How to Ask About Suicide? A Question in Need of an Empirical Answer
There are over a dozen papers published between 2001 and 2017 that have unanimously found that asking patients and/or research subjects about suicide ideation (SI) does not have an iatrogenic effect, such as leading to an increase in SI (Dazzi, Gribble, Wessely, & Fear, 2014; Law et al., 2015). By contrast, Dazzi et al. (2014) reported that acknowledging and talking about suicide may, in fact, reduce rather than increase SI, a finding consistent with qualitative studies of both pediatric and adult medical inpatients who remain supportive of suicide risk screening after they themselves had been screened (Ross et al., 2016; Snyder et al., 2016).
That's the good news.
What's not such good news is that to date we do not really know how to ask "the ask"; moreover, there is significant confusion about just what it is we are asking. Both our research scales and our methods of clinical enquiry about SI remain lost in the weeds of unaddressed questions about just what it is we want our patients and our research subjects to tell us when we ask and just what is the meaning of what they tell us when they respond affirmatively to our questions about it. Some define SI as strictly containing only the thought of intentionally ending one's life, while others define SI to include the desire to actively end one's life (suicide intent), the reasons for ending one's life (suicide motivation), the means to end one's life (suicide methods), and the method, date, and place to end one's life (suicide plans).
The bottom line is that the term SI has no consistent operational definition. The result is that we cannot reasonably compare the incidence and prevalence of SI across patient groups, in the general population, or compare one group with another. Furthermore, we cannot reasonably expect the affirmation of SI to have any positive predictive value relative to future suicidal behavior, and we cannot reasonably know the meaning of SI when a patient responds affirmatively to our enquiry about it. For, in fact, SI has been reported as being at times "fleeting," and context specific.
Research/Screening Scales
Valtonen and colleagues (Valtonen, et al., 2009) asked 191 bipolar patients whether they had ever seriously considered suicide during their current bipolar episode and further evaluated SI via the Scale for Suicide Ideation (SSI), the Beck Depression Inventory (BDI, Item 9) and the Hamilton Depression Scale (HAM-D, Item 3). In all, 74% of patients were suicide ideators as measured by at least one of the three measures; but only 29% of patients met the criteria for having SI on all measures. These researchers concluded that, "Who is classified as having suicide ideation depends strongly on the definition and means of measurement of suicide ideation" (Valtonen et al., 2009, p. 53).
Similarly, another Finnish group of researchers (Vourilehto et al., 2014) measured SI in six different ways among primary care patients diagnosed with major depressive disorder (MDD) in order to assess the prevalence of SI. These researchers found that the prevalence of SI among these patients was strongly influenced by the method of its assessment. Of 153 MDD patients seen in primary care, only 8% tested positive for SI across all measures.
Indeed, screening scale items ask the ask in very different ways, and therefore affirmative responses have to be considered within the narrow frames of each specific scale question.
Elsewhere, Silverman and Berman (2014) have commented on the Patient Health Questionnaire (PHQ-9) that addresses SI via a compound question: "Have you had thoughts that you would be better off dead or of hurting yourself in some way for at least several days in the last two weeks?" By addressing both passive and active SI (or does hurting yourself refer to nonsuicidal self-injury [NSSI]?) in the same question, affirmative responses demand follow-up questions to ascertain exactly what thoughts the respondent has been having. In addition, of course, if these are active thoughts (the latter part of the question), even further follow-up is indicated regarding frequency, duration, intensity, controllability, planning, etc. Even the term for at least several days may be understood and responded to differently by the patient.
Moreover, we also discussed the difference that makes no difference, that is, that active SI has no greater predictive value with regard to future suicidal behavior compared with passive SI (Silverman & Berman, 2014). Since that publication, Berman (2017) has added data to this argument in finding that of 43 patients in or recently in clinical care who responded positively to having SI when last asked before their deaths by suicide, almost equal proportions affirmed having active versus passive SI. Findings such as these bring into question the underpinnings of scales such as the Columbia Suicide Severity Rating Scale (C-SSRS) that establishes a Likert scale for assessing SI with active ideation having greater clinical and predictive import than passive ideation. Notably, the positive predictive value of the PHQ-9 against the C-SSRS has been found to be only 22.5%, providing evidence that these two measures are essentially measuring different constructs (Giddens & Sheehan, 2014).
Clinical Enquiry
There are myriad ways to ask the ask and no agreement as to how best to ask the ask, what time frame best defines the ask, and/or what terms to use to assert clarity in the ask. Consider the following combinations and permutations of the ask (which are not offered as exhaustive):
In the past (1) have you had (2) ____ thoughts about (3) _____?
(1) Several days; 2 weeks; 1 month; 12 months; 2 years
(2) Any; serious; frequent; uncontrollable
(3) Being better off dead; going to sleep and not wanting to wake up in the morning; not wanting to live; dying; just giving up; life not being worth living; killing yourself; harming yourself; hurting yourself; dying by suicide
Thoughts of Suicide or Self-Harm
In the United States, the question asked typically focuses on thoughts of suicide or killing oneself. In the UK and elsewhere across the globe, the preferred term is self-harm. In the United States, the Centers for Disease Control (CDC) makes these terms synonymous as mortality data are offered under "intentional self-harm (suicide)" (https://www.cdc.gov/nchs/fastats/death.htm).
Of course, we do not know whether our patients understand that these are meant to be synonymous, especially if they interpret a question about self-harm to refer to what otherwise would be labeled as NSSI. No study has been published to date that helps us to understand the connotations of these terms in our patients' minds, to determine the meaning of their responses if either in the affirmative or negative. To confound this, NSSI and suicide attempts frequently co-occur (Burke, Hamilton, Cohen, Stange, & Alloy, 2016; Nock, Joiner, Gordon, Lloyd-Richardson, & Prinstein, 2006; Whitlock et al., 2013).
Thoughts of Suicide, Self-Harm, Being Dead, or Wanting Not to Wake
As noted earlier, the PHQ-9 asks both about active SI or NSSI ("hurting oneself") and passive SI ("being better off dead"), and clinicians, as well, have no consistent guidance as to which question best affirms a patient's suicidal thoughts that are most associated with near-term potential to enact a suicidal behavior. Just how readily passive thoughts of suicide may shift to active thoughts, planning, and/or impulsive action has simply not been established by research nor sufficiently explored to date.
How to Ask the Ask
We know little about how questions about SI are asked by clinicians. A recent study using real-time data examined how UK psychiatrists initiated enquiries about suicide risk (when they did, as the majority did not) and found that the overwhelming majority (75%) of questions asked were phrased in the negative, as in, "You don't have thoughts of harming yourself?". Not surprisingly, when the question was negatively framed, patients were significantly more likely to say they were not suicidal than when the question was positively framed (McCabe, Sterno, Priebe, Barnes, & Byng, 2017).
A second UK study (O'Reilly, Kiyimba, & Karim, 2016) looked at how mental health practitioners addressed self-harm risk in young people in actual practice and found that of those practitioners who addressed the question (the majority did not), the two most common approaches were (a) incremental – a building-up style beginning with questions about emotions such as how the patient behaviorally dealt with frustration or sadness, then moving into specific questions about the link between emotion and self-harm; and (b) normalizing and externalizing – relying on being required by an external authority to ask the question, as in, "There is a question we have to ask everybody… have you ever thought about…?"
While clinical wisdom historically has dictated that clinicians should ask about SI in at least two different ways, there is a dearth of empirical study of the relative effectiveness of different approaches to asking questions.
The Time Frame
Additionally, empirical study has yet to sufficiently address the time period in which to frame the questions clinicians (or screening scales) should ask. Which of the following options has the greatest predictive value: "Are you currently thinking…?"; "In the past 2 weeks have you been thinking…?"; "In the past month have you had thoughts…?"; "Have you recently been having thoughts…?"; "Have you ever had thoughts…?" The best answer to date is that none of these does! Beck, Brown, Steer, Dahlsgaard, and Grisham (1999) found that a retrospective report of SI at the worst point in a patient's life was a better predictor of eventual death by suicide than was current (or presumably recent) SI. Joiner and colleagues (2003) replicated these findings.
It should not surprise any of us that in a typical clinical chart assessment a psychiatric patient has little more than a circled yes or no response to the stimulus item suicide ideation or self-harm.
Conclusion
If a question about SI is the gateway to assessing a patient's risk for suicide (Silverman & Berman, 2014), suicide researchers have significant opportunities to help improve the quality and predictive value of how we address this question. Are there synonymous ways of asking the question? Does the patient's understanding of the question agree with the clinician's understanding of the question? How do patients who self-injure differentiate between self-harm and dying by suicide when asked if they have been thinking of harming (or hurting) themselves? What is the smallest number of questions a clinician needs to ask to reasonably understand what is needed to be understood of a patient's affirmative response to the index question about SI? How should culture be taken into consideration when asking the question such that cross-cultural differences in responses and the prevalence of these responses can be minimized? The context of when to ask the question is also paramount. For example, this enquiry should not be asked when there is any indication that the patient is cognitively impaired (e.g., under the influence of alcohol or another drug).
In the meantime, the field of suicidology would benefit greatly from establishing a consensus best practice to asking the ask and to operationalizing the language we use in addressing the question. Doing so would improve our data and our understanding of that data across time and cultures/countries/risk groups and improve our communications with our clinical and research colleagues, our patients, and our patients' families and significant others. Doing so might very well also spur more mental health clinicians, emergency department personnel, and primary care clinicians/general practitioners to ask questions they now too infrequently ask, in part because they lack such guidance on what and how to ask, with the result that those harboring suicidal thoughts will respond affirmatively, hopefully leading to better treatment planning and nonsuicidal outcomes. These issues, as well as related topics, are the focus of the IASP Special Interest Group on Nomenclature and Classification (Silverman & DeLeo, 2016).
Alan L. Berman, PhD, is Adjunct Professor of Psychiatry, Department of Psychiatry and Behavioral Sciences, Johns Hopkins School of Medicine, USA. He is Past-President of the International Association for Suicide Prevention (2009–2013) and served as Executive Director of the American Association of Suicidology between 1995 and 2014.
Morton M. Silverman, MD, is Senior Science Advisor to the Suicide Prevention Resource Center (USA), Assistant Clinical Professor of Psychiatry at The University of Colorado Denver School of Medicine, USA, and Distinguished Life Fellow of the American Psychiatric Association. He was the editor-in-chief of Suicide and Life-Threatening Behavior between 1996 and 2009.
References
(1999). Suicide ideation at its worst point: A predictor of eventual suicide in psychiatric outpatients. Suicide and Life-Threatening Behavior, 29, 1–9.
(2017). Risk factors proximate to suicide and suicide risk assessment in the context of denied suicide ideation. Suicide and Life-Threatening Behavior. Advance online publication. 10.1111/sltb.12351
(2016). Identifying a physical indicator of suicide risk: Non-suicidal self-injury scars predict suicidal ideation and suicide attempts. Comprehensive Psychiatry, 65, 79–87.
(2014). Does asking about suicide and related behaviours induce suicidal ideation? What is the evidence? Psychological Medicine, 44, 3361–3363.
(2014). Is there value in asking the question "Do you think you would be better off dead?" in assessing suicidality? A case study. Innovations in Clinical Neuroscience, 11, 182–190.
(2003). Worst-point suicidal plans: A dimension of suicidality predictive of past suicide attempts and eventual death by suicide. Behaviour Research and Therapy, 41, 1469–1480.
(2015). Does assessing suicidality frequently and repeatedly cause harm? A randomized control study. Psychological Assessment, 27, 1171–1181.
(2017). How do healthcare professionals interview patients to assess suicide risk? BMC Psychiatry, 17, 122.
(2006). Non-suicidal self-injury among adolescents: Diagnostic correlates and relation to suicide attempts. Psychiatry Research, 144, 65–72.
(2016). "This is a question we have to ask everyone": Asking young people about self-harm and suicide. Journal of Psychiatric and Mental Health Nursing, 23, 479–488.
(2016). To ask or not to ask? Opinions of pediatric medical inpatients about suicide risk screening in the hospital. Journal of Pediatrics, 170, 295–300.
(2014). Suicide risk assessment and risk formulation. Part I: A focus on suicide ideation in assessing suicide risk. Suicide and Life-Threatening Behavior, 24, 420–431.
(2016). Why there is a need for an international nomenclature and classification system for suicide. Crisis, 37(2), 83–87.
(2016). Patient opinions about screening for suicide risk in the adult medical inpatient unit. Journal of Behavioral Health Services and Research, 44, 364–372.
(2009). How suicidal bipolar patients are depends on how suicidal ideation is defined. Journal of Affective Disorders, 118, 48–54.
(2014). Method of assessment determines prevalence of suicide ideation among patients with depression. European Psychiatry, 29, 338–344.
(2013). Nonsuicidal self-injury as a gateway to suicide in young adults. Journal of Adolescent Health, 52, 486–492.
