Controllability of Suicidal Thoughts Scale (CoST)

Abstract

The Controllability of Suicidal Thoughts Scale (CoST) is a specialized psychometric instrument developed in 2023 by Brianna Meddaoui, Bianca C. Iddiols, and Erin A. Kaufman at the University of Western Ontario. Designed to capture an individual’s metacognitive and self-regulatory evaluation of their suicidal cognitions, the instrument systematically quantifies the degree to which a person perceives their suicidal ideation as manageable, malleable, and subject to personal agency versus autonomous, intrusive, and unalterable. Grounded in contemporary ideation-to-action frameworks of suicidology, cognitive-behavioral theories of rumination, and metacognitive models of mental distress, the scale isolates perceived control over suicidal thinking—a critical, historically under-measured psychological buffer against suicide risk escalation.

Structurally, the CoST comprises self-report items evaluated on a 7-point Likert scale ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”). Exploratory and confirmatory factor analyses establish a parsimonious two-dimensional architecture comprising two distinct yet interrelated factors: General Control (reflecting the broader executive capacity to attenuate, down-regulate, or shift attention away from suicidal intrusions) and Changing Thought Content (measuring cognitive reappraisal and the perceived ability to alter or reframe the internal narrative and trajectory of suicidal ideas). The instrument displays exceptional psychometric properties across community and clinical adult samples in North America, boasting an internal consistency McDonald’s omega (ω) coefficient of .92, robust one-month test–retest reliability ($r = .83$ to $.86$), and strong structural fit ($ ext{CFI} = 0.94$,$ ext{TLI} = 0.92$,$ ext{SRMR} = 0.05$,$ ext{RMSEA} = 0.07$). Moreover, the CoST demonstrates remarkable incremental validity, predicting prospective suicidal ideation severity over a one-month interval beyond standard clinical indicators such as hopelessness, general coping efficacy, and suicide-specific self-efficacy.

Keywords

Controllability of Suicidal Thoughts Scale, CoST, Suicidal Ideation, Perceived Control, Metacognition, Suicidology, Psychological Assessment, Cognitive Reappraisal, Attentional Control, Incremental Validity

Authors

The Controllability of Suicidal Thoughts Scale (CoST) was conceptualized, constructed, and empirically validated by clinical psychology researchers at the University of Western Ontario in London, Ontario, Canada:

• Brianna Meddaoui, M.Sc., Ph.D. Candidate: Department of Psychology, University of Western Ontario, London, Ontario, Canada. ORCID: 0000-0002-8813-5844. Primary correspondence contact: [email protected]. Mailing Address: Department of Psychology, University of Western Ontario, 361 Windermere Road, London, Ontario, Canada, N6A 3K7.
• Bianca C. Iddiols, M.Sc.: Department of Psychology, University of Western Ontario, London, Ontario, Canada. Specializes in developmental psychopathology, emotion regulation, and self-injurious thoughts and behaviors.
• Erin A. Kaufman, Ph.D.: Associate Professor, Department of Psychology, University of Western Ontario, London, Ontario, Canada. Director of the Emotion and Personality Development Lab. ORCID: 0000-0002-7636-6602.

Purpose

The primary clinical and empirical objective of the Controllability of Suicidal Thoughts Scale (CoST) is to systematically quantify a person’s perceived agency and capacity to control, down-regulate, redirect, and modify their suicidal cognitions. In the landscape of psychiatric risk evaluation, clinicians and researchers have traditionally focused on assessing the presence, frequency, and severity of suicidal ideation. However, empirical suicidology has increasingly demonstrated that the sheer presence of suicidal thoughts does not uniformly correlate with imminent suicidal behavior. Rather, qualitative dynamics—specifically how an individual appraises, responds to, and experiences mastery over their thoughts—act as critical catalysts or inhibitors in the progression toward suicidal crisis.

In clinical practice, when individuals perceive their suicidal thoughts as uncontrollable, immutable, and autonomous, cognitive exhaustion and psychological paralysis rapidly ensue. The thought patterns are interpreted as inevitable mandates rather than transient mental events. Conversely, an individual who maintains a robust sense of control over their ideation retains the perceived capacity to dismiss intrusions, reframe cognitive content, or direct focus toward protective coping mechanisms. The CoST was devised to address a critical assessment gap: prior inventories routinely aggregated perceived control within broader coping batteries or collapsed it into single items within severity scales (such as the Columbia-Suicide Severity Rating Scale [C-SSRS]), failing to offer an isolated, multidimensional psychometric evaluation of the controllability construct.

The applications of the CoST span both acute risk assessment and treatment monitoring:

• Stratification of Suicide Risk: By measuring perceived controllability, clinicians can distinguish individuals experiencing passive, manageable ideation from those undergoing severe cognitive dyscontrol who face an elevated risk of progressing along the ideation-to-action spectrum.
• Targeted Psychotherapeutic Interventions: Cognitive-behavioral therapies (e.g., Cognitive Therapy for Suicide Prevention [CT-SP]) and Dialectical Behavior Therapy (DBT) emphasize distress tolerance and cognitive reappraisal. CoST subscale scores clarify whether an intervention should prioritize attentional shifting techniques (General Control) or cognitive restructuring and narrative reframing (Changing Thought Content).
• Empirical Research and Ecological Monitoring: The scale enables investigators to longitudinally model how micro-fluctuations in cognitive control interact with acute life stressors, affective dysregulation, and neurocognitive deficits in executive functioning.

Psychological Construct

The psychological construct captured by the CoST is Perceived Controllability of Suicidal Thoughts. This construct is situated at the intersection of metacognitive appraisal, cognitive control, and suicide-specific self-efficacy. Metacognition refers to knowledge, awareness, and beliefs about one’s own thinking processes. When applied to suicidal ideation, perceived controllability reflects the subjective appraisal that one can intentionally initiate, modulate, inhibit, or alter the course of thoughts centered on self-harm, death, or suicide.

Dimension 1: General Control

The General Control subscale taps the broad subjective perception of attentional flexibility and executive dominance over suicidal cognitions. Clinically, when suicidal ideation emerges, it often acts as an invasive cognitive loop characterized by mental stickiness, perseveration, and involuntary cognitive fixation. The General Control dimension measures:

• Attentional Shifting: The subjective capability to disengage focus from suicidal imagery and direct mental resources toward alternative external activities or benign internal representations.
• Resistance to Entrapment: The perceived power to prevent oneself from feeling dominated, powerless, or overwhelmed by dark intrusive thoughts.
• Inhibition and Dissipation: The internal belief that one can actively attenuate the emotional intensity or persistence of thoughts without being swept into cognitive despair.

Low scores on this dimension represent severe perceived dyscontrol—a psychological state where the patient feels that the suicidal thought process has achieved full autonomy and cannot be dislodged.

Dimension 2: Changing Thought Content

While General Control reflects attentional regulation and mental disengagement, the Changing Thought Content dimension evaluates the subjective ability to actively manipulate, dispute, and reframe the narrative substance of suicidal ideation. Drawing heavily upon cognitive restructuring principles, this dimension captures:

• Cognitive Malleability: The belief that suicidal thoughts are not rigid, absolute truths, but mutable mental representations that can be actively altered.
• Reframing Competence: The subjective capacity to confront cognitive distortions underlying suicidal desires (e.g., perceived burdensomeness, thwarted belongingness) and substitute constructive perspectives.
• Prevention of Catastrophic Escalation: The perceived agency to interrupt negative cognitive spirals and prevent suicidal thoughts from progressively deteriorating into hopelessness and despair.

Taken together, the two dimensions illustrate that perceived control is not monolithic: individuals may retain the skill to reappraise their thoughts while struggling with attentional disengagement, or they may possess distraction skills while remaining unable to alter the fatalistic internal narrative.

Theoretical Framework

The theoretical foundations of the CoST are anchored in three major psychological paradigms: the Ideation-to-Action Framework, Metacognitive Theory, and the Cognitive Model of Suicidal Behavior.

The Ideation-to-Action Framework

Modern suicidology models—such as the Interpersonal Theory of Suicide (Joiner, 2005; Van Orden et al., 2010), the Integrated Motivational-Volitional (IMV) Model (O’Connor, 2011), and the Three-Step Theory (Klonsky & May, 2015)—explicitly differentiate the mechanisms governing the development of suicidal ideation from those governing the transition from ideation to lethal action. The CoST operates as a key psychometric operationalization within the motivational and volitional phases. In the IMV model, feelings of defeat and entrapment catalyze suicidal thoughts. If an individual experiences total uncontrollability over those thoughts, the internal sense of entrapment deepens drastically, eroding motivational barriers and accelerating movement toward suicide-specific capability.

Metacognitive and Attentional Control Theories

Adrian Wells’ Self-Regulatory Executive Function (S-REF) model posits that psychological suffering is perpetuated by the Cognitive Attentional Syndrome (CAS), which consists of perseverative thinking (worry, rumination), attentional focusing on threat, and maladaptive coping behaviors. Central to the S-REF model are metacognitive beliefs—specifically negative metacognitive beliefs concerning the uncontrollability and dangerousness of thoughts. When applied to suicidology, individuals who hold the negative metacognitive belief that “my suicidal thoughts are uncontrollable” experience heightened distress, panic, and demoralization when suicidal thoughts arise. The CoST provides a direct metric of these negative metacognitive beliefs, assessing whether individuals view their suicidal ideation as an unmanageable crisis or a controllable mental event.

The Cognitive Model of Suicidal Behavior

According to Aaron T. Beck’s cognitive model of suicide, suicidal individuals operate under a hyperactivated “suicide mode” characterized by cognitive rigidity, selective abstraction, and severe hopelessness. In this state, suicidal ideation is perceived as the sole viable problem-solving mechanism. Developing perceived controllability disrupts this mode: when a patient recognizes that suicidal thoughts are changeable and amenable to self-regulation, the cognitive lock of hopelessness is weakened, empowering cognitive flexibility and adaptive problem solving.

Validity

The construct, convergent, discriminant, concurrent, predictive, and incremental validity of the CoST have been rigorously documented in both clinical and community cohorts across Canada and the United States (Meddaoui, Iddiols, & Kaufman, 2023).

Convergent and Discriminant Validity

The CoST exhibits a clear, predictable pattern of convergent and discriminant associations:

• Convergent Validity: CoST scores display moderate to strong inverse correlations with measures of suicidal ideation severity, depressive symptom burden, generalized hopelessness, and negative metacognitions. Crucially, the CoST correlates strongly with suicide-specific self-efficacy (e.g., self-efficacy to avoid suicidal action) and suicide-related coping inventories, verifying that it directly captures cognitive self-regulation within suicidal crises.
• Discriminant Validity: The scale’s associations with general, non-suicide-specific constructs—such as general problem-solving ability, general coping mechanisms, and generic self-esteem—are consistently and significantly weaker than its associations with suicide-specific constructs. This confirms that the CoST does not merely capture global positive affect or generalized optimism, but rather a distinct, domain-specific metacognitive capability.

Concurrent and Predictive Validity

Empirical evaluations confirm the powerful predictive utility of the CoST in both cross-sectional and prospective longitudinal designs:

• Concurrent Ideation Severity: At baseline assessment, higher total CoST scores substantially predicted lower suicidal ideation (SI) severity ($eta = -0.52, SE = 0.14, p < .001$).
• Prospective Predictive Power: In a one-month prospective follow-up, baseline CoST total scores robustly predicted reductions in suicidal ideation severity ($eta = -0.44, SE = 0.18, p < .001$).
• Subscale Predictive Effects: Both individual subscales independently demonstrated significant prospective predictive validity for SI severity: the General Control subscale ($eta = -0.38, SE = 0.25, p < .01$) and the Changing Thought Content subscale ($eta = -0.44, SE = 0.51, p < .01$).

Incremental Validity

A pivotal psychometric benchmark of any new psychological instrument is incremental validity—its capacity to explain unique variance above and beyond established clinical gold standards. Hierarchical multiple regression analyses demonstrate that total CoST scores explain substantial, statistically significant variance in prospective suicidal ideation severity even after strictly controlling for:

• Hopelessness: Explained unique incremental variance of $f^2 = .12$ beyond the Beck Hopelessness Scale.
• Self-Efficacy to Avoid Suicidal Action: Explained unique incremental variance of $f^2 = .05$ beyond specialized self-efficacy inventories.
• Suicide-Related Coping: Accounted for unique variance of $f^2 = .07$ beyond established coping indices (e.g., Batterham et al., 2015).

These findings substantiate that measuring perceived controllability provides vital clinical data unobtainable from conventional risk factors alone.

Reliability

The CoST exhibits outstanding reliability across diverse adult samples, satisfying the most stringent psychometric requirements for both research paradigms and individual clinical evaluations.

Internal Consistency

Internal consistency analyses reveal exceptional item homogeneity and precision across the measurement spectrum:

• Full Scale: The scale yields a McDonald’s omega coefficient ($\omega$) of .92, demonstrating superior structural reliability that surpasses typical research thresholds ($\omega ge .80$) and meets clinical cutoffs for individual diagnostics ($\omega ge .90$).
• Subscale Consistency: Both the General Control and Changing Thought Content subscales demonstrate excellent reliability estimates, with Cronbach’s alpha ($lpha$) and omega ($\omega$) values ranging reliably between .84 and .90.

Test–Retest Reliability

The temporal stability of the CoST was evaluated across a prospective one-month retest window among adult community and at-risk participants. Pearson product-moment correlation coefficients demonstrated robust temporal consistency:

• Total CoST Scale: $r = 0.86$, reflecting excellent stability over a 30-day period while retaining enough sensitivity to track clinically meaningful psychological change.
• General Control Subscale: $r = 0.83$, confirming that perceived executive capacity to down-regulate thoughts is a stable yet dynamic metacognitive trait.
• Changing Thought Content Subscale: $r = 0.83$, highlighting the enduring nature of perceived reappraisal competence over time.

Factor Analysis

The structural validity of the CoST was established through comprehensive exploratory factor analysis (EFA) followed by confirmatory factor analysis (CFA) across independent split samples of North American adults.

Exploratory Factor Analysis (EFA)

Initial exploratory factor extraction utilized parallel analysis and examination of the empirical scree plot. Although raw mathematical criteria identified five factors with initial eigenvalues exceeding 1.0, multi-factor models beyond two factors proved substantively unstable, in line with methodological guidelines outlined by Costello and Osborne (2005). Specifically, models specifying three or more factors accounted for only an additional 3% to 4% of total variance and were plagued by severe cross-loadings, uninterpretable factor splits, and weak item loadings ($< .35$).

In contrast, an oblique (Promax) rotated two-factor solution emerged as theoretically coherent, parsimonious, and mathematically optimal, accounting for approximately 48% of the total scale variance:

• Factor 1: General Control: Subsumed items capturing the perceived ability to shift attention away, prevent perseveration (“getting stuck”), and maintain executive governance over the presence of suicidal thoughts.
• Factor 2: Changing Thought Content: Subsumed items addressing cognitive malleability, narrative reframing, and interrupting negative thought escalation.

Confirmatory Factor Analysis (CFA)

A confirmatory factor analysis was conducted on an independent validation sample to formally evaluate the goodness-of-fit of the oblique two-factor measurement model. Standard goodness-of-fit indices demonstrated strong empirical alignment:

• Comparative Fit Index (CFI): 0.94 (surpassing the standard $ge .90$ acceptable threshold).
• Tucker-Lewis Index (TLI): 0.92 (indicating robust model parsimony).
• Standardized Root Mean Square Residual (SRMR): 0.05 (well below the $.08$ cutoff for excellent fit).
• Root Mean Square Error of Approximation (RMSEA): 0.07 (90% Confidence Interval [0.05, 0.09]), confirming adequate to good population fit.

All standardized factor loadings on designated latent dimensions were statistically significant ($p < .001$) and moderate to large in magnitude, confirming that each item functions as a robust indicator of its underlying construct.

Instrument / Measurement Tool

The operational characteristics and administrative protocol of the Controllability of Suicidal Thoughts Scale are summarized below:

• Test Type: Original Psychological Inventory / Self-Report Questionnaire.
• Target Population: Human adults (aged 18 years and older); validated in Canadian and American clinical and community populations.
• Administration Format: Paper-and-pencil questionnaire or digital computer-based / mobile administration.
• Estimated Completion Time: Approximately 2 to 4 minutes.
• Item Count: 7 core validated assessment items reflecting primary structural dimensions.
• Response Scale: Items are scored on a 7-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). Sixteen items are reverse-scored to mitigate response bias. Scores are summed to generate a total score, with higher scores indicating a greater perceived sense of control.
• Scoring and Directionality: Items expressing perceived dyscontrol, helplessness, or cognitive entrapment are reverse-coded prior to computing composite totals. Higher aggregated scores reflect greater perceived controllability, superior self-regulatory capacity, and enhanced metacognitive mastery over suicidal ideation.

Permissions & Fee and Test Year

• Publication Year: 2023.
• Primary Reference Publication: Published in the American Psychological Association journal Psychological Assessment (Meddaoui, Iddiols, & Kaufman, 2023).
• Fee: Free of charge ($0.00). The CoST is an open-access empirical instrument.
• Permissions & Terms of Use: The scale may be utilized freely without formal prior written permission for academic research, education, non-commercial clinical evaluation, and teaching purposes, provided that appropriate scholarly attribution and citation are given to the original developers. Commercial exploitation, incorporation into proprietary commercial platforms, or re-distribution behind paid paywalls without express written authorization from the authors and the American Psychological Association is strictly prohibited.

References

• Batterham, P. J., Ftanou, M., Pirkis, J., Brewer, J. L., Mackinnon, A. J., Mountford, H., Christet, L., & Christensen, H. (2015). A systematic review and evaluation of measures for suicidal ideation and behaviors in population-based studies. Psychological Assessment, 27(2), 501–512. https://doi.org/10.1037/pas0000053
• Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. Psychological Corporation. https://doi.org/10.1037/t00742-000
• Costello, A. B., & Osborne, J. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research, and Evaluation, 10(7), 1–9. https://doi.org/10.7275/jyj1-4868
• Joiner, T. E. (2005). Why people die by suicide. Harvard University Press.
• Klonsky, E. D., & May, A. M. (2015). The Three-Step Theory (3ST): A new theory of suicide rooted in the “ideation-to-action” framework. International Journal of Cognitive Therapy, 8(2), 114–129. https://doi.org/10.1521/ijct.2015.8.2.114
• Meddaoui, B., Iddiols, B. C., & Kaufman, E. A. (2023). The Controllability of Suicidal Thoughts (CoST) Scale: Development, factor structure, and initial validation. Psychological Assessment, 35(10), 880–887. https://doi.org/10.1037/pas0001271
• O’Connor, R. C. (2011). The Integrated Motivational-Volitional model of suicidal behavior. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 32(6), 295–298. https://doi.org/10.1027/0227-5910/a000120
• Van Orden, K. A., Witte, T. K., Cukrowicz, K. C., Braithwaite, S. R., Selby, E. A., & Joiner, T. E. (2010). The interpersonal theory of suicide. Psychological Review, 117(2), 575–600. https://doi.org/10.1037/a0018697
• Wells, A. (2009). Metacognitive therapy for anxiety and depression. Guilford Press.

Items of the Scale