Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Feb 15.
Published in final edited form as: J Affect Disord. 2022 Nov 22;323:320–326. doi: 10.1016/j.jad.2022.11.041

Deconstructing Resilience in Patients at High Risk for Suicidal Behavior

Rachel Hendricks 1, John G Keilp 1, Mohammad Lesanpezeshki 1, Rina Muqkurtaj 1, Steven P Ellis 1, Hanga Galfalvy 1, Ainsley K Burke 1, Nadine Melhem 1, J John Mann 1
PMCID: PMC9945908  NIHMSID: NIHMS1855109  PMID: 36427650

Abstract

Background:

Resilience represents coping abilities to overcome exposure to psychopathological risk. In the context of risk factors for suicidal behavior, it is unknown if this attribute is deficient in suicide attempters, how it relates to other measures of risk, and where it may overlap with other risk factors associated with suicidal behavior.

Methods:

The present study examined the performance on the Connor-Davidson Resilience Scale (CD-RISC) in three groups of individuals with familial risk for both mood disorder and suicidal behavior, as well as a healthy comparison group. Other risk factors for suicidal behavior, such as depression severity, hopelessness, and lifetime impulsiveness were examined as well to determine if these mediated group differences in CD-RISC scores.

Results:

CD-RISC scores differed between groups, with lowest scores in the past attempter group. However, CD-RISC scores were strongly correlated with other common risk factors for suicide attempt, including hopelessness, subjective depression, and reasons for living, which together explained 68% of the CD-RISC variance. Group differences in CD-RISC scores were eliminated when the model included these covariates.

Limitations:

Sample sizes were modest, and depression severity was low overall and significantly higher in the past suicide attempter group.

Conclusions:

The CD-RISC has demonstrated utility for predicting risk for depression, but appears to overlap with other known risk factors for suicidal behavior, especially hopelessness and subjective depression. Though it encapsulates variance from multiple risk factors in a single scale, it may not provide additional predictive power above and beyond these other risk factors for suicidal behavior.

Keywords: resilience, depression, suicide, risk for suicide

Introduction

Suicide attempt risk has been found to be partly familial. Offspring at risk for mood disorders are five times more likely to attempt suicide themselves when a parent has a history of a suicide attempt (Brent et al., 2015). Several studies have explored the risk factors for suicide and protective factors (Bostwick et al., 2016; Choi et al., 2013; Franklin et al., 2017; Grendas et al., 2019; Lawrence et al., 2016; Pawlak et al., 2018; Rasouli et al., 2019). Few studies have examined psychological resilience as a potential protective factor among suicide attempters (Roy et al., 2007, 2011; You & Park, 2017; Youssef et al., 2013).

Resilience is more than the converse of risk and may comprise other factors that enable individuals to overcome risk. Definitions of resilience vary, however, and work on the concept has evolved over time (Bonanno, 2004; Richardson et al., 2013; Rolin et al., 2018; Rutter, 1985, 2012). Herman et al. (2011) provided the following definition: “fundamentally, resilience is understood as referring to positive adaptation, or the ability to maintain or regain mental health, despite experiencing adversity.” Connor & Davidson (2003) define resilience as “the personal qualities that enable one to thrive in the face of adversity.” Resilience may represent independent coping abilities or psychobiological resources independent of risk that enable individuals exposed to various risk factors to manage or overcome them.

Psychological resilience has been measured via rating scales, such as the Connor- Davidson Resilience scale (CD-RISC). This scale measures stress-coping ability in the face of risk or adversity and has been found to be both reliable and valid in clinical populations (Connor & Davidson, 2003). The content of the scale was derived in part from the work of Kobasa (1979) and Rutter (1985) with items reflecting hardiness and traits such as strong self-esteem and adaptability. Items assessing patience and tolerance in the face of stress were drawn from the work of Lyons (1991). Thus, the scale assesses several aspects of resilience, including a sense of personal competence, tenacity, trust in one’s instincts, tolerance of negative affect, positive acceptance of change, a sense of secure relationships, spiritual faith and recognition of one’s successes and achievements (Connor & Davidson, 2003). Resilience, as measured by the CD-RISC, has been found to be influenced by health status (Connor & Davidson, 2003). Individuals with psychiatric illnesses have lower levels of resilience than the general population, however, resilience appears to be modifiable and can improve with treatment. The scale has not been validated against any behavioral or biological measure of resilience, including neuropeptide Y responses to extreme stress (Connor & Davidson, 2003). Furthermore, the scale does not address the potential directional aspects of resilience; resilience may shape response, or it may be an effect of exposure to stress (Connor & Davidson, 2003).

Several studies have explored the relationship between depressive symptoms and resilience using the CD-RISC. Resilience factors were consistently negatively correlated with depression (McGillivray & Ho, 2016; Liu et al., 2020; Song et al., 2021; Tahereh et al., 2012). While resilience is negatively associated with depression, its relationship to suicidal thinking and suicide attempt is less clear. Previous studies have linked lower CD-RISC resilience scores to suicide risk (Roy et al., 2007, 2011, Carli et al., 2010). Multiple attempters were also found to have lower resilience scores on the CD-RISC in comparison with single attempters (Roy et al., 2014). In the general population, one study found resilience was inversely related to suicide, however only among men (You & Park, 2017). Finally, resilience was found to be a protective factor for risk for suicide in military veterans in a longitudinal study (Youssef et al, 2013). A limitation of these studies, however, is the absence of control for the severity of depression and other risk factors for suicidal behavior.

Lower resilience scores correlated with higher levels of hopelessness, for example, as measured by the Beck Hopelessness Scale (Gambaro et al., 2020). A study of suicidality among incarcerated men found that resilience was positively correlated to impulsivity (Carli et al., 2010; Carli et al., 2014). Another study of resilience and psychological hardiness (ability to adapt to unexpected changes as well as a sense of purpose and control) among women with breast cancer found a significant inverse relationship between resilience scores (CD-RISC) and depression severity (BDI), such that women with more severe depression had lower resilience scores (Tadayon et al., 2018).

The present study was designed to investigate the relationship of resilience to suicide attempt, as well as its potential association to other risk factors for suicidal behavior, in samples enriched for familial risk for such behavior. The study population included a group of healthy volunteers with no familial history of mood disorder or suicidal behavior, and three groups of individuals with a first or second degree relative who had early onset mood disorder and a history of a suicide attempt. These latter three “high-risk” groups included individuals all over the age of 25 (i.e., beyond the typical age of risk of first episode of depression) who had (a) never experienced a mood disorder, (b) had a history of a mood disorder but no history of suicidal behavior, and (c) had both a history of a mood disorder and a history of a past suicide attempt.

Our aim was to compare resilience scores on the CD-RISC across these samples, and – if differences were found – to determine if other factors such as depression, severity, hopelessness, or impulsiveness mediated any group differences. A central goal of these analyses was to determine whether resilience, as measured via the CD-RISC, provided independent information about the susceptibility to risk in suicide attempters or overlapped with other known risk factors.

We expected resilience to be highest among both healthy volunteers and the high-risk sample of individuals with no history of depression or suicidal behavior. We expected that resilience would be lower in those high-risk individuals with a history of depression, and lowest overall in those high-risk participants with both a history of depression and a history of a suicide attempt. We hypothesize as well that some of these differences will be explained by other known risk factors for suicidal behavior, but that resilience will remain lower in past suicide attempters even after adjustment for these other measures.

Methods

Sample:

Four groups of participants over the age of 25 were recruited from two sites: Western Psychiatric Institute and Clinic in Pittsburgh and at New York State Psychiatric Institute in New York City. Three of these groups were at high-risk because they had a first or second degree relative with a history of suicide attempt and early onset mood disorder, with one group having no history of psychiatric illness or suicidal behavior (high risk resilient individuals, HR-R) one having a history of major depressive disorder or bipolar disorder but no prior suicidal behavior (high risk individuals with mood disorders only, HR-MOOD), and one having a history of major depression or bipolar disorder and at least one prior suicide attempt (HR-SA/MOOD) (Melham et al., 2022). A comparison sample of healthy individuals over the age of 25 with no family or personal history of psychiatric illness or suicide attempt (HV) was included as well (Melham et al., 2022).

Instruments:

Participants were assessed diagnostically via the SCID (Spitzer et al., 1995) and Columbia Suicide History Form (Oquendo et al., 2003). Basic demographic information, developmental, and clinical history were assessed via a structured interview. Depression severity, hopelessness, and the severity of suicidal thoughts were assessed with standard rating scales (Hamilton Rating Scale for Depression, Beck Depression Inventory, Beck Hopelessness Scale, Beck Scale for Suicidal Ideation; Hamilton, 1960; Beck et al., 1961; Beck et al., 1974; Beck et al., 1979). Impulsiveness and aggression were assessed via the Barratt Impulsiveness Scale (Barratt, 1965) and Buss-Perry Aggression Questionnaire (Buss & Perry, 1992), respectively. Social and cultural protective factors against suicidal behavior were assessed with the Reasons for Living Inventory (Linehan et al., 1983).

Psychological resilience was measured via the CD-RISC. The scale includes 25 items, each rated on a 5-point Likert scale ranging from 0 (not true at all) to 4 (true nearly all the time), with the total score ranging from 0–100; higher scores reflecting greater resilience. The scale reflects participants’ feelings over the past month. The original scale includes five sub-factors, though subsequent factor analytic studies in both clinical and nonclinical populations have not supported this initial factor structure (Green et al., 2014; Pietrzak et al., 2010; Velickovic et al., 2020; Yu & Zhang, 2007). Thus, only the total score for the scale was used here.

Procedures:

Participants were recruited by advertisement, by referral from individual physicians, from inpatient services, or from hospital emergency rooms. All participants also participated in imaging procedures that constituted an additional component of this study, including both Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) scans.

Recruitment of participants and the informed consent or assent process were conducted in accordance with the guidelines of the institutional review boards of University of Pittsburgh School of Medicine and New York State Psychiatric Institute, both of which approved the study. Interviews and ratings were conducted at the study’s intake, prior to imaging procedures, and included up to three clinical follow-up visits in the subsequent two years.

Statistical analysis:

Demographic and clinical information was compared among the groups using One-Way Analysis of Variance (ANOVA) for continuous variables and Chi-Squared analyses for categorical variables. This included unadjusted comparison of CD-RISC score among the groups.

Correlations (Pearson) were then computed between demographic and clinical correlates of suicidal behavior and CD-RISC score, both across the entire sample as well as within the patients with depression (HR-MOOD and HR-SA/MOOD groups) alone. Correlations were examined in the depressed groups alone to determine if associations between clinical variables and CD-RISC scores were driven by group differences in these measures or were a reflection of increasing clinical severity.

Stepwise regressions were run in the next step of the analyses with significant correlates of the CD-RISC score, again in both the total sample and the two groups with a history of depression (HR-MOOD and HR-SA/MOOD), to determine if and to what degree some combination of these other clinical variables could predict resilience.

Analyses of Covariance (ANCOVA) was then conducted, with site and any other variables that distinguished the groups as covariates, to determine if group differences in resilience were mediated by these other characteristics.

Finally, a stepwise logistic regression in the HR-MOOD and HR-SA/MOOD groups was run to determine the best predictors of classification as a suicide attempter in this sample, to determine if CD-RISC score contributed to the classification of attempter status independent of other risk factors.

Results

Group Comparisons:

Groups were similar in age, sex distribution, racial composition, and educational attainment (Table 1.).

Table 1.

Demographic and Clinical Characteristics of Groups

Variable HV HR-R HR-MOOD HR-SA/MOOD p value Contrast
Demographics and Clinical History
N 26 33 32 30
Age (years) 41.4 ± 9.7 38.3 ± 12.2 37.3 ± 10.0 39.7 ± 11.3 .522
Sex (%F) 69.2% (18) 68.8% (22) 65.6% (21) 63.3% (19) .959
Race .602
 Black or African American 34.6% (9) 30.3% (10) 21.9% (7) 23.3% (7) –
 White 61.5% (16) 60.6% (20) 75.0% (24) 63.3% (19) –
 Other 3.8% (1) 9.1% (3) 3.1% (1) 13.3% (4) –
Ethnicity (Hispanic) 7.7% (2) 6.1% (2) 9.4% (3) 23.3% (7) .121
Education (years) 16.6 ± 2.5 15.9 ± 2.2 16.2 ± 2.0 15.0 ± 2.8 .070
Current Major Depression (MDE, %) – – 18.8% (6) 46.7% (14) .034 HR-SA/MOOD > HR-MOOD
Comorbid BPD (%) – – 6.3%% (2) 30.0% (9) .014 HR-SA/MOOD > HR-MOOD
Past Substance Abuse 0% (0) 12.1% (4) 43.8% (14) 53.3% (16) <.001 HR-SA/MOOD, HR-MOOD > HR-R, HV
Clinical Ratings
Hamilton Depression Scale HDRS, 17-item) 0.7 ± 1.0 1.2 ± 2.0 5.4 ± 6.2 10.9 ± 6.7 <.001 HR-SA/MOOD > HR-MOOD > HR-R, HV
Beck Depression Inventory (BDI) 2.0 ± 2.5 2.1 ± 3.3 9.3 ± 9.6 18.9 ± 12.6 <.001 HR-SA/MOOD > HR-MOOD > HR-R, HV
Beck Hopelessness
Scale (BHS)		 2.0±2.1 1.4±2.3 5.9±6.4 8.4±6.6 <.001 HR-SA/MOOD > HR-MOOD > HR-R, HV
Scale for Suicidal Ideation (SSI, Current) – – 0.7±2.2 1.6±4.0 .273
Scale for Suicidal Ideation (SSI, Prior to Admission) – – 0.8±2.5 2.2±4.7 .170
Reasons for Living Scale (RFL) 207.7 ± 26.2 201.2 ± 28.7 188.6± 42.4 185.6 ± 42.4 .076
Barratt Impulsiveness Scale (BIS-11) 55.0 ± 9.4 56.8 ± 12.1 60.5 ± 12.0 71.6 ± 13.8 <.001 HR-SA/MOOD > HR-MOOD, HR-R, HV
Buss-Perry Aggression Questionnaire (BPAQ) 52.2 ± 14.5 54.9 ± 20.9 63.8 ± 21.7 74.8 ± 22.9 <.001 HR-SA/MOOD > HR-MOOD, HR-R > HV
Brown-Goodwin Aggression History (BGAH) 12.1 ± 2.7 13.5 ± 3.7 16.4 ± 5.6 19.7 ± 5.7 <.001 HR-SA/MOOD > HR-MOOD > HR-R, HV
Connor-Davidson Resilience Scale (CD-RISC) 82.5 ± 13.3 80.9 ± 10.4 69.2 ± 21.7 59.5 ± 21.1 <.001 HR-SA/MOOD < HR-MOOD < HR-R, HV
Attempters Only
Number of Past Suicide Attempts - - - 1.8 ± 1.1
Lethality – Most Lethal Attempt _
Lethality – Most Recent Attempt _ _ _ 1.5 ± 1.5
Suicide Intent – Most Lethal Attempt – – – 15.3 ± 4.8
Suicide Intent – Most Recent Attempt – – – 13.6 ± 6.9
Open in a new tab

HV = Healthy Volunteers; HR-R = High Risk Resilient; HR-MOOD = High Risk with History of Mood Disorder only; HR – SA/MOOD = High Risk with History of both Mood Disorder and Suicide Attempt

The HR-SA/MOOD high risk attempters were more likely to be in a current Major Depressive Episode (MDE) compared with HR-MOOD non-attempters, with just under half meeting current diagnostic criteria for depression, while less than one-fifth of HR-MOOD non-attempters were currently depressed. The percentage of patients with comorbid Borderline Personality Disorder was also higher in HR-SA/MOOD past attempters than HR-MOOD non-attempters.

Severity of depression and hopelessness was mild overall in both patient groups with a history of depression, but higher in HR-SA/MOOD past attempters than in HR-MOOD non-attempters and in both depression groups compared with HV or HR-R. Suicidal ideation was low in both past depressed groups (over 85% of HR-MOOD non-attempters, and over 76% of HR-SA/MOOD attempters reported no ideation either just before enrollment or at the time of assessment) and ideation severity did not differ between groups. Reasons for Living was comparable across the four groups. Impulsiveness, aggression, and aggressive behavior history, however, were highest in those with past suicide attempt.

Scores on the CD-RISC, consistent with our initial hypotheses, were lowest in the HR-SA/MOOD past attempters, followed by HR-MOOD non-attempters, with no difference between HR-R and HV in this unadjusted univariate analysis.

Correlations:

Within the patient sample, CD-RISC scores were not correlated with either age (r = .16, p = .218) or education (r = −.04, p = .737), nor were there any differences between sexes (t[60] = .52, p = .604) or between those with and without past substance abuse (t[60] = 2.62, p = .111). There was no difference in CD-RISC score by presence or absence of comorbid BPD (t[60] = .86, p =.393), but there was a clear difference whether individuals met criteria for a current MDE (t[60] = 3.38, p = .001).

In all subjects across all groups, CD-RISC correlated with HDRS (r = −.53, p<.001), BDI (r = −.74, p<.001), BHS (r = −.79, p<.001), SSI current (r = −.41, p<.001), SSI prior (r = −.31, p=.001), RFL (r = .54, p<.001), BIS-11 (r = −.62, p<.001), BPAQ (r = −.44, p=.014), and BGAH (r = −.28, p=.003).

In the two patient groups alone (HR-MOOD and HR-SA/MOOD), the CD-RISC score correlated with HDRS (r = −.43, p=.001), BDI (r = −.72, p<.001), BHS (r = −.78, p<.001), SSI current (r = −.40, p=.002), RFL (r = .61, p<.001), BIS-11 (r = −.65, p<.001), and BPAQ (r = −.32, p=.014).

Stepwise Regressions:

Given the strong association between many of these psychopathology ratings and CD-RISC scores, a series of stepwise regressions were run attempting to predict CD-RISC scores in (a) all subjects, and (b) patients with a history of depression alone (HR-MOOD and HR- SA/MOOD). Variables available for selection included those most strongly correlated with the CD-RISC in both the full sample and the patients alone: HDRS, BDI, BHS, SSI (current), RFL, BIS-11, and BPAQ.

In the full sample with all participants (Figure 1a.), three variables were included in the final significant equation (F[3,117] = 82.24, p < .001) which explained 67.8% of the variance in the CD-RISC score (R2 = .678). These included the BHS (Beta = −.48, t = −5.18, p < .001), BDI (Beta = −.28, t = −3.10, p = .002), and RFL (Beta = .18, t = 3.01, p = .003).

Figures 1a. & 1b.:

Figures 1a. & 1b.:

Open in a new tab

Stepwise regression results predicting CD-RISC score from a linear combination of Beck Hopelessness Scale (BHS), Beck Depression Inventory (BDI), and Reasons for Living Scale (RFL) in full sample of all subjects across all groups (1a.) and those with past history of depression, with or without past suicide attempt (1b.). HV = Healthy Volunteer; HR-R = High-Risk Resilient; HR-MOOD = High-Risk with Mood Disorder; HR-SA = High-Risk with Mood Disorder and Prior Suicide Attempt.

A comparable equation was obtained in the patient sample (HR-MOOD and HR-SA/MOOD) alone (Figure 1b.). This equation (F[3,58] = 41.17, p < .001) explained 68.0% of the variance in the CD-RISC score (R2 = .680). The same set of predictors were included: BHS (Beta = −.46, t = −3.75, p < .001), BDI (Beta = −.26, t = −2.21, p = .031), and RFL (Beta = .24, t = 2.76, p = .008).

Both of these regressions, then - in the total sample as well as in the depressed patients alone - produced comparable weightings of the BHS, BDI, and RFL, and predicted approximately 68% of the variance in the CD-RISC score, with the largest contribution in both regressions coming from the BHS hopelessness score.

Covariance Analysis:

When CD-RISC scores were compared among the groups including covariates for BHS, BDI, and RFL, all three covariates were significant (for BHS: F[1,113]= 23.81, p < .001; for BDI: F[1,113]= 5.95, p = .016; for RFL: F[1,113]= 9.15, p = .003), and the overall main effect of group on the CD-RISC score was now non-significant (F[3,113] = 0.15, p = .928).

Logistic Regression:

In a stepwise logistic regression including the CD-RISC score, all the variables predictive of the CD-RISC score (BHS, BDI, RFL), and categorical variables that differentiated groups for the presence of current MDE, substance abuse, and Borderline Personality Disorder, only the BDI score entered the equation (Wald[1] = 8.53, OR = 1.08, p = .003). The BDI depression score alone correctly classified 78.1% of non-attempters and 63.3% of attempters (71.0% overall correct classification).

Discussion

CD-RISC scores differed between groups and – as predicted – were lowest in the past attempter group. However, the CD-RISC score was strongly correlated with several other common risk factors for suicide attempt, including hopelessness, subjective depression, and reasons for living. In a stepwise regressions, hopelessness, subjective depression, and reasons for living explained 68% of the variance in these CD-RISC scores, suggesting significant overlap in these scales. When these factors were controlled in covariance analyses, CD-RISC scores no longer differentiated the groups. Though the concept of resilience is hypothesized to be a mediating factor in the expression of other psychopathological risk factors on the risk for suicidal behavior, it appears to be heavily confounded with these risk factors. It would likely be a significant mediating factor in any multivariate classification or prediction model, but in large part because it shares so much variance with these other measures.

The strong correlation between the CD-RISC score and the Beck Hopelessness Scale is at least partially due to a similarity of content in specific items. Review of items on the CD-RISC reveals that many ask directly, in a positive manner, about hopefulness and optimism about life. These include “I believe I can achieve my goals, even if there are obstacles” (Item 11), “Even when hopeless, I do not give up” (Item 12), and “I am not easily discouraged by failure” (Item 16). These are concordant with similar items on the Beck Hopelessness Scale that are scaled the opposite way: “There’s no use in really trying to get something I want because I probably won’t get it” (BHS Item 20), “Things just won’t work out the way I want them to” (BHS Item 14), and “I might as well give up because I can’t make things better for myself” (BHS Item 2). The BHS also contains positively scaled items nearly identical to those on the CD-RISC, but which are reverse scored on the BHS. For example, one BHS item asks if “My past experiences have prepared me well for my future” (Item 10, reverse scored) while the CD-RISC asks if “Past successes give me confidence” (Item 5, positively scored). Similarly, the general positive orientation of the items on the CD-RISC, as well as specific items such as “I am able to handle unpleasant or painful feelings like sadness, fear, and anger” (Item 19) represent the inverse of depressive feelings assessed on the BDI. Items such as “I have a strong sense of purpose in life” (Item 21) ask directly about beliefs assessed on the RFL.

In our final logistic regression, the most important measure distinguishing the past attempter group from the non-attempter group was the BDI. Neither the CD-RISC score, nor its strongest correlate, the BHS, entered this final equation. Though sample sizes here are small, this finding is consistent with our prior work in which subjective depression is one of the strongest correlates of past attempt status (Keilp et al., 2012; Keilp et al., 2018; Mann et al., 1999) and the severity of subjective depression relative to clinician rated depression is linked to familial transmission of suicidal behavior risk (Mann et al., 2016).

While the CD-RISC appears to have demonstrated utility for characterizing risk for depression, it may not be sufficiently independent of other known risk factors for suicidal behavior to be an effective risk measure for suicide attempt. Resilience as a predictor of suicide attempt may be too confounded with other factors that that are well established markers of risk for suicidal behavior. As noted above, resilience should reflect factors other than the inverse of risk, such as specific behaviors that are initiated in the face of risk that enable the individual to surmount pathological urges or feelings. In the context of attempting to predict suicidal behavior, there may be too little additional variance in CD-RISC scores beyond that contained in known risk factors for this behavior.

Limitations:

There were several limitations in this study. First, sample sizes were small, and presence of current depression as well as depression severity were confounded to a degree with past attempt status. This was managed statistically, but larger samples with better balance between the groups would provide a better test of these associations. Second, depression severity scores of the attempter and non-attempter groups were low, suicidal ideation scores were generally low, and severity of past suicidal behavior (average lethality score) in the past attempter group was relatively low compared to past studies. Attenuation of the range of these scores may have affected relationships among various clinical variables and the ability of our analyses to detect independent group differences.

Associations between the CD-RISC score and other clinical variables, however, were very similar in the two stepwise regressions including all participants and the two groups with a history of depression. Moreover, the association of CD-RISC scores with both depression severity and hopelessness is consistent with prior studies (Gambaro et al., 2020; Tadayon et al., 2018), suggesting these associations have some stability and consistency across studies. Further studies of individuals at risk for suicidal behavior with larger and more clinically heterogenous samples are needed, though, to determine if additional information from the CD-RISC might be extracted for distinguishing resilience factors relevant to overcoming risk for self-harm. At present, though, it appears that the CD-RISC captures variance that overlaps with known risk factors for suicidal behavior and may not provide sufficient additional information about the hypothesized trait proposed to mitigate the effects of these risk factors.

Highlights.

  • Participants with a history of suicide attempt had the lowest resilience scores

  • Resilience scores were strongly correlated with other common risk factors

  • Differences in resilience scores were eliminated when the model included covariates

Acknowledgements:

We would like to acknowledge the individuals who enrolled in and participated in this study. The authors also acknowledge the contributions of the clinical staff of the Molecular Imaging and Neuropathology Division who assisted in data collection for this project.

Funding/Support:

This study was supported by 5R01MH108032 from NIMH.

Role of Funding Source:

Funding for this project was derived from a governmental agency. The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest: Drs. Mann and Burke receive royalties from the Columbia Suicide Severity Scale, which was not used in analyses reported here. Other authors report no conflicts of interest with the material presented in this manuscript.

References:

  1. Barratt ES, (1965). Factor analysis of some psychometric measures of impulsiveness and anxiety. Psychol Rep, 16, 547–554. 10.2466/pr0.1965.16.2.547 [DOI] [PubMed] [Google Scholar]
  2. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J, (1961). An inventory for measuring depression. Arch Gen Psychiatry, 4, 561–571. 10.1001/archpsyc.1961.01710120031004 [DOI] [PubMed] [Google Scholar]
  3. Beck AT, Weissman A, Lester D, Trexler L, (1974). The measurement of pessimism: the hopelessness scale. J Consult Clin Psychol, 42(6), 861–865. 10.1037/h0037562 [DOI] [PubMed] [Google Scholar]
  4. Beck AT, Kovacs M, Weissman A, (1979). Assessment of suicidal intention: the Scale for Suicide Ideation. J Consult Clin Psychol, 47(2), 343–352. 10.1037//0022-006x.47.2.343 [DOI] [PubMed] [Google Scholar]
  5. Bonanno GA, (2004). Loss, trauma, and human resilience: have we underestimated the human capacity to thrive after extremely aversive events?. Am Psychol, 59(1), 20–28. 10.1037/0003-066X.59.1.20 [DOI] [PubMed] [Google Scholar]
  6. Bostwick JM, Pabbati C, Geske JR, McKean AJ, (2016). Suicide attempt as a risk factor for completed suicide: even more lethal than we knew. Am J Psychiatry, 173(11), 1094–1100. 10.1176/appi.ajp.2016.15070854 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brent DA, Melhem NM, Oquendo M, Burke A, Birmaher B, Stanley B, Biernesser C, Keilp J, Kolko D, Ellis S, Porta G, Zelazny J, Iyengar S, Mann JJ, (2015). Familial pathways to early-onset suicide attempt: a 5.6-year prospective study. JAMA Psychiatry, 72(2), 160–168. 10.1001/jamapsychiatry.2014.2141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Buss AH, Perry M, (1992). The aggression questionnaire. J Pers Soc Psychol, 63(3), 452–459. 10.1037//0022-3514.63.3.452 [DOI] [PubMed] [Google Scholar]
  9. Carli V, Jovanović N, Podlesek A, Roy A, Rihmer Z, Maggi S, Marusic D, Cesaro C, Marusic A, Sarchiapone M, (2010). The role of impulsivity in self-mutilators, suicide ideators and suicide attempters - a study of 1265 male incarcerated individuals. J Affect Disord, 123(1– 3), 116–122. 10.1016/j.jad.2010.02.119 [DOI] [PubMed] [Google Scholar]
  10. Carli V, Mandelli L, Zaninotto L, Alberti S, Roy A, Serretti A, Sarchiapone M, (2014). Trait-aggressiveness and impulsivity: role of psychological resilience and childhood trauma in a sample of male prisoners. Nord J Psychiatry, 68(1), 8–17. 10.3109/08039488.2012.756061 [DOI] [PubMed] [Google Scholar]
  11. Choi KH, Wang SM, Yeon B, Suh SY, Oh Y, Lee HK, Kweon YS, Lee CT, Lee KU (2013). Risk and protective factors predicting multiple suicide attempts. Psychiatry Res, 210(3), 957–961. 10.1016/j.psychres.2013.09.026 [DOI] [PubMed] [Google Scholar]
  12. Connor KM, & Davidson JR (2003). Development of a new resilience scale: the Connor-Davidson Resilience Scale (CD-RISC). Depress Anxiety, 18(2), 76–82. 10.1002/da.10113 [DOI] [PubMed] [Google Scholar]
  13. Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, Musacchio KM, Jaroszewski AC, Chang BP, Nock MK, (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychol Bull, 143(2), 187–232. 10.1037/bul0000084 [DOI] [PubMed] [Google Scholar]
  14. Gambaro E, Mastrangelo M, Sarchiapone M, Marangon D, Gramaglia C, Vecchi C, Airoldi C, Mirisola C, Costanzo G, Bartollino S, Baralla F, Zeppegno P (2020). Resilience, trauma, and hopelessness: protective or triggering factor for the development of psychopathology among migrants?. BMC Psychiatry, 20(1), 358. 10.1186/s12888-020-02729-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Green KT, Hayward LC, Williams AM, Dennis PA, Bryan BC, Taber KH, Mid-Atlantic Mental Illness Research, Education and Clinical Center Workgroup, Davidson JR, Beckham JC, Calhoun PS (2014). Examining the factor structure of the Connor-Davidson Resilience Scale (CD-RISC) in a post-9/11 U.S. military veteran sample. Assessment, 21(4), 443–451. 10.1177/1073191114524014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Grendas LN, Rojas SM, Puppo S, Vidjen P, Portela A, Chiapella L, Rodante DE, Daray FM (2019). Interaction between prospective risk factors in the prediction of suicide risk. J. Affect Disord, 258, 144–150. 10.1016/j.jad.2019.07.071 [DOI] [PubMed] [Google Scholar]
  17. Hamilton M (1960). A rating scale for depression. J Neurol Neurosurg Psychiatry 23(1), 56–62. 10.1136/jnnp.23.1.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Herrman H, Stewart DE, Diaz-Granados N, Berger EL, Jackson B, Yuen T (2011). What is resilience?. Can J Psychiatry 56(5), 258–265. 10.1177/070674371105600504 [DOI] [PubMed] [Google Scholar]
  19. Keilp JG, Grunebaum MF, Gorlyn M, LeBlanc S, Burke AK, Galfalvy H, Oquendo MA, Mann JJ (2012). Suicidal ideation and the subjective aspects of depression. J Affect Disord, 140(1), 75–81. 10.1016/j.jad.2012.01.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Keilp JG, Ellis SP, Gorlyn M, Burke AK, Oquendo MA, Mann JJ, Grunebaum MF (2018). Suicidal ideation declines with improvement in the subjective symptoms of major depression. J Affect Disord, 65–70. 10.1016/j.jad.2017.09.018 [DOI] [PubMed]
  21. Kobasa SC (1979). Stressful life events, personality, and health: an inquiry into hardiness. J Pers Soc Psychol, 37(1), 1–11. 10.1037//0022-3514.37.1.1 [DOI] [PubMed] [Google Scholar]
  22. Lawrence RE, Oquendo MA, & Stanley B (2016). Religion and Suicide Risk: A Systematic Review. Arch Suicide Res, 20(1), 1–21. 10.1080/13811118.2015.1004494 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Linehan MM, Goodstein JL, Nielsen S L, Chiles, J. A. (1983) Reasons for staying alive when you are thinking of killing yourself: The Reasons for Living Inventory. J Consult Clin Psychol, 51(2), 276–286. 10.1037//0022-006x.51.2.276 [DOI] [PubMed] [Google Scholar]
  24. Lyons JA (1991). Strategies for assessing the potential for positive adjustment following trauma. J Trauma Stress, 4(1), 93–111. 10.1002/jts.2490040108 [DOI] [Google Scholar]
  25. Mann JJ, Oquendo M, Underwood MD, Arango V (1999). The neurobiology of suicide risk: a review for the clinician. J Clin Psychiatry, 60 Suppl 2, 7–116. [PubMed] [Google Scholar]
  26. Mann JJ, Ellis SP, Currier D, Zelazny J, Birmaher B, Oquendo MA, Kolko DJ, Stanley B, Melhem N, Burke A, & Brent DA, (2016). Self-rated depression severity relative to clinician-rated depression severity: Trait stability and potential role in familial transmission of suicidal behavior. Arch Suicide Res, 20(3), 412–425. 10.1080/13811118.2015.1033504 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McGillivray K, Ho R, (2016). Validation of the Connor Davidson Resilience Scale (CD-RISC) as applied within the Thai context. Scholar: Human Sciences, 8(2). [Google Scholar]
  28. Melhem NM, Zhong Y, Miller JM, Zanderigo F, Ogden RT, Sublette ME, Newell M, Burke A, Keilp JG, Lesanpezeshki M, Bartlett E, Brent DA, & Mann JJ, (2022). Brain 5-HT1A Receptor PET Binding, Cortisol Responses to Stress, and the Familial Transmission of Suicidal Behavior. Int. J. Neuropsychopharmacol, 25(1), 36–45. 10.1093/ijnp/pyab060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Oquendo MA, Halberstam B, Mann JJ: Risk factors for suicidal behavior: the utility and limitations of research instruments. In: First MB, ed. Standardized Evaluation in Clinical Practice Washington, DC: American Psychiatric Publishing; 2003:103–129. [Google Scholar]
  30. Pawlak JM, Miechowicz I, Dmitrzak-Węglarz M, Szczepankiewicz A, Zaremba D, Kapelski P, Rajewska-Rager A, Czerski P, Skibińska M, Hauser J, (2018). Are suicide risk factors gender specific. Czy czynniki związane z ryzykiem samobójstwa są specyficzne dla płci?. Psychiatr Pol, 52(1), 21–32. 10.12740/PP/78974 [DOI] [PubMed] [Google Scholar]
  31. Pietrzak RH, Goldstein MB, Malley JC, Rivers AJ, Johnson DC, & Southwick SM (2010). Risk and protective factors associated with suicidal ideation in veterans of Operations Enduring Freedom and Iraqi Freedom. J Affect Disord, 123(1–3), 102–107. 10.1016/j.jad.2009.08.001 [DOI] [PubMed] [Google Scholar]
  32. Ran L, Wang W, Ai M, Kong Y, Chen J, & Kuang L, (2020). Psychological resilience, depression, anxiety, and somatization symptoms in response to COVID-19: A study of the general population in China at the peak of its epidemic. Soc Sci Med. (1982), 262, 113261. 10.1016/j.socscimed.2020.113261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rasouli N, Malakouti SK, Rezaeian M, Saberi SM, Nojomi M, De Leo D, Ramezani- Farani A, (2019). Risk Factors of Suicide Death Based on Psychological Autopsy Method; a Case-Control Study. Arch Acad Emerg Med, 7(1), e50. [PMC free article] [PubMed] [Google Scholar]
  34. Richardson GE, Neiger BL, Jensen S, Kumpfer KL, (1990). The Resiliency Model. Am J Health Educ, 21, 33–39. [Google Scholar]
  35. Robinson JS, Larson CL, Cahill SP, (2014). Relations between resilience, positive and negative emotionality, and symptoms of anxiety and depression. Psychol Trauma, 6(Suppl 1), S92–S98. 10.1037/a0033733 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rolin H, Fossion P, Kotsou I, Leys C, (2018). Considérations sur la résilience : trait ou aptitude? [Perspectives on resilience: trait or aptitude?]. Rev Med Brux, 39(1), 22–28. 10.30637/2018.17-050 [DOI] [PubMed] [Google Scholar]
  37. Roy A, Sarchiapone M, Carli V, (2007). Low resilience in suicide attempters. Arch Suicide Res, 11(3), 265–269. 10.1080/13811110701403916 [DOI] [PubMed] [Google Scholar]
  38. Roy A, Carli V, & Sarchiapone M (2011). Resilience mitigates the suicide risk associated with childhood trauma. J Affect Disord, 133(3), 591–594. 10.1016/j.jad.2011.05.006 [DOI] [PubMed] [Google Scholar]
  39. Roy A, Carli V, Sarchiapone M, & Branchey M (2014). Comparisons of prisoners who make or do not make suicide attempts and further who make one or multiple attempts. Arch Suicide Res, 18(1), 28–38. 10.1080/13811118.2013.801816 [DOI] [PubMed] [Google Scholar]
  40. Rutter M, (1985). Resilience in the face of adversity. Protective factors and resistance to psychiatric disorder. Br J Psychiatry, 147, 598–611. 10.1192/bjp.147.6.598 [DOI] [PubMed] [Google Scholar]
  41. Rutter M, (2012). Resilience as a dynamic concept. Dev Psychopathol, 24(2), 335–344. 10.1017/S0954579412000028 [DOI] [PubMed] [Google Scholar]
  42. Song S, Yang X, Yang H, Zhou P, Ma H, Teng C, Chen H, Ou H, Li J, Mathews CA, Nutley S, Liu N, Zhang X, Zhang N, (2021) Psychological Resilience as a Protective Factor for Depression and Anxiety Among the Public During the Outbreak of COVID-19. Front Psychol 11, 618509. 10.3389/fpsyg.2020.618509 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Spitzer RL, Williams J, Gibbon M, et al. : Structured Clinical Interview for DSM-IV (SCID) New York State Psychiatric Institute, Biometrics Research Department, New York: (1995) [Google Scholar]
  44. Tadayon M, Dabirizadeh S, Zarea K, Behroozi N, & Haghighizadeh MH (2018). Investigating the relationship between psychological hardiness and resilience with depression in women with breast cancer. Gulf J Oncolog, 1(28), 23–30. [PubMed] [Google Scholar]
  45. Ziaian T, de Anstiss H, Antoniou G, Baghurst P, Sawyer M, (2012). Resilience and its association with depression, emotional and behavioural problems, and Mental Health Service utilisation among refugee adolescents living in South Australia. Int J Popul Res, 2012, 1–9. 10.1155/2012/485956 [DOI] [Google Scholar]
  46. Velickovic K, Rahm Hallberg I, Axelsson U, Borrebaeck C, Rydén L, Johnsson P, & Månsson J, (2020). Psychometric properties of the Connor-Davidson Resilience Scale (CD- RISC) in a non-clinical population in Sweden. Health Qual Life Outcomes, 18(1), 132. 10.1186/s12955-020-01383-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. You S, Park M, (2017). Resilience protected against suicidal behavior for men but not women in a community sample of older adults in Korea. Front Psychol, 8, 401. 10.3389/fpsyg.2017.00401 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Youssef NA, Green KT, Beckham JC, Elbogen EB, (2013). A 3-year longitudinal study examining the effect of resilience on suicidality in veterans. Ann Clin Psychiatry, 25(1), 59–66. [PMC free article] [PubMed] [Google Scholar]
  49. Yu X, & Zhang J (2007). Factor analysis and psychometric evaluation of the Connor-Davidson Resilience Scale (CD-RISC) with Chinese people. Soc Behav Pers, 35(1), 19–30. 10.2224/sbp.2007.35.1.19 [DOI] [Google Scholar]

RESOURCES