, Omerov and colleagues (2013) aim to allay some of the concerns commonly held in psychology that enquiring about suicidality can increase suicidal tendencies. Given the potential vulnerability of subjects included in psychological research, it is only right that concerns around their involvement in research are addressed. However, this relationship is not absolute. As Omerov and colleagues demonstrate, asking recently suicide-bereaved parents about their child's death indicates that rather than being harmed by their participation in the study, the majority found the experience to be a valuable and positive experience.

Apprehensions about conducting studies in this area create a Catch-22 situation. Ethics committees require evidence that the proposed study will not cause participant distress or suicidal ideation (Lakeman & Fitzgerald, 2009) yet there is a lack of published research for investigators to use in addressing these concerns (Petrie et al. 2013). Omerov and colleagues discuss this tension and highlight the fact that gaining ethical approval to conduct studies into suicide is a major hurdle in suicide-related research.

Concerns about questions on suicidal thoughts are not limited to research. In psychiatric settings, performing a risk assessment is an important prerequisite for patient safety (Wu et al. 2012). Psychiatric interviews, which may include questions on suicidality, are often seen as appropriate for patients as they are accessing psychiatric support and treatment. However, there is a reluctance to enquire about suicidality in other healthcare settings. In a survey of 170 German primary-care physicians, 23% expressed that they would not assess risk of suicide of elderly depressed patients over worries that it might encourage thoughts of suicide (Stoppe et al. 1999). Similar results were seen among 103 general practitioners based in England (Bajaj et al. 2008); one-third believed that questions about suicidal behaviour could induce thoughts of self-harm.

Are these concerns evidence-based?

A review of the recent literature (PsycINFO and Ovid Medline) revealed a total of 13 papers published between 2001 and 2013 that have examined whether asking about suicide induces suicidal ideation. Conducted using samples of both adolescents and adults and in general and at-risk populations, none found a statistically significant increase in suicidal ideation in participants as a result of being asked about their suicidal thoughts.

Gould et al. (2005) screened high-school students (n = 1172) for suicidal ideation and behaviour 2 days after an initial suicide questionnaire, while the control group (n = 1170) was not asked any sensitive questions. The study found that high-school students asked about suicidality had significantly decreased participant distress scores compared to controls (p = 0.01). Likewise, Aseltine et al. (2007) reported that exposure to an educational high-school suicide programme significantly reduced self-reported suicide attempts 3 months after programme completion (n = 2100, p < 0.01).

A similar finding has been found in interventions for adolescents and adults seeking treatment for suicidal intent or those with previous or existing mental health problems. Mathias et al. (2012) recruited adolescents aged 12–17 years who had previously received in-patient psychiatric care and screened them for suicidal ideation every 6 months for a period of up to 2 years (n = 54 completed all five assessments). Participants who had higher levels of suicidality at the start of the study demonstrated larger reductions in suicidal ideation by the final assessment (p < 0.0001). King et al. (2003) examined the effectiveness of telephone counselling sessions for adolescents seeking treatment for suicidal intent (n = 101) and reported significant improvements in mental state (anger, sadness, hopelessness, agitation, guilt, slowed speech) (p < 0.0005) and decreases in suicide urgency (p < 0.001). Cedereke et al. (2002) interviewed participants a month and then at 12 months after a failed suicide attempt. At the 12-month follow-up interview, there were significant improvements in global functioning in the group that received two telephone interviews (aimed at increasing motivation for professional treatment) at 4 and 8 months, compared to no calls (n = 107, p < 0.001). In a similar study where participants were also recruited after a failed suicide attempt, the intervention group, which was contacted by telephone at 1 month, was significantly less likely to re-attempt suicide than the control group (not contacted) over a 13-month period (n = 605, p = 0.03) (Vaiva et al. 2006). Lang et al. (2010) screened outpatients (n = 719) with serious mental health problems for a 6-monthly period (mean number of screening tests 3.6) and found no increase in suicide attempts. In a study of 63 female patients with borderline personality disorder and a history of suicide attempts over a period of 2 years, Reynolds et al. (2006) evaluated participants every 4 months for self-reported ratings of suicidality both at the beginning and end of assessment sessions. While 77% (n = 44) reported at least one increase in suicidality following assessment, out of the total 764 participant sessions, 16.4% (n = 125) showed an increase and 17.5% (n = 134) showed decreases in suicidality.

Using data from four qualitative interview studies of adult and adolescent participants (n = 63) with a history of self-harm or suicidality, Biddle et al. (2013) measured emotional state at the beginning and end of interviews and asked participants about suicidal behaviour or self-harm. Improvements in scores were reported for 57% (n = 36) of participants, who found talking about their experiences beneficial and helpful in providing new insight. Of the 22% (n = 14) with a decrease in scores, most described how this was due to them being reminded of, or forced to focus on, previous or current difficulties they might be having or perceptions of shame and embarrassment. Several felt these feelings would be temporary. Regardless of scores, taking part in the research was seen as being necessary to help people understand suicidality and self-harm and gave people a positive outcome even when scores did not improve.

The findings of this review suggest that in both adolescent and adult populations, acknowledging and talking about suicide may in fact reduce, rather than increase suicidal ideation, with a suggestion that repeat questioning may benefit long-term mental health. Studies in treatment-seeking populations suggest that asking people who are or have been suicidal about suicidality can lead to improvements in mental health.

The study by Omerov et al. (2013) also indicates that there are benefits for research participants when they are asked about suicide. After initially being sent a participation letter, 666 suicide-bereaved and 377 non-bereaved parents were surveyed about their perceptions regarding being contacted for the study. Non-bereaved parents and bereaved parents were matched for age, gender, living area, marital status, number of children and having had a child born in same year as the deceased child. Ninety-four percent (633 of bereaved and 347 of non-bereaved) found the study a valuable and positive experience and 91% of suicide-bereaved and 76% of non-bereaved parents would recommend other parents to participate. While a small minority of bereaved (70, 11%) and non-bereaved (3, 1%) parents were negatively affected by the questions due to recalling painful memories and feelings of sadness, only two out of the 1043 felt that the negative effects would last – only one of whom was bereaved.

Although suicidality was not directly assessed in Omerov et al.'s paper, it is known that close family members of suicide completers are themselves at higher risk of suicide (Ness & Pfeffer, 1990; Mościcki, 1995) and that this population have poorer self-reported psychiatric and general health than families who have experienced a natural death (Groot et al. 2006). That suicide-bereaved parents in similar studies (Dyregov, 2004) have also reported positive experiences as a result of participation in research suggests that it possible to conduct research into sensitive topics without causing harm to potentially vulnerable participants.

Conclusion

The findings of our review and the study by Omerov and colleagues suggest that recurring ethical concerns about enquiring about suicidality could be relaxed in order to encourage and improve research into suicidal ideation and related behaviours without negatively affecting the well-being of participants.