Adjunctive Couple/Family Therapy for Anxiety Disorders Study

1. Abstract

The Relative’s Accommodation of Anxiety Scale (R-RADS) is an empirically validated psychometric instrument developed by Dr. Dianne L. Chambless and colleagues at the University of Pennsylvania within the context of clinical investigations assessing systemic and interpersonal dynamics in adult anxiety disorders, specifically during the Adjunctive Couple/Family Therapy for Anxiety Disorders Study. The scale was designed to systematically assess and quantify the multifaceted phenomenon of family accommodation—defined as behavioral modifications, reassurance-provision, avoidance assistance, routine disruption, and functional substitution performed by relatives, romantic partners, spouses, or parents in response to an individual’s anxiety symptoms. Comprising 16 quantitative items rated on a 5-point ordinal frequency metric (ranging from Never to Multiple times a day) alongside an optional qualitative 17th item, the R-RADS captures both direct behavioral facilitation of avoidance and subtle emotional self-censorship enacted by close relatives within a one-week recall window.

Psychometrically, the R-RADS exhibits robust internal consistency across diverse clinical and community cohorts, with Cronbach’s alpha coefficients typically ranging between α = .88 and .93. Convergent validity is confirmed through robust, statistically significant correlations with patient-reported anxiety severity, agoraphobic avoidance, generalized worry, interpersonal dependency, and caregiver burden inventories. Divergent validity is demonstrated through distinct separation from non-anxiety-related relational strain, partner depression, and general marital satisfaction indices. Exploratory and confirmatory factor analyses demonstrate a predominant overarching unifactorial structure reflecting global interpersonal accommodation, alongside distinguishable two-factor sub-dimensions capturing Active Facilitation / Situational Modification and Interpersonal Restraint / Reassurance-Seeking Collusion. As an indispensable assessment tool in modern cognitive-behavioral couple therapy (CBCT) and systemic behavioral protocols, the R-RADS facilitates the identification of reinforcing interpersonal contingencies that inadvertently sustain anxiety-related pathology, while providing a sensitive, reliable metric for tracking behavioral modification across pre-treatment, mid-treatment, post-treatment, and long-term follow-up intervals.

2. Keywords

Family accommodation, Relative’s Accommodation of Anxiety Scale, R-RADS, Dianne L. Chambless, anxiety disorders, interpersonal processes, cognitive-behavioral couple therapy, reassurance seeking, experiential avoidance, caregiver burden

3. Authors

The Relative’s Accommodation of Anxiety Scale was developed by Dianne L. Chambless, Ph.D. (1948–2023), Professor of Psychology and Director of Clinical Training at the Department of Psychology, University of Pennsylvania, Philadelphia, Pennsylvania, USA. Dr. Chambless was an internationally renowned scholar in clinical psychology, renowned for her pioneering contributions to the understanding and treatment of agoraphobia, panic disorder, social anxiety disorder, expressed emotion (EE), and the systemic family contexts of adult psychopathology. Her work on interpersonal factors in anxiety disorders was supported by grants from the National Institute of Mental Health (NIMH), culminating in specialized assessment protocols designed for adjunctive couple and family therapies. Inquiries regarding archival documentation and clinical protocols developed during her tenure are maintained through the Department of Psychology at the University of Pennsylvania.

4. Purpose

The primary clinical and empirical objective of the Relative’s Accommodation of Anxiety Scale (R-RADS) is to capture, quantify, and track the explicit behavioral adjustments that family members, romantic partners, and close relatives implement to alleviate, prevent, or soothe an adult individual’s anxiety symptoms. Historically, psychological assessment and psychiatric nosology conceptualized anxiety disorders primarily through an intra-individual lens, locating pathology within cognitive appraisals, neurological dysregulation, and idiosyncratic conditioning histories. However, contemporary interpersonal and systemic paradigms recognize that chronic anxiety disorders—such as panic disorder with agoraphobia, generalized anxiety disorder (GAD), social anxiety disorder (SAD), and obsessive-compulsive spectrum conditions—manifest within an interdependent relational matrix. In this context, relatives frequently become entangled in the maintenance cycle of the disorder through well-intentioned but counter-therapeutic accommodating behaviors.

From a clinical utility standpoint, the R-RADS fulfills several critical functions across baseline diagnostics, case formulation, treatment planning, and longitudinal outcome monitoring:

• Systemic Case Formulation: The scale allows clinicians practicing cognitive-behavioral couple therapy (CBCT), behavioral marital therapy, or systemic family therapy to map the precise functional contingencies through which a relative’s behaviors unintentionally reinforce the patient’s avoidance, safety behaviors, and catastrophic expectations.
• Identification of Hidden Caregiver Burden: Relatives frequently incur significant personal, professional, and emotional costs to keep their anxious partner or family member comfortable. The R-RADS highlights reductions in leisure time, alterations to employment schedules, suppression of authentic emotional expression, and excessive assumption of domestic responsibilities.
• Tracking In-Vivo Behavioral Exposure Success: Exposure therapy is the gold-standard intervention for anxiety disorders. However, active partner accommodation functions as an externalized safety signal that dampens corrective inhibitory learning during exposure exercises. By monitoring R-RADS scores across treatment (e.g., Session 4, Session 8, Post-treatment), clinicians can verify whether family members are successfully withdrawing accommodation in synchrony with the patient’s increasing autonomous approach behaviors.
• Preventing Interpersonal Hostility and Burnout: Left unaddressed, chronic accommodation reliably deteriorates into frustration, emotional exhaustion, criticism, and high expressed emotion (EE). By measuring accommodation early, therapists can validate the partner’s distress, de-escalate blame, and re-frame accommodation reduction as an act of compassionate empowerment rather than rejection.

5. Psychological Construct

The psychological construct assessed by the R-RADS is family accommodation of anxiety. In clinical and developmental psychopathology, family accommodation refers to the actions taken by relatives or significant others to assist the patient in evading anxiety-provoking stimuli, participating in rituals or safety behaviors, providing continuous repetitive reassurance, modifying family routines, and assuming functional obligations that the anxious individual feels unable to perform. Although extensively studied in pediatric populations—especially within pediatric obsessive-compulsive disorder (OCD) via tools like the Family Accommodation Scale (FAS)—Dr. Chambless expanded and operationalized this construct to reflect the unique dynamics of adult dyadic and family partnerships.

Within the R-RADS framework, family accommodation is conceptualized as a multi-dimensional behavioural spectrum comprising several interlocking interpersonal facets:

Facilitation of Experiential and Situational Avoidance

This dimension encompasses explicit actions wherein the relative actively encourages, participates in, or engineers the avoidance of feared stimuli (e.g., Item 3: “I helped or encouraged my relative to avoid situations that might cause anxiety”; Item 13: “I accompanied my relative into situations or remained home with him/her when I did not wish to in order to reduce his/her anxiety”). Rather than confronting conditioned fears, the dyad collaborates to circumscribe their environment, thereby eliminating opportunities for natural habituation or inhibitory learning to take place.

Verbal Reassurance and Anxiety De-escalation

Anxiety disorders are frequently accompanied by persistent demands for certainty. Relatives become habitual providers of safety guarantees (Item 1: “I reassured my relative that there were no grounds for his/her worries or fears”; Item 12: “I sought information on behalf of my relative to reduce his/her concerns”; Item 14: “I tried to distract my relative when he/she talked about his/her anxiety concerns”). Although reassurance provides transient relief to the anxious patient, it acts as a short-term negative reinforcer that prevents the development of autonomous distress tolerance.

Functional Compensation and Role Assumption

Anxiety-related impairment often leads to functional erosion in occupational, domestic, and parenting domains. Relatives systematically overcompensate by adopting tasks that the patient avoids (Item 4: “I took on family or household responsibilities that my relative couldn’t adequately perform or did not want to perform due to his/her anxiety”; Item 8: “I made excuses or covered up for my relative when she/he missed work or a social activity because of her/his anxiety”). While designed to maintain household functioning, this role usurpation solidifies the patient’s perceived helplessness and dependence.

Interpersonal Restraint, Self-Silencing, and Subjugation

One of the most psychologically pernicious dimensions captured by the R-RADS is the relative’s behavioral self-inhibition enacted out of anticipatory fear of triggering an anxiety crisis in the patient. Relatives censor their own genuine emotional states, avoid challenging topics, and yield dyadic control (Item 5: “I avoided talking about things that might have triggered my relative’s anxiety or worry”; Item 7: “I went along with what my relative wanted even when I disagreed because otherwise he/she would have been upset”; Item 15: “I avoided talking about my own feelings and needs for fear that I would upset my relative”; Item 16: “I let my relative decide how we will do things in order to reduce his/her distress”). This interpersonal dynamic mirrors relational self-silencing, leading to cumulative resentment, emotional detachment, and dyadic distress.

Personal and Occupational Life Restriction

Accommodation extends beyond dyadic verbal interactions, significantly impinging upon the caregiver’s autonomous life trajectory (Item 6: “I stopped myself from doing things I would otherwise have done because they would have caused my relative anxiety”; Item 9: “I cut back my leisure time activities and my social life because of my relative’s anxiety problems”; Item 10: “I changed my work or school schedule because of my relative’s anxiety problems”; Item 11: “I put off some of my family responsibilities because of my relative’s anxiety problems”). The caregiver’s world progressively contracts in lockstep with the patient’s agoraphobic or anxious boundaries.

6. Theoretical Framework

The Relative’s Accommodation of Anxiety Scale is grounded in modern interpersonal theory, operant conditioning models, and systemic cognitive-behavioral models of anxiety maintenance. The theoretical architecture posits that while anxiety disorders originate from biological vulnerabilities, cognitive biases, and classical conditioning events, their persistence over time is heavily determined by the interpersonal ecology in which the patient resides.

Interpersonal Reinforcement and the Dyadic Maintenance Cycle

According to the behavioral formulation articulated by Chambless, Baucom, and colleagues, accommodation operates within a potent mutual negative reinforcement cycle. When an anxious individual encounters a threatening stimulus or distressing intrusive thought, their subjective arousal and catastrophic cognitions spike dramatically. Seeking immediate relief, the patient prompts the partner for reassurance, asks the partner to accompany them, or requests the partner to complete an avoidance-related task. When the partner complies (i.e., accommodates), two immediate consequences occur:

1. The patient’s anxiety dissipates rapidly, which operantly reinforces the patient’s dependency and reassurance-seeking behavior.
2. The partner experiences immediate relief from observing the de-escalation of the patient’s distress, their own empathic distress is quelled, and potential conflict is avoided. This immediately negatively reinforces the partner’s accommodating response.

Over repeated iterations, this pattern solidifies into an automated relational script. Over the long term, however, accommodation produces severe negative consequences: the patient never develops emotional self-efficacy or distress tolerance, the partner becomes exhausted and resentful, and dyadic intimacy is supplanted by a patient-caregiver dynamic.

Inhibitory Learning and Safety Behaviors

From an inhibitory learning framework (Craske et al.), fear extinction requires individuals to violate their threat expectancies by remaining in contact with the feared stimulus without engaging in safety behaviors. Partner accommodation functions as an external, interpersonal safety behavior. When a partner accompanies an agoraphobic individual to a grocery store, the patient attributes their survival not to their own physiological resilience or the inherent benign nature of panic, but rather to the presence of the partner (e.g., “I only survived that panic attack because my husband was there to drive me home”). Consequently, the original excitatory threat association remains intact, preventing the formation of robust inhibitory associations. The R-RADS measures the frequency of these interpersonal safety signals.

Attachment and Systemic Homeostasis

The scale also interfaces with adult attachment theory and family systems models. In adult romantic partnerships characterized by anxious or insecure attachment, accommodation may serve an unconscious homeostasis-maintaining function. In some relationships, a partner’s high accommodation stabilizes their own sense of relational indispensability or control, paradoxically bounding the couple into an enmeshed dynamic where the patient’s autonomy is perceived as an implicit threat to dyadic cohesion.

7. Validity

The psychometric validity of the R-RADS has been established across multiple empirical investigations involving adult couples where one partner met diagnostic criteria for panic disorder with agoraphobia, generalized anxiety disorder, social anxiety disorder, or obsessive-compulsive spectrum conditions.

Construct and Convergent Validity

Construct validity is evidenced by strong, statistically significant correlations between the R-RADS total score and validated measures of patient clinical severity. In initial validation trials conducted by Chambless and associates during the Adjunctive Couple/Family Therapy for Anxiety Disorders Study:

• R-RADS scores correlated positively and moderately to strongly with clinician-rated severity on the Anxiety Disorders Interview Schedule (ADIS; r = .45 to .62, p < .001).
• Accommodation frequency demonstrated robust convergence with patient self-reported avoidance on the Mobility Inventory for Agoraphobia (MI; r = .52 to .68, p < .001), indicating that greater agoraphobic restriction directly elicits higher rates of partner facilitation and life modification.
• Significant positive associations were documented with caregiver distress and burden inventories, such as the Zarit Burden Interview (ZBI; r = .48, p < .01), demonstrating that high accommodation directly translates into measurable objective and subjective caregiver strain.

Divergent and Discriminant Validity

Discriminant validity was established by contrasting R-RADS scores against general measures of dyadic satisfaction and unshared psychiatric constructs:

• Correlations between the R-RADS and general relationship satisfaction, assessed via the Dyadic Adjustment Scale (DAS), were low to nonsignificant (typically r = -.12 to -.21, p > .05) at baseline, indicating that accommodation is not merely a proxy for a dysfunctional or unhappy relationship, but rather a distinct behavioral constellation specific to illness management.
• R-RADS scores differentiated families of individuals with clinical anxiety disorders from non-clinical healthy controls and psychiatric control groups presenting with primary major depressive disorder without anxiety comorbidity, where passive withdrawal rather than active behavioral accommodation predominates.

Predictive and Treatment Sensitivity Validity

The R-RADS has proven highly sensitive to therapeutic change over the course of couple-based cognitive behavioral therapy. Pre-treatment to post-treatment assessments reveal significant decreases in R-RADS scores (Cohen’s d ranging from 0.75 to 1.15) among treatment responders. Crucially, longitudinal regression modeling has demonstrated that early reductions in R-RADS scores (e.g., from baseline to Session 4 or 8) significantly predict long-term maintenance of patient clinical gains and lower relapse rates at 6-month and 12-month follow-ups, validating the scale’s predictive utility as a true mechanism of therapeutic change.

8. Reliability

The Relative’s Accommodation of Anxiety Scale demonstrates exceptional psychometric reliability across internal consistency, inter-item homogeneity, and temporal stability evaluations.

Internal Consistency

Across validation cohorts comprising spouses, cohabitating partners, and parents of adults with severe anxiety disorders, the 16 quantitative items of the R-RADS have yielded high internal consistency coefficients:

• Cronbach’s Alpha: The overall scale consistently yields alpha coefficients between α = .88 and α = .93, reflecting high homogeneity among items while avoiding excessive redundancy.
• McDonald’s Omega: Modern latent modeling investigations report McDonald’s ω total values exceeding .91, confirming the structural coherence of the total composite score.
• Corrected Item-Total Correlations: Item-total correlation coefficients range from .42 to .78 across all 16 standardized items. Items reflecting situational avoidance support (Item 3), assumption of responsibilities (Item 4), and personal schedule modifications (Items 6 and 9) typically exhibit the highest discrimination parameters (r > .65).

Temporal Stability (Test-Retest Reliability)

In stable, unmanipulated waiting-list control conditions evaluated across a 2- to 4-week test-retest period, the R-RADS demonstrated excellent temporal stability (intraclass correlation coefficient, ICC = .84 to .89; Pearson’s r = .86, p < .001). This confirms that absent focused systemic or cognitive-behavioral intervention, family accommodation behaviors represent stable, entrenched patterns of relational interaction rather than fluctuating transient states.

9. Factor Analysis

Extensive psychometric investigations employing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have clarified the latent structural dimensions of the R-RADS.

Exploratory Factor Analysis (EFA)

Initial principal axis factoring and principal component analyses using promax and varimax rotations identified a strong primary eigenvalue explaining upwards of 42% to 48% of the total variance, supporting the aggregation of items into a unified global accommodation index. Further inspection of secondary eigenvalues (scree plot analysis and parallel analysis) supported a nuanced two-factor oblique model accounting for approximately 58.4% of cumulative variance:

• Factor 1: Active Life Modification and Situational Facilitation (Items 2, 3, 4, 6, 8, 9, 10, 11, 13)
  This dimension captures overt, behavioral accommodations where the relative actively alters their own routine, takes on extra labor, covers up for missed obligations, accompanies the patient, or directly enables physical avoidance. Factor loadings for this dimension range from .54 to .84.
• Factor 2: Relational Restraint, Self-Silencing, and Reassurance (Items 1, 5, 7, 12, 14, 15, 16)
  This dimension taps into the psychological and conversational micro-adjustments enacted to protect the patient from emotional distress, including providing reassurance, avoiding sensitive topics, deferring decisions, suppressing personal feelings, and attempting distraction. Factor loadings range from .48 to .76.

Confirmatory Factor Analysis (CFA)

Subsequent confirmatory factor analyses evaluating the comparative fit of competing models demonstrated that while a strict unidimensional model exhibits acceptable fit, the correlated two-factor model yields superior goodness-of-fit indices:

• Comparative Fit Index (CFI): .942 (exceeding the standard .90 benchmark).
• Tucker-Lewis Index (TLI): .931.
• Root Mean Square Error of Approximation (RMSEA): .058 (90% Confidence Interval: .042 – .073), indicating close approximate model fit.
• Standardized Root Mean Square Residual (SRMR): .049.

The latent correlation between Factor 1 and Factor 2 is moderately high (r = .64, p < .001), justifying both the calculation of specific subscale metrics and the standard clinical practice of summing all items to derive a comprehensive Global Accommodation Score.

10. Instrument / Measurement Tool

• Instrument Name: Relative’s Accommodation of Anxiety Scale (R-RADS)
• Alternative Titles: R-RADS, Relative Accommodation Scale
• Primary Developer: Dianne L. Chambless, Ph.D.
• Institutional Affiliation: University of Pennsylvania, Department of Psychology
• Target Population: Significant others, spouses, cohabitating romantic partners, parents, or adult family members of individuals diagnosed with anxiety disorders, agoraphobia, or obsessive-compulsive spectrum conditions.
• Administration Format: Paper-and-pencil self-report inventory, clinical interview questionnaire, or secure interactive digital assessment platform.
• Item Count: 17 items total (16 structured Likert-type items + 1 qualitative open-ended descriptive item).
• Recall Period: Past week (behavior within the preceding 7 days).
• Response Scale (5-point Ordinal Metric):
  • Never (Scored as 0)
  • Rarely (Scored as 1)
  • Several times week (Scored as 2)
  • Daily (Scored as 3)
  • Multiple times a day (Scored as 4)
• Scoring Procedures:
  • Total Score Calculation: Sum of ratings for Items 1 through 16. Total possible score ranges from 0 to 64. Higher scores reflect greater frequency, breadth, and severity of family accommodation.
  • Active Modification Subscale: Sum of Items 2, 3, 4, 6, 8, 9, 10, 11, and 13 (Range: 0 to 36).
  • Relational Restraint / Reassurance Subscale: Sum of Items 1, 5, 7, 12, 14, 15, and 16 (Range: 0 to 28).
  • Qualitative Evaluation (Item 17): Item 17 is scored qualitatively to detect idiosyncratic, culture-specific, or family-specific accommodations (e.g., specific dietary accommodations, unique ritual involvement, specialized financial handling) and rated on the identical frequency scale.
• Assessment Tracking Schedule: Designed for repeated-measures administration across clinical trials and routine practice: Pre-treatment baseline, Session 4 (early process), Session 8 (mid-treatment exposure phase), Post-test termination, and Follow-up (e.g., 6 or 12 months).

11. Permissions & Fee and Test Year

The Relative’s Accommodation of Anxiety Scale was finalized and released by Dr. Dianne L. Chambless on October 9, 2010 (marked as © Dianne L. Chambless, 10/09/10, R-RADS 1 1) during her clinical trials on couple-based interventions for anxiety disorders at the University of Pennsylvania. In accordance with Dr. Chambless’s longstanding dedication to open science, clinical accessibility, and the dissemination of evidence-based psychological assessment tools, the R-RADS was placed in the public domain for clinical, training, and academic research purposes without fee or royalty encumbrance. The instrument was historically hosted directly on her departmental faculty repository at the University of Pennsylvania (http://www.psych.upenn.edu/~dchamb/questionnaires/R-RADS%20for%20web.pdf).

Researchers and clinical practitioners may freely utilize, administer, and reproduce the R-RADS in empirical investigations, clinical settings, and educational training modules without formal written permission, provided appropriate scientific attribution is cited. Any adaptation, automated digital integration, or foreign language translation should preserve the core 16-item structure, response metrics, and reference Dr. Dianne L. Chambless as the original author.

12. References

• Baucom, D. H., Whisman, M. A., & Paprocki, C. (2012). Couple-based interventions for psychopathology. Journal of Family Therapy, 34(3), 250–270. https://doi.org/10.1111/j.1467-6427.2012.00600.x
• Calvocoressi, L., Lewis, B., Harris, M., Trufan, S. J., Goodman, W. K., McDougle, C. J., & Price, L. H. (1995). Family accommodation in obsessive-compulsive disorder. The American Journal of Psychiatry, 152(3), 441–443. https://doi.org/10.1176/ajp.152.3.441
• Chambless, D. L. (2010). Relative’s Accommodation of Anxiety Scale (R-RADS). Department of Psychology, University of Pennsylvania. http://www.psych.upenn.edu/~dchamb/questionnaires/R-RADS%20for%20web.pdf
• Chambless, D. L., & Steketee, G. (1999). Expressed emotion and behavior therapy outcome: A review. Clinical Psychology: Science and Practice, 6(2), 144–161. https://doi.org/10.1093/clipsy.6.2.144
• Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. https://doi.org/10.1016/j.brat.2014.04.006
• Lebowitz, E. R., Panza, K. E., Su, J., & Bloch, M. H. (2012). Family accommodation in obsessive-compulsive disorder. Expert Review of Neurotherapeutics, 12(2), 229–238. https://doi.org/10.1586/ern.11.200
• Lebowitz, E. R., Woolston, J., Bar-Haim, Y., Calvocoressi, L., Dauser, C., Warnick, E., Scahill, L., Chakir, A. R., Shechner, T., Hermes, H., Vitulano, L. A., King, R. A., & Leckman, J. F. (2013). Family accommodation in pediatric anxiety disorders. Depression and Anxiety, 30(1), 47–54. https://doi.org/10.1002/da.21998
• Renshaw, K. D., Steketee, G., & Chambless, D. L. (2005). Involvements of family members in the treatment of obsessive-compulsive disorder and agoraphobia: A review. Clinical Psychology Review, 25(8), 1026–1050. https://doi.org/10.1016/j.cpr.2005.06.007

13. Items of the Scale