A clinical measure of suicidal ideation, suicidal behavior, and associated symptoms in bipolar disorder: Psychometric properties of the Concise Health Risk Tracking Self-Report (CHRT-SR)

Abstract

Objective:

People with bipolar disorder are at high risk of suicide, but no clinically useful scale has been validated in this population. The aim of this study was to evaluate the psychometric properties in bipolar disorder of the 7- and 12-item versions of the Concise Health Risk Tracking Self-Report (CHRT-SR), a scale measuring suicidal ideation, suicidal behavior, and associated symptoms.

Methods:

The CHRT was administered to 283 symptomatic outpatients with bipolar I or II disorder who were randomized to receive lithium plus optimized personalized treatment (OPT), or OPT without lithium in a six month longitudinal comparative effectiveness trial. Participants were assessed using structured diagnostic interviews, clinician-rated assessments, and self-report questionnaires.

Results:

The internal consistency (Cronbach α) was 0.80 for the 7-item CHRT-SR and 0.90 for the 12-item CHRT-SR with a consistent factor structure, and three independent factors (current suicidal thoughts and plans, hopelessness, and perceived lack of social support) for the 7-item version. CHRT-SR scores are correlated with measures of depression, functioning, and quality of life, but not with mania scores.

Conclusions:

The 7- and 12-item CHRT-SR both had excellent psychometric properties in a sample of symptomatic subjects with bipolar disorder. The scale is highly correlated with depression, functioning, and quality of life, but not with mania. Future research is needed to determine whether the CHRT-SR will be able to predict suicide attempts in clinical practice.

1. Introduction

Suicidal behavior, which incorporates the act of suicide, is an unfortunately too common outcome for patients with bipolar disorder, causing disproportionate mortality for these patients (Costa et al., 2015; Jamison, 2000; Leverich et al., 2003). Approximately one third to one half of bipolar patients will attempt suicide at least once (Gonda et al., 2012) with the rate of completed suicide among bipolar patients estimated to be within the range of 10–19% (Nierenberg et al., 2014). Suicidal behavior among patients with bipolar disorder may also be more lethal. Within the general population of the United States, completed suicide occurs one time for every 30 attempts. Among bipolar patients, completed suicide occurs one time for every 3–4 attempts (Simon et al., 2007a,b). Suicidal behavior and ideation scales have been developed to monitor patients during clinical trials, but are not used widely in clinical practice (Bongiovi-Garcia et al., 2009; FDA, 2012; Posner et al., 2007, 2011). While there have been scales whose psychometric properties have been examined in populations that include subjects with bipolar disorder, there are no published studies to date of any suicide scale that has been validated specifically for use in bipolar disorder (Pinninti et al., 2002; Saunders and Hawton, 2013).

Instruments currently used to measure suicidal ideation or behavior require interviews that typically evaluate both state- and trait-like ideation (Bongiovi-Garcia et al., 2009; Posner et al., 2007). Scales assessing suicidal ideation or behavior, which a clinician can then use to attempt to assess suicidal risk, should not only accurately monitor suicide-related factors to identify suicidal ideation and related symptoms clinically, but also be able to be used to detect changes in risk over time (Thompson et al., 2014). Because of the need for a simple scale to detect and monitor suicidal risk, the Concise Health Risk Tracking Self-Report (CHRT-SR) and Clinician Rating (CHRT-C) scales were developed to measure suicidal ideation and related symptoms in unipolar major depressive disorder (Trivedi et al., 2011a,b). These scales include items that inquire about hopelessness, self-worth, pessimism, perception of social support, and current suicidal plans. Following the model set forth by Beck and associates, eight items were developed to encompass stages of escalating suicidal ideation, from hopelessness about the future and lack of perceived support, to nihilistic thinking or passive thoughts of death to active plans about death and suicide (Beck et al., 1979). Items were included in the questionnaire only if there was good face validity for the construct being assessed. To maximize clinical and research utility, the Concise Health Risk Tracking (CHRT) scale was designed to be used as both self-rated (CHRT-SR) and clinician-rated (CHRT-C) instruments. Although previously studied in unipolar major depressive disorder, prior to the current study, the psychometric properties of the CHRT have not been examined in subjects with bipolar disorder (Trivedi et al., 2011a,b).

Presently, measures such as the Columbia Suicide Severity Rating Scale (C-SSRS; Posner et al., 2011), Suicidal Behaviors Questionnaire (SBQ; Linehan and Nielsen, 1981) and the Beck Scale for Suicide Ideation (SSI; Beck et al., 1979) are used to assess risk factors for suicidal behavior. The C-SSRS is a clinician-administered scale that extensively assesses past and recent suicidal ideation and/or behavior. The 4-item self-report SBQ assesses for current and past suicidal behaviors and ideation. A more extensive version of the questionnaire, the Suicidal Behaviors Questionnaire (SBQ-14), consists of items that assess past suicidal ideation and suicidal attempts, as well as likelihood of future suicidal ideation and suicidal attempts, among other behaviors (Linehan, 1996). The SSI (Beck et al., 1979) is a 19-item clinician-administered scale that helps to quantify suicidal ideation by assessing the presence, intensity, frequency, and severity of ideation.

The CHRT-SR presents some noteworthy advantages relative to the C-SSRS, SBQ and the SSI. First, the CHRT-SR allows for a comprehensive assessment of an individual’s risk for suicidal behavior. While the SBQ and the SSI largely focus on suicidal ideation as a measure of suicidal risk, the CHRT-SR considers suicidal ideation in conjunction with other factors (e.g., hopelessness, perceived lack of social support) to allow for a more holistic measurement of suicidal behavior. The self-report nature of the CHRT-SR may contribute to its ease of administration relative to the C-SSRS and the SSI, which are clinician-administered measures. Though the SBQ is also a self-report measure, the 4-item version of this questionnaire may be too short to accurately assess an individual’s suicidal behavior.

There are some limitations to the CHRT-SR. Clinician-rated measures such as the C-SSRS and the SSI allow for a rater-based assessment of an individual’s suicidal behavior and potentially could more accurately describe past behavior than a self-report measure.

This report presents psychometric data on the self-rated version of the CHRT-SR scale in the Lithium Treatment - Moderate dose Use Study (LiTMUS). LiTMUS was an NIMH-funded 6 month, randomized effectiveness trial examining the effectiveness of lithium as a component of optimized personalized treatment (OPT; guideline-informed, personalized treatment) versus OPT without lithium (Nierenberg et al., 2009, 2013). CHRT-SR data were collected at baseline and at follow-up visits to examine the psychometric properties of the scale including internal consistency, construct validity, criterion validity, and concurrent validity.

2. Materials and methods

2.1. Study design

The CHRT-SR was administered as part of the LiTMUS trial, which is described in detail elsewhere (Nierenberg et al., 2009, 2013). Participants were 283 subjects with bipolar I or II disorder who were randomized to receive open-label lithium plus OPT or OPT without lithium. While clinicians and participants were aware of treatment assignments, raters who measured the primary outcomes were blinded to treatment group. The study was approved by the Institutional Review Boards at each site and was monitored by the Data Safety and Monitoring Board at the National Institute of Mental Health. Subjects provided both verbal and written informed consent prior to participating in this research. The rationale, design, and methods of LiTMUS have been detailed elsewhere (Nierenberg et al., 2009).

2.2. Participating sites

LiTMUS was conducted by the Bipolar Trials Network. This was a network of six sites: Stanford University, Massachusetts General Hospital, the University of Pittsburgh School of Medicine, Case Western Reserve University, University of Texas Health Science Center at San Antonio, and University of Pennsylvania.

2.3. Participants

Participants were included in the current study if they: 1) Met DSM-IV criteria for bipolar disorder (Type I or II) American Psychiatric Association, 1994; 2) Were able to give informed consent (e.g., mentally able to make an independent, conscious choice to participate in the study); 3) Were at least 18 years of age; 4) If a woman of child-bearing age, agreed to inform their prescriber as early as possible of their plans to conceive and to use adequate contraception, and understood the risks of lithium to the fetus and infant; 5) Were symptomatic at study entry, defined as Clinical Global Impression for Bipolar Disorder-Severity Index (CGI-BP-S; Spearing et al., 1997) ≥ 3; and 6) Had not taken lithium for at least 30 days.

Participants were excluded if they: 1) Were unwilling or unable to comply with the study requirements; 2) Had renal impairment (serum creatinine >1.5 mg/dL); 3) Had thyroid stimulating hormone over 20% above the upper normal limit; 4) Had any other contraindication to lithium; 5) Were currently in crisis such that inpatient hospitalization or other crisis management should take priority; 6) Were in need of medically-supervised withdrawal from substances; 7) Were pregnant or breastfeeding; 8) Were women of child-bearing potential who were not able to agree to the requirements specified above; 9) Were currently prescribed lithium within the past 30 days; 10) Had participated in a clinical trial of an investigational drug within the past month; 11) Were unable to agree to comply with the visit schedule or study procedures; or 12) Had a history of lithium intolerability.

2.4. Measures

All participants were assessed using structured diagnostic interviews, clinician-rated assessments, and self-reported questionnaires. Diagnoses were confirmed using the Extended Mini-International Neuropsychiatric Interview (MINI), a validated, structured, clinician-administered diagnostic interview to determine current and lifetime Diagnostic and Statistical Manual-Version IV diagnoses (Sheehan et al., 1998). The Structured Clinical Interview for DSM-IV Substance Use Disorder Module was used to assess substance use disorders because it provided more detail regarding substance use course specifiers than the MINI (First et al., 2002).

For the current study, overall severity of bipolar disorder was measured by the CGI-BP-S (Spearing et al., 1997). The CGI-BP-S is a modified version of the CGI designed specifically for use in assessing global illness severity in patients with bipolar disorder. Mood symptoms were measured with the Montgomery-Asberg Depression Rating Scale (MADRS), the Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR), the Young Mania Rating Scale (YMRS), the Clinician-Administered Rating Scale for Mania (CARS-M), the LIFE-Range of Impaired Functioning Tool (LIFE-RIFT) and the Quality of Life; Enjoyment and Satisfaction Questionnaire e Short Form (Q-LES-Q-SF) (Altman et al., 1994; Bernstein et al., 2009; Endicott et al., 1993; Leon et al., 2000; Montgomery and Asberg, 1979; Rush et al., 2003; Young et al., 1978). Suicidal ideation during the study was assessed using the Beck Scale for Suicide Ideation (SSI) and the Concise Health Risk Tracking self-report form (CHRT-SR) (Beck et al., 1979; Trivedi et al., 2011a). The CHRT-SR was analyzed separately as a 12-item scale and as a 7-item abbreviated scale. Originally developed by Trivedi et al. (2011b) as a 12-item measure, the 7-item scale of the CHRTSR represents a shorter version of the measure. Trivedi et al. (2011b) found the 7-item CHRT-SR to have good psychometric properties and be a useful measure for assessing suicidal risk in both clinical practice and research environments. For the present study, we were interested in assessing the psychometric properties of the 12-item and 7-item versions of the CHRT-SR.

The CHRT-SR was originally developed to include 12 items reflecting the following domains: (1) active suicidal ideation and behavior, (2) perceived lack of social support, and (3) hopelessness. Each of the items is presented as a statement and the extent to which it is applicable to the preceding 24 h is rated on a 5-point, fully anchored, Likert scale, with responses ranging from “strongly agree” to “strongly disagree” for all items (see Supplementary data).

2.5. Statistical analysis

Analyses were completed on both the 7-item and 12-item scales. Cronbach’s coefficient alpha was calculated to assess internal consistency reliability between the scale items. A confirmatory factor analysis was performed to assess construct validity, assuming each of the items loaded onto the corresponding factor(s), which have been previously reported (Trivedi et al., 2011a). Model fit was assessed using the Goodness of Fit Index (GFI), adjusted GFI, root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), and the Bentler comparative fit index (CFI). We calculated Pearson Correlation Coefficients between CHRT-SR scores and SSI, MADRS suicide item, and QIDS suicide item to assess criterion validity (Beck et al., 1979; Montgomery and Asberg, 1979; Rush et al., 2003) and MADRS, YMRS, CARS-M, QIDS, LIFE-RIFT, and Q-LES-Q-SF to assess concurrent validity (DiazGranados et al., 2010; Morris et al., 2010; Ponizovsky et al., 2003; Trivedi et al., 2011a; Umamaheswari et al., 2014; Weiss et al., 2005). Two-sample t-tests were used to determine whether CHRT-SR scores were associated with treatment group and clinical/demographic variables at baseline (substance use disorder, anxiety disorder, smoking status, history of suicide attempts, bipolar disorder subtype, and gender) known to be associated with suicide risk (Berlin et al., 2015; Hawton et al., 2005; Ness et al., 2015; Sareen et al., 2005; Saunders and Hawton, 2013).

A significance level of 0.05 was used to test each hypothesis and, due to the exploratory nature of this paper, no adjustment for multiple hypothesis testing was made.

3. Results

Mean (SD) scores at baseline on the CHRT 7-item, 12-item, and SSI were as follows: 14.0 (5.1), 25.3 (9.4), 2.5 (4.4). Baseline scores on these three scales did not differ between randomized treatment groups (p > 0.05). A full description of the sample is provided in Table 1.

Table 1.

Sample baseline demographic and illness characteristics.

Patient characteristic	Mean ± SD (N) or % (N/N)
Demographic Characteristics
Age (years)	39.1 ± 12.3 (283)
Female Gender	56.5% (160/283)
Hispanic or Latino	13.8% (39/282)
Race
White	75.3% (213/283)
African American/Black	17.0% (48/283)
Other	7.7% (22/283)
Marital status
Single/never married	49.1% (139/283)
Married/living as married	26.1% (74/283)
Divorced/separated/widowed	24.7% (70/283)
Education
High school or less	27.2% (77/283)
Some college/Technical school/Associates Degree	37.1% (106/283)
College diploma/Graduate or professional degree	35.7% (101/283)
Employment status
Employed	40.6% (115/283)
Unemployed	32.9% (93/283)
Disability recipient	16.6% (47/283)
Student/retired/other	9.9% (28/283)
Illness Characteristics
Bipolar I	76.3% (216/283)
Age depressive symptoms first interfered with function (years)	16.6 ± 8.2 (276)
Age mood elevation symptoms first interfered with function (years)	19.0 ± 8.3 (260)
History of psychosis	9.1% (24/263)
History of psychiatric hospitalization	43.3% (119/275)
History of suicide attempt	41.1% (115/280)
Number of suicide attempts	1.15 ± 2.33 (279)
Rapid cycling in prior year	60.7% (150/247)
Current Anxiety Disorder	49.8% (141/283)
Cigarette smoking – Former or Current	53.4% (150/281)
History of Alcohol/Drug use Disorder	49.5% (140/283)
Current Alcohol/Drug use Disorder	11.7% (33/283)
Q-LES-Q-SF
Overall/Total (% of Total Score)	46.2 ± 18.6 (280)
Work (Item#3)	2.6 ± 1.1 (259)
Ability to Function (Item#8)	2.9 ± 1.1 (279)
Overall Sense of Well-being (Item#14)	2.8 ± 1.1 (279)
Overall Life Satisfaction (Item#16)	2.8 ± 1.0 (279)
LIFE-RIFT
Total (Sum of 4 scores: maximum of items 1–3 + max. of items 4–7 + item 8 + item 9)	13.4 ± 3.6 (282)
Work (maximum score of items 1–3)	3.5 ± 1.3 (279)
Interpersonal (maximum score of items 4–7)	3.5 ± 1.2 (282)
Satisfaction (Item#8)	3.2 ± 1.0 (281)

Reliability of both scales was strong, with Cronbach’s coefficient alphas of 0.80 (7-item) and 0.90 (12-item). Since Cronbach’s alpha is a lower bound of reliability and it is thought to be more meaningful with fewer scale items, this measure supports that the items of both scales have little within-item variability, and thus have strong inter-correlation at baseline (Cronbach, 1951).

Our confirmatory factor analysis models suggested that the 7-item scale fit the given 3 factor construct better than the 12-item scale, which is consistent with previous results (Trivedi et al., 2011a). All model fit diagnostics of the 7-item scale (GFI 0.97, RMSEA = 0.07, SRMR 0.05, CFI = 0.98) were as good as or better than those of the 12-item scale (see Table 2). Factor loadings for the 7-item scale were similar to those reported in Trivedi et al., though the loading of item 9 (The people I care most for are gone) on factor 3 was markedly lower (0.45 vs. 0.85). Inter-item correlations are displayed in Table 3 for the 7-item scale. Correlations were low between factor 1 (active suicidal ideation/behavior) and factors 2 and 3 (0.38 and 0.29; hopelessness and interpersonal attachment/social support, respectively). We found a moderately high correlation between hopelessness and perceived lack of social support (0.69).

Table 2.

Factor loadings and goodness of fit indices for the three-factor model of Concise Health Risk Tracking self-report (CHRT-SR) by items at baseline.

	Item	12-Items			7-Items
#		F1a	F2	F3	F1	F2	F3
1	Things are never going to get better.	–	0.76	–	–	–	–
2	There is no one I can depend on.	–	–	0.79	–	–	0.81
3	I have no future.	–	0.81	–	–	–	–
4	I can do nothing right.	–	0.79	–	–	0.84	–
5	I wish it could just all be over.	0.23	0.66	–	–	–	–
6	Everything I do turns out wrong.	–	0.80	–	–	0.93	–
7	There is no reason to live.	0.19	0.66	–	–	–	–
8	I wish I would not wake up.	0.27	0.60	–	–	–	–
9	The people I care most for are gone.	–	–	0.46	–	–	0.45
10	I think of killing myself.	0.88	–	–	0.87	–	–
11	I think of how I might kill myself.	0.89	–	–	0.91	–	–
12	I have a plan to kill myself.	0.77	–	–	0.75	–	–
GFI (adjusted)		0.87 (0.79)			0.97 (0.93)
RMSEA		0.12			0.07
SRMR		0.05			0.05
Bentler CFI		0.91			0.98

DF: degrees of freedom, GFI: goodness of fit index, RMSEA: root mean square error of approximation, SRMR: standardized root mean square residual, CFI: comparative fit index.

^aF1 (factor 1) is active suicidal ideation and behavior, F2 is hopelessness, F3 perceived lack of social support.

Table 3.

Inter-item correlations for the 7-item Concise Health Risk Tracking self-report (CHRT-SR) at baseline.

	Item	Item no.
Item no.		2	4	6	9	10	11	12
2	There is no one I can depend on.	1.00	0.47	0.52	0.37	0.20	0.14	0.21
4	It seems as if I can do nothing right.	–	1.00	0.78	0.21	0.36	0.26	0.29
6	Everything I do turns out wrong.	–	–	1.00	0.30	0.32	0.27	0.29
9	The people I care the most for are gone.	–	–	–	1.00	0.25	0.23	0.21
10	I have been having thoughts of killing myself.	–	–	–	–	1.00	0.79	0.64
11	I have thoughts about how I might kill myself.	–	–	–	–	–	1.00	0.68
12	I have a plan to kill myself.	–	–	–	–	–	–	1.00

^*all correlations are significant (ie. p < 0.05).

There were statistically significant correlations of the CHRT-SR and each of the three factor scores with the SSI, MADRS suicide item, QIDS suicide item, MADRS, QIDS, LIFE-RIFT, and Q-LES-Q-SF (Table 4). Directionality of the associations was such that higher CHRT-SR and factor scores (i.e. worse) were associated with worse clinical scores. There were no significant associations between the CHRT-SR scales or factors and YMRS or CARS-M scores. Patients with current anxiety had significantly higher scores on the CHRT-12 (mean + SD = 27.7 + 9.0 vs. 22.9 + 9.1), CHRT-7 (15.4 5.1 vs.12.6 + 4.8), hopelessness factor 2 (5.7 2.2 vs. 4.3+2.0) and perceived lack of social support factor 3 (4.8 + 2.2 vs. 3.9 + 1.8) at baseline compared to those without anxiety (all p < 0.001). Patients with a history of suicide attempts had higher scores on CHRT factor 1 (suicidal ideation and behavior) compared to those without a history of suicide attempts (6.1 + 2.7 vs. 4.6 + 2.4, p 0.03). We also found that current smokers reported higher scores on the perceived lack of social support factor 3 (4.7+2.0 vs. 4.1+2.1, p = 0.02) compared to non-smokers. We found no associations of substance use disorder, bipolar disorder subtype, and gender with any of the CHRT-SR scales or factors (all p > 0.05).

Table 4.

Baseline Correlations of CHRT-SR-12, CHRT-SR-7, and three factors with other clinical variables.

Scale	CHRT-SR-7	CHRT-SR-12	F1a	F2	F3
SSI	0.53*	0.58*	0.63*	0.33*	0.22*
MADRS	0.52*	0.56*	0.37*	0.49*	0.34*
MADRS suicide item	0.51*	0.60*	0.52*	0.39*	0.24*
YMRS	0.04	0.02	0.03	0.03	0.04
CARS-M mania sub-scale	0.04	0.01	0.02	0.03	0.04
QIDS	0.60*	0.65*	0.40*	0.58*	0.38*
QIDS suicide item	0.57*	0.59*	0.66*	0.37*	0.24*
LIFE-RIFT	0.51*	0.53*	0.34*	0.47*	0.37*
Q-LES-Q-SF	−0.57*	−0.62*	−0.31*	−0.58*	−0.42*

^*p-value < 0.05.

^aF1 (factor 1) is active suicidal ideation and behavior, F2 is hopelessness, F3 perceived lack of social support.

4. Discussion

This study demonstrates that the psychometric properties of the CHRT-SR are excellent in a sample of symptomatic subjects with bipolar disorder. The self-report version of the 7-item CHRT is well defined by a 3-factor structure consisting of indices of active suicidal ideation and plans (items 10, 11, and 12), hopelessness (items 4 and 6), and perceived lack of social support (items 2 and 9). Consistent with what was found in a major depressive disorder cohort, these findings suggest that the CHRT-7 may have better psychometric properties than the CHRT-12 among bipolar patients.

Both the 7- and 12-item CHRT are highly correlated with other measures of suicidal ideation, including the SSI, the MADRS suicide item, and the QIDS suicide item. All three individual factors of the CHRT e active suicidal ideation and plan, hopelessness, and perceived lack of social support – are also significantly correlated with the other measure of suicidal ideation. As expected, the correlation is largest between factor 1, active suicidal ideation and plan, and the other scales, although the correlation is still significant for factors 2 and 3.

Several clinical variables were correlated with CHRT scores. Subjects with current anxiety had significantly higher scores on the CHRT-12, CHRT-7, hopelessness factor 2 and perceived lack of social support factor 3 compared to those without anxiety. These findings are consistent with those of Simon et al. in STEP-BD, although because this is an exploratory analysis such results should be interpreted with caution (Simon et al., 2007a,b). Patients with a history of suicide attempts had higher scores on CHRT factor 1 (suicidal ideation and behavior) compared to those without a history of suicide attempts. We found no associations between substance use disorder, bipolar disorder subtype, and gender with any of the CHRT-SR scales or factors, but this lack of association is limited by the exploratory nature of the analysis and the size of the sample.

We found no correlation between the CHRT-7 or CHRT-12 with mania scale scores. While we anticipated that there would be strong correlations between depression scale scores and the CHRT, some prior evidence suggested that a positive association might remain for suicidal behavior and mania, especially mixed states (Dilsaver et al., 1994; Goldberg et al., 1999; Sato et al., 2004). The lack of any correlation between the YMRS or CARS-M with the CHRT-SR scores suggests that any relationship between mixed symptoms and suicidal thoughts might be driven by the association with concurrent depressive symptoms and not with manic symptoms per se, a finding that has not been previously reported. Although there is not empirical evidence for this, it is possible that individuals with bipolar disorder who have suicidal ideation might be more likely to act on such thoughts while in a manic or mixed state due to increased impulsivity as a symptom of mania, so it is of course not possible from our findings to state that manic symptoms play no role in self-directed violence. Further examination of this finding in future studies is merited.

The three-factor model of the CHRT-7 contributes to the scale’s psychometric value as all three factors (e.g. current suicidal ideation and plans, hopelessness/pessimism, and perceived lack of social support) have consistently demonstrated an association with suicidal behavior (Beck et al., 1976; Trivedi et al., 2011b; Wetzel, 1976). Suicidal ideation is a logical companion to suicidal behavior and precursor to an act of attempted suicide (Beck et al., 1979). Perceived lack of social support has been linked to the development of suicidal behavior and thoughts (Beck et al., 1976; Wetzel, 1976). In a study of 1958 outpatients, Beck et al. (1990) found hopelessness (as measured by the 20-item Beck Hopelessness Scale; Beck et al., 1974) to be significantly associated with eventual suicide.

It is important to note that the CHRT-SR is a patient-rated scale that can be used in clinical practice, in contrast to the Columbia Scale, which was designed primarily to be used in research settings such as clinical trials and requires significant training to use (Posner et al., 2007). It is pertinent that the CHRT-SR 7-item scale has excellent psychometric properties and produces results, as expected, that are consistent with mood scale scores, as shorter forms completed by patients themselves may be more practical to use in practice. As discussed prior, the CHRT-SR also differs from other self-report measures such as the SBQ (Linehan, 1996; Linehan and Nielsen, 1981) in that it considers a collection of factors (through its three-factor model) that contribute to suicidal behavior.

This study was conducted in academic medical centers in a relatively large sample of treatment seeking subjects who were moderately ill at baseline, with a high degree of psychiatric comorbidity. The sample of subjects in this study were a diverse group with multiple psychiatric, substance use, and medical comorbidities that is likely representative of clinical practice – and likely quite different from subjects who would qualify for an efficacy study with narrow inclusion and broad exclusion criteria. The study was intended to be a “practical clinical trial” that would in many ways mimic practice, and the subjects were primarily drawn from patients seeking treatment for bipolar disorder rather than subjects seeking to participate primarily in research. Patients could enter if they had bipolar I or II disorder and were at least mildly symptomatic, without past insensitivity to lithium, and in equipoise about being randomized to treatment with lithium or not (Nierenberg et al., 2013). Therefore, the results reported here might be generalizable to patients with bipolar disorder seeking routine outpatient care.

A limitation of this study is that the sample size and the number of suicide attempts, although adequate for determining the psychometric properties of the scale, were not sufficient to determine whether the scale is predictive of suicidal behavior over time: additionally, those patients who were most at risk of suicide – those requiring hospitalization, for example – were excluded from the analyses. The sample as a whole improved symptomatically during the study (Nierenberg et al., 2013), which limited the predictive value of the analyses.

It is possible that response bias had an impact on the present study and may have impacted how participants completed assessments. Both participants and clinicians were aware of their treatment conditions; accordingly, participants’ self-reports may have been influenced by what they thought they should be experiencing or by what they thought their clinician expected. More broadly, the methodological flaws inherent to self-report instruments may limit the generalizability of the study findings. Self-report measures are often inherently inaccurate due to their sole reliance on the respondents’ report, which can be tainted by multiple factors (e.g., lack of introspective ability, effects of the social desirability bias; Okada and Oltmanns, 2009). In the LiTMUS trial, however, no differences were found between treatment groups (Nierenberg et al., 2013). This suggests, however, that response bias may note have had a significant role in how participants completed study assessments.

Cultural considerations may pose an additional limitation to the present study; specifically, the manifestations of suicidal ideation may be different across cultures. For instance, Chu et al. (2010) found that one-third of Asian Americans reported suicidal ideation without a concurrent presentation of depression. The present study found the CHRT-SR to be strongly correlated with depression scores. Given that suicidal ideation may not be associated with depression in some cultures, it is possible that the CHRT-SR may not accurately assess suicidal behavior among individuals within some cultures. Further, the self-report nature of the CHRT-SR may limit its ability to accurately assess the extent of suicidal behavior among individuals within certain cultures – for instance, Raskin et al. (1992) found that some Asian-American individuals tend to under-report on self-report measures.

This is the first study to specifically examine the psychometric properties of any suicide rating scale in subjects with bipolar I and II disorder. It is especially important to have such scales validated in this illness given the high lifetime suicide rates in bipolar disorder and the need for scales that can be easily implemented in clinical practice. Because the scale correlates strongly with depression scores and not with mania scores, this study suggests that the contribution of mania symptom severity to suicidal ideation is minimal and that careful attention be paid to depressive symptoms during all phases of the illness. Further study is warranted of this simple scale for monitoring suicidal thoughts, suicidal behavior, and associated symptoms in a much larger sample of patients in treatment to determine whether it can be of predictive value in assessing suicide risk of suicidal or self-directed self-injurious behavior in bipolar disorder.

Biography

Dr. Ketter had the following financial interests/arrangements or affiliations that could be perceived as real or apparent conflicts of interest in the past three years: Grant/Research Support from the AstraZeneca Pharmaceuticals LP, Cephalon Inc., Eli Lilly and Company, Pfizer Inc., and Sunovion Pharmaceuticals; Consultant Fees from Allergan, Inc., Avanir Pharmaceuticals, Bristol-Myers Squibb Company, Cephalon Inc., Forest Pharmaceuticals, Janssen Pharmaceutica Products, LP, Merck & Co., Inc., Sunovion Pharmaceuticals, Teva Pharmaceuticals; Lecture Honoraria from Abbott Laboratories, Inc., AstraZeneca Pharmaceuticals LP, GlaxoSmithKline, and Otsuka Pharmaceuticals; and Publication Royalties from American Psychiatric Publishing, Inc. In addition, Dr. Ketter’s spouse is an employee of and holds stock in Janssen Pharmaceuticals.

Dr. Calabrese Dr. Calabrese has received federal funding from the Department of Defense, Health Resources Services Administration and National Institute of Mental Health and research support from Abbott, AstraZeneca, Bristol-Myers Squibb, Cephalon, Cleveland Foundation, Eli Lilly, GlaxoSmithKline, Janssen, Lund-beck, NARSAD, Repligen, Stanley Medical Research Institute, Takeda, and Wyeth. Dr. Calabrese consulted to or served on advisory boards of Abbott, AstraZeneca, Bristol-Myers Squibb, Cephalon, Dainippon Sumitomo, EPI-Q, Inc., Forest, France Foundation, GlaxoSmithKline, Janssen, Johnson and Johnson, Lundbeck, Merck, Neurosearch, OrthoMcNeil, Otsuka, Pfizer, Repligen, Schering-Plough, Servier, Solvay, Supernus, Synosia, Takeda, and Wyeth. He has provided CME lectures supported by AstraZeneca, Bristol-Myers Squibb, France Foundation, GlaxoSmithKline, Janssen, Johnson and Johnson, Merck, Sanofi Aventis, Schering-Plough, Pfizer, Solvay, and Wyeth.

Mr. Rabideau, Ms. Gold, and Ms. Shesler have no relationships to disclose.

Dr. Nierenberg is a consultant for the Abbott Laboratories, American Psychiatric Association, Appliance Computing Inc. (Mindsite), Basliea, Brain Cells, Inc., Brandeis University, Bristol Myers Squibb, Clintara, Corcept, Dey Pharmaceuticals, Dainippon Sumitomo (now Sunovion), Eli Lilly and Company, EpiQ, L.P./Mylan Inc., Forest, Genaissance, Genentech, GlaxoSmithKline, Hoffman LaRoche, Infomedic, Lundbeck, Janssen Pharmaceutica, Jazz Pharmaceuticals, Medavante, Merck, Methylation Sciences, Naurex, Novartis, PamLabs, Pfizer, PGx Health, Ridge Diagnostics Shire, Schering-Plough, Somerset, Sunovion, Takeda Pharmaceuticals, Targacept, and Teva; consulted through the MGH Clinical Trials Network and Institute (CTNI) for Astra Zeneca, Brain Cells, Inc, Dianippon Sumitomo/Sepracor, Johnson and Johnson, Labopharm, Merck, Methylation Science, Novartis, PGx Health, Shire, Schering-Plough, Targacept and Takeda/Lundbeck Pharmaceuticals. He receives grant/research support from American Foundation for Suicide Prevention, AHRQ, Brain and Behavior Research Foundation, Bristol-Myers Squibb, Cederroth, Cephalon, Cyberonics, Elan, Eli Lilly, Forest, GlaxoSmithKline, Janssen Pharmaceutica, Lichtwer Pharma, Marriott Foundation, Mylan, NIMH, PamLabs, PCORI, Pfizer Pharmaceuticals, Shire, Stanley Foundation, Takeda, and Wyeth-Ayerst. Honoraria include Belvoir Publishing, University of Texas Southwestern Dallas, Brandeis University, Bristol-Myers Squibb, Hillside Hospital, American Drug Utilization Review, American Society for Clinical Psychopharmacology, Baystate Medical Center, Columbia University, CRICO, Dartmouth Medical School, Health New England, Harold Grinspoon Charitable Foundation, IMEDEX, International Society for Bipolar Disorder, Israel Society for Biological Psychiatry, Johns Hopkins University, MJ Consulting, New York State, Medscape, MBL Publishing, MGH Psychiatry Academy, National Association of Continuing Education, Physicians Postgraduate Press, SUNY Buffalo, University of Wisconsin, University of Pisa, University of Michigan, University of Miami, University of Wisconsin at Madison, APSARD, ISBD, SciMed, Slack Publishing and Wolters Klower Publishing. He was currently or formerly on the advisory boards of Appliance Computing, Inc., Brain Cells, Inc., Eli Lilly and Company, Johnson and Johnson, Takeda/Lundbeck, Targacept, and InfoMedic. He owns stock options in Appliance Computing, Inc., Brain Cells, Inc, and Medavante; has copyrights to the Clinical Positive Affect Scale and the MGH Structured Clinical Interview for the Montgomery Asberg Depression Scale exclusively licensed to the MGH Clinical Trials Network and Institute (CTNI).

Dr. Leon served on independent Data and Safety Monitoring Boards for AstraZeneca, Sunovion, and Pfizer and as a Consultant/Advisor to: FDA, NIMH, MedAvante and Roche. He had equity in MedAvante.

Dr. Friedman receives grant support from Novartis, St. Jude Medical, Medtronics, Repligen, Astra-Zeneca, Roche, and Takeda. He receives royalties from Springer.

Dr. Iosifescu has had research funding from Aspect Medical, Forest and Ortho-McNeil and has received speaker honoraria from Reed Medical (sponsor of the MGH Psych Academy).

Dr. Sylvia was a shareholder in Concordant Rater Systems and serves as a consultant for United Biosource Corporation and Clin-tara. She receives royalties from New Harbinger.

Dr. Thase has been an advisor/consultant: to Alkermes, Astra-Zeneca, Bristol-Myers Squibb, Eli Lilly, Forest Laboratories, GlaxoSmithKline, Janssen Pharmaceuticals, Lundbeck, MedAvante, Merck, Mylan, Neuronetics, Otsuka, Pamlab, PharmaNeuroboost, Pfizer, Rexahn, Roche, Shire, Sunovion, Supernus, Takeda, and Teva, as well as the US Food and Drug Administration and the National Institute of Mental Health. During the same time frame, Dr. Thase has received honoraria for talks from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Merck, and Pfizer and he has received research grants from Alkermes, AstraZeneca, Eli Lilly, Forest, GlaxoSmithK-line, Otsuka, PharmaNeuroboost, and Roche, as well as the National Institute of Mental Health and the Agency for Healthcare Research and Quality.

Dr. Reilly-Harrington was a shareholder in Concordant Rater Systems and serves as a consultant for Clintara and United Bio-source Corporation. She receives royalties from the American Psychological Association, New Harbinger, and Oxford University Press.

Dr. Bowden is a research collaborator with Elan and a consultant with Teva, He has no participation with speaker bureaus, nor does he or his wife hold any equity position in any biomedical or pharmaceutical corporation.

Dr. Trivedi has been a speaker or received consultant honoraria from Abbott Laboratories, Inc., Abdi Ibrahim, Akzo (Organon Pharmaceuticals Inc.), Alkermes, AstraZeneca, Axon Advisors, Bristol-Myers Squibb Company, Cephalon, Inc., Cerecor, Concert Pharmaceuticals, Inc., Eli Lilly & Company, Evotec, Fabre Kramer Pharmaceuticals, Inc., Forest Pharmaceuticals, GlaxoSmithKline, Janssen Global Services, LLC, Janssen Pharmaceutica Products, LP, Johnson & Johnson PRD, Libby, Lundbeck, Meade Johnson, MedAvante, Medtronic, Merck, Mitsubishi Tanabe Pharma Development America, Inc., Naurex, Neuronetics, Otsuka Pharmaceuticals, Pamlab, Parke-Davis Pharmaceuticals, Inc., Pfizer Inc., PgxHealth, Phoenix Marketing Solutions, Rexahn Pharmaceuticals, Ridge Diagnostics, Roche Products Ltd., Sepracor, SHIRE Development, Sierra, SK Life and Science, Sunovion, Takeda, Tal Medical/Puretech Venture, Targacept, Transcept, VantagePoint, Vivus, and Wyeth-Ayerst Laboratories. In addition, he has received research support from Agency for Healthcare Research and Quality (AHRQ), Corcept Therapeutics, Inc., Cyberonics, Inc., National Alliance for Research in Schizophrenia and Depression, National Institute of Mental Health, National Institute on Drug Abuse, Novartis, Pharmacia & Upjohn, Predix Pharmaceuticals (Epix), and Solvay Pharmaceuticals, Inc.