- Journal List
- HHS Author Manuscripts
- PMC9321173
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
J Affect Disord. Author manuscript; available in PMC 2023 Jun 15.
Published in final edited form as:
J Affect Disord. 2022 Jun 15; 307: 171–177.
Published online 2022 Mar 22. doi:10.1016/j.jad.2022.03.046
PMCID: PMC9321173
NIHMSID: NIHMS1796460
PMID: 35331824
Marsal Sanches, MD, PhD,1 Linh K Nguyen, PhD.,2 Tong Han Chung, PhD.,2 Paul Nestadt, MD,3 Holly C. Wilcox, PhD,4 William H. Coryell, MD, PhD,5 Jair C. Soares, MD, PhD.,1 and Sudhakar Selvaraj, MD, PhD.1
Author information Copyright and License information PMC Disclaimer
The publisher's final edited version of this article is available at J Affect Disord
This article has been corrected. See J Affect Disord. 2022 December 23; : .
Associated Data
- Supplementary Materials
Abstract
Background:
Though the association between anxiety disorders and suicidal behavior is well-described, the impact of anxiety symptoms on suicidal thoughts and behaviors (STB) across different mood disorders is still unclear.
Methods:
We performed a registry-based retrospective study utilizing outcome measure data collected by the National Network of Depression Centers (NNDC), a nationwide nonprofit consortium of 26 leading clinical and academic member centers in the United States. The sample consisted of 2,607 outpatients with mood disorders (major depressive disorder or bipolar disorders). Demographic and clinical variables were compared based on the presence or absence of STB and severity of anxiety symptoms (minimal, mild, moderate, and severe). Univariate and multivariable logistic regressions were conducted to examine the correlations of STB, considering multicollinearity.
Results:
Patients with mild, moderate, and severe anxiety symptoms had higher odds of STB than those with minimal symptoms. Gender, marital status, age, and depressive symptoms were other strong predictors of STB. There was no difference in the odds of STB between patients with major depressive disorder (MDD) and those with bipolar disorders (BD). However, the odds of suicidal ideation were slightly lower among patients with BD than those with MDD.
Limitations:
Our sample was comprised only of outpatients, limiting the generalization of our findings. Other limitations include the lack of structured interviews for diagnostic characterization of the patients and the utilization of data on anxiety and mood obtained solely through self-report scales.
Conclusions:
We found a cross-sectional association between the severity of anxiety symptoms and STB among patients with mood disorders. This study demonstrates the need for a suicide risk assessment in patients with mood disorders reporting anxiety symptoms.
Keywords: Anxiety, Suicidal Ideation, Suicidal Attempts, Depression, Major Depressive Disorder, Bipolar Disorder
INTRODUCTION
Suicide is the 10th leading cause of death (2nd among those under age 40) and is associated with over 47,000 deaths annually in the United States (Stone et al., 2021). In the last two decades, a steady increase in suicide rates in the United States has been observed in all geographic areas and age ranges (Baldessarini, 2019). The human and economic cost of suicidal behavior for individuals, families, communities, and society is substantial. Therefore the accurate recognition of underlying factors that can contribute to suicide risk in those in treatment for behavioral health concerns remains a clinical and health policy priority, given its importance and potential role in optimizing the identification of individuals at risk and preventing suicide.
The association between anxiety disorders and suicidal behavior is well-described (Fawcett, 1997). Epidemiological data from the National Comorbidity Survey show that a lifetime history of generalized anxiety disorder and posttraumatic stress disorder is associated with increases in the risk of suicide attempts (Cougle et al., 2009). The risks of suicide are significantly higher among patients with anxiety disorders (Pfeiffer et al., 2009). Comorbid anxiety disorders seem to increase the risk of suicide attempts in patients with mood disorders (Goldberg and Fawcett, 2012; Malhi et al., 2018; Oude Voshaar et al., 2016; Sareen et al., 2005). In a recently published National Network of Depression Centers (NNDC) mood outcomes data involving a total of 3,998 patients, we showed that suicidal ideation was highly prevalent in patients with unipolar depression and bipolar disorder (27-33%) and that comorbid anxiety was associated with the poorer longitudinal course (Zandi et al., 2020). However, it is not clear whether there are differences in the impact of anxiety symptoms on suicidal thoughts and behaviors (STB) across different mood disorders. A more refined characterization of the role of anxiety in STB among mood disorder patients is of critical importance for suicide prevention.
The main goal of this study was to evaluate the relationship between anxiety symptoms and STB among patients with unipolar and bipolar disorders. We hypothesized that patients with major depressive disorder and bipolar disorders would show similar associations between STB and severity of anxiety symptoms.
METHODS
Dataset
The study utilized data collected by the Mood Outcomes Program, which was implemented by The National Network of Depression Centers (NNDC) in 2011. The purpose and structure of the Mood Outcomes Program have been described in detail elsewhere (Zandi et al., 2020). The program utilizes a standard protocol to collect patient-reported outcomes during routine clinical care across the different participating sites. The data obtained are combined into a central repository that can be utilized for population health analyzes and quality improvement initiatives. In the present study, the following data were included: 1) patient demographic information (age, gender, race, ethnicity, and marital status); 2) mood disorder diagnosis; 3) presence and severity of anxiety symptoms, as measured by the 7-item Generalized Anxiety Disorder Scale (GAD-7); 4) STB as measured by the 7-item patient-rated screener version of the Columbia-Suicide Severity Rating Scale (C-SSRS); and 5) presence and severity of depressive symptoms, as measured by the 9-item Patient Health Questionnaire (PHQ- 9). The data included in the present analysis were collected between January 1, 2012, and December 31, 2020. The clinical data included in the present study was collected as part of the standard of care and are fully deidentified; therefore, patient informed consent was waived. The study was approved by the respective Institutional Review Board.
Study Design and Population
This is a retrospective outpatient clinical data study. The study population was categorized into patients with bipolar disorders (BD) and patients with major depressive disorder (MDD), based on the primary diagnosis included at the time of their earliest visit. Patients who had a primary diagnosis of any anxiety disorder were excluded from the analysis. In our secondary analysis, we considered a secondary diagnosis of anxiety and therefore categorized patients into four groups: 1) BD; 2) MDD; 3) BD but no secondary diagnosis of anxiety; 4) MDD but no secondary diagnosis of anxiety. This approach aimed to assess the impact of anxiety as a secondary diagnosis in addition to the primary diagnosis of mood disorders.
Measures
Suicidal Thoughts and Behaviors
For purposes of the present analysis, only baseline assessments were included. STB status was dichotomized as either “positive” or “negative.” Positive STB was defined as a positive answer to item number 5, 6b, or 7b of the C-SSRS, which include current (over the past month) suicidal ideation with intent to act and recent (within the last three months) preparatory actions, suicide attempts, and aborted or interrupted suicidal attempts (Posner et al., 2011). For the secondary analyses, we defined suicidal ideation as a positive answer to item 2 (non-specific suicidal thoughts without thinking of suicide methods, intent, or plan) and past suicidal attempts as a positive answer to item number 6 of the C-SSRS.
Anxiety symptoms
Anxiety symptoms were evaluated using the GAD-7 questionnaire (Spitzer et al., 2006). Only complete records and non-missing answers to any of the GAD-7 questions were included in the analyses. If multiple assessments are on the same date, we kept the record with the highest score per NNDC guidelines. Based on the GAD-7 scores, we categorized anxiety symptoms into minimal anxiety (<5), mild anxiety (5-9), moderate anxiety (10-14), and severe anxiety (≥ 15).
Depressive symptoms
Depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9) (Kroenke et al., 2001). Only complete records and non-missing answers to any of the PHQ-9 questions 1-9 were included in the analysis. We also followed the NNDC guideline to keep the record with the highest score if multiple assessments are on one date. Depressive symptom severity was categorized into three groups: cut points <10 representing no/minimal depression, 10-19 moderate depression, and ≥ 20 severe depression.
Demographic characteristics
Demographic characteristics at the first baseline assessment include age, gender at birth (male or female), race (White, non-White, or unknown), ethnicity (Hispanic, non-Hispanic or unknown), and marital status (single/never married, married/in a committed relationship, or separated/widowed/divorced).
Analysis
If there are multiple baseline assessments, the earliest baseline assessment for each patient was included in the analyses. Demographic and clinical variables were compared based on C-SSRS and GAD-7 responses. Chi-square test and Fisher’s exact test were performed for categorical variables; t-test and one-way ANOVA were performed for continuous variables. We conducted univariate and multivariable logistic regressions using the earliest baseline assessment to examine associations with STB, considering multicollinearity. Statistical Analysis was conducted using STATA14 (StataCorp, 2015).
RESULTS
Among 12,653 patients with records in the mood outcomes dataset, we identified 7,136 patients with mood disorders as having a primary diagnosis of either major depressive disorder (MDD) or bipolar disorders (BD). The final analytical population consisted of 2,607 patients with mood disorders who had valid, non-missing assessments and demographic information. Figure 1 shows the flow chart of the selection of the study population. Corresponding estimates of STB (C-SSRS), anxiety symptoms (GAD-7), and depressive symptoms (PHQ-9), as well as patient demographic information, were obtained for a visit.
Figure 1 –
Study Population Flowchart
Table 1 presents the patient’s demographic characteristics and key clinical constructs by STB status (positive/negative) and severity of anxiety symptoms (minimal, mild, moderate, and severe). Out of 2,607 patients, 304 (11.66%) were found to have positive STB status. There was no significant difference in the distribution of race, ethnicity, and primary mood disorder diagnosis between patients who reported STB compared with those who did not. Flowever, we found significant differences in the distribution of age, marital status, and severity of depressive symptoms by STB status. A higher proportion of men (14.61% vs. 10.35%), single (17.62% vs. 6.9% & 7.9%), having severe depressive symptoms (21.88% vs. 5.7% & 12.5), and younger (33.39 years old vs. 40.21 years old) reported STB (p<0.001). Also, at baseline, 1,990 patients (76.33%) reported at least mild anxiety symptoms (592 (22.71%) mild, 579 (22.21%) moderate, and 819 (31.42%) severe). There was no significant difference in the distribution of race, ethnicity, and marital status by the severity of anxiety symptoms. The distribution of gender, severity of depressive symptoms, primary mood disorder diagnosis, and age were statistically different by the severity of anxiety symptoms.
Table 1.
Clinical and Demographic Characteristics by STB Status and GAD-7 Score
| Variables | C-SSRS Positive Status | GAD-7 Anxiety Level | ||||||
|---|---|---|---|---|---|---|---|---|
| (N, %) | (N, %) | |||||||
| C-SSRS (+) | C-SSRS (−) | P- value | Minimal Anxiety | Mild Anxiety | Moderate Anxiety | Severe Anxiety | P- value | |
| Total N= 2,607 | N= 304 (11.66) | N= 2,303 (88.34) | N= 617 (23.67) | N= 592 (22.71) | N= 579 (22.21) | N= 819 (31.42) | ||
| Gender | 0.002 | |||||||
| Male | 117 (14.61) | 684 (85.39) | 229 (28.59) | 190 (23.72) | 158 (19.73) | 224 (27.97) | 0.000 | |
| Female | 187 (10.35) | 1,619 (89.65) | 388 (21.48) | 402 (22.26) | 421 (23.31) | 595 (32.95) | ||
| Race | 0.566 | 0.171 | ||||||
| White | 252 (11.4) | 1,958 (88.6) | 539 (24.39) | 511 (23.12) | 482 (21.81) | 678 (30.68) | ||
| Non-White | 50 (13.16) | 330 (86.84) | 75 (19.74) | 78 (20.53) | 94 (24.74) | 133 (35) | ||
| Unknown | 2 (11.76) | 15 (88.24) | 3 (17.65) | 3 (17.65) | 3 (17.65) | 8 (47.06) | ||
| Ethnicity | 0.676 | 0.520 | ||||||
| Hispanic | 16 (13.79) | 100 (86.21) | 18 (15.52) | 29 (25) | 30 (25.86) | 39 (33.62) | ||
| Non-Hispanic | 285 (11.61) | 2,169 (88.39) | 591 (24.08) | 556 (22.66) | 540 (22) | 767 (31.26) | ||
| Unknown | 3 (8.11) | 34 (91.89) | 8 (21.62) | 7 (18.92) | 9 (24.32) | 13 (35.14) | ||
| Marital status | 0.000 | 0.281 | ||||||
| Single (Never married) | 200 (17.62) | 935 (82.38) | 250 (22.03) | 250 (22.03) | 257 (22.64) | 378 (33.3) | ||
| Married or in a committed relationship | 81 (6.86) | 1,100 (93.14) | 289 (24.47) | 283 (23.96) | 260 (22.02) | 349 (29.55) | ||
| Separated, widowed or divorced | 23 (7.9) | 268 (92.1) | 78 (26.8) | 59 (20.27) | 62 (21.31) | 92 (31.62) | ||
| PHQ–9 score categories | 0.000 | 0.000 | ||||||
| No/Minimal depression (score ≤ 9) | 57 (5.73) | 937 (94.27) | 527 (53.02) | 302 (30.38) | 121 (12.17) | 44 (4.43) | ||
| Moderate depression (score 10–19) | 140 (12.46) | 984 (87.54) | 83 (7.38) | 260 (23.13) | 381 (33.9) | 400 (35.59) | ||
| Severe depression (score ≥ 20) | 107 (21.88) | 382 (78.12) | 7 (1.43) | 30 (6.13) | 77 (15.75) | 375 (76.69) | ||
| Primary mood disorder diagnosis | 0.234 | 0.000 | ||||||
| Major depressive disorder | 259 (12) | 1,899 (88) | 477 (22.1) | 496 (22.98) | 496 (22.98) | 689 (31.93) | ||
| Bipolar disorders | 45 (10.02) | 404 (89.98) | 140 (31.18) | 96 (21.38) | 83 (18.49) | 130 (28.95) | ||
| Age, years - Mean (SD) | 33.39 (13.98) | 40.21 (15.72) | 0.000 | 45.0 (16.82) | 39.48 (15.92) | 37.34 (15.27) | 36.63 (13.72) | 0.000 |
Open in a separate window
Using baseline assessment at the earliest visit, both the univariate and multivariable logistic regression revealed associations between anxiety symptoms and STB among patients with mood disorders. Table 2 displays the relationships between severity of anxiety symptoms, demographic and clinical factors, and STB. Controlling for other factors, patients with mild, moderate, and severe anxiety symptoms had higher odds of STB than those with minimal anxiety symptoms (p<0.001). Gender, marital status, age, and severity of depressive symptomatology (measured through the PHQ-9) were also strongly associated with STB risk. Older patients, female, or married/committed relationships had lower STB odds than their younger, male, single counterparts (p <0.001). In addition, compared with patients with no/minimal depressive symptoms, patients with moderate and severe depressive symptoms had higher odds of STB (p<0.01). The adjusted odds of STB were lower among patients with bipolar disorders than those who had major depressive disorder; however, that difference was not statistically significant.
Table 2 –
Baseline Anxiety Symptom Severity and STB: Logistic Regression
| Correlates of STB | ||||||
|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariable Analysis | |||||
| Variables | Odds Ratio | Std. Err. | P- value | Odds Ratio | Std. Err. | P- value |
| Anxiety Level (using GAD-7) | ||||||
| Minimal Anxiety | Reference | Reference | ||||
| Mild Anxiety | 2.65 | 0.63 | 0.000 | 1.94 | 0.49 | 0.009 |
| Moderate Anxiety | 3.71 | 0.85 | 0.000 | 2.15 | 0.57 | 0.004 |
| Severe Anxiety | 4.09 | 0.89 | 0.000 | 1.69 | 0.47 | 0.056 |
| Sex | ||||||
| Male | Reference | Reference | ||||
| Female | 0.68 | 0.09 | 0.000 | 0.63 | 0.09 | 0.001 |
| Race | ||||||
| White | Reference | Reference | ||||
| Non-White | 1.18 | 0.2 | 0.33 | 0.96 | 0.18 | 0.817 |
| Unknown | 1.04 | 0.78 | 0.960 | 0.86 | 0.74 | 0.863 |
| Ethnicity | ||||||
| Hispanic | Reference | Reference | ||||
| Non-Hispanic | 0.82 | 0.23 | 0.48 | 0.98 | 0.30 | 0.948 |
| Unknown | 0.55 | 0.36 | 0.37 | 0.61 | 0.46 | 0.514 |
| Marital status | ||||||
| Single (Never married) | Reference | Reference | ||||
| Married or in a committed relationship | 0.34 | 0.05 | 0.000 | 0.46 | 0.07 | 0.000 |
| Separated, widowed or divorced | 0.40 | 0.09 | 0.000 | 0.6 | 0.15 | 0.047 |
| PHQ–9 score categories | ||||||
| No/Minimal depression (score ≤ 9) | Reference | Reference | ||||
| Moderate depression (score 10–19) | 2.34 | 0.38 | 0.000 | 1.72 | 0.34 | 0.005 |
| Severe depression (score ≥ 20) | 4.60 | 0.81 | 0.000 | 3.50 | 0.80 | 0.000 |
| Primary mood disorder diagnosis | ||||||
| Major depressive disorder | Reference | Reference | ||||
| Bipolar disorders | 0.82 | 0.14 | 0.24 | 0.91 | 0.16 | 0.586 |
| Age | 0.97 | 0.004 | 0.000 | 0.98 | 0.01 | 0.000 |
| Intercept | 0.19 | 0.08 | 0.000 | |||
Open in a separate window
In the secondary analysis (Table 3), the patients’ diagnoses were categorized into four groups (BD, MDD, BD with secondary diagnosis of anxiety, and MDD with secondary diagnosis of anxiety). The associations described in the primary analysis (with just two diagnostic categories), related to anxiety symptoms, sex, marital status, age at the time of baseline assessment, and depressive symptoms, did not change significantly.
Table 3 –
Baseline Anxiety Symptom Severity and STB: Logistic Regression with Four Diagnostic Categories
| Correlates of STB | ||||||
|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariable Analysis | |||||
| Variables | Odds Ratio | Std. Err. | P- value | Odds Ratio | Std. Err. | P- value |
| Anxiety Level (using GAD-7) | ||||||
| Minimal Anxiety | Reference | Reference | ||||
| Mild Anxiety | 2.65 | 0.63 | 0.000 | 1.95 | 0.5 | 0.009 |
| Moderate Anxiety | 3.71 | 0.85 | 0.000 | 2.16 | 0.58 | 0.004 |
| Severe Anxiety | 4.09 | 0.89 | 0.000 | 1.68 | 0.47 | 0.061 |
| Sex | ||||||
| Male | Reference | Reference | ||||
| Female | 0.68 | 0.09 | 0.002 | 0.63 | 0.09 | 0.001 |
| Race | ||||||
| White | Reference | Reference | ||||
| Non-White | 1.18 | 0.20 | 0.325 | 0.95 | 0.18 | 0.800 |
| Unknown | 1.04 | 0.78 | 0.963 | 0.86 | 0.73 | 0.854 |
| Ethnicity | ||||||
| Hispanic | Reference | Reference | ||||
| Non-Hispanic | 0.82 | 0.23 | 0.476 | 0.98 | 0.30 | 0.947 |
| Unknown | 0.55 | 0.36 | 0.367 | 0.62 | 0.47 | 0.525 |
| Marital status | ||||||
| Single (Never married) | Reference | Reference | ||||
| Married or in a committed relationship | 0.34 | 0.05 | 0.000 | 0.46 | 0.07 | 0.000 |
| Separated, widowed or divorced | 0.4 | 0.09 | 0.000 | 0.60 | 0.15 | 0.048 |
| PHQ–9 score categories | ||||||
| No/Minimal depression (score ≥ 9) | Reference | Reference | ||||
| Moderate depression (score 10–19) | 2.34 | 0.38 | 0.000 | 1.71 | 0.33 | 0.006 |
| Severe depression (score ≥ 20) | 4.6 | 0.81 | 0.000 | 3.52 | 0.81 | 0.000 |
| Primary mood disorder diagnosis | ||||||
| Major depressive disorder | Reference | Reference | ||||
| Bipolar disorders | 0.66 | 0.15 | 0.070 | 0.69 | 0.16 | 0.119 |
| Major depressive disorder + Anxiety (Secondary dx) | 1.19 | 0.16 | 0.183 | 0.89 | 0.13 | 0.417 |
| Bipolar disorders + Anxiety (Secondary dx) | 1.51 | 0.39 | 0.113 | 1.23 | 0.34 | 0.446 |
| Age | 0.97 | 0.004 | 0.000 | 0.98 | 0.01 | 0.000 |
| Intercept | 0.20 | 0.08 | 0.000 | |||
Open in a separate window
In addition, we analyzed correlates of the risk of current suicidal ideation (non-specific suicidal thoughts without thinking of suicide methods, intent, or plan). Notably different from examining STB risk, the odds of having suicidal ideation significantly differed between the two diagnostic categories (MDD and BD). The adjusted odds of having suicidal ideation among patients with bipolar disorder were 35% lower than those with MDD (Table 4). We also looked at possible relationships between past suicidal attempts and current suicidal ideation, considering the severity of anxiety symptoms (Table 5, Supplement 1). We found that compared with patients with minimal anxiety symptoms and no past suicidal attempts, patients with more severe anxiety symptoms but no past suicide attempts had higher odds of having current suicidal ideation. The adjusted ORs for suicidal ideation among patients with mild, moderate, and severe anxiety symptoms were 1.49, 1.98, and 2.11, respectively. Patients who had attempted suicide in the past had strikingly higher odds of having recent suicidal ideation. Among those individuals, the adjusted ORs for suicidal ideation in patients with mild, moderate, and severe anxiety symptoms were 8.85, 8.82 and 7.01, respectively (p-value <0.05).
Table 4 –
Baseline Anxiety Symptom Severity and Suicidal Ideation: Logistic Regression
| Correlates of Suicidal Ideation | ||||||
|---|---|---|---|---|---|---|
| Univariate Analysis | Multivariable Analysis | |||||
| Variables | Odds Ratio | Std. Err. | P-value | Odds Ratio | Std. Err. | P-value |
| Anxiety Level (GAD-7) | ||||||
| Minimal Anxiety | Reference | Reference | ||||
| Mild Anxiety | 2.84 | 0.53 | < 0.001 | 1.79 | 0.37 | 0.004 |
| Moderate Anxiety | 4.4 | 0.8 | < 0.001 | 1.97 | 0.42 | 0.001 |
| Severe Anxiety | 6.75 | 1.16 | < 0.001 | 1.91 | 0.42 | 0.003 |
| Sex | ||||||
| Male | Reference | Reference | ||||
| Female | 0.68 | 0.07 | < 0.001 | 0.58 | 0.07 | < 0.001 |
| Race | ||||||
| White | Reference | Reference | ||||
| Non-White | 1.05 | 0.14 | 0.714 | 0.8 | 0.12 | 0.142 |
| Unknown | 1.45 | 0.78 | 0.486 | 0.64 | 0.42 | 0.502 |
| Ethnicity | ||||||
| Hispanic | Reference | Reference | ||||
| Non-Hispanic | 0.75 | 0.16 | 0.169 | 0.79 | 0.20 | 0.357 |
| Unknown | 1.11 | 0.46 | 0.801 | 1.38 | 0.72 | 0.536 |
| Marital status | ||||||
| Single (Never married) | Reference | Reference | ||||
| Married or in a committed relationship | 0.41 | 0.04 | < 0.001 | 0.54 | 0.06 | < 0.001 |
| Separated, widowed or divorced | 0.48 | 0.08 | < 0.001 | 0.71 | 0.14 | 0.085 |
| Depressive symptoms (PHQ-9 score) | ||||||
| No/Minimal depression (score ≤ 9) | Reference | Reference | ||||
| Moderate depression (score 10–19) | 3.41 | 0.46 | < 0.001 | 2.5 | 0.40 | < 0.001 |
| Severe depression (score ≥ 20) | 9.86 | 1.44 | < 0.001 | 7.03 | 1.33 | < 0.001 |
| Primary mood disorder diagnosis | ||||||
| Major depressive disorder | Reference | Reference | ||||
| Bipolar disorders | 0.65 | 0.09 | 0.001 | 0.71 | 0.11 | 0.022 |
| Age | 0.97 | 0.003 | < 0.001 | 0.98 | 0.004 | < 0.001 |
| Intercept | 0.41 | 0.14 | <0.001 | |||
Open in a separate window
Furthermore, we investigated the association between STB risk and comorbid anxiety and depression (Table 6, Supplement 2). We observed that, compared with patients who had minimal anxiety symptoms and no/minimal depressive symptoms, the odds of STB were 5.97 (patients with severe anxiety and severe depression), 6.26 (patients with minimal anxiety and severe depression), 2.92 (patients with severe anxiety and no/minimal depression). Patients with any level of anxiety symptoms and severe depression tended to have the highest odds of STB.
DISCUSSION
Our findings indicate a strong association between the severity of anxiety symptomatology and current STB in a sample of individuals with MDD and BD. There were no differences between mood disorder categories (MDD and BD) regarding STB risk. However, the odds of experiencing current suicidal ideation were slightly lower among individuals with BD than those with MDD. Finally, our findings indicate that, among individuals with no previous history of suicide attempts, the odds of current suicidal ideation were proportional to the severity of anxiety symptoms, with higher anxiety indicating higher odds of current suicidal ideation. In contrast, among patients with a history of suicide attempts, the odds of current suicidal ideation were higher among individuals with moderate anxiety than mild anxiety but smaller among those with severe anxiety. These findings have important implications for the identification of individuals at increased risk of suicide.
The fact that anxiety symptomatology is associated with higher risks of STB is well-described. It has been an object of great interest since the publication of the findings by Weissman et al (Weissman et al., 1989). They identified increased odds of suicidal ideation and attempts among patients from the Epidemiologic Catchment Area (ECA) Study suffering from panic disorder/panic attacks. Several possible mechanisms to explain the association between anxiety and STB have been proposed, including the possibility of direct effects of anxiety symptoms in triggering suicidal ideation and attempts, perceptions of powerlessness when facing an aversive status/role, and common factors associated with both anxiety symptoms and suicidal behavior, such as childhood adversity, genetic predisposition, medical issues, and social isolation (Hawton and van Heeringen, 2009; Taylor et al., 2011). Our findings, pointing to associations between anxiety severity and current STB, agree with literature findings and support the hypothesis mentioned above.
On the other hand, other studies analyzing data from the ECA and the National Comorbidity Survey have found conflicting results regarding the association between anxiety disorders and suicide, leading to some questioning whether anxiety disorders represent an independent risk factor for suicidal behavior (Sareen, 2011). While, in some studies, anxiety was associated with increased suicide risk even in the absence of a positive screening for depression (Shepardson et al., 2019), other groups did not find such association. For example, in the Netherlands Study of Depression and Anxiety, comorbid depression and anxiety disorders, in contrast with a sole diagnosis of an anxiety disorder, increased the odds of suicidal behavior (Wiebenga et al., 2021). Some of these discrepancies could be, at least in part, related to methodological differences in the adjustment for other comorbid disorders. While it has been proposed that indirect effects of anxiety on other psychiatric conditions associated with elevated suicide risk, such as mood disorders, could contribute to increased rates of suicide, it is not clear how specific those indirect effects are. In our sample, anxiety severity seemed to have different impact patterns on current suicidal ideation. While among individuals with no recent history of suicide attempts, higher anxiety indicated higher odds of current suicidal ideation, in recent suicide attempters, the odds of current suicidal ideation were significantly elevated among patients with mild, moderate, or severe anxiety, but there was no direct correlation between anxiety severity and those odds, which was slightly higher among patients with moderate anxiety compared to those with mild anxiety but proportionally lower among individuals with severe anxiety. Those findings suggest that a possible modulatory effect of anxiety on suicidal ideation may be distinct among individuals with and without a recent history of suicidal attempts. Since a history of suicide attempts may be considered an indicator of higher severity of mental disorders, particularly depression (Melhem et al., 2019), one can hypothesize that, among patients with more severe mental illness, the impact of anxiety on suicidal ideation, albeit significant, might be less specific and less dependent on the severity of the anxiety symptoms. At least one study has demonstrated complex interactions between suicide risk factors and comorbid anxiety and depression, with possible synergic effects of anxiety and depressive symptoms regarding suicide risk (Zhang et al., 2019). In that study, compared with individuals with both low levels of depression and anxiety, the odds ratio for completed suicide were 54.77 (for individuals with high anxiety and high depression), 26.32 (for those with low anxiety and high depression), and 2.46 (for those with high anxiety and low depression).
We also found evidence of a possible effect of other demographic factors, such as age, gender, and marital status, on the interaction between anxiety symptomatology and STB among patients with mood disorders. Specifically, young age, single marital status, and male gender were associated with higher rates of STB. There were no statistically significant differences between patients with unipolar and bipolar disorder regarding those effects. While young age, as well as non-married marital status and male gender, are associated with a higher risk of suicide among patients with a mood disorders (Miller and Black, 2020), these findings may indicate the impact of anxiety disorders on suicidality among younger and older patients is distinct. Further research is necessary to clarify that issue.
Furthermore, there was no difference between patients with MDD and BD regarding the odds of suicidal ideation. These results do not seem to agree with literature findings, which point to higher rates of suicide among patients with BD (Holma et al., 2014). These conflicting findings may be related to the nature of our sample, comprised of patients with possibly less severe forms of BD.
Finally, in our secondary analysis, we attempted to verify the impact of a secondary diagnosis of an anxiety disorder (among patients with MDD and BD) compared to those without a formal diagnosis of comorbid anxiety disorders. The differences in the risk for STB were not statistically significant among these groups, suggesting that a comorbid diagnosis of anxiety is not necessarily implicated in elevations in the odds of suicidal behavior. The elevated risk of suicide associated with the primary psychiatric conditions mentioned above might override the increase in risk associated with a diagnosis of anxiety disorder. In the Lundby Study, a longitudinal cohort study carried out for 50 years, individuals diagnosed with an anxiety disorder had a higher suicide risk than those with no diagnosis. However, that risk was not as elevated as the one observed associated with other psychiatric conditions(Anderberg et al., 2016). Another study offered similar conclusions using data from the 2012 Canadian Community Health Survey–Mental Health, which described identical rates of suicide ideation among patients with subthreshold and threshold generalized anxiety disorders (Gilmour, 2016).
Limitations
Our study has some limitations that need to be acknowledged. We analyzed a mood disorders registry sample comprised of individuals currently engaged in psychiatric care. It is unclear whether our findings and conclusions would also apply to patients in the community. In addition, our diagnoses were clinically established without the administration of a structured interview. Further, we did not include data on other comorbidities that could potentially impact the relationship between anxiety and suicidality in the sample studied, such as personality disorders and substance use disorders. Last, the data included in the present analysis were obtained through self-report instruments, with the limitations and biases inherent to self-report instruments.
In summary, our findings support the strong association between current anxiety symptoms and STB. Clinicians need to continue to be educated about the critical role anxiety symptoms play in developing suicidal ideation and behavior and the need to screen patients with anxiety regarding their risk of suicide.
Highlights
We identified a strong association between severity of anxiety symptoms and suicidal thoughts and behaviors among patients with mood disorders
This association was similar across different subtypes of mood disorder (unipolar depression versus bipolar disorder)
in the specific case of suicidal ideation, patients with bipolar disorder were found to have slightly lower odds when compared to those with unipolar depression
Supplementary Material
1
Click here to view.(22K, docx)
2
Click here to view.(22K, docx)
Acknowledgements
This work was made possible, in part, by a research collaboration supported by the National Network of Depression Centers (NNDC), a consortium of academic medical centers united to advance research and care delivery for mood disorders. We thank the member centers and patients of the National Network of Depression Centers Mood Outcomes Program.
Special thanks to Pat Rinvelt, Diana Burnett, Dane Larsen from the NNDC, and Ryan Callahan from the Altarum Mood Outcomes team for facilitating the project
This work was made possible by a research collaboration supported by the National Network of Depression Centers (NNDC), an inter-dependent consortium of academic depression centers. Operational support was provided by: Edie Douglas, MPH (Medical University of South Carolina), Kara Glazer (Johns Hopkins University), Marjorie Gresbrink (Mayo Clinic), Deanna Hofschulte, CCRP (Mayo Clinic), Dane Larsen (NNDC), Martha Shaw (University Iowa), Dana Steidtmann, PhD (University of Colorado Anschutz Medical Campus), and Carol Wahl (University of Louisville).
Altarum, nonprofit research and consulting organization that creates and implements solutions to advance health, provided technical expertise to build and maintain the Mood Outcomes system. Operational support from Altarum was provided by: Rachelle May-Gentile, Mike Grim, and Ryan Callahan.
Special thanks also go to Katie Hurtis and other members of the Epic staff for their ongoing assistance and support during this effort.
Role of Funding
This study received no specific funding from any agency in the public, commercial, or not-for-profit sectors
Role of Funder
The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or SAMHSA. The UTHealth institution played 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.
Conflicts of Interest
Dr. Selvaraj has received speaking honoraria from Global Medical Education and honoraria from British Medical Journal Publishing Group and owns shares at Flow Med Tech and site-investigator for a clinical trial by COMPASS Pathways Limited and Liva Nova.
Dr. Soares has received grants/research support from BMS, Forrest, J&J, Merck, Compass pathways, Stanley Medical Research Institute, NIH and has been a speaker for Pfizer and Abbott.
All other authors have no conflicts of interest to report
REFERENCES
- Anderberg J, Bogren M, Mattisson C, Brådvik L, 2016. Long-Term Suicide Risk in Anxiety-The Lundby Study 1947-2011. Arch Suicide Res20, 463–475. 10.1080/13811118.2015.1057663 [PubMed] [CrossRef] [Google Scholar]
- Baldessarini RJ, 2019. Epidemiology of suicide: recent developments. Epidemiol Psychiatr Sci29, e71. 10.1017/S2045796019000672 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Cougle JR, Keough ME, Riccardi CJ, Sachs-Ericsson N, 2009. Anxiety disorders and suicidality in the National Comorbidity Survey-Replication. J Psychiatr Res43, 825–829. 10.1016/j.jpsychires.2008.12.004 [PubMed] [CrossRef] [Google Scholar]
- Fawcett J, 1997. The detection and consequences of anxiety in clinical depression. J Clin Psychiatry58Suppl 8, 35–40. [PubMed] [Google Scholar]
- Gilmour H, 2016. Threshold and subthreshold Generalized Anxiety Disorder (GAD) and suicide ideation. Health Reports27, 11. [PubMed] [Google Scholar]
- Goldberg D, Fawcett J, 2012. The importance of anxiety in both major depression and bipolar disorder. Depress Anxiety29, 471–478. 10.1002/da.21939 [PubMed] [CrossRef] [Google Scholar]
- Hawton K, van Heeringen K, 2009. Suicide. Lancet373, 1372–1381. 10.1016/S0140-6736(09)60372-X [PubMed] [CrossRef] [Google Scholar]
- Holma KM, Haukka J, Suominen K, Valtonen HM, Mantere O, Melartin TK, Sokero TP, Oquendo MA, Isometsä ET, 2014. Differences in incidence of suicide attempts between bipolar I and II disorders and major depressive disorder. Bipolar Disord16, 652–661. 10.1111/bdi.12195 [PubMed] [CrossRef] [Google Scholar]
- Kroenke K, Spitzer RL, Williams JB, 2001. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med16, 606–613. 10.1046/j.1525-1497.2001.016009606.x [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Malhi GS, Outhred T, Das P, Morris G, Hamilton A, Mannie Z, 2018. Modeling suicide in bipolar disorders. Bipolar Disord20, 334–348. 10.1111/bdi.12622 [PubMed] [CrossRef] [Google Scholar]
- Melhem NM, Porta G, Oquendo MA, Zelazny J, Keilp JG, Iyengar S, Burke A, Birmaher B, Stanley B, Mann JJ, Brent DA, 2019. Severity and Variability of Depression Symptoms Predicting Suicide Attempt in High-Risk Individuals. JAMA Psychiatry76, 603–613. 10.1001/jamapsychiatry.2018.4513 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Miller JN, Black DW, 2020. Bipolar Disorder and Suicide: a Review. Curr Psychiatry Rep22, 6. 10.1007/s11920-020-1130-0 [PubMed] [CrossRef] [Google Scholar]
- Oude Voshaar RC, van der Veen DC, Hunt I, Kapur N, 2016. Suicide in late-life depression with and without comorbid anxiety disorders. Int J Geriatr Psychiatry31, 146–152. 10.1002/gps.4304 [PubMed] [CrossRef] [Google Scholar]
- Pfeiffer PN, Ganoczy D, Ilgen M, Zivin K, Valenstein M, 2009. Comorbid anxiety as a suicide risk factor among depressed veterans. Depress. Anxiety26, 752–757. 10.1002/da.20583 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA, Currier GW, Melvin GA, Greenhill L, Shen S, Mann JJ, 2011. The Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry168, 1266–1277. 10.1176/appi.ajp.2011.10111704 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Sareen J, 2011. Anxiety disorders and risk for suicide: why such controversy?Depress. Anxiety28, 941–945. 10.1002/da.20906 [PubMed] [CrossRef] [Google Scholar]
- Sareen J, Cox BJ, Afifi TO, de Graaf R, Asmundson GJG, ten Have M, Stein MB, 2005. Anxiety disorders and risk for suicidal ideation and suicide attempts: a population-based longitudinal study of adults. Arch Gen Psychiatry62, 1249–1257. 10.1001/archpsyc.62.11.1249 [PubMed] [CrossRef] [Google Scholar]
- Shepardson RL, Kosiba JD, Bernstein LI, Funderburk JS, 2019. Suicide risk among Veteran primary care patients with current anxiety symptoms. Fam Pract36, 91–95. 10.1093/fampra/cmy088 [PubMed] [CrossRef] [Google Scholar]
- Spitzer RL, Kroenke K, Williams JBW, Löwe B, 2006. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med166, 1092–1097. 10.1001/archinte.166.10.1092 [PubMed] [CrossRef] [Google Scholar]
- StataCorp, 2015. Stata Statistical Software: Release 14. StataCorp LP, College Station, TX. [Google Scholar]
- Stone DM, Jones CM, Mack KA, 2021. Changes in Suicide Rates - United States, 2018-2019. MMWR Morb Mortal Wkly Rep70, 261–268. 10.15585/mmwr.mm7008a1 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
- Taylor PJ, Gooding P, Wood AM, Tarrier N, 2011. The role of defeat and entrapment in depression, anxiety, and suicide. Psychol Bull137, 391–420. 10.1037/a0022935 [PubMed] [CrossRef] [Google Scholar]
- Weissman MM, Klerman GL, Markowitz JS, Ouellette R, 1989. Suicidal ideation and suicide attempts in panic disorder and attacks. N Engl J Med321, 1209–1214. 10.1056/NEJM198911023211801 [PubMed] [CrossRef] [Google Scholar]
- Wiebenga JX, Eikelenboom M, Heering HD, van Oppen P, Penninx BW, 2021. Suicide ideation versus suicide attempt: Examining overlapping and differential determinants in a large cohort of patients with depression and/or anxiety. Aust N Z J Psychiatry55, 167–179. 10.1177/0004867420951256 [PubMed] [CrossRef] [Google Scholar]
- Zandi PP, Wang Y-H, Patel PD, Katzelnick D, Turvey CL, Wright JH, Ajilore O, Coryell W, Schneck CD, Guille C, Saunders EFH, Lazarus SA, Cuellar VA, Selvaraj S, Dill Rinvelt P, Greden JF, DePaulo JR, 2020. Development of the National Network of Depression Centers Mood Outcomes Program: A Multisite Platform for Measurement-Based Care. PS71, 456–464. 10.1176/appi.ps.201900481 [PubMed] [CrossRef] [Google Scholar]
- Zhang J, Liu X, Fang L, 2019. Combined effects of depression and anxiety on suicide A case-control psychological autopsy study in rural China. Psychiatry Research271, 370–373. 10.1016/j.psychres.2018.11.010 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
