Content uploaded by Megan L. Rogers
Author content
All content in this area was uploaded by Megan L. Rogers on Mar 20, 2017
Content may be subject to copyright.
Review of General Psychology
Rumination, Suicidal Ideation, and Suicide Attempts: A
Meta-Analytic Review
Megan L. Rogers and Thomas E. Joiner
Online First Publication, March 13, 2017. http://dx.doi.org/10.1037/gpr0000101
CITATION
Rogers, M. L., & Joiner, T. E. (2017, March 13). Rumination, Suicidal Ideation, and Suicide Attempts:
A Meta-Analytic Review. Review of General Psychology. Advance online publication.
http://dx.doi.org/10.1037/gpr0000101
Rumination, Suicidal Ideation, and Suicide Attempts:
A Meta-Analytic Review
Megan L. Rogers and Thomas E. Joiner
Florida State University
Rumination has been implicated as a risk factor for suicidal ideation and attempts, yet the literature to
date has not been synthesized. We conducted a meta-analysis of the association between rumination and
both suicidal ideation and attempts to consolidate the existing literature (k⫽29). Results indicated that
the relationships between global rumination (k⫽13; Hedge’s g⫽.74, p⬍.001, 95% CI [.45, 1.04]),
brooding (k⫽12; Hedge’s g⫽.63, p⬍.001, 95% CI [.35, .90]), and reflection (k⫽12; Hedge’s g⫽
.38, p⫽.002, 95% CI [.10, .65]) with suicidal ideation were significant. Associations between global
rumination (k⫽3; Hedge’s g⫽.26, p⬍.001, 95% CI [.08, .44]) and brooding (k⫽4; Hedge’s g⫽
.47, p⫽.004, 95% CI [.02, .91]) and suicide attempts were significant, but reflection (k⫽4; Hedge’s
g⫽.09, p⫽.646, 95% CI [⫺.54, .72]) was unrelated. However, given the limited studies included in
suicide attempt analyses—and the exclusive use of cross-sectional designs and heterogeneity with regard
to samples and measures—these parameters should be taken with caution. Generally, age, gender,
race/ethnicity, and year of publication were not moderators, and there was little evidence for publication
bias across effects, with the exception of the effect of global rumination on suicidal ideation. Several
future research directions are discussed.
Keywords: rumination, brooding, reflection, suicidal ideation, suicide attempts
Over 800,000 people die by suicide around the world each year
(World Health Organization, 2014), making suicide a global public
health concern. Though substantial suicide-related research has
been conducted, a significant shortcoming of work to date is a
failure to differentiate between risk factors for suicidal ideation
and risk factors for suicidal behaviors (i.e., suicide attempts, death
by suicide; Klonsky & May, 2014; see also Nock, Kessler, &
Franklin, 2016). The majority of identified risk factors for suicide,
including depression, hopelessness, and most psychiatric disor-
ders, relate only to suicidal ideation, as opposed to suicide attempts
or death by suicide specifically (May & Klonsky, 2016). This
highlights a need to identify factors that predict suicidal behaviors
(i.e., suicide attempts, death by suicide).
Cognitive risk factors, particularly rumination (Miranda &
Nolen-Hoeksema, 2007; Smith, Alloy, & Abramson, 2006), have
garnered attention as factors that may increase vulnerability to
suicidal ideation and attempts (Abramson et al., 1998; Ellis, 2006).
Rumination refers to a tendency to repetitively and passively focus
on one’s distress and the causes, meanings, and consequences of
this distress (Nolen-Hoeksema, 1991). Unlike other forms of cop-
ing that result in active problem-solving, individuals who are
ruminating remain fixated on the problems and on their feelings
about them without taking action (Nolen-Hoeksema, Wisco, &
Lyubomirsky, 2008). Rumination has been associated with sui-
cidal ideation (Ahrens & Linden, 1996; Fairweather, Anstey, Rod-
gers, Jorm, & Christensen, 2007; Miranda & Nolen-Hoeksema,
2007) and suicide attempts (Grassia & Gibb, 2009; Surrence,
Miranda, Marroquín, & Chan, 2009), both concurrently and pro-
spectively (see Morrison & O’Connor, 2008 for a systematic
review). In addition to being related to suicidal ideation and
behavior, rumination is also consistently associated with symp-
toms of depression, anxiety, alcohol abuse, and nonsuicidal self-
injury (Nolen-Hoeksema et al., 2008).
Two subsets of ruminative thinking have been proposed: brood-
ing and reflection (Treynor, Gonzalez, & Nolen-Hoeksema, 2003).
Brooding involves a tendency to dwell on the negative conse-
quences of one’s distress, whereas reflection involves attempts to
understand the reasons for one’s distress. Most studies have
found a significant positive association between brooding and
suicidal ideation, cross-sectionally (e.g., Cole et al., 2015;
Tucker et al., 2013) and longitudinally (e.g., Miranda & Nolen-
Hoeksema, 2007; D. B. O’Connor, O’Connor, & Marshall,
2007; R. C. O’Connor & Noyce, 2008); however, findings with
regard to reflection have been mixed. For instance, some studies
found that reflection was predictive of suicidal ideation cross-
sectionally and at 1-year follow-up (e.g., Miranda & Nolen-
Hoeksema, 2007; Tucker et al., 2013), whereas others demon-
strated no association (R. C. O’Connor & Noyce, 2008).
Moreover, Crane, Barnhofer, and Williams (2007) found higher
levels of reflection in never-suicidal individuals than in suicide
Megan L. Rogers and Thomas E. Joiner, Department of Psychology,
Florida State University.
This work was in part supported by the Military Suicide Research
Consortium (MSRC), an effort supported by the Department of Defense
(Grant W81XWH-10-2– 0181). Opinions, interpretations, conclusions, and
recommendations are those of the authors and are not necessarily endorsed
by the Military Suicide Research Consortium or the Department of De-
fense.
Correspondence concerning this article should be addressed to Megan L.
Rogers, Department of Psychology, Florida State University, 1107 West
Call Street, Tallahassee, FL 32306-4301. E-mail: rogers@psy.fsu.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Review of General Psychology © 2017 American Psychological Association
2017, Vol. 0, No. 999, 000 1089-2680/17/$12.00 http://dx.doi.org/10.1037/gpr0000101
1
attempters, suggesting that this association may be more nu-
anced. Nonetheless, there appears to be consensus to date that
rumination and negative repetitive thinking are related to sui-
cidal ideation to at least some degree (Kerkhof & van Spijker,
2011; Morrison & O’Connor, 2008).
It is also worth considering past conceptualizations of the rela-
tionship between rumination and suicidal thoughts and behaviors;
for instance, some types of rumination have been shown to account
for associations between other risk factors and suicidal ideation.
Previous research has demonstrated indirect effects of negative life
events (Chan, Miranda, & Surrence, 2009), emotion dysregula-
tion (Miranda, Tsypes, Gallagher, & Rajappa, 2013), cognitive
vulnerability (Smith et al., 2006) and inflexibility (Miranda,
Valderrama, Tsypes, Gadol, & Gallagher, 2013), perfectionism
(D. B. O’Connor et al., 2007), and self-criticism (R. C.
O’Connor & Noyce, 2008) on suicidal ideation through rumi-
nation. This suggests that rumination may be a proximal factor to
suicidal ideation and an ideal target for clinical intervention to
reduce the frequency and impact of this pernicious thought pattern.
Mediators of the associations between rumination and suicidal
thoughts and behaviors have also been identified, however. Spe-
cifically, depression (Chan et al., 2009), hopelessness (Antypa,
Van der Does, & Penninx, 2010; Smith et al., 2006), and entrap-
ment (Teismann & Forkmann, 2017) have each been found to
account for the link between rumination and suicidal thoughts,
whereas agitation and nightmares have been found to account for
the relations between rumination and both suicidal thoughts and
past suicide attempts (Rogers et al., 2017). Overall, the complexity
of these findings, together, indicates a need for synthesis of the
extant literature to clarify these associations.
However, although rumination has been positively associated
with both suicidal ideation (Miranda & Nolen-Hoeksema, 2007)
and suicide attempts (Grassia & Gibb, 2009), rumination has not
been included in recent meta-analyses of suicide risk factors (May
& Klonsky, 2016). Due to the discrepancy of previous findings, as
well as the need to understand how suicide-related risk factors
confer risk for suicidal ideation versus suicidal behavior, there
exists a need to quantitatively synthesize the existing literature on
facets of rumination and suicidal ideation and behaviors. As such,
the purpose of the present study was to meta-analytically examine
the nature and strength of the relationship between rumination
(including its subsets— brooding and reflection) and suicidal ide-
ation and behaviors. We included retrospective, cross-sectional,
and prospective studies in our review in an effort to determine
whether rumination represents a correlate or risk factor of
suicidal ideation and/or behaviors. A meta-analysis also allows
us to examine whether a number of potential moderator vari-
ables influence the magnitude of these associations; thus, we
aimed to determine whether the strength of the association
between rumination and suicidality was influenced by age,
gender, race/ethnicity, publication year, study design, or sample
type. Moreover, a meta-analysis allows us to examine whether
the findings reported in the literature to date have been influ-
enced by publication bias. Finally, by examining the extant
literature on the association between rumination and suicidal
thoughts/behaviors, we aimed to highlight areas in which future
research is warranted.
Method
Study Selection
We identified studies by searching on PsycINFO and PubMed.
Search terms included “ruminat
ⴱ
AND suicid
ⴱ
,” “brood
ⴱ
AND
suicid
ⴱ
,” and “reflect
ⴱ
AND suicid
ⴱ
.” Using these search terms
enabled us to get a comprehensive list of articles, as the asterisk (
ⴱ
)
includes all words that contain the prefix (e.g., “ruminat
ⴱ
” returns
both “ruminate” and “rumination”; “suicid
ⴱ
” returns both “suicidal
ideation” and “suicide attempts”). Reference sections of relevant
articles were also examined for further sources. Articles published
or available online before March 2016 were included in the meta-
analysis.
A total of 188 unique published reports were identified. Titles
and abstracts of these 188 studies were reviewed by the first author
(MR) and a research assistant to determine eligibility for inclusion;
127 studies clearly did not meet criteria for inclusion. Inclusion
criteria required that studies (a) were original empirical articles
(i.e., not theoretical, systematic reviews, or meta-analyses), (b)
were written in English, (c) included a quantitative measure of
both rumination and suicidality and examined a direct association
between these two variables in the published manuscript, and (d)
included necessary statistical information to calculate effect sizes.
Efforts were made to contact authors in cases where necessary
statistical information was missing. Importantly, studies examin-
ing retrospective, cross-sectional, and prospective associations be-
tween rumination and suicidal ideation and behavior were all
included in the meta-analysis. Studies examining characteristics of
suicidal behavior (e.g., medical lethality) or self-injury without
suicidal intent (i.e., nonsuicidal self-injury), and studies that com-
bined suicidal ideation and behavior into a single outcome (e.g.,
suicide risk) were excluded. Of the 61 remaining studies, 25
studies comprising 29 samples met all eligibility criteria for inclu-
sion (⫽.92). Figure 1 presents the study selection process;
included studies are described in Table 1.
Subsets of the same data were reanalyzed across several publi-
cations meeting inclusion criteria. To ensure independence of
effects, the most inclusive study was retained, and duplicate cases
were removed (n⫽2), resulting in 23 studies comprising 27
unique samples. Several studies also reported multiple effects,
Figure 1. Flowchart for study selection process.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2ROGERS AND JOINER
either because (a) both cross-sectional and longitudinal associa-
tions were tested, or (b) multiple measures of rumination were
utilized. In the former scenario, we aimed to use the most conser-
vative test of the association between rumination and suicidality;
thus, we included the longitudinal effect. In the latter case, we
followed recommendations to choose the most representative ef-
fect from each study to avoid artificial inflation of our effect sizes
(Cooper, 1998)—in the present meta-analysis, we used the Rumi-
native Responses Scale (RRS; Nolen-Hoeksema & Morrow,
1991). When subscales of rumination (Brooding, Reflection) were
included in the study alongside a rumination total score, we used
the subscale scores instead of the total score to provide a more
nuanced examination of the association between different facets of
rumination and suicidality. When multiple effects on the same
measure were reported from multiple comparisons (e.g., attempters
vs. ideators vs. nonsuicidal individuals), we calculated combined
Table 1
Included Study Information
Study
Sample
size
Attempter
(N) Sample type Study design
Follow-up
period
Mean
age
Female
(%)
White
(%)
Predictor
variable
Outcome
variable
Batterham & Christensen
(2012) 6,557 Community Longitudinal 4 years 51.3 Rumination Suicidal ideation
Burke et al. (2015) 324 Community Longitudinal 2 years 12.90 53.5 47.53 Rumination Suicidal ideation
Chan et al. (2009) 1,011 College student Cross-sectional N/A 19.10 34.0 Brooding,
reflection
Suicidal ideation
Cheref et al. (2015) 690 College student Cross-sectional N/A 19.28 73.0 0 Brooding,
reflection
Suicidal ideation
Cole et al. (2015) 360 College student Cross-sectional N/A 19.72 66.4 79.4 Brooding,
reflection
Suicidal ideation
Crane et al. (2007) 32 10 Community Cross-sectional N/A 31.65 65.6 93.75 Brooding,
reflection
Suicidal ideation
Suicide attempts
Eshun (2000)—Sample 1 105 College student Cross-sectional N/A 18.93 59.05 100 Rumination Suicidal ideation
Eshun (2000)—Sample 2 89 College student Cross-sectional N/A 21.61 41.6 0 Rumination Suicidal ideation
Fairweather et al. (2007) 7,313 Community Cross-sectional N/A 52.1 Rumination Suicidal ideation
Fazakas-Dehoog (2008)—
Sample 1 397 College student Cross-sectional N/A 18.69 79.1 Rumination Suicidal ideation
Fazakas-Dehoog (2008)—
Sample 2
61 11 Community Cross-sectional N/A 29.82 70.5 Rumination Suicidal ideation
Outpatient Suicide attempts
Grassia & Gibb (2009) 31 16 Inpatient Cross-sectional N/A 35.61 54.84 83.9 Brooding,
reflection
Suicide attempts
Krajniak et al. (2013) 143 32 College student Longitudinal
a
2–3 years 18.50 79.72 29.0 Rumination Suicidal ideation
Suicide attempts
Lyness et al. (1997) 124 Inpatient Cross-sectional N/A 71.31 Rumination Suicidal ideation
Miranda & Nolen-
Hoeksema (2007)
1,134 Community Longitudinal 1 year 47.80 53.5 72.0 Brooding,
reflection
Suicidal ideation
Niu (2015)—Sample 1 57 Outpatient Cross-sectional N/A 78.85 62.7 0 Brooding,
reflection
Suicidal ideation
Niu (2015)—Sample 2 29 Outpatient Cross-sectional N/A 80.11 85.2 24.14 Brooding,
reflection
Suicidal ideation
O’Connor & Noyce (2008) 153 Community Longitudinal 3 months 27.43 73.3 Brooding,
reflection
Suicidal ideation
Polanco-Roman Jurska,
Quiñones, & Miranda
(2015)
354 17 College student Cross-sectional N/A 19.08 74.0 30.0 Brooding,
reflection
Suicide attempts
Smith et al. (2006) 127 College student Longitudinal 2.5 years 20.05 64.1 63.0 Rumination Suicidal ideation
Stange et al. (2015) 72 College student Longitudinal 3 years 19.72 65.0 Brooding,
reflection
Suicidal ideation
Surrence et al. (2009) 96 37 College student Longitudinal
a
18 days 18.90 76.0 29.0 Brooding,
reflection
Suicidal ideation
Suicide attempts
Teismann & Forkmann
(2015)—Sample 1 142 Community Cross-sectional N/A 35.20 75.4 100 Rumination Suicidal ideation
Teismann & Forkmann
(2015)—Sample 2 226 Outpatient Cross-sectional N/A 36.40 56.6 100 Rumination Suicidal ideation
Tucker et al. (2013) 298 College student Cross-sectional N/A 19.61 69.1 80.5 Brooding,
reflection
Suicidal ideation
Tucker, Wingate, O’Keefe,
Hollingsworth, & Cole
(2015)
140 College student Cross-sectional N/A 21.09 69.3 0 Brooding,
reflection
Suicidal ideation
Yaseen et al. (2012) 183 141 Inpatient Cross-sectional N/A 37.2 37.7 Rumination Suicidal ideation
Suicide attempts
a
The rumination-suicidal ideation association was assessed longitudinally; however, the rumination-suicide attempt association was cross-sectional/
retrospective in nature (i.e., baseline rumination to past attempts).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
3
RUMINATION AND SUICIDALITY META-ANALYSIS
means, standard deviations, and sample sizes of the comparison
group (i.e., ideators and nonsuicidal individuals) to compare
groups that differ primarily on the outcome of interest (i.e., suicide
attempts). Finally, we included measures of both suicidal ideation
and suicide attempts, when available.
Data Extraction and Statistical Analyses
For each sample, data relevant to our meta-analysis was ex-
tracted and coded. When suicide attempts were available as an
outcome variable, we recorded means, standard deviations (SDs),
and sample sizes for each group (i.e., suicide attempt vs. no suicide
attempt). Odds ratios were coded when these data were unavail-
able. For suicidal ideation, we coded Pearson’s r, odds ratios, or
means, SDs, and sample sizes of each group (i.e., suicide ideators
vs. nonsuicidal individuals). Effects were converted to the same
metric (Cohen’s dfor independent groups). We also coded a
number of other variables to examine as potential moderator
variables, including mean sample age, percentage of the sample
that identified as female, percentage of the sample that identified
as White/Caucasian, study design, sample type, and publication
year. These variables were selected due to their potential theoret-
ical relevance to rumination, suicidality, and/or the association
between these constructs (e.g., age: Centers for Disease Control
and Prevention, 2016; Ricarte, Ros, Serrano, Martínez-Lorca, &
Latorre, 2016; gender: Eshun, 2000; Johnson & Whisman, 2013;
race/ethnicity: Cheref, Lane, Polanco-Roman, Gadol, & Miranda,
2015). Following recommendations presented by Borenstein,
Hedges, Higgins, and Rothstein (2009), subgroups for categorical
moderator analysis had to include at least six effect sizes; as such,
study design and sample type were unable to be included in
moderation analyses. When more than three studies of each study
design were available for a particular analysis, we conducted
follow-up sensitivity analyses to determine whether effects dif-
fered depending on cross-sectional or longitudinal design.
Effect sizes were then converted to Hedge’s gfor independent-
group designs to calculate the standardized mean difference on
suicide-related outcomes. Hedge’s gwas used in place of Cohen’s
dbecause Cohen’s dslightly overestimates the population effect
size in small samples, whereas Hedge’s gis an unbiased correc-
tion. Effect sizes were weighted, and we examined the significance
of the Q-test to determine whether substantial heterogeneity ex-
isted across effect sizes. We also report the I
2
statistic, which
indicates how much of the variability across studies results from
heterogeneity as opposed to chance due to sampling error. Values
from 0 –25% represent low heterogeneity, 26 –50% medium het-
erogeneity, and 51–100% large heterogeneity (Higgins, Thomp-
son, Deeks, & Altman, 2003). Because we expected that effect
sizes would vary across studies due to the diverse study samples
and designs, we used a random effects model. Systematic between-
study variance (i.e., heterogeneity) is common and accounted for
in a random effects model in the weighting and calculation of each
case. Effect sizes were interpreted according to standards outlined
by Cohen (1988), such that approximately .2 represents a small
effect, .5 a moderate effect, and .8 a large effect.
Moderator analyses were conducted to determine whether vari-
ation in effect sizes depended on selected covariates. Specifically,
we employed a random effects meta-regression to examine the
associations between Hedge’s geffect sizes and each potential
moderator to determine the unique role of each of these factors,
with corrections made to standard errors and significance testing.
As noted previously, the following moderators were examined:
age, gender, race/ethnicity, and publication year.
Finally, because significant findings are more likely to be pub-
lished, multiple indicators of publication bias were assessed.
Namely, we calculated Egger’s test and Fail-safe N. We also report
the number of studies with nonsignificant effects that would be
required to result in nonsignificant findings.
Results
Study Description
A total of 23 studies and 27 unique study samples were included
in the present meta-analysis (see Table 1 for study characteristics).
The earliest study was published in 1997 (Lyness, Conwell, King,
Cox, & Caine, 1997), and the most recent studies (n⫽9) were
published in 2015. Seventeen studies were cross-sectional in de-
sign, whereas eight studies were longitudinal. Follow-up periods
ranged from 18 days to 4 years (M⫽26.16 months, SD ⫽15.82).
Across these studies, there were 20,673 participants with a mean
age of 30.08 years (SD ⫽18.95). On average, 64.09% (SD ⫽
11.88) of each sample identified as female. There were 12 samples
of college students, seven general community samples, two sam-
ples of psychiatric outpatients, four samples of psychiatric inpa-
tients, and two mixed community/outpatient samples. Most studies
examined suicidal ideation as an outcome (n⫽26), whereas 10
studies examined suicide attempts; seven studies examined both
suicidal ideation and attempts.
The vast majority of studies used the RRS (Nolen-Hoeksema &
Morrow, 1991) to assess rumination, with one study using the
Children’s Response Styles Questionnaire (Abela, Aydin, & Au-
erbach, 2007), one using the Perseverative Thinking Questionnaire
(Ehring et al., 2011), one using the Ruminative Flooding subscale
of the Suicide Triggers Scale–3 (Yaseen, Gilmer, Modi, Cohen, &
Galynker, 2012), and one assessing rumination via clinical obser-
vation (Lyness et al., 1997). A wide variety of measures of suicidal
ideation were utilized, including the Beck Scale for Suicide Ide-
ation (Beck & Steer, 1991), Columbia Suicide Severity Rating
Scale (Posner et al., 2011), Depressive Symptoms Inventory—
Suicidality Subscale (Metalsky & Joiner, 1997), Geriatric Suicidal
Ideation Scale (Heisel & Flett, 2006), Item 9 of the Beck Depres-
sion Inventory—II (Beck, Steer, & Brown, 1996), Item 9 of the
Children’s Depression Inventory (Kovacs, 1985), Item 9 of the
Patient Health Questionnaire (Kroenke, Spitzer, & Williams,
2001), Self-Harm Behavior Questionnaire (Gutierrez, Osman, Bar-
rios, & Kopper, 2001), Structured Clinical Interview for DSM–IV
Disorders (First, Spitzer, Gibbon, & Williams, 2002), Item 1 of the
Suicide Behaviors Questionnaire (Osman et al., 2001), Suicide
Behaviors Screening (Krajniak, Miranda, & Wheeler, 2013), Sui-
cide Probability Scale (Cull & Gill, 1988), Suicidal Ideation Ques-
tionnaire (Reynolds, 1991), suicidal ideation item of the Psychi-
atric Symptom Frequency Scale (Lindelow, Hardy, & Rodgers,
1997), suicidal ideation item of the Hamilton Depression Rating
Scale (Hamilton, 1960), clinical interviews, and individual items
not associated with a validated scale.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4ROGERS AND JOINER
Suicidal Ideation Prediction and Publication Bias
Rumination. A total of 13 studies comprising 15,907 unique
participants examined the relationship between global rumination
and suicidal ideation. Six samples were composed of college
students, four of community members, and four of psychiatric
patients. Nine studies were cross-sectional, and four were longi-
tudinal. Effect sizes and related 95% CIs are presented in a forest
plot (Figure 2). The Q-test was significant (226.77), and the I
2
value (94.71) indicated that a large amount of variability was due
to heterogeneity rather than chance. The test of the null was
significant, with a large effect size (Hedge’s g⫽.74, p⬍.001,
95% CI [.45, 1.04]). This indicates that the association between
global rumination and suicidal ideation is significant and large in
magnitude. Follow-up sensitivity analyses indicated that the effect
for cross-sectional studies was significant and large in magnitude
(Hedge’s g⫽.82, p⬍.001, 95% CI [.40, 1.24]). Similarly, the
longitudinal effect was significant and large in magnitude
(Hedge’s g⫽.62, p⬍.001, 95% CI [.17, 1.08]), though slightly
smaller than the cross-sectional effect. Neither age (Z⫽⫺.76, p⫽
.448), gender (Z⫽1.11, p⫽.266), race/ethnicity (Z⫽1.19, p⫽
.234), nor year of publication (Z⫽.16, p⫽.872) moderated this
association. Egger’s regression test demonstrated the presence of
publication bias (B⫽5.75, t(1,11) ⫽4.57, p⫽.001); however, the
Fail-safe Ntest indicated that 1,107 studies with an effect size of
0 would be required to bring the overall average weighted effect
size to a nonsignificant value.
Brooding. Twelve studies, composed of 3,051 unique partic-
ipants, tested the association between brooding and suicidal ide-
ation. Seven studies were composed of college students, whereas
three were community samples and two were composed of psy-
chiatric patients. Eight studies were cross-sectional, and four were
longitudinal. Effect sizes and 95% CIs are displayed in Figure 3.
The Q-test was significant (37.24), and the I
2
value (70.46) indi-
cated that a large amount of variability was due to heterogeneity as
opposed to chance. Hedge’s gwas equal to .63 (p⬍.001, 95% CI
[.35, .90]), suggesting that the relationship between brooding and
suicidal ideation is significant and moderate-to-large in magnitude.
Follow-up sensitivity analyses indicated that both the cross-
sectional (Hedge’s g⫽.66, p⬍.001, 95% CI [.27, 1.05]) and
longitudinal (Hedge’s g⫽.57, p⫽.007, 95% CI [.03, 1.11])
effects were significant and moderate-to-large in magnitude. Age
was a significant moderator of this relationship (Z⫽2.24, p⫽
.026), such that the association between brooding and suicidal
ideation was strongest at older ages. Neither gender (Z⫽.97, p⫽
.332), race/ethnicity (Z⫽⫺1.69, p⫽.091), nor publication year
(Z⫽.37, p⫽.712) moderated this relationship. Egger’s regres-
sion test did not indicate publication bias (B⫽1.46, t(1,10) ⫽
1.68, p⫽.124), and the Fail-safe Nsuggested that a total of 761
studies with an effect size of 0 would be required to reduce the
average weighted effect to a nonsignificant value.
Reflection. A total of 12 studies with 3,051 participants ex-
amined the association between reflection and suicidal ideation.
Similar to the brooding and suicidal ideation association, seven
studies were composed of college students, whereas three were
community samples and two were composed of psychiatric pa-
tients. Eight studies were cross-sectional, and four were longitu-
dinal. Figure 4 presents effect sizes and 95% CIs for all effects.
Heterogeneity was high (Q⫽31.71, I
2
⫽65.31). The test of the
null was significant (Hedge’s g⫽.38, p⫽.002, 95% CI [.10, .65])
and indicated that the relationship between reflection and suicidal
ideation is significant and small-to-moderate in magnitude.
Follow-up sensitivity analyses indicated that the association be-
tween reflection and suicidal ideation was nonsignificant cross-
sectionally (Hedge’s g⫽.33, p⫽.067, 95% CI [⫺.10, .76]), but
was significant and moderate in magnitude longitudinally
(Hedge’s g⫽.46, p⫽.002, 95% CI [.04, .87]). Neither age (Z⫽
.31, p⫽.756), gender (Z⫽1.29, p⫽.198), race/ethnicity
(Z⫽⫺.52, p⫽.603), nor publication year (Z⫽1.10, p⫽.272)
moderated the magnitude of the association between reflection and
suicidal ideation. Egger’s regression test did not indicate signifi-
cant publication bias (B⫽1.91, t(1,10) ⫽2.18, p⫽.055); the
Fail-safe Ntest indicated that 330 studies with an effect size of 0
would be required to reduce the average weighted effect to a
nonsignificant value.
Suicide Attempts Prediction and Publication Bias
Rumination. Three studies with 362 unique participants ex-
amined the relationship between global rumination and suicide
attempts. All three studies were cross-sectional; two examined
psychiatric patients, and one examined college students. Figure 5
displays effect sizes and related 95% CIs. The Q-test (3.43) and I
2
value (41.65) indicated that a moderate amount of the variability
was due to heterogeneity rather than chance. The test of the null
was significant (Hedge’s g⫽.26, p⬍.001, 95% CI [.08, .44]).
This indicates that the association between global rumination and
suicide attempts is significant and small in magnitude. Publication
year (Z⫽.00, p⫽1.00) did not moderate this association; there
were not enough data to examine age, gender, or race/ethnicity as
moderators. With regard to publication bias, Egger’s regression
test was nonsignificant (B⫽.18, t(1,1) ⫽2.01, p⫽.294), but the
Figure 2. Forest plot of all effect sizes included in the rumination-suicidal ideation meta-analysis.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
5
RUMINATION AND SUICIDALITY META-ANALYSIS
Fail-safe Nindicated that only two studies with an effect size of 0
would be required to bring the average weighted effect to a
nonsignificant value.
Brooding. Four studies with 481 unique participants have
tested the association between brooding and suicide attempts. All
four studies were cross-sectional; two were college student sam-
ples, one was a community sample, and one was a psychiatric
patient sample. Effect sizes and related 95% CIs are displayed in
Figure 6. The Q-test was significant (57.76), and the I
2
value
(94.81) indicated that a large amount of the variability was due to
heterogeneity rather than chance. Hedge’s gwas equal to .47
(Hedge’s g⫽.47, p⫽.004, 95% CI [.02, .91]). This indicates that
the relationship between brooding and suicide attempts is positive
and moderate in magnitude. Neither age (Z⫽⫺.02, p⫽.984),
gender (Z⫽.03, p⫽.976), race/ethnicity (Z⫽⫺.05, p⫽.960),
nor publication year (Z⫽.02, p⫽.984) moderated this associa-
tion. Egger’s regression test did not suggest the influence of
publication bias (B⫽⫺.93, t(1,2) ⫽⫺2.24, p⫽.155); the
Fail-safe Ntest indicated that 70 studies with an effect size of 0
would be required to bring the overall average weighted effect to
a nonsignificant value.
Reflection. A total of four studies comprising 481 unique
participants examined the association between reflection and sui-
cide attempts. All four studies were cross-sectional; two were
college student samples, one was a community sample, and one
was a psychiatric patient sample. Effect sizes and 95% CIs are
presented in Figure 7. A large amount of the variability was due to
heterogeneity rather than chance (Q⫽13.36, I
2
⫽77.54). The test
of the null was nonsignificant (Hedge’s g⫽.09, p⫽.646, 95% CI
[⫺.54, .72]). This suggests that reflection and suicide attempts are
not significantly related to each other. Age (Z⫽⫺2.60, p⫽.009),
race/ethnicity (Z⫽⫺2.71, p⫽.007), and publication year (Z⫽
2.34, p⫽.019) were significant moderators, such that the rela-
tionship between reflection and suicide attempts was stronger at
younger ages, those who did not identify as White/Caucasian, and
in later publications. Gender (Z⫽1.37, p⫽.171) was not a
significant moderator. Regarding publication bias, Egger’s regres-
sion test was nonsignificant (B⫽3.72, t(1,2) ⫽2.71, p⫽.114),
and the Fail-safe Nindicated that three studies with an effect size
of 0 would be required to bring the average weighted effect to 0.
Discussion
The present meta-analysis aimed to synthesize the existing
literature on rumination and suicidality, with the goal of identify-
ing the magnitude of effects of specific types of rumination on
both suicidal ideation and attempts. Results indicated that global
rumination, brooding, and reflection were each associated with
suicidal ideation, whereas only global rumination and brooding
were associated with suicide attempts. Effect sizes ranged from
moderate to large and were typically stronger in the prediction of
suicidal ideation, as opposed to suicide attempts. Despite previous
research finding age (Ricarte et al., 2016), gender (Johnson &
Whisman, 2013), and racial/ethnic (Cheref et al., 2015) differences
Figure 3. Forest plot of all effect sizes included in the brooding-suicidal ideation meta-analysis.
Figure 4. Forest plot of all effect sizes included in the reflection-suicidal ideation meta-analysis.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
6ROGERS AND JOINER
in rumination, effects were generally consistent across ages, gen-
ders, and race/ethnicity, with the exception of age moderating the
association between brooding and suicidal ideation. The influence
of publication bias was minimal; publication bias was only indi-
cated in the association between global rumination and suicidal
ideation. However, due to the small number of studies included in
several analyses, particularly those involving suicide attempts,
interpretations should be considered with caution.
Overall, synthesis of studies examining the association between
rumination and suicidal ideation yielded moderate-to-large effect
sizes, with the global rumination effect being largest in magnitude
(Hedge’s g⫽.74). In interpreting this effect, however, it is
important to consider the measurement source: the majority of
studies utilized the 22-item RRS (Nolen-Hoeksema & Morrow,
1991). Factor analysis of the 22-item RRS indicated that 12 items
were conflated with depressive symptoms, and that, instead, the
five-item Brooding and Reflection subscales were nonredundant
with depression and appropriate for use as measures of rumination
(Treynor et al., 2003). Given recent evidence that suggests strong
conceptual and empirical covariation between depression and sui-
cidal ideation (Rogers et al., 2016), the magnitude of the global
rumination and suicidal ideation association may be artificially
inflated. Nevertheless, that the effect between brooding and sui-
cidal ideation was moderate-to-large in magnitude (Hedge’s g⫽
.63) suggests the robustness of this association.
Further, the difference in the magnitude of the effects of brood-
ing (.63) and reflection (.38) on suicidal ideation suggests impor-
tant differences in how risk may be conferred for suicidal ideation.
Items comprising the Brooding subscale capture the construct of
passively pondering one’s mood and situations, whereas items
comprising the Reflection subscale are neutrally valenced, focus
on contemplation, and, importantly, capture the concept of coping
with one’s problems (Treynor et al., 2003). That both brooding and
reflection demonstrate positive associations with suicidal ideation
indicate that continuous, repetitive thinking about one’s mood or
problems may result in negative outcomes; however, important
differences exist between passively focusing on one’s mood and
problems as opposed to problem-solving and reflection—with the
former being a more pernicious form of overthinking.
Interestingly, global rumination and brooding were associated
with suicidal ideation both cross-sectionally and longitudinally;
however, reflection was only significantly associated with suicidal
ideation longitudinally. It may be possible that individuals with
suicidal ideation engage in reflection less often when experiencing
suicidal ideation (rather, brooding or general overthinking); how-
ever, contemplating one’s problems and emotions over time may
lead to the development of future suicidal ideation. Namely, re-
flective rumination may contribute to suicidal ideation in the long
term. As suggested by Miranda and Nolen-Hoeksema (2007), the
relation between reflection and suicidal ideation might be moder-
ated by other factors, such as poor coping or problem-solving
abilities (Joormann, Dkane, & Gotlib, 2006; Watkins & Moulds,
2005). Given that most of the literature examining mechanisms of
the rumination and suicidal ideation to date has focused on global
rumination or brooding, future research that investigates the nu-
ances and potential mechanisms of the association between reflec-
tion and suicidal ideation, particularly over time, would be infor-
mative. Alternatively, large standard errors may have precluded a
statistically significant cross-sectional effect, given the small-to-
moderately positive effect size in cross-sectional analyses. Future
research should examine these associations further.
On the contrary, very few studies examined rumination in rela-
tion to suicide attempts or compared levels of rumination in
Figure 5. Forest plot of all effect sizes included in the rumination-suicide attempts meta-analysis.
Figure 6. Forest plot of all effect sizes included in the brooding-suicide
attempts meta-analysis.
Figure 7. Forest plot of all effect sizes included in the reflection-suicide
attempts meta-analysis.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
7
RUMINATION AND SUICIDALITY META-ANALYSIS
suicide attempters versus nonattempters. Global rumination and
brooding were both positively associated with suicide attempts, but
reflection was not. However, because so few studies were included
in the present meta-analysis, parameter estimates are unstable and
findings should be considered with caution. Moreover, there was
significant heterogeneity between the samples with regard to age,
gender, race/ethnicity, and setting (i.e., clinical vs. nonclinical),
and none of the studies examined relations between rumination
and suicidal behavior longitudinally. Although none of the tested
moderators were significant in the current study, the limited re-
search done to date highlights a need for further research to
determine the generalizability and applicability of our synthesized
suicide attempt findings, particularly with regard to whether rumi-
nation (global, brooding, reflection) is a risk factor for suicide
attempts, prospectively. Further, we were unable to test differences
in rumination between suicide attempters and those with suicidal
ideation who had not attempted suicide, as only two studies com-
pared these two groups (Crane et al., 2007; Fazakas-Dehoog,
2008)—and findings in those two studies were mixed with regard
to reflection. This further underscores the need for future research
that compares individuals at all points on the suicide continuum
(nonsuicidal, ideation, plans, attempts, death by suicide) on rumi-
nation.
Finally, it is worth considering the potential clinical implications
of these findings, particularly with regard to the associations
between all types of rumination and suicidal ideation. Namely, the
results of the present meta-analysis suggest that incorporating
rumination as a treatment focus among individuals with or who are
at risk for suicidal ideation may be beneficial. For instance,
rumination-focused cognitive-behavioral therapies have been uti-
lized to treat depression by focusing on changing the process and
sequence of thinking, rather than individual thoughts (Watkins,
2016). Evidence thus far suggests that standard cognitive-
behavioral therapy interventions are less effective in high rumina-
tors than low ruminators (Ciesla & Roberts, 2002; Schmaling,
Dimidjian, Katon, & Sullivan, 2002), indicating that a focus on
ruminative thinking processes may improve patient outcomes,
although this has not yet been tested in individuals with heightened
risk for suicidal thoughts or behaviors. Similarly, other therapies that
emphasize changes in thinking processes, such as mindfulness-
focused cognitive therapy and dialectical behavioral therapy, have
been found to reduce the incidence of suicidal thoughts and behaviors
(Forkmann et al., 2014; Neacsiu, Rizvi, & Linehan, 2010). As a
whole, the integration of interventions to reduce ruminative thinking
patterns may result in reductions of suicidal ideation and is worth
examining.
Limitations and Future Directions
Based on the findings of the present meta-analysis, a number of
limitations and future research directions are indicated. First, as
noted previously, more research is needed to better understand the
role of rumination in conferring risk for suicidal behaviors. Given
how few studies to date have examined rumination in relation to
suicide attempts, effect size estimates from the present meta-
analysis should be taken with caution. Furthermore, a particularly
beneficial avenue for future research would be the differentiation
of suicide attempters and ideators by levels of rumination to
determine if rumination is a factor that facilitates the transition
from suicidal thoughts to actions (Klonsky & May, 2014).
Second, although a sizable minority of the studies in the present
meta-analysis had prospective designs—particularly in predicting
suicidal ideation—more longitudinal work is needed. The current
meta-analysis had insufficient cases to determine whether study
design moderated the association between rumination and suicidal
ideation; however, we did conduct sensitivity analyses to examine
whether rumination reflected a correlate or risk factor of suicidal
ideation. Moreover, all studies examining suicide attempts as an
outcome variable were retrospective/cross-sectional in design;
thus, while global rumination and brooding were associated with
the presence of a past suicide attempt, no conclusions can be drawn
regarding their respective predictive abilities. Along these lines,
longitudinal designs with short follow-up periods may be espe-
cially informative, as suicide risk is dynamic (Bryan & Rudd,
2016) and changes in rumination may impact short-term changes
in suicidal ideation or behaviors.
Third, it may be worthwhile for future research to examine
different types of rumination that may be particularly pernicious in
relation to suicidal thoughts and behaviors. As noted previously,
the cognitive process of negative repetitive thinking appears to
contribute to, at minimum, desire for suicide. Other lines of recent
research have demonstrated that suicide attempters have atten-
tional biases toward suicide-related stimuli (Adler et al., 2015;
Cha, Najmi, Park, Finn, & Nock, 2010). For instance, recent
suicide attempters took significantly longer to name the color of
suicide-related words relative to both neutral and negatively va-
lenced words in comparison with control participants (Becker,
Strohbach, & Rinck, 1999; Williams & Broadbent, 1986). Further,
suicide-specific attentional biases have been shown to predict
future suicide attempts over a 6-month follow-up period, above
and beyond other clinical predictors (Cha et al., 2010). This led to
the development of a new, potentially relevant, construct—
suicide-specific rumination, or a mental fixation on one’s thoughts,
intentions, and plans—thereby combining these two previously
unrelated lines of research (Kerkhof & van Spijker, 2011; Rogers
& Joiner, 2016). Preliminary evidence suggests that suicide-
specific rumination demonstrates strong correlations with number
of past suicide attempts, above and beyond suicidal ideation and
depressive rumination (Rogers & Joiner, 2016). Relatedly, persis-
tence of suicidal thoughts has been associated with occurrence of
future suicide attempts, particularly within a year (Miranda, Ortin,
Scott, & Shaffer, 2014). It follows that suicide-specific rumination
may facilitate further attention toward suicide-relevant cues and
result in an inability to disengage from suicide-related cognitions
and cues. Combined with other relevant risk factors, suicide-
specific rumination might potentially lead to suicidal behavior. As
such, this promising line of research may lead to a better under-
standing of particularly pernicious types of rumination in relation
to suicide attempts, specifically.
Further, as noted previously, it may be worthwhile to further
examine conceptual models of the relation between rumination and
both suicidal thoughts and attempts to better understand the nature
of these associations. Although a number of factors—negative life
events, emotion dysregulation, cognitive vulnerability and inflex-
ibility, perfectionism, self-criticism— have been found to lead to
ruminative thinking (Chan et al., 2009; Miranda, Tsypes, et al.,
2013; Miranda, Valderrama, et al., 2013; D. B. O’Connor et al.,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
8ROGERS AND JOINER
2007; R. C. O’Connor & Noyce, 2008; Smith et al., 2006), and
other factors, including depression, hopelessness, entrapment, ag-
itation, and nightmares have been shown to account for the asso-
ciation between rumination and suicidal thoughts and behaviors
(Antypa et al., 2010; Chan et al., 2009; Rogers et al., 2017; Smith
et al., 2006; Teismann & Forkmann, 2017), these constructs have
primarily been examined independently. It would be beneficial for
future research to examine the nature of these variables in a single
model to determine the relative impact of each variable on rumi-
nation and on the relationship between rumination and suicide risk.
Finally, the majority of the studies included in the present
meta-analysis examined rumination and suicidal ideation and at-
tempts in nonclinical, particularly college student, samples. This
limitation should be addressed in future work by including samples
at heightened risk for suicidal thoughts and behaviors, including
psychiatric outpatients and inpatients, military personnel and vet-
erans, and those with a prior history of self-injurious behaviors.
Further, the mean age of these samples was relatively young;
future research should also examine older participants to ensure
that these effects are generalizable across age and other demo-
graphic groups, particularly given the significant moderation of
age in the relationship between brooding and suicidal ideation. For
instance, older individuals who brood may be more prone to
suicidal ideation in the context of other age-relevant risk factors.
Conclusions
In sum, results of the present meta-analysis indicated that there
exists a strong positive association between global rumination/
brooding and suicidal ideation, a moderate positive association
between reflection and suicidal ideation, a moderate positive as-
sociation between global rumination/brooding and suicide at-
tempts, and no association between reflection and suicide attempts.
Heterogeneity was high, but in general, there was little evidence of
publication bias. Future research should focus on short-term, pro-
spective designs, prediction of suicide attempts, examination of
additional variants of rumination, and include more diverse sam-
ples to foster greater understanding of these relationships.
References
Abela, J. R. Z., Aydin, C. M., & Auerbach, R. P. (2007). Responses to
depression in children: Reconceptualizing the relation among response
styles. Journal of Abnormal Child Psychology, 35, 913–927. http://dx
.doi.org/10.1007/s10802-007-9143-2
Abramson, L. Y., Alloy, L. B., Hogan, M. E., Whitehouse, W. G., Cornette,
M., Akhavan, S., & Chiara, A. (1998). Suicidality and cognitive vulner-
ability to depression among college students: A prospective study.
Journal of Adolescence, 21, 473– 487. http://dx.doi.org/10.1006/jado
.1998.0167
Adler, A., Jager-Hyman, S., Green, K. L., Brown, G. K., Beck, A. T., &
Wenzel, A. (2015). Initial psychometric properties of the Attentional
Fixation on Suicide Experiences Questionnaire. Cognitive Therapy and
Research, 39, 492– 498. http://dx.doi.org/10.1007/s10608-015-9683-7
Ahrens, B., & Linden, M. (1996). Is there a suicidality syndrome indepen-
dent of specific major psychiatric disorder? Results of a split half
multiple regression analysis. Acta Psychiatrica Scandinavica, 94, 79 –
86. http://dx.doi.org/10.1111/j.1600-0447.1996.tb09829.x
Antypa, N., Van der Does, A. J. W., & Penninx, B. W. J. H. (2010).
Cognitive reactivity: Investigation of a potentially treatable marker of
suicide risk in depression. Journal of Affective Disorders, 122, 46 –52.
http://dx.doi.org/10.1016/j.jad.2009.06.013
Batterham, P. J., & Christensen, H. (2012). Longitudinal risk profiling for
suicidal thoughts and behaviours in a community cohort using decision
trees. Journal of Affective Disorders, 142, 306 –314. http://dx.doi.org/
10.1016/j.jad.2012.05.021
Beck, A. T., & Steer, R. A. (1991). Manual for the Beck Scale for Suicide
Ideation. San Antonio, TX: Psychological Corporation.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck
Depression Inventory-II. San Antonio, TX: Psychological Corporation.
Becker, E. S., Strohbach, D., & Rinck, M. (1999). A specific attentional
bias in suicide attempters. Journal of Nervous and Mental Disease, 187,
730 –735. http://dx.doi.org/10.1097/00005053-199912000-00004
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009).
Introduction to meta-analysis. New York, NY: Wiley. http://dx.doi.org/
10.1002/9780470743386
Bryan, C. J., & Rudd, M. D. (2016). The importance of temporal dynamics
in the transition from suicidal thought to behavior. Clinical Psychology:
Science and Practice, 23, 21–25. http://dx.doi.org/10.1111/cpsp.12135
Burke, T. A., Connolly, S. L., Hamilton, J. L., Stange, J. P., Abramson,
L. Y., & Alloy, L. B. (2016). Cognitive risk and protective factors for
suicidal ideation: A two year longitudinal study in adolescence. Journal
of Abnormal Child Psychology, 44, 1145–1160.
Centers for Disease Control and Prevention. (2016). WISQARS: Web-based
inquiry statistics query and reporting system. Retrieved from http://
www.cdc.gov/ncipc/wisqars/default.htm
Cha, C. B., Najmi, S., Park, J. M., Finn, C. T., & Nock, M. K. (2010).
Attentional bias toward suicide-related stimuli predicts suicidal behav-
ior. Journal of Abnormal Psychology, 119, 616 – 622. http://dx.doi.org/
10.1037/a0019710
Chan, S., Miranda, R., & Surrence, K. (2009). Subtypes of rumination in
the relationship between negative life events and suicidal ideation.
Archives of Suicide Research, 13, 123–135. http://dx.doi.org/10.1080/
13811110902835015
Cheref, S., Lane, R., Polanco-Roman, L., Gadol, E., & Miranda, R. (2015).
Suicidal ideation among racial/ethnic minorities: Moderating effects of
rumination and depressive symptoms. Cultural Diversity and Ethnic
Minority Psychology, 21, 31– 40. http://dx.doi.org/10.1037/a0037139
Ciesla, J. A., & Roberts, J. E. (2002). Self-directed thought and response to
treatment for depression: A preliminary investigation. Journal of Cog-
nitive Psychotherapy, 16, 435– 453. http://dx.doi.org/10.1891/
088983902780935650
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Routledge.
Cole, A. B., Wingate, L. R., Tucker, R. P., Rhoades-Kerswill, S., O’Keefe,
V. M., & Hollingsworth, D. W. (2015). The differential impact of
brooding and reflection on the relationship between perceived stress and
suicide ideation. Personality and Individual Differences, 83, 170 –173.
http://dx.doi.org/10.1016/j.paid.2015.04.013
Cooper, H. (1998). Synthesizing research: A guide for literature reviews
(3rd ed.). Beverly Hills, CA: Sage.
Crane, C., Barnhofer, T., & Williams, J. M. G. (2007). Reflection, brood-
ing, and suicidality: A preliminary study of different types of rumination
in individuals with a history of major depression. British Journal of
Clinical Psychology, 46, 497–504. http://dx.doi.org/10.1348/
014466507X230895
Cull, J. G., & Gill, W. S. (1988). Suicide Probability Scale (SPS) manual.
Los Angeles, CA: Western Psychological Services.
Ehring, T., Zetsche, U., Weidacker, K., Wahl, K., Schönfeld, S., & Ehlers,
A. (2011). The Perseverative Thinking Questionnaire (PTQ): Validation
of a content-independent measure of repetitive negative thinking. Jour-
nal of Behavior Therapy and Experimental Psychiatry, 42, 225–232.
http://dx.doi.org/10.1016/j.jbtep.2010.12.003
Ellis, T. E. (Ed.), (2006). Cognition and suicide: Theory, research, and
therapy. Washington, DC: American Psychological Association. http://
dx.doi.org/10.1037/11377-000
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
9
RUMINATION AND SUICIDALITY META-ANALYSIS
Eshun, S. (2000). Role of gender and rumination in suicide ideation: A
comparison of college samples from Ghana and the United States.
Cross-Cultural Research: The Journal of Comparative Social Science,
34, 250 –263. http://dx.doi.org/10.1177/106939710003400303
Fairweather, A. K., Anstey, K. J., Rodgers, B., Jorm, A. F., & Christensen,
H. (2007). Age and gender differences among Australian suicide ide-
ators: Prevalence and correlates. Journal of Nervous and Mental Dis-
ease, 195, 130 –136. http://dx.doi.org/10.1097/01.nmd.0000254746
.15325.fa
Fazakas-Dehoog, L. L. (2008). An integrated cognitive-affective model of
suicidal thinking and behavior (Doctoral dissertation). Retrieved from
ProQuest Dissertations and Theses database. (UMI No. 304319597)
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002).
Structured Clinical Interview for DSM–IV–TR Axis I Disorders, Re-
search Version, Patient Edition. (SCID-I/P). New York, NY: Biometrics
Research, New York State Psychiatric Institute.
Forkmann, T., Wichers, M., Geschwind, N., Peeters, F., van Os, J., Mainz,
V., & Collip, D. (2014). Effects of mindfulness-based cognitive therapy
on self-reported suicidal ideation: Results from a randomised controlled
trial in patients with residual depressive symptoms. Comprehensive
Psychiatry, 55, 1883–1890. http://dx.doi.org/10.1016/j.comppsych.2014
.08.043
Grassia, M., & Gibb, B. E. (2009). Rumination and lifetime history of
suicide attempts. International Journal of Cognitive Therapy, 2, 400 –
406. http://dx.doi.org/10.1521/ijct.2009.2.4.400
Gutierrez, P. M., Osman, A., Barrios, F. X., & Kopper, B. A. (2001).
Development and initial validation of the Self-Harm Behavior Question-
naire. Journal of Personality Assessment, 77, 475– 490. http://dx.doi.org/
10.1207/S15327752JPA7703_08
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology,
Neurosurgery & Psychiatry, 23, 56 – 62. http://dx.doi.org/10.1136/jnnp
.23.1.56
Heisel, M. J., & Flett, G. L. (2006). The development and initial validation
of the geriatric suicide ideation scale. The American Journal of Geriatric
Psychiatry, 14, 742–751. http://dx.doi.org/10.1097/01.JGP.0000218699
.27899.f9
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003).
Measuring inconsistency in meta-analyses. BMJ: British Medical Jour-
nal, 327, 557–560. http://dx.doi.org/10.1136/bmj.327.7414.557
Johnson, D. P., & Whisman, M. A. (2013). Gender differences in rumina-
tion: A meta-analysis. Personality and Individual Differences, 55, 367–
374. http://dx.doi.org/10.1016/j.paid.2013.03.019
Joormann, J., Dkane, M., & Gotlib, I. H. (2006). Adaptive and maladaptive
components of rumination? Diagnostic specificity and relation to de-
pressive biases. Behavior Therapy, 37, 269 –280. http://dx.doi.org/10
.1016/j.beth.2006.01.002
Kerkhof, A., & van Spijker, B. (2011). Worrying and rumination as
proximal risk factors for suicidal behaviour. In R. C. O’Connor, S. Platt,
& J. Gordon (Eds.), International handbook of suicide prevention (pp.
199 –209). Chichester, UK: Wiley. http://dx.doi.org/10.1002/9781
119998556.ch12
Klonsky, E. D., & May, A. M. (2014). Differentiating suicide attempters
from suicide ideators: A critical frontier for suicidology research. Sui-
cide and Life-Threatening Behavior, 44, 1–5. http://dx.doi.org/10.1111/
sltb.12068
Kovacs, M. (1985). The Children’s Depression Inventory (CDI). Psycho-
pharmacology Bulletin, 21, 995–998.
Krajniak, M., Miranda, R., & Wheeler, A. (2013). Rumination and pessi-
mistic certainty as mediators of the relation between lifetime suicide
attempt history and future suicidal ideation. Archives of Suicide Re-
search, 17, 196 –211. http://dx.doi.org/10.1080/13811118.2013.805638
Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9. Journal
of General Internal Medicine, 16, 606 – 613. http://dx.doi.org/10.1046/j
.1525-1497.2001.016009606.x
Lindelow, M., Hardy, R., & Rodgers, B. (1997). Development of a scale to
measure symptoms of anxiety and depression in the general UK popu-
lation: The psychiatric symptom frequency scale. Journal of Epidemi-
ology and Community Health, 51, 549 –557. http://dx.doi.org/10.1136/
jech.51.5.549
Lyness, J. M., Conwell, Y., King, D. A., Cox, C., & Caine, E. D. (1997).
Ruminative thinking in older inpatients with major depression. Journal
of Affective Disorders, 46, 273–277. http://dx.doi.org/10.1016/S0165-
0327(97)00068-2
May, A. M., & Klonsky, E. D. (2016). What distinguishes suicide attempt-
ers from suicide ideators? A meta-analysis of potential factors. Clinical
Psychology: Science and Practice, 23, 5–20. http://dx.doi.org/10.1111/
cpsp.12136
Metalsky, G. I., & Joiner, T. E., Jr. (1997). The Hopelessness Depression
Symptom Questionnaire. Cognitive Therapy and Research, 21, 359 –
384. http://dx.doi.org/10.1023/A:1021882717784
Miranda, R., & Nolen-Hoeksema, S. (2007). Brooding and reflection:
Rumination predicts suicidal ideation at 1-year follow-up in a commu-
nity sample. Behaviour Research and Therapy, 45, 3088 –3095. http://
dx.doi.org/10.1016/j.brat.2007.07.015
Miranda, R., Ortin, A., Scott, M., & Shaffer, D. (2014). Characteristics of
suicidal ideation that predict the transition to future suicide attempts in
adolescents. Journal of Child Psychology and Psychiatry, 55, 1288 –
1296. http://dx.doi.org/10.1111/jcpp.12245
Miranda, R., Tsypes, A., Gallagher, M., & Rajappa, K. (2013). Rumination
and hopelessness as mediators of the relation between perceived emotion
dysregulation and suicidal ideation. Cognitive Therapy and Research,
37, 786 –795. http://dx.doi.org/10.1007/s10608-013-9524-5
Miranda, R., Valderrama, J., Tsypes, A., Gadol, E., & Gallagher, M.
(2013). Cognitive inflexibility and suicidal ideation: Mediating role of
brooding and hopelessness. Psychiatry Research, 210, 174 –181. http://
dx.doi.org/10.1016/j.psychres.2013.02.033
Morrison, R., & O’Connor, R. C. (2008). A systematic review of the
relationship between rumination and suicidality. Suicide and Life-
Threatening Behavior, 38, 523–538. http://dx.doi.org/10.1521/suli.2008
.38.5.523
Neacsiu, A. D., Rizvi, S. L., & Linehan, M. M. (2010). Dialectical behavior
therapy skills use as a mediator and outcome of treatment for borderline
personality disorder. Behaviour Research and Therapy, 48, 832– 839.
http://dx.doi.org/10.1016/j.brat.2010.05.017
Niu, G. C. (2015). The use of ruminative coping among Chinese older
adults: A pathway to depression and suicidal ideation (Doctoral disser-
tation). Retrieved from ProQuest Dissertations and Theses database.
(UMI No. 3615574)
Nock, M. K., Kessler, R. C., & Franklin, J. C. (2016). Risk factors for
suicide ideation differ from those for the transition to suicide attempt:
The importance of creativity, rigor, and urgency in suicide research.
Clinical Psychology: Science and Practice, 23, 31–34. http://dx.doi.org/
10.1111/cpsp.12133
Nolen-Hoeksema, S. (1991). Responses to depression and their effects on
the duration of depressive episodes. Journal of Abnormal Psychology,
100, 569 –582. http://dx.doi.org/10.1037/0021-843X.100.4.569
Nolen-Hoeksema, S., & Morrow, J. (1991). A prospective study of depres-
sion and posttraumatic stress symptoms after a natural disaster: The
1989 Loma Prieta Earthquake. Journal of Personality and Social Psy-
chology, 61, 115–121. http://dx.doi.org/10.1037/0022-3514.61.1.115
Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking
rumination. Perspectives on Psychological Science, 3, 400 – 424. http://
dx.doi.org/10.1111/j.1745-6924.2008.00088.x
O’Connor, D. B., O’Connor, R. C., & Marshall, R. (2007). Perfectionism
and psychological distress: Evidence of the mediating effects of rumi-
nation. European Journal of Personality, 21, 429 – 452. http://dx.doi.org/
10.1002/per.616
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
10 ROGERS AND JOINER
O’Connor, R. C., & Noyce, R. (2008). Personality and cognitive processes:
Self-criticism and different types of rumination as predictors of suicidal
ideation. Behaviour Research and Therapy, 46, 392– 401. http://dx.doi
.org/10.1016/j.brat.2008.01.007
Osman, A., Bagge, C. L., Gutierrez, P. M., Konick, L. C., Kopper, B. A.,
& Barrios, F. X. (2001). The Suicidal Behaviors Questionnaire-Revised
(SBQ-R): Validation with clinical and nonclinical samples. Assessment,
8, 443– 454. http://dx.doi.org/10.1177/107319110100800409
Polanco-Roman, L., Jurska, J., Quiñones, V., & Miranda, R. (2015).
Brooding, reflection, and distraction: Relation to non-suicidal self-injury
versus suicide attempts. Archives of Suicide Research, 19, 350 –365.
http://dx.doi.org/10.1080/13811118.2014.981623
Posner, K., Brown, G. K., Stanley, B., Brent, D. A., Yershova, K. V.,
Oquendo, M. A.,...Mann, J. J. (2011). The Columbia-Suicide Severity
Rating Scale: Initial validity and internal consistency findings from three
multisite studies with adolescents and adults. The American Journal of
Psychiatry, 168, 1266 –1277. http://dx.doi.org/10.1176/appi.ajp.2011
.10111704
Reynolds, W. (1991). ASIQ, Adult Suicidal Ideation Questionnaire: Pro-
fessional manual. Odessa, FL: Psychological Assessment Resources.
Ricarte, J., Ros, L., Serrano, J. P., Martínez-Lorca, M., & Latorre, J. M.
(2016). Age differences in rumination and autobiographical retrieval.
Aging & Mental Health, 20, 1063–1069.
Rogers, M. L., & Joiner, T. E. (2016). Suicide-specific rumination as a
predictor of acute risk factors for suicide. Paper presented at the Amer-
ican Association of Suicidology, Chicago, IL.
Rogers, M. L., Schneider, M. E., Tucker, R. P., Law, K. C., Anestis, M. D.,
& Joiner, T. E. (2017). Overarousal as a mechanism of the relation
between rumination and suicidality. Manuscript submitted for publica-
tion.
Rogers, M. L., Stanley, I. H., Hom, M. A., Chiurliza, B., Podlogar, M. C.,
& Joiner, T. E. (2016). Conceptual and empirical scrutiny of covarying
depression out of suicidal ideation. Assessment. Advance online publi-
cation. http://dx.doi.org/10.1177/1073191116645907
Schmaling, K. B., Dimidjian, S., Katon, W., & Sullivan, M. (2002).
Response styles among patients with minor depression and dysthymia in
primary care. Journal of Abnormal Psychology, 111, 350 –356. http://dx
.doi.org/10.1037/0021-843X.111.2.350
Smith, J. M., Alloy, L. B., & Abramson, L. Y. (2006). Cognitive vulner-
ability to depression, rumination, hopelessness, and suicidal ideation:
Multiple pathways to self-injurious thinking. Suicide and Life-
Threatening Behavior, 36, 443– 454. http://dx.doi.org/10.1521/suli.2006
.36.4.443
Stange, J. P., Hamilton, J. L., Burke, T. A., Kleiman, E. M., O’Garro-
Moore, J. K., Seligman, N. D.,...Alloy, L. B. (2015). Negative
cognitive styles synergistically predict suicidal ideation in bipolar spec-
trum disorders: A 3-year prospective study. Psychiatry Research, 226,
162–168. http://dx.doi.org/10.1016/j.psychres.2014.12.042
Surrence, K., Miranda, R., Marroquín, B. M., & Chan, S. (2009). Brooding
and reflective rumination among suicide attempters: Cognitive vulnera-
bility to suicidal ideation. Behaviour Research and Therapy, 47, 803–
808. http://dx.doi.org/10.1016/j.brat.2009.06.001
Teismann, T., & Forkmann, T. (2017). Rumination, entrapment and suicide
ideation: A mediational model. Clinical Psychology & Psychotherapy,
24, 226 –234.
Treynor, W., Gonzalez, R., & Nolen-Hoeksema, S. (2003). Rumination
reconsidered: A psychometric analysis. Cognitive Therapy and Re-
search, 27, 247–259. http://dx.doi.org/10.1023/A:1023910315561
Tucker, R. P., Wingate, L. R., O’Keefe, V. M., Hollingsworth, D. W., &
Cole, A. B. (2016). An examination of historical loss thinking frequency
and rumination on suicide ideation in American Indian young adults.
Suicide and Life-Threatening Behavior, 46, 213–222.
Tucker, R. P., Wingate, L. R., O’Keefe, V. M., Mills, A. C., Rasmussen,
K., Davidson, C. L., & Grant, D. M. (2013). Rumination and suicidal
ideation: The moderating roles of hope and optimism. Personality and
Individual Differences, 55, 606 – 611. http://dx.doi.org/10.1016/j.paid
.2013.05.013
Watkins, E. (2016). Rumination-focused cognitive-behavioral therapy for
depression (1st ed.). New York, NY: Guilford Press.
Watkins, E., & Moulds, M. (2005). Distinct modes of ruminative self-
focus: Impact of abstract versus concrete rumination on problem solving
in depression. Emotion, 5, 319 –328. http://dx.doi.org/10.1037/1528-
3542.5.3.319
Williams, J. M., & Broadbent, K. (1986). Autobiographical memory in
suicide attempters. Journal of Abnormal Psychology, 95, 144 –149.
http://dx.doi.org/10.1037/0021-843X.95.2.144
World Health Organization. (2014). Preventing suicide: A global impera-
tive. Geneva, Switzerland: Author.
Yaseen, Z. S., Gilmer, E., Modi, J., Cohen, L. J., & Galynker, I. I. (2012).
Emergency room validation of the revised Suicide Trigger Scale (STS-
3): A measure of a hypothesized suicide trigger state. PLoS ONE, 7,
e45157. http://dx.doi.org/10.1371/journal.pone.0045157
Received October 19, 2016
Revision received February 2, 2017
Accepted February 6, 2017 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
11
RUMINATION AND SUICIDALITY META-ANALYSIS
