represent a good example of the coming together of theory and
treatment. Beck and many other people working from a cognitive
206 ● WHY PEOPLE DIE BY SUICIDE
viewpoint have developed theories of psychopathology wherein a
maladaptive schema for the understanding and making sense of one’s
role in the world confers vulnerability to various forms of psycho-
pathology. The treatment revolves around correcting the maladap-
tive schema.
My goal in this chapter is to try to emulate Beck in a useful, rele-
vant, and productive bringing together of theory and practice. The
previous chapters have laid out the theory and its anecdotal and sci-
entific support. This chapter attempts to use the theory to inform
clinical practice regarding suicidal behavior, starting with the impor-
tant area of suicide risk assessment.
Risk Assessment
Recall the distinction between the alarmist and the dismissive ap-
proach to suicide risk assessment. The alarmist position involves the
idea that whenever someone mentions suicide, it is a life-threatening
situation and alarms should be sounded. Those who take a dis-
missive approach make a mistake in the opposite direction. They be-
come blasé about suicidal behavior, often attributing it to manipula-
tion or gesturing on the part of the potentially suicidal person. A
compromise is thus needed between the alarmist and dismissive
approaches—one that is efficient and clinically useful, scientifically
supported, and conceptually consistent with the model developed in
this book.
Any risk assessment system has to grapple with the fact that there
are dozens of suicide risk factors, some of which are associated with
imminent risk (e.g., severe agitated anxiety), and others of which are
important but are more distal and not clearly tied to imminent risk
(e.g., family history of suicide). In fact, a quick survey of the websites
of organizations like the American Association of Suicidology, the
American Foundation for Suicide Prevention, and the American Psy-
Assessment, Intervention, Treatment, and Prevention ● 207
chological Association, among many others, shows that over seventy-
five factors are listed as suicide risk factors or warning signs, includ-
ing things as diverse and questionable as “loss of religious faith,”
“neurotransmitters,” “perfectionism,” and “loss of security.” Given
limited time, clinicians cannot thoroughly assess all of these various
factors, and even if they could, how are they to organize the resulting
mass of data?
Of all the numerous risk factors and warning signs, do any stand
out as particularly important? If so, then a risk assessment approach
might be built around them. Of course, based on the model devel-
oped in this book, the acquired ability to enact lethal self-injury de-
serves emphasis, as do perceived burdensomeness and low belong-
ingness. To assess acquired ability, my colleagues and I argued that
two factors deserve particular weight: a history of multiple suicide
attempts and the specific nature of current suicidal symptoms, with
specific reference to whether the symptoms include resolved plans
and preparations or suicidal desire.
Multiple attempt status is emphasized because it is perhaps the
clearest marker of the acquired ability for lethal self-injury. The dis-
tinction between resolved plans and preparations and suicidal desire
is important too. Resolved plans and preparations includes the fol-
lowing symptoms: a sense of courage to make an attempt; a sense
of competence to make an attempt; availability of means to and op-
portunity for attempt; specificity of plan for attempt; preparations
for attempt; duration of suicidal ideation; and intensity of suicidal
ideation. Suicidal desire includes a different set of symptoms: reasons
for living, wish to die, frequency of ideation, wish not to live, passive
attempt, desire for attempt, and talk of death or suicide.
Symptoms of the resolved plans and preparations cluster are evi-
dence of the person’s ability to lethally injure themselves, because the
symptoms require a fearlessness and sense of resolve in order to for-
mulate clear and actionable plans about death. My theory also sug-
208 ● WHY PEOPLE DIE BY SUICIDE
gests a broadening of the focus in suicide risk assessment from mul-
tiple attempters to anyone who has, through various means, acquired
the capability for lethal self-injury. This capability is acquired by
means of repeated practice or repeated exposure to self-injury. Prac-
titioners should assess for instances in which a patient may have been
able to practice self-harm, including aborted suicide attempts. Expe-
riences like multiple surgeries and repeated tattooings and piercings
represent other possible areas of inquiry. Various forms of exposure
to violence, as well as other provocative experiences like self-injecting
drug use, are other possible areas of investigation.
Another implication of the theory is that the desire for suicide
may be most pernicious when it contains themes of both burden-
someness and thwarted belongingness. If suicidal desire in general is
endorsed, it should be explored as to whether burdensomeness and
thwarted belongingness undergird it. If so, risk may be more ele-
vated; if not, risk may be more moderate. One complexity is that
feelings of burdensomeness and low belongingness are not necessar-
ily static; they are fluid and may vacillate in some people. A patient
who has mild feelings of burdensomeness one week may have intense
feelings the next day or the next week. A person who genuinely
professes strong belongingness on one day may subsequently de-
velop a sense of disconnectedness on another day, perhaps as a func-
tion of relationship conflict. Clinicians therefore have to monitor
risk regularly, even in previously low-risk patients, and in particular
need to monitor variables like perceived burdensomeness and failed
belongingness, which are both central and fluctuating.
Returning to the overall risk assessment framework, it emphasizes
two general domains—multiple attempt status and the two factors of
suicidal symptoms described earlier (resolved plans and preparations
and suicidal desire). For multiple attempters and those who indicate
they have resolved plans and preparations for suicide, risk assess-
ment proceeds differently than for everybody else. In these cases, risk
Assessment, Intervention, Treatment, and Prevention ● 209
is automatically viewed as elevated, especially in the presence of at
least one other risk factor (e.g., burdensomeness, low belongingness,
current and serious substance abuse, or severe negative life events).
For those who are neither multiple attempters nor endorse symp-
toms of the resolved plans and preparations factor but who do have
symptoms of suicidal desire, the threshold to establish elevated risk is
set higher.
More specifically, here is how the framework is used: For multi-
ple attempters, mos
t any other additional risk factor (e.g., substance abuse) translates into at least moderate suicide risk. For nonmultiple attempters, those with resolved plans and preparations and most
any other additional risk factor are at moderate suicide risk at
least. For nonmultiple attempters with no resolved plans and preparations but who do voice suicidal desire, the presence of two or more additional risk factors translates into at least moderate suicide risk.
The framework is not a completely automated statistical predic-
tion rule, but provides a relatively objective starting point for clinical
decision-making in risk assessment. In Chapter 1, I mentioned my
patient “Gayle,” who had recurrent depressions and who had de-
veloped ideas about dying by severing her hand with a machete.
She had acquired the ability to enact lethal self-injury not through
previous suicidal behavior—she had never attempted suicide—but
through severe substance abuse and an array of associated painful
and provocative experiences in her past. When I saw her, she had
been sober for many years, but there were residues of this past, and
one was the acquired ability to lethally injure herself.
I wanted to hospitalize Gayle because of her clear and detailed sui-
cide plan and perhaps especially because of her sense of calm and her
lack of fear about the plan. But she was not at particularly high risk
for suicide, because she did not report thwarted belongingness and
perceived burdensomeness. On the contrary, Gayle was very con-
210 ● WHY PEOPLE DIE BY SUICIDE
nected to her son and had many friends. Also, she was a particularly
capable woman, and there was no evidence that she felt ineffective,
certainly not to the point that she believed she burdened others. The
risk assessment framework described above clarifies clinical deci-
sion-making regarding a situation that would otherwise be very dif-
ficult to handle.
When combined with the current theory, an assessment approach
like the one just described encourages scientifically and theoretically
informed assessment and relatively routinized clinical decision-mak-
ing and activity. This assessment approach also represents a satisfy-
ing integration of theory and application. This same kind of integra-
tion can be seen in the important area of crisis management and
resolution, to be discussed next.
Crisis Intervention
As demonstrated by the case of Gayle, the acquired ability to enact
lethal self-injury, once in place, does not fade quickly. It is a relatively static quality that does not come and go over time. It therefore would
not be a particularly useful focus for crisis intervention, where the
goal is to take the edge off the pain of the current crisis, so that it is
within a tolerable range. Since acquired ability is unlikely to change
much in the short-term, it does not provide any leverage to accom-
plish short-term reduction of distress.
By contrast, professionals who deal with suicidal crises would do
well to focus on burdensomeness and belongingness. Unlike the ac-
quired capability for serious suicidal behavior, burdensomeness and
belongingness may be more malleable and thus more amenable to
short-term crisis intervention. My colleagues and I have described
techniques for in-session diminution of distress.6 For example, tech-
niques such as the symptom-matching hierarchy and development
of a crisis card often take the edge off of intense negative moods.
Each of these techniques is described next.
Assessment, Intervention, Treatment, and Prevention ● 211
The symptom-matching hierarchy simply involves listing disrup-
tive symptoms and feelings. The patient ranks these in terms of
which are most upsetting (e.g., as rated on a one-to-ten scale). For
the top two or three symptoms or feelings, very concrete recommen-
dations are made (e.g., sleep hygiene for insomnia; relaxation for
general emotional distress; pleasant activities for depressive symp-
toms). These recommendations are not intended or expected to solve
the problem or to even change it very much; rather, they are intended
to just take the edge off of the problem, so that the person is some-
what more comfortable and thus better able to tolerate the crisis and
to start working toward solving the underlying problems.
Feelings of burdensomeness and low belongingness should rou-
tinely be targeted within this straightforward crisis-resolution ap-
proach. For example, a clinician might say, “I see that you perceive
yourself a burden on your family, but do they see it the same way?”;
or “Let’s briefly review the relationships and groups, not just right
now but in the past too, to which you felt a sense of belonging”; or
“Let’s review the ways you have contributed to people or society, not
just right now but in the past too.” The therapist could summarize
the products of this discussion, perhaps in bulleted form on an index
card, and give the summary to the patient, with instructions to ex-
pand and elaborate the list at home. Again, the point of this exercise
is not to fully undo underlying feelings of burdensomeness and low
belongingness, but to destabilize and reduce them slightly, so that the
patient will be in a better position to handle the current crisis and
build skills through therapy that will eventually target long-standing
problems.
The crisis card is another simple technique designed to lessen the
intensity of a crisis so that more clear-headed approaches can emerge.
The crisis card simply involves the development of a straightforward
crisis plan that can be written down on an index card or a sheet of
paper. An example would be:
212 ● WHY PEOPLE DIE BY SUICIDE
When I’m upset and thinking of suicide, I’ll take the following steps:
1. Use what I’ve learned in therapy to try to identify what is upset-
ting me, focusing especially on feeling I’m a burden on others and
like I don’t belong;
2. Write down and review some reasonable, nonsuicidal responses
to what is bothering me;
3. Try to do things that, in the past, have made me feel better (e.g.,
music, exercise, etc.);
4. If the suicidal thoughts continue and get specific, or I find my-
self preparing for suicide, I’ll call the emergency call person at
(phone number);
5. If I feel that I cannot control my suicidal behavior, I’ll go to the
emergency room or call 911.
Both the symptom-matching hierarchy and the in-session devel-
opment of a crisis card may dilute intense distress then and there in
the session. Both techniques should be focused on relieving feelings
of burdensomeness and low belongingness. This, in turn, may de-
crease discomfort—not completely, but enough so that difficulties
can be better tolerated and tackled with skill-based therapeutic tech-
niques (to be discussed in the next section). A slight decrease in dis-
comfort, incidentally, may also facilitate the occurrence of general
positive moods, which my colleagues and I have shown improves
treatment outcome
in suicidal individuals.7
I recommend the use of a crisis card, as described above, instead of
what are known as “no-suicide contracts.” No-suicide contracts are
written agreements that patients will not attempt suicide while under
treatment. They usually are signed by both patient and therapist.
One reason that I do not recommend them is that they apparently do
not work very well. For example, a survey of Minnesota psychiatrists
found that of those who used no-suicide contracts, over 40 percent
reported that they had patients die by suicide or make a near-lethal
Assessment, Intervention, Treatment, and Prevention ● 213
attempt while under a contract.8 In a study of self-harm incidents
among psychiatric inpatients, some of whom were on no-suicide
contracts, there was some indication that being under contract was
associated with more self-harm.9 This could be because the most seriously ill patients were put under contract, but nevertheless, it does
not represent a ringing endorsement of no-suicide contracts.
Another reason that I do not use no-suicide contracts is that they
only tell patients what not to do and neglect telling patients what
they should do instead. Relatedly, no-suicide contracts ignore impor-
tant aspects of the model developed in this book. Instead of advising
patients not to try suicide, a better approach would be to advise them
what to do in the event that suicidality in general escalates and that
perceived burdensomeness and feelings of failed belongingness in
particular intensify. The crisis card accomplishes this.
I borrow one last thought on crisis intervention from William
James. He wrote that to persuade a suicidal person to live, one could
“appeal—and appeal in the name of the very evils that make his
heart sick—to wait and see his part in the battle out.”10 In their 1933
book on suicide, Dublin and Bunzel expand the point: “the consent
to live on is a resignation based on manliness and pride” (pardon
their politically incorrect use of “manliness,” which I suppose was
not politically incorrect in 1933). This perspective neglects the pain
of perceived burdensomeness and failed belongingness; when these
feelings are very intense, people do not want to continue living re-
gardless of pride or manliness. However, it is an interesting idea to
Why People Die By Suicide Page 27