Assessing and Treating Different Suicidal States in a Danish Outpatient Sample

Abstract

The studies presented compare two methodologies for categorizing suicidal patients based on clinical data. Discussion follows regarding implications for risk assessment and treatment. In these studies, 52 outpatient subjects were placed into different groups based on coding their "suicidal motivation" (Study 1) and their "internal struggle" ratings (Study 2) using data collected at intake. Self-report ratings of 6 Suicide Status Form (SSF) Core Constructs (Psychological Pain, Stress, Agitation, Hopelessness, Self-Hate, and Overall Risk of Suicide) recorded both at intake and at completion of treatment were then compared to determine differences in Core Construct ratings among groups at different time points. In Study 1, overall differences among motivation groups (Life-motivated, Ambivalent, and Death-motivated) were significant for ratings at treatment completion of Overall Risk of Suicide, Self-Hate, and Psychological Pain. In Study 2, overall differences among groups (Wish to live, Ambivalent, and Wish to die) were significant for ratings at intake of Overall Risk of Suicide. At completion of treatment, overall differences among groups were significant for ratings of Overall Risk of Suicide, Hopelessness, and Self-Hate. In addition, significant interactions were found between test time and group for Overall Risk of Suicide and Self-Hate. Results suggest that categorizing suicidal patients by motivation and by the nature of their internal struggle could be beneficial to differential risk assessment with implications for clinical treatment.

Full text

USUI #777002, VOL 17, ISS 3
Assessing and Treating Different
Suicidal States in a Danish
Outpatient Sample
Christopher D. Corona, David A. Jobes, Ann C. Nielsen,
Christian M. Pedersen, Keith W. Jennings, Rene
´
M. Lento, and
Katherine A. Brazaitis
QUERY SHEET
This page lists questions we have about your paper. The numbers displayed at left can be found in the
text of the paper for reference. In addition, please review your paper as a whole for correctness.
"There are no Editor Queries for this paper"
TABLE OF CONTENTS LISTING
The table of contents for the journal will list your paper exactly as it appears below:
Assessing and Treating Different Suicidal States in a Danish Outpatient Sample
Christopher D. Corona, David A. Jobes, Ann C. Nielsen, Christian M. Pedersen, Keith W.
Jennings, Rene
´
M. Lento, and Katherine A. Brazaitis

Assessing and Treating
Different Suicidal States
in a Danish Outpatient
Sample
5
Christopher D. Corona, David A. Jobes, Ann C. Nielsen,
Christian M. Pedersen, Keith W. Jennings, Rene
´
M. Lento, and
Katherine A. Brazaitis
The studies presented compare two methodologies for categorizing suicidal patients
based on clinical data. Discussion follows regarding implications for risk assessment
10
and treatment. In this study, 52 outpatient subjects were placed into different groups
based on coding their ‘‘suicidal motivation’’ (Study 1) and their ‘‘internal struggle’’
ratings (Study 2) using data collected at intake. Self-report ratings of 6 Suicide Status
Form (SSF) Core Constructs (Psychological Pain, Stress, Agitation, Hopelessness,
Self-Hate, and Overall Risk of Suicide) recorded both at intake and at completion
15
of treatment were then compared to determine differences in Core Construct ratings
among groups at different time points. In Study 1, overall differences among motivation
groups (Life-motivated, Ambivalent, and Death-motivated) were significant for rat-
ings at intake of Overall Risk of Suicide, Self-Hate, and Psychological Pain. In Study
2, overall differences among motivation groups (Wish to live, Ambivalent, and Wish
20
to die) were significant for ratings at intake of Overall Risk of Suicide. At completion
of treatment, overall differences among motivation groups were significant for ratings of
Overall Risk of Suicide, Hopelessness, and Self-Hate. In addition, significant inter-
actions were found between test time and motivation group for Overall Risk of Suicide
and Self-Hate. Results suggest that categorizing suicidal patients by motivation and by
25
the nature of their internal struggle could be beneficial to differential risk assessment
with implications for clinical treatment.
Keywords internal struggle, motivation, suicide, typology
INTRODUCTION
The prevalence of suicidal ideation and
30
behavior both in the United States and
abroad supports the argument that suicide
is a legitimate public health concern. The
Substance Abuse and Mental Health
Services Administration (SAMHSA, 2009)
35
reported that, in 2008, an estimated 8.3
million adults (3.7% of the population aged
18 years and older) experienced suicidal
ideation, 2.3 million (1.0%) made a plan
to attempt suicide, and 1.1 million (0.5%)
40
actually attempted suicide. Moreover,
the Centers for Disease Control and
Prevention (CDC, 2010) reported that
Archives of Suicide Research, 17:1–11, 2013
Copyright # International Academy for Suicide Research
ISSN: 1381-1118 print=1543-6136 online
DOI: 10.1080/13811118.2013.777002
3b2 Version Number : 7.51c/W (Jun 11 2001)
File path : p:/Santype/Journals/TandF_Production/Usui/v17n3/USUI777002/USUI777002.3d
Date and Time : 07/06/13 and 20:32
1

almost 37,000 suicides were completed in
2009, making suicide the tenth leading
45
cause of death across all age groups in
the United States that year.
Worldwide, the numbers are just as
striking. The World Health Organization
(WHO, 2002) reported that approximately
50
849,000 people died from suicide in 2001.
According to Nordentoft (2007), in
addition to being among the top ten causes
of death worldwide, suicide is also the
second leading cause of non-illness-related
55
death. In Denmark specifically, the suicide
rate was among the highest in Europe in
1980. Though the suicide rate in Denmark
has declined since then, it is still higher
than that of other Scandinavian countries
60
and than that of most other Western
European nations (Nordentoft, 2007).
These statistics underscore the need
for effective methods that can both identify
suicide risk and provide treatment for those
65
most at risk for completing suicide. Nock,
Borges, Bromet et al. (2008) identified
consistencies among risk factors within a
sample drawn from 17 different nations.
These risk factors include being female,
70
being younger, receiving fewer years of
education, being unmarried, and being
diagnosed with a mental disorder. Diagnos-
tically, the risk factor most associated
with suicide in high-income countries was
75
the presence of a mood disorder, while in
low-income countries the presence of an
impulse control disorder was the strongest
risk factor. Additionally, Nock and collea-
gues found that, among all countries
80
included in their study, approximately
60% of the progressions of suicidal beha-
vior from ideation to attempt occurred
during the first year after onset of ideation.
Despite these consistencies in world-
85
wide risk factors for suicidal behavior, the
ways in which suicidal thoughts manifest
(i.e., their specific content) can be quite
nuanced. Jobes et al. (2004) highlighted
this point in a study that compared
90
qualitative data from two samples of
suicidal outpatients, one comprised of
college students and the other consisting
of active duty United States Air Force per-
sonnel. Participants were asked to provide
95
qualitative responses to five suicide-relevant
stimulus prompts, and significant between-
group differences were found when parti-
cipants reported on their experience of
psychological pain and perturbation. The
100
student responses were suggestive of
developmental struggles that pertained to
identity and relationships, while Air Force
personnel responses were suggestive of
situation-specific crises that centered on
105
hardships associated with military life. What
is notable in this finding is that the same
stimulus prompt elicited different qualitat-
ive content in each sample, suggesting
that the etiology of suicidal behavior can
110
vary widely.
Variability has also been found when
examining the extent to which suicidal indi-
viduals have either a wish to live or a wish to
die. Kovacs and Beck (1977) used ratings of
115
these wishes as the foundation for develop-
ing the ‘‘Internal Struggle Hypothesis,’’
which posited that the risk for suicide is
lower in individuals who have simultaneous
desires to live and to die when compared to
120
those individuals who have only a wish to
die. This hypothesis was based on a study
that analyzed suicidal risk among parti-
cipants hospitalized in the wake of a suicide
attempt. Approximately 50% of the sample
125
experienced an internal debate between life
and death as indicated by two self-report
items that asked participants to rate their
wish to live and their wish to die on interval
scales. Patients who expressed both a wish
130
to live and a wish to die scored lower on a
measure of suicidal intent than patients
who expressed only a wish to die. Brown,
Steer, Henriques, and Beck (2005) used a
similar methodology to assess suicidal risk
135
among psychiatric outpatients. They com-
bined interval scale ratings of two items
(the Wish to Live and the Wish to Die) to
create an ambivalence index score for each
Assessing and Treating Different Suicidal States
2 VOLUME 17

NUMBER 3

2013

patient, and found that this index score was
140
significantly associated with prospective
suicidal risk. More specifically, those with
a higher index score (indicating a stronger
wish to die) were at a higher risk for suicide.
This finding supports the Internal Struggle
145
Hypothesis, and suggests that patient
wishes as they pertain to living and dying
may hold information that could be key to
accurately identifying future suicidal risk.
Given the growing body of literature
150
that supports suicidality as a multidimen-
sional construct that operates differently
in different individuals, the development
of more nuanced methods for assessing sui-
cidal risk is indicated. The present studies
155
attempt to further explore the relationship
between motivation, internal struggle, and
suicidal risk in a Danish sample. Two separ-
ate methodologies were used to categorize
participants seen at two outpatient com-
160
munity mental health centers in Denmark
based on their so called ‘‘suicidal motiv-
ation’’ and the nature of their wish to live
vs. wish to die internal struggle. To this
end, various analyses were run to assess
165
differences between groups on measures
of suicidal risk both before and after treat-
ment. Results will be discussed with an
emphasis on the potential utility of these
categorization methods in relation to asses-
170
sing risk with potential implications for
treating suicidal patients therein.
METHOD
Participants
The studies presented were conducted
175
with a sample of 52 outpatients seen at
one of two community mental health
centers in Aarhus and Copenhagen. The
mean age of the sample was 28.33 years,
with a range of 15 to 54 years. 73.1% of
180
the sample was female; 90.4% was of
Danish ethnicity; 69.2% was single; 63.5%
were either employed or working towards
a degree; 67.3% had not received previous
mental health treatment of any kind at any
185
time; and 59.6% had not made a suicide
attempt at the time of referral. It should
be noted that being without a suicide
attempt at referral does not imply that a
suicide attempt had never been made.
190
Demographic data about suicidal behavior
prior to referral was not available. Subjects
included in the study were those that had
either attempted suicide or expressed suici-
dal ideation at the time of referral. Subjects
195
excluded from the study were those that
were already receiving mental health treat-
ment elsewhere, as well as those with major
or chronic mental disorders (including
major depressive disorder, bipolar disorder,
200
and schizophrenia), substance abuse or
dependence, personality disorders, and
attention deficit hyperactivity disorder.
Instruments
The instrument used in both studies
205
was the Suicide Status Form (SSF), which
is the primary assessment, treatment plan-
ning, and tracking tool used in the ‘‘Colla-
borative Assessment and Management of
Suicidality’’ (CAMS). Developed by Jobes
210
(2006), CAMS is a therapeutic framework
that clinicians from any theoretical back-
ground can adapt to their treatment style
when addressing suicidality. The approach
is founded upon the formation of a strong
215
alliance between clinician and patient, which
encourages the clinician to acknowledge
the patient’s suicidal wish as well and
encourages the patient to be an active par-
ticipant in his or her own risk assessment,
220
treatment planning, and outcomes tracking.
There is an evolving evidence base showing
solid support for the effectiveness of CAMS
in reducing suicidal constructs (Arkov,
Rosenbaum, Christiansen, Jonsson, &
225
Munchow, 2008; Ellis, Green, Allen, Jobes,
& Nadorff, 2012; Jobes, Kahn-Greene,
Greene, & Goeke-Morey, 2009; Nielsen,
C. D. Corona et al.
ARCHIVES OF SUICIDE RESEARCH 3

Alberdi, & Rosenbaum, 2011) and improv-
ing the therapeutic experience (Arkov,
230
Rosenbaum, Christiansen et al., 2008;
Comtois, Jobes, O’Connor et al., 2011; Ellis,
Green, Allen et al., 2012). Additionally,
comparison control studies have shown
CAMS to be superior to treatment-as-usual
235
with regard to resolving suicidality in both
randomized (Comtois, Jobes, O’Connor
et al., 2011) and non-randomized clinical
trials (Jobes, Wong, Conrad et al., 2005).
The SSF is used throughout CAMS to
240
assess and track useful clinical data regard-
ing a patient’s suicidality, and it has been
psychometrically studied and found to be
a valid and reliable clinical assessment
measure for suicidal risk (Conrad, Jacoby,
245
Jobes et al., 2009; Jobes, Jacoby, Cimbolic
et al., 1997). Previous research also supports
the ability of the SSF to predict treatment-
related changes in suicidality over the
course of clinical treatment (Jobes, Kahn-
250
Greene, Greene et al., 2009), and categorize
suicidal patients with implications for
treatment outcomes (Jobes, Jacoby,
Cimbolic et al., 1997; Jobes, Nelson,
Peterson et al., 2004).
255
Three key elements of the SSF are the
focus of the present studies. The first
element consists of written Reasons for
Living (RFL) and Reasons for Dying
(RFD) responses. At intake, patients are
260
asked to write down up to five Reasons
for Living and up to five Reasons for
Dying, and then asked to rank each RFL
and RFD from 1 to 5 (a score of 1 indicates
the reason that is of most importance to the
265
patient). The second element (also com-
pleted at intake) consists of two items that
ask the patient to rate on a Likert scale
(from 0–8 with higher scores indicated
higher intensity) their respective Wish to
270
Live and their Wish to Die.
The third element consists of six self-
rated ‘‘Core SSF Constructs’’ (Psychologi-
cal Pain, Stress, Agitation, Hopelessness,
Self-Hate, and Overall Risk of Suicide) that
275
serve to assist in the assessment of suicidal
risk and in tracking outcomes. Patients are
asked to rate the severity of each construct
on a Likert scale (from 1 to 5 with higher
scores indicating higher intensity). Then,
280
with the exception of Overall Risk of
Suicide, patients are asked to rank the
importance of the first five constructs from
1to5(1¼ most important;5¼ least important).
Traditionally, these SSF Core Constructs
285
are assessed at intake, at the start of each
CAMS tracking session, and in the final
CAMS session. In the present studies, the
SSF Core Construct ratings were only avail-
able for each subject from intake and from
290
their final CAMS session. The number of
CAMS sessions for all subjects ranged from
2 to 11, with the average treatment cycle
lasting 5.52 sessions.
Study 1
295
The first study used a ‘‘macro-coding’’
methodology to categorize subjects by sui-
cidal motivation according to the number
of written RFL and RFD responses
recorded on the SSF at intake. Support for
300
this methodology comes from Jobes and
Mann (1999), who found significant
quantitative differences in RFL and RFD
responses in a sample of suicidal university
counseling center patients, and posited that
305
suicidal risk may be higher for those with
fewer Reasons for Living. For each subject,
the total number of RFD responses was
first subtracted from the total number of
RFL responses. Subjects were then categor-
310
ized based on the value obtained using
this calculation. Those subjects with a posi-
tive value (indicating more written RFL
responses on the SSF) were place in the
‘‘Life-motivated’’ (Life-M) group; those
315
subjects with a negative value (indicating
more written RFD responses) were
placed in the ‘‘Death-motivated’’ (Death-M)
group; and those with a value of zero
(indicating an equal number of written
320
RFL and RFD responses) were placed in
the ‘‘Ambivalent’’ (Ambiv) group.
Assessing and Treating Different Suicidal States
4 VOLUME 17

NUMBER 3

2013

The primary research question in this
study asked whether categorizing patients
according to suicidal motivation could be
325
related to differences in SSF Core Construct
ratings at intake and at completion of treat-
ment. Comparisons were made using motiv-
ation group as the independent variable and
each of the six SSF Core Constructs as
330
dependent variables. First, six one-way
Analysis of Variance (ANOVA) tests were
run to assess differences among motivation
groups on each Core Construct rating at
intake. Six additional one-way ANOVAs
335
were then run comparing construct ratings
by motivation group at completion of treat-
ment. Finally, six repeated-measures ANO-
VAs were run to detect any interactions
between assessment time (intake and com-
340
pletion of treatment) and motivation group
when comparing Core Construct ratings.
Follow-up contrasts were run on significant
overall models to test for specific between-
group differences.
345
Study 2
The second study categorized subjects
by the nature of their internal struggle
according to responses to SSF items asking
patients to rate their Wish to Live and Wish
350
to Die at intake. Each subject’s responses
were first converted to a 3-point scale.
Ratings from 0–2 were assigned a value of
0, ratings from 3–5 were assigned a value
of 1, and ratings from 6–8 were assigned a
355
value of 2. This conversion was done to
adapt the Wish to Live and Wish to Die
scales found on the SSF to be more aligned
with those found on the Scale for Suicide
Ideation (Beck, Kovacs, & Weissman,
360
1979). The converted Wish to Die score
was then subtracted from the converted
Wish to Live score, allowing for the
calculation of a ‘‘Suicide Index Score’’ with
values ranging from 2 to 2. Support for
365
this methodology comes from Brown,
Steer, Henriques et al. (2005), who found
that an index score with similar intervals
was associated with suicidal risk. A notable
difference between their methodology and
370
the one presented here is that positive SIS
values in this study indicated a stronger
wish to live. Finally, subjects were placed
into one of three motivation categories
based on SIS. Those with scores of 2
375
or 1 were placed in the ‘‘Wish-to-die’’
(WTD) group; those with a score of 0 were
placed in the ‘‘Ambivalent’’ (AMB) group;
and those with scores of 1 or 2 were place
in the ‘‘Wish-to-live’’ (WTL) group.
380
The same analyses conducted in Study
1 were also run in Study 2. In addition to
asking whether categorizing patients
according to either suicidal motivation or
the nature of their internal struggle is
385
related to differences in SSF Core Con-
struct ratings, another research goal was
to compare and contrast results garnered
from each methodology.
RESULTS
390
Study 1
This categorization methodology
resulted in a frequency distribution of 36
(69.2%) subjects in the Life-M group, 9
(17.3%) subjects in the Ambiv group, and
395
7 (13.5%) subjects in the Death-M group.
One-way ANOVAs found no significant
differences among motivation groups on
any of the SSF Core Constructs at intake.
Additionally, there were no significant
400
interactions between assessment time and
motivation group. When comparing SSF
Core Construct ratings at completion of
treatment (Table 1), overall differences
among motivation groups were significant
405
for ratings of Overall Risk of Suicide
(F ¼ 3.61, p ¼ .035), Self-Hate (F ¼ 3.68,
p ¼ .032), and Psychological Pain
(F ¼ 3.22, p ¼ .048).
Post-hoc pairwise comparisons were
410
run for both Overall Risk of Suicide and
Self-Hate using a Bonferroni test because
C. D. Corona et al.
ARCHIVES OF SUICIDE RESEARCH 5

of its control of Type I Error rate and
relative power when testing few compari-
sons (Field, 2009). Overall Risk of Suicide
415
ratings at completion of treatment were
significantly different between the Life-M
(1.31) and Death-M (M ¼ 2.14) groups
(p ¼ .032). When comparing Self-Hate rat-
ings at completion of treatment, significant
420
differences were found between the Ambiv
(M ¼ 1.89) and Death-M (M ¼ 3.57)
groups (p ¼ .032). Because Levene’s Test
of Equality of Error Variances was signifi-
cant for Psychological Pain (F ¼ 4.61,
425
p ¼ .015), post-hoc pairwise comparisons
were run for this construct using
Tamhane’s T2 test. This test is known for
its consistent conservatism when variances
are unequal (Jaccard, Becker, & Wood,
430
1984). No significant differences were
found between pairs of motivation groups
when comparing Psychological Pain ratings
at completion of treatment.
Study 2
435
This methodology produced a fre-
quency distribution of 33 (63.5%) subjects
in the WTL group, 13 (25%) subjects in
the AMB group, and 6 (11.5%) subjects
in the WTD group. When comparing
440
Core Construct ratings at intake, overall
differences among motivation groups were
significant for Overall Risk of Suicide
(F ¼ 13.61, p ¼ < .001). At completion of
treatment (Table 2), overall differences
445
among motivation groups were significant
for ratings of Overall Risk of Suicide (F ¼
10.77, p < .001), Hopelessness (F ¼ 3.94,
p ¼ .026), and Self-Hate (F ¼ 4.83,
p ¼ .012). Unique to Study 2 were signifi-
450
cant findings when using repeated-
measures ANOVAs to compare construct
ratings across test times by motivation
group. Significant interactions were found
between test time and motivation group
455
for Overall Risk of Suicide (F ¼ 4.31,
p ¼ .019) and Self-Hate (F ¼ 4.40, p ¼ .018).
A post-hoc Bonferroni test found
significant differences in Overall Risk of
Suicide ratings at intake between the WTL
460
(M ¼ 1.76) group and both the AMB
(M ¼ 3.04) group (p < .001) and the WTD
(M ¼ 3.33) group (p ¼ .001). Post-hoc com-
parisons were run using Tamhane’s T2 test
for Overall Risk of Suicide at treatment
465
completion due to a significant Levene’s
test (F ¼ 9.91, p < .001), and a significant
difference was found between the WTL
(M ¼ 1.12) and AMB (M ¼ 2.15) groups
(p ¼ .014). For ratings of Hopelessness
470
at completion of treatment, a post-hoc
Bonferroni test found a significant difference
TABLE 1. Study 1
Ratings at treatment completion (Mean)
SSF core construct Life-M Ambiv Death-M
Psychological Pain

2.44 1.89 3.57
Stress 2.67 2.33 3.00
Agitation 2.61 1.89 3.43
Hopelessness 2.31 1.56 3.14
Self-Hate

2.42 1.89
a
3.57
a
Overall Risk of Suicide

1.31
a
1.33 2.14
a
Note. Groups presented in this table were derived by subtracting the total number of Reasons For Dying
responses from the total number of Reasons For Living responses for each subject.
a
Significant difference from one other group within the same Suicide Status Form (SSF) Core Construct.

Overall model significant at p < .05.
Assessing and Treating Different Suicidal States
6 VOLUME 17

NUMBER 3

2013

between the WTL (1.91) and the AMB
(3.00) groups (p ¼ .039). Significant
differences in Self-Hate ratings at treatment
475
completion between the WTL (2.09) and
the AMB (3.31) groups (p ¼ .012) were
also found after conducting a post-hoc
Bonferonni test.
DISCUSSION
480
Both methodologies described were used in
an attempt to categorize suicidal individuals
and to determine whether such categoriza-
tions were able to detect differences in
initial assessments and in treatment out-
485
comes. The first study categorized subjects
by suicidal motivation based on qualitative
written Reasons for Living and Reasons
for Dying as recorded at intake. Results
indicated that self-rated Overall Risk of
490
Suicide at treatment completion increases
in intensity across the three constructs,
from the Life-motivated to Ambivalent to
Death-motivated groups. For the other five
SSF Core Constructs (Psychological Pain,
495
Stress, Agitation, Hopelessness, and Self-
Hate), subjects in the Ambivalent group gave
the lowest ratings at treatment completion,
whereas those in the Death-motivated group
gave the highest ratings. This suggests that
500
the magnitude of emotional upset within
subjects in the different motivation groups
does not mirror their own interpretations
of suicidal risk. This finding supports
results from O’Connor, Jobes, Lineberry,
505
and Bostwick (2010) suggesting that, in
addition to the magnitude of emotional
upset, the specific ways in which this upset
is experienced in suicidal individuals could
also be indicative of risk. Also suggested
510
by these results is the protective role of
Reasons for Living both when they pre-
dominate and whey they are serving as a
buffer against an equal number of Reasons
for Dying. This interpretation is aligned
515
with findings from Linehan, Goodstein,
Nielsen, and Chiles (1983) indicating that
self-assessment of suicidal risk is negatively
correlated with a connection to reasons for
staying alive. Similarly, Jobes and Mann
520
(1999) posited that Reasons for Living
could serve as a critical focus point for
clinicians when treating suicidal patients.
Study 2 categorized subjects by the nat-
ure of their internal struggle using quantitat-
525
ive ratings of their Wish to Live vs. their
Wish to Die as recorded at intake. For
Overall Risk of Suicide, ratings at intake
followed the same progression as those at
TABLE 2. Study 2
Ratings at treatment completion (Mean)
SSF core construct WTL AMB WTD
Psychological Pain 2.21 3.00 3.00
Stress 2.42 3.08 3.00
Agitation 2.27 3.31 2.83
Hopelessness

1.91
a
3.00
a
2.83
Self-Hate

2.09
a
3.31
a
2.83
Overall Risk of Suicide

1.12
a
2.15
a
1.50
Note. Groups presented in this table were derived by calculating a Suicide Index Score (subtracting
the converted Wish to Die score from the converted Wish to Live score) for each subject.
a
Significant difference from one other group within the same Suicide Status Form (SSF) Core
Construct.

Overall model significant at p < .05.

Overall model significant at p < .001.
C. D. Corona et al.
ARCHIVES OF SUICIDE RESEARCH 7

treatment completion in Study 1 (i.e., an
530
increase in intensity as one either is domi-
nated by Reasons for Dying or expresses
more of a Wish to Die). At treatment com-
pletion in Study 2, however, subjects in the
Ambivalent group gave the highest ratings,
535
while subjects in the Wish-to-Live group
gave the lowest. For the other five SSF Core
Constructs, ratings at treatment completion
mirrored this trend (ratings were lowest in
the Wish-to-Live group and highest in the
540
Ambivalent group), which corresponds
with results found by O’Connor, Jobes,
Yeargin et al. (2011) suggesting that
Ambivalent patients fare the worst on cer-
tain measures related to suicidality. Also
545
unique to Study 2 was a significant interac-
tion between motivation group and assess-
ment time when measuring Overall Risk
of Suicide. At intake, the Wish-to-die group
provided the highest ratings, whereas the
550
Ambivalent group rated this construct high-
est at treatment completion. This finding
suggests that these two groups responded
differently to treatment. More specifically,
the implication is that suicidal risk in the
555
Ambivalent group dissipated less over time
than it did in the Wish-to-die group.
A possible interpretation of the results
from Study 2 suggests that a Wish to Die
is most associated with risk when operating
560
in conjunction with a Wish to Live of the
same magnitude, and that risk as measured
by these items is mitigated as the Wish to
Die decreases and the Wish to Live
increases. This interpretation seems to con-
565
tradict the Internal Struggle Hypothesis by
suggesting that the existence of a debate
between the Wish to Live and the Wish to
Die could be a risk factor for suicide in
and of itself. This implication is supported
570
by findings from Harris, McLean, Sheffield,
and Jobes (2010) indicating that, among
survey respondents, the vast majority
(94.5%) of those most suicidal reported
experiencing a debate between the wish to
575
live and the wish to die. These findings also
suggest that the presence of such a debate is
correlated with increased suicidal risk.
While the exact mechanism for contradic-
tion with the internal struggle hypothesis
580
is not clear, it is worth addressing the
potential existence of a sample more at risk
for a non-fatal suicide attempt than for a
completed suicide. Existing literature sug-
gests the relatively low lethality of suicide
585
attempts made by females compared to
those made by men (Moscicki, 1994). Given
that the sample studied here was predomi-
nantly female (73.1%), it could be posited
that the ambivalence expressed by this
590
group is potentially more indicative of risk
for a non-fatal suicide attempt.
In general, suicidal risk as it pertains to
the sample studied was highest in those
dominated by Reasons for Dying and strug-
595
gling with the debate between a Wish to
Live and a Wish to Die. This conclusion
further validates existing literature suggest-
ing the protective potential of Reasons for
Living (Linehan, Goodstein, Neilen et al.,
600
1983; Jobes & Mann, 1999; Malone,
Oquendo, Haas et al., 2000; Lizardi, Currier,
Galfalvy et al., 2007). Furthermore, the
methodologies presented here highlight
means through which these constructs can
605
be efficiently assessed in a clinical setting,
and subsequently written into treatment
plans as an important focus. For example,
a possible interpretation suggests that an
emphasis on cultivating Reasons for Living
610
could not only buffer against the presence
of Reasons for Dying, but could also serve
as a potential avenue for mitigating both a
proclivity to death and the internal struggle
experienced by ambivalent patients. As the
615
evidence base grows regarding acute risk
factors for suicide, developing methods
for translating these findings into effective
clinical practice remains imperative.
Given the differences in results, it is
620
reasonable to posit that the two methodol-
ogies presented here could be measuring
overlapping but still separate constructs as
it pertains to what is driving suicidal desire,
and that each construct could provide
Assessing and Treating Different Suicidal States
8 VOLUME 17

NUMBER 3

2013

625
information that is useful in a clinical
context. It is worth nothing that one
fundamental difference between these two
constructs is the means through which data
pertaining to each is collected. Reasons for
630
Living and Reasons for Dying are both
reported qualitatively by patients, while
the Wish to Live and the Wish to Die are
both reported quantitatively. While the
methodology used in Study 1 presented a
635
means through which to quantify Reasons
for Living and Reasons for Dying, it is
not clear what might be lost in such a trans-
lation or what the implications of such a
loss might be for comparison of these con-
640
structs. Given the widespread collection of
both qualitative and quantitative data in
clinical settings, further research should
continue to investigate the ways in which
these types of data interact with specific
645
emphasis on assessing suicide risk. Such
research should also include special atten-
tion to known differences among those
are likely to make non-fatal suicide attempts
versus those who are likely to complete sui-
650
cide, with an emphasis on the ways in
which risk for different outcomes might
present similarly in clinical settings.
Limitations of the studies include the
relatively small sample size (limiting statisti-
655
cal power) and a skewed distribution of
subjects across motivation categories.
Specifically with regard to statistical power,
it is possible that a larger sample size would
have yielded additional differences in SSF
660
Core Construct ratings at intake across
studies. It is important to differentiate
between an underpowered study and the
conceptual implication that the groups
studied do not differ at intake. A larger
665
sample size would be necessary to further
elucidate the nature of these differences.
Taken from outpatient community men-
tal health centers, the sample analyzed was
weighted heavily towards Life-motivated
670
and Wish-to-live subjects. Furthermore,
existing evidence suggests that previous hos-
pitalization for a major psychiatric illness is
among the most predominant risk factors
for suicide in both men and women in
675
Denmark (Qin, Agerbo, Westergard-Nielsen
et al., 2000). Our studies, in turn, excluded
those who would likely present with the
highest risk for suicide, which has the poten-
tial to skew results pertaining to the assess-
680
ment of such risk. Additionally, a lack of
background information pertaining to sub-
jects’ history of suicidal ideation and beha-
vior before referral limited the ability to
control for potential confounding factors in
685
this regard. These factors limit the external
validity of the studies presented.
There is an evidence base that supports
the mitigation of suicidal risk both through
the development of meaningful connec-
690
tions to Reasons for Living and through a
reduction in the Wish to Die, however the
explicit connection between these avenues
is not fully understood. Further research
into both the nature of these constructs as
695
they are measured today and their interplay
is necessary before the nuances of the suici-
dal mind can be fully understood, and
before such knowledge can provide maxi-
mal clinical benefit to those most in need.
700
AUTHOR NOTE
The authors have no potential conflicts of
interest to disclose. Additionally, no finan-
cial agreements or affiliations exist with
any institution, product, or service that
705
could be interpreted as influencing this
research, nor is there any bias expressed
against another institution, product, or
service.
Christopher D. Corona and David A.
710
Jobes, The Catholic University of America,
Washington, D.C., USA.;
Ann C. Nielsen, Psykiatrisk Center
København, Kompetencecenter for Selv-
mordsforebyggelse, Copenhagen, Denmark.
715
Christian M. Pedersen, Klinik for
Selvmordsforebyggelse, Aarhus Universi-
tetshospital Risskov, Aarhus, Denmark.
C. D. Corona et al.
ARCHIVES OF SUICIDE RESEARCH 9

Keith W. Jennings, Rene´ M. Lento, and
Katherine A. Brazaitis, The Catholic Uni-
720
versity of America, Washington, D.C., USA.
Correspondence concerning this article
should be addressed to Christopher D.
Corona, M.A. Department of Psycholo gy,
The Catholic University of America, O’Boyle
725
Hall, Room 314 Washington, DC 20064.
E-mail: 18corona@cardinalmail.cua.edu
REFERENCES
Arkov, K., Rosenbaum, B., Christiansen, L., Jonsson,
H., & Munchow, M. (2008). Treatment of suicidal
730
patients: The collaborative assessment and
management of suicidality. Ugeskrift for Laeger,
170(3), 149–153.
Beck, A. T., Kovacs, M., & Weissman, A. (1979).
Assessment of suicidal intention: The scale for
735
suicide ideation. Journal of Consulting and Clinical
Psychology, 47(2), 343–352.
Brown, G. K., Steer, R. A., Henriques, G. R., & Beck,
A. T. (2005). The internal struggle between the
wish to die and the wish to live: A risk factor for
740
suicide. The American Journal of Psychiatry, 162(10),
1977–1979. doi:10.1176/appi.ajp.162.10.1977
Centers for Disease Control, & Prevention, National
Center for Injury Prevention, & Control. (2010).
Web-based Injury Statistics Query and Reporting
745
System (WISQARS): Leading causes of death
1999–2009, National or regional. Retrieved April
15, 2012, from http://webappa.cdc.gov/sasweb/
ncipc/leadcaus10_us.html
Comtois, K. A., Jobes, D. A., S. O’Connor, S., Atkins,
750
D. C., Janis, K., E. Chessen, C., ...Yuodelis-
Flores, C. (2011). Collaborative assessment and
management of suicidality (CAMS): Feasibility trial
for next-day appointment services. Depression and
Anxiety, 28(11), 963–972. doi:10.1002/da.20895
755
Conrad, A. K., Jacoby, A. M., Jobes, D. A., Lineberry,
T. W., Shea, C. E., Arnold Ewing, T. D., .. . Kung,
S. (2009). A psychometric investigation of the
suicide status form II with a psychiatric inpatient
sample. Suicide & Life-Threatening Behavior, 39(3),
760
307–320. doi:10.1521/suli.2009.39.3.307
Ellis, T. E., Green, K. L., Allen, J. G., Jobes, D. A., &
Nadorff, M. R. (2012). Collaborative assessment
and management of suicidality in an inpatient set-
ting: Results of a pilot study. Psychotherapy, 49(1),
765
72–80. doi:10.1037/a0026746
Field, A. (2009). Discovering statistics using SPSS.London,
England: Sage Publications.
Harris, K. M., McLean, J. P., Sheffield, J., & Jobes,
D. (2010). The internal suicide debate hypothesis:
770
Exploring the life versus death struggle. Suicide &
Life-Threatening Behavior, 40(2), 181–192. doi:10.1521/
suli.2010.40.2.181
Jaccard, J., Becker, M. A., & Wood, G. (1984).
Pairwise multiple comparison procedures: A review.
775
Psychological Bulletin, 96(3), 589–596. doi:10.1037/
0033-2909.96.3.589
Jobes, D. A. (2006). Managing suicidal risk: A collabora-
tive approach. New York, NY: Guilford Press.
Jobes, D. A., Jacoby, A. M., Cimbolic, P., & Hustead,
780
L. A. T. (1997). Assessment and treatment of
suicidal clients in a university counseling center.
Journal of Counseling Psychology, 44(4), 368–377.
doi:10.1037/0022-0167.44.4.368
Jobes, D. A., Kahn-Greene, E., Greene, J. A., &
785
Goeke-Morey, M. (2009). Clinical improvements
of suicidal outpatients: Examining suicide status
form responses as predictors and moderators.
Archives of Suicide Research, 13(2), 147–159.
doi:10.1080/13811110902835080
790
Jobes, D. A., & Mann, R. E. (1999). Reasons for
living versus reasons for dying: Examining the
internal debate of suicide. Suicide and Life-Threatening
Behavior, 29(2), 97–104.
Jobes, D. A., Nelson, K. N., Peterson, E. M., Pentiuc,
795
D., Downing, V., Francini, K., & Kiernan, A.
(2004). Describing suicidality: An investigation of
qualitative SSF responses. Suicide & Life-Threatening
Behavior, 34(2), 99–112. doi:10.1521/suli.34.2.99.
32788
800
Jobes, D. A., Wong, S. A., Conrad, A. K., Drozd,
J. F., & Neal-Walden, T. (2005). The collaborative
assessment and management of suicidality versus
treatment as usual: A retrospective study with
suicidal outpatients. Suicide and Life-Threatening
805
Behavior, 35(5), 483–497. doi:10.1521/suli.2005.
35.5.483
Kovacs, M., & Beck, A. T. (1977). The wish to die
and the wish to live in attempted suicides. Journal
of Clinical Psychology, 33(2), 361–365.
810
Linehan, M. M., Goodstein, J. L., Nielsen, S. L., &
Chiles, J. A. (1983). Reasons for staying alive when
you are thinking of killing yourself: The reasons
for living inventory. Journal of Consulting and Clinical
Psychology, 51(2), 276–286.
815
Lizardi, D., Currier, D., Galfalvy, H., Sher, L., Burke,
A., Mann, J., & Oquendo, M. (2007). Perceived
reasons for living at index hospitalization and
Assessing and Treating Different Suicidal States
10 VOLUME 17

NUMBER 3

2013

future suicide attempt. The Journal of Nervous and
Mental Disease, 195(5), 451–455. doi:10.1097/
820
NMD.0b013e3180522661
Malone, K. M., Oquendo, M. A., Haas, G. L., Ellis,
S. P., Li, S., & Mann, J. J. (2000). Protective factors
against suicidal acts in major depression: Reasons
for living. The American Journal of Psychiatry, 157(7),
825
1084–1088.
Moscicki, E. K. (1994). Gender differences in
completed and attempted suicides. Annals of
Epidemiology, 4(2), 152–158.
Nielsen, A. C., Alberdi, F., & Rosenbaum, B. (2011).
830
Collaborative assessment and management of
suicidality method shows effect. Danish Medical
Bulletin, 58(8), A4300.
Nock, M. K., Borges, G., Bromet, E. J., Alonso, J.,
Angermeyer, M., Beautrais, A., ...Williams, D.
835
(2008). Cross-national prevalence and risk factors
for suicidal ideation, plans and attempts. British
Journal of Psychiatry, 192(2), 98–105. doi:10.1192/
bjp.bp.107.040113
Nordentoft, M. (2007). Prevention of suicide and
840
attempted suicide in Denmark: epidemiological
studies of suicide and intervention studies in
selected risk groups. Danish Medical Bulletin, 54(4),
306–369.
O’Connor, S. S., Jobes, D. A., Lineberry, T. W., &
845
Michael Bostwick, J. (2010). An investigation
of emotional upset in suicide ideation. Archives of
Suicide Research, 14(1), 35–43. doi:10.1080/
13811110903479029
O’Connor, S. S., Jobes, D. A., Yeargin, M. K.,
850
Fitzgerald, M. E., Rodriguez, V. M., Conrad, A.
K., & Lineberry, T. W. (2011). A cross-sectional
investigation of the suicidal spectrum: Typologies
of suicidality based on ambivalence about living
and dying. Comprehensive Psychiatry, doi:10.1016/
855
j.comppsych.2011.09.007
Qin, P., Agerbo, E., Westergard-Nielsen, N.,
Eriksson, T., & Mortensen, P. B. (2000). Gender
differences in risk factors for suicide in Denmark.
The British Journal of Psychiatry: The Journal of Mental
860
Science, 177, 546–550.
Substance Abuse and Mental Health Services Admini-
stration, Office of Applied Studies. (2009). The
NSDUH Report: Suicidal thoughts and behaviors
among adults. Retrieved April 15, 2012, from
865
http://oas.samhsa.gov/2k9/165/suicide.cfm
World Health Organization. (2002). The World
Health Report, 2002: Statistical annex 2. Retrieved
April 15, 2012, from http://www.who.int/whr/
2002/annex/en/index.html
C. D. Corona et al.
ARCHIVES OF SUICIDE RESEARCH 11