Clinically Useful Anxiety Outcome Scale (CUXOS)

1. Abstract

The Clinically Useful Anxiety Outcome Scale (CUXOS) is a psychometrically robust, brief, 20-item self-report questionnaire engineered specifically to quantify the severity of anxiety symptomatology across routine clinical settings, measurement-based care paradigms, and psychiatric outcome studies. Developed in 2010 by Mark Zimmerman and colleagues at Rhode Island Hospital and the Warren Alpert Medical School of Brown University as part of the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project, the CUXOS addresses the acute need for an assessment instrument that balances diagnostic rigor with minimal administrative and respondent burden. Designed to capture symptomatic manifestations derived from the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R and DSM-IV) criteria for Generalized Anxiety Disorder (GAD) and Panic Disorder, the instrument operationalizes anxiety into two primary dimensions: a 6-item Psychic Anxiety subscale and a 14-item Somatic Anxiety subscale. Respondents rate the frequency and severity of each symptom over the preceding week using a 5-point Likert scale ranging from 0 (“not at all true”) to 4 (“almost always true”), yielding a cumulative score spanning 0 to 80 points.

Psychometric evaluations demonstrate exemplary performance across diverse psychiatric cohorts. In its initial validation involving 963 adult psychiatric outpatients, the CUXOS demonstrated high internal consistency across the total scale (α = .95), psychic subscale (α = .90), and somatic subscale (α = .93), paired with exceptional test-retest reliability over a 24- to 48-hour window (r = .90). Confirmatory investigations, including cross-cultural adaptation studies, have replicated these strong parameters (e.g., Cronbach’s α = .90; test-retest r = .74). Convergent validity is confirmed via robust correlations with established anxiety benchmarks (such as the Hamilton Rating Scale for Anxiety and Beck Anxiety Inventory; median r = .54 to .74), maintaining clear discriminant differentiation from depressive and general psychological distress indices. With an average patient completion time of under two minutes and a clinician scoring time of approximately 15 seconds, the CUXOS represents a benchmark measurement-based care instrument designed to optimize clinical decision-making, monitor treatment response trajectories, and enhance longitudinal psychiatric research.

2. Keywords

Clinically Useful Anxiety Outcome Scale, CUXOS, anxiety assessment, psychometrics, measurement-based care, Generalized Anxiety Disorder, Panic Disorder, psychic anxiety, somatic anxiety, symptom severity, rating scale, MIDAS project, clinical outcome monitoring, self-report questionnaire, psychiatric evaluation

3. Authors

The Clinically Useful Anxiety Outcome Scale was developed by a team of clinical psychiatric researchers affiliated with the Department of Psychiatry and Human Behavior at the Warren Alpert Medical School of Brown University and the Outpatient Psychiatry Department at Rhode Island Hospital in Providence, Rhode Island, USA.

• Mark Zimmerman, M.D. — Professor of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University; Director of Outpatient Psychiatry and the MIDAS Project, Rhode Island Hospital, Providence, RI. Dr. Zimmerman is a leading authority in psychiatric diagnostic methodology, clinical psychometrics, and the implementation of measurement-based care in routine clinical practice.
• Iwona Chelminski, Ph.D. — Associate Professor (Research), Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University; Staff Psychologist, Rhode Island Hospital, Providence, RI.
• Diane Young, Ph.D. — Clinical Assistant Professor, Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University; Clinical Psychologist, Rhode Island Hospital, Providence, RI.
• Kristy Dalrymple, Ph.D. — Associate Professor (Research), Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University; Director of Adult Psychology, Lifespan Physicians Group, Providence, RI.

Primary Institutional Affiliation: Rhode Island Hospital, Department of Psychiatry, Ambulatory Care Center, 593 Eddy Street, Providence, RI 02903, USA.
Corresponding Inquiries: Inquiries regarding the MIDAS project assessment battery can be directed to the Department of Psychiatry at Rhode Island Hospital / Brown University.

4. Purpose

The fundamental impetus behind the construction of the Clinically Useful Anxiety Outcome Scale (CUXOS) was the pervasive chasm separating academic clinical trials from routine real-world psychiatric care. For decades, psychiatric research relied upon comprehensive, clinician-administered instruments such as the Hamilton Rating Scale for Anxiety (HAM-A). While psychometrically venerable, clinician-administered scales require substantial administrative time (typically 15 to 30 minutes), specialized interviewer calibration, and extensive training to ensure inter-rater reliability. Consequently, standardized, longitudinal quantification of anxiety severity has historically been neglected in high-volume psychiatric clinics, community mental health centers, and primary care environments. Clinicians have predominantly relied on unstructured subjective clinical impressions, which are susceptible to diagnostic drift, recall bias, and unstandardized documentation.

To resolve this systemic challenge, Zimmerman and colleagues instituted the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project. Having previously demonstrated that depression severity could be tracked efficiently via the Clinically Useful Depression Outcome Scale (CUDOS), the authors engineered the CUXOS to serve as a structurally identical, companion assessment module for anxiety. The clinical and empirical purposes of the CUXOS are multifaceted:

• Facilitating Measurement-Based Care (MBC): Measurement-based care—the systematic administration of validated symptom measures to guide clinical decisions, pharmacotherapy titration, and psychotherapeutic adjustments—significantly enhances patient outcomes. The CUXOS provides a standardized metric that can be embedded into routine intake and follow-up appointments without disrupting clinical workflow.
• Rapid Screening and Severity Stratification: The tool reliably discriminates between individuals with and without clinical anxiety disorders and grades clinical presentations into empirically validated severity tiers: non-anxious, minimal, mild, moderate, and severe anxiety.
• Longitudinal Treatment Tracking and Outcome Evaluation: Designed to be highly sensitive to treatment-induced clinical shifts, the CUXOS captures subtle symptomatic improvements or deteriorations across psychopharmacological interventions, cognitive-behavioral therapies (CBT), and interdisciplinary psychiatric programs.
• Minimizing Patient and Administrative Burden: Unlike lengthy diagnostic batteries, the CUXOS requires an average of under two minutes for patients to complete in clinic waiting rooms or via digital patient portals, while clinical staff can reliably score and interpret the instrument within 15 seconds.
• Research Scalability: The instrument affords clinical researchers an economical, standardized, continuous measure of anxiety severity that correlates strongly with legacy interview schedules without incurring the overhead costs and scheduling delays inherent to rater-administered protocols.

5. Psychological Construct

The CUXOS evaluates pathological anxiety, operationalized as an emotional state characterized by persistent apprehension, autonomic arousal, subjective tension, and somatic reactivity that disrupts functional adaptation. Drawing upon the phenomenological taxonomy formalized within the DSM-III-R and DSM-IV, as well as the historical clinical observations of Max Hamilton, the scale delineates anxiety into two distinct yet interrelated dimensions: Psychic Anxiety and Somatic Anxiety.

1. Psychic Anxiety Subscale (6 Items)

The Psychic Anxiety subscale quantifies the cognitive, subjective, and affective facets of anxiety states. Rather than focusing merely on transient stress responses, this dimension measures the chronic cognitive processing biases and subjective mental distress characteristic of Generalized Anxiety Disorder and the anticipatory fear related to Panic Disorder. Key phenomenological domains evaluated within this subscale include:

• Apprehensive Expectation and Pervasive Worry: Uncontrollable, excessive, and generalized cognitive rumination regarding future catastrophes, daily responsibilities, and unforeseen difficulties.
• Subjective Terror and Mental Apprehension: Acute episodes of cognitive panic, dread, and feeling overwhelmed by impending danger (exemplified by items reflecting subjective fear, such as feeling terrified or scared).
• Psychological Tension and Inability to Relax: An internal sensation of mental strain, subjective restlessness, and cognitive exhaustion resulting from unremitting vigilance.
• Irritability and Emotional Fragility: Heightened emotional reactivity, frustration intolerance, and a lowered threshold for stress-induced cognitive distress.

2. Somatic Anxiety Subscale (14 Items)

The Somatic Anxiety subscale measures the peripheral, physiological, and neuromuscular manifestations triggered by autonomic nervous system hyperarousal. Pathological anxiety involves substantial sympathovagal dysregulation, leading to sustained physiological activation. The CUXOS captures this physiological cascade across several physiological systems:

• Cardiovascular Reactivity: Tachycardia, palpitations, perceived chest tightness, and pounding heart rhythms resulting from beta-adrenergic stimulation.
• Respiratory Symptoms: Sensations of dyspnea, shortness of breath, hyperventilation, and choking sensations frequently tied to panic pathology.
• Neuromuscular and Musculoskeletal Tension: Generalized muscular rigidity, muscle aches, motor tremors, shakiness, physical instability, and subjective restlessness.
• Autonomic and Vasomotor Hyperarousal: Excessive diaphoresis (sweating), hot or cold flashes, clamminess, dry mouth, and cutaneous flushing.
• Gastrointestinal Distress: Visceral autonomic hyperactivity, including nausea, abdominal discomfort, epigastric churning, and digestive distress.
• Sensory and Vestibular Distortions: Dizziness, lightheadedness, faintness, unsteady gait, and paresthesias (tingling or numbness in extremities).

By segregating these symptom patterns across 20 focused items, the CUXOS enables clinicians to distinguish between patients whose presentation is driven primarily by cognitive rumination versus those experiencing disabling autonomic storming and somatic reactivity, thereby informing differential therapeutic selection (e.g., targeted cognitive restructuring versus somatic downregulation techniques or pharmacotherapy).

6. Theoretical Framework

The conceptual architecture of the CUXOS is grounded in classical psychometric theory, nosological psychiatric taxonomy, and cognitive-affective models of psychopathology. Primarily, the scale reflects the convergence of three dominant frameworks: the Clark and Watson Tripartite Model of Anxiety and Depression, the empirical nosology of the American Psychiatric Association (DSM), and the historical clinical measurement paradigm established by Max Hamilton.

The Tripartite Model and Affective Differentiation

A central dilemma in clinical psychometrics is the substantial diagnostic and symptomatic comorbidity observed between major depressive episodes and anxiety disorders. David Clark and Lee Anna Watson (1991) formulated the Tripartite Model to explain both the shared variance and the unique features distinguishing anxiety from depression:

• General Distress / Negative Affectivity: A non-specific factor shared between anxiety and depression, comprising general dysphoria, irritability, demoralization, and subjective distress.
• Anhedonia / Low Positive Affectivity: A unique factor characteristic of depression, marked by a loss of pleasure, diminished energy, and emotional blunting.
• Physiological Hyperarousal: A specific factor distinctive to anxiety states (particularly panic and somatic tension), characterized by autonomic discharge, somatic tremors, cardiovascular acceleration, and physical vigilance.

The CUXOS was engineered specifically to capture this distinct physiological hyperarousal domain along with the pathognomonic cognitive dread of anxiety, avoiding items heavily saturated with general anhedonic depression. In doing so, the CUXOS reliably disentangles clinical anxiety from primary affective disorders when utilized in tandem with its counterpart, the Clinically Useful Depression Outcome Scale (CUDOS).

DSM Phenomenological Integration and Hamilton Tradition

Historically, the Hamilton Rating Scale for Anxiety (HAM-A) established the foundational dichotomy between psychic and somatic anxiety. However, the HAM-A predated modern categorical diagnostic frameworks and contained items that overlap substantially with medical illnesses and depressive neurovegetative symptoms. Zimmerman and colleagues integrated the psychic/somatic structural distinction of the HAM-A while aligning item criteria directly with modern operational definitions of Generalized Anxiety Disorder (characterized by persistent, diffuse apprehensive expectation and motor tension) and Panic Disorder (characterized by discrete surges of autonomic panic and autonomic hyperarousal) as delineated in the DSM-III-R and DSM-IV. Consequently, the CUXOS serves as a modern self-report operationalization of established nosological criteria, translated into a brief, psychometrically sound format.

7. Validity

The construct, convergent, discriminant, and criterion validity of the CUXOS have been rigorously evaluated in large-scale clinical cohorts. The empirical evidence demonstrates that the scale functions as an accurate, sensitive, and diagnostically discerning index of anxiety severity.

Construct and Discriminant Validity

In the landmark validation study conducted by Zimmerman et al. (2010), involving 963 psychiatric outpatients presenting for comprehensive evaluation (556 diagnosed with a primary non-comorbid anxiety disorder and 407 without an anxiety disorder based on structured diagnostic interviews), the CUXOS demonstrated pronounced construct validity. Patients presenting with diagnosed anxiety disorders scored significantly higher on the CUXOS compared to non-anxious clinical controls (mean = 38.6 vs. 15.3, p < .001). Receiver Operating Characteristic (ROC) analyses demonstrated excellent diagnostic discrimination, yielding an Area Under the Curve (AUC) exceeding .88 for detecting clinical anxiety disorders.

Discriminant validity was established by evaluating the degree of divergence between the CUXOS and measures of non-anxiety constructs. The CUXOS correlated significantly more strongly with other validated anxiety scales (median r = .54) than with measures of distinct psychopathological domains, such as depressive severity (via the CUDOS and the Hamilton Depression Rating Scale; median r = .32) and psychosocial dysfunction. In an analysis of variance (ANOVA) examining social impairment across severity gradients measured by the Social Avoidance and Distress Scale (SADS), higher SADS severity tiers mapped onto monotonically higher CUXOS scores, confirming the scale’s sensitivity to social and agoraphobic distress.

Convergent Validity

Convergent validity has been replicated across independent clinical evaluations. In a large-scale transcultural investigation conducted by Jeon et al. (2017) involving 838 psychiatric outpatients across multi-center psychiatric departments utilizing a validated Korean adaptation, the CUXOS demonstrated strong convergent alignment with gold-standard anxiety metrics:

• Beck Anxiety Inventory (BAI): r = .78 (p < .001)
• State-Trait Anxiety Inventory (STAI – State subscale): r = .72 (p < .001)
• Hamilton Rating Scale for Anxiety (HAM-A): r = .71 (p < .001)

Critically, Jeon et al. verified that the CUXOS demonstrated significantly stronger correlations with composite anxiety instruments (mean r = .74) than with depressive symptom scales (mean r = .53), confirming that the instrument maintains construct distinctiveness even within highly comorbid clinical populations.

Sensitivity to Change and Criterion Validity

Investigations into measurement-based care implementation (e.g., Beidas et al., 2015; D’Avanzato et al., 2013) demonstrated that the CUXOS exhibits high sensitivity to change throughout psychiatric and psychological interventions. Reductions in CUXOS scores parallel clinical global impressions of improvement (CGI-I), with clinically recovered individuals demonstrating dramatic drops across both psychic and somatic subscales. Furthermore, longitudinal outcome evaluations revealed that residual elevations on the CUXOS among remitted depressed patients predicted an elevated risk of functional impairment and subsequent affective relapse.

8. Reliability

The reliability of the CUXOS has been documented across internal consistency and temporal stability metrics, confirming that the tool provides consistent measurement across varied psychiatric populations.

Internal Consistency

In the initial cohort of 963 psychiatric outpatients evaluated by Zimmerman et al. (2010), internal consistency coefficients were exceptionally high, reflecting minimal measurement error and substantial item homogeneity:

• Total CUXOS Scale (20 items): Cronbach’s α = .95
• Psychic Anxiety Subscale (6 items): Cronbach’s α = .90
• Somatic Anxiety Subscale (14 items): Cronbach’s α = .93

Item-total correlations ranged from .58 to .81, indicating that each item contributed reliably to the total construct without redundant collinearity. Cross-cultural validation by Jeon et al. (2017) confirmed this high internal reliability within an independent East Asian psychiatric sample, yielding a total Cronbach’s alpha of .90 (with psychic and somatic subscales exhibiting alphas of .86 and .88, respectively).

Test-Retest Reliability

Temporal stability was evaluated within outpatient subsets re-assessed across short test-retest intervals (24 to 48 hours) prior to the initiation of active pharmacological or psychological interventions, ensuring clinical stability:

• Zimmerman et al. (2010): The test-retest reliability coefficient for the total CUXOS score was r = .90 (p < .001), indicating remarkable stability across brief retest intervals without systematic recall attenuation.
• Jeon et al. (2017): In a clinical retest cohort evaluated over a 1- to 2-week baseline period, the intraclass correlation coefficient remained substantial at r = .74 (p < .001), demonstrating stable baseline measurement over broader routine intake intervals.

9. Factor Analysis

Structural evaluations through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) substantiate the theoretical architecture of the CUXOS, validating both a multidimensional two-factor formulation and a higher-order overarching anxiety severity factor.

Exploratory Factor Structure

Initial principal components and exploratory factor analyses conducted on the 20 CUXOS items using promax (oblique) rotations yielded a robust two-factor solution accounting for the preponderance of total variance:

• Factor 1: Somatic Anxiety: Comprising 14 items tapping physiological autonomic reactivity, neuromuscular tension, cardiovascular symptoms, and visceral sensations. Factor loadings for individual somatic items ranged from .52 to .84, with negligible cross-loadings onto the psychic dimension.
• Factor 2: Psychic Anxiety: Comprising 6 items capturing psychological apprehension, subjective dread, cognitive rumination, mental tension, and irritability. Factor loadings for these cognitive-affective items ranged from .61 to .88.

The inter-factor correlation between the Psychic and Somatic dimensions was moderate to high (r ≈ .62 to .68), confirming that while the two dimensions represent distinct symptomatic manifestations, they are bound by a shared core of generalized anxiety severity.

Confirmatory Factor Analysis and Structural Fit

Confirmatory factor analytic investigations examining the fit of alternative theoretical models (unidimensional vs. correlated two-factor vs. bifactor models) have been conducted across diverse datasets (e.g., Jeon et al., 2017). Structural equation modeling parameters yielded superior goodness-of-fit indices for the correlated two-factor model and a hierarchical bifactor model:

• Comparative Fit Index (CFI): .93 to .96, indicating good structural congruence with observed data.
• Tucker-Lewis Index (TLI): .92 to .95, confirming model parsimony.
• Root Mean Square Error of Approximation (RMSEA): .052 to .068 (90% CI [.048, .072]), reflecting acceptable to close fit.
• Standardized Root Mean Square Residual (SRMR): .038 to .044, within strict conventional thresholds for model acceptance.

The bifactor structural model confirms that while psychic and somatic subscales can be interpreted separately to tailor clinical interventions, calculating a single composite summary score (0 to 80) is psychometrically justified due to the strong general anxiety factor that underpins all 20 items.

10. Instrument / Measurement Tool

The Clinically Useful Anxiety Outcome Scale (CUXOS) is formatted as an ultra-brief, standardized, pencil-and-paper or digitally administered self-report rating scale. Its structural parameters and administrative rules are summarized below:

• Construct Measured: Overall anxiety symptom severity, differentiated into cognitive/affective (psychic) and physiological (somatic) dimensions.
• Item Count: 20 items total.
  • Psychic Anxiety Subscale: 6 items (Items assessing fear, uncontrollable worry, inner tension, irritability, and cognitive panic).
  • Somatic Anxiety Subscale: 14 items (Items assessing diaphoresis, palpitations, tremors, shortness of breath, dizziness, GI disturbance, muscle tension, and autonomic hyperarousal).
• Recall Period: The preceding week (“Past 7 days, including today”).
• Response Format: 5-point Likert rating scale:
  • 0 = Not at all true
  • 1 = Rarely true
  • 2 = Sometimes true
  • 3 = Often true
  • 4 = Almost always true
• Administration Time: Approximately 90 to 120 seconds for patient self-completion.
• Scoring Time: Approximately 15 seconds for manual clerical or clinician scoring; instantaneous via electronic medical record (EMR) entry.
• Scoring Procedure: Item scores are summed directly to produce the total score and subscale scores.
  • Total CUXOS Score Range: 0 to 80 points.
  • Psychic Anxiety Score Range: 0 to 24 points.
  • Somatic Anxiety Score Range: 0 to 56 points.
• Empirical Severity Cutoff Ranges:
  • ≤ 10: Non-anxious (Symptom levels equivalent to community normative benchmarks).
  • 11 – 20: Minimal anxiety (Subthreshold symptoms not indicative of functional impairment).
  • 21 – 30: Mild anxiety (Clinically detectable anxiety warranting monitoring or watchful waiting).
  • 31 – 40: Moderate anxiety (Clinically significant anxiety disorder typically requiring therapeutic intervention).
  • ≥ 41: Severe anxiety (Marked to disabling anxiety symptomatology requiring comprehensive clinical management).

11. Permissions & Fee and Test Year

The Clinically Useful Anxiety Outcome Scale was first published in 2010 by Mark Zimmerman, Iwona Chelminski, Diane Young, and Kristy Dalrymple following clinical development under the Rhode Island Hospital Methods to Improve Diagnostic Assessment and Services (MIDAS) project.

• Licensing and Clinical Accessibility: Consistent with the MIDAS project’s commitment to disseminating measurement-based care tools, the CUXOS was placed into clinical and non-profit research availability without commercial royalty fees. Clinicians and researchers in non-profit or hospital environments are generally granted permission to reproduce and administer the scale free of financial charges, provided appropriate citation is maintained and the content is unaltered.
• Commercial and Digital Redistribution: Commercial entities, pharmaceutical clinical trial sponsors, for-profit technology platforms, and electronic medical record vendors seeking to license or integrate the scale into commercial software products should seek explicit permission from the original developers at Rhode Island Hospital and the Brown Medical School research group.
• Original Publication Source: The Journal of Clinical Psychiatry (Zimmerman et al., 2010).

12. References

• Beidas, R. S., Stewart, R. E., Walsh, L., Lucas, S., Downey, M. M., Jackson, K., Fernandez, T., & Mandell, D. S. (2015). Free, brief, and validated: Standardized instruments for low-resource mental health settings. Cognitive and Behavioral Practice, 22(1), 5–19. https://doi.org/10.1016/j.cbpra.2014.02.002
• Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100(3), 316–336. https://doi.org/10.1037/0021-843X.100.3.316
• D’Avanzato, C., Martinez, J., Attiullah, N., Friedman, M., Toba, C., Boerescu, D. A., & Zimmerman, M. (2013). Anxiety symptoms among remitted depressed outpatients: Prevalence and association with quality of life and psychosocial functioning. Journal of Affective Disorders, 151(1), 401–404. https://doi.org/10.1016/j.jad.2013.06.040
• Hamilton, M. (1959). The assessment of anxiety states by rating. British Journal of Medical Psychology, 32(1), 50–55. https://doi.org/10.1111/j.2044-8341.1959.tb00467.x
• Jeon, S. W., Han, C., Ko, Y.-H., Yoon, S., Pae, C.-U., Choi, J., Kim, S.-G., & Zimmerman, M. (2017). A Korean validation study of the Clinically Useful Anxiety Outcome Scale: Comorbidity and differentiation of anxiety and depressive disorders. PLoS ONE, 12(6), Article e0179247. https://doi.org/10.1371/journal.pone.0179247
• Zimmerman, M., Chelminski, I., McGlinchey, J. B., & Posternak, M. A. (2008). A clinically useful depression outcome scale. Comprehensive Psychiatry, 49(2), 131–140. https://doi.org/10.1016/j.comppsych.2007.10.006
• Zimmerman, M., Chelminski, I., Young, D., & Dalrymple, K. (2010). A Clinically Useful Anxiety Outcome Scale. The Journal of Clinical Psychiatry, 71(5), 534–542. https://doi.org/10.4088/JCP.09m05313blu

13. Items of the Scale