Adult Separation Anxiety Questionnaire (ASA 27)

1. Abstract

The Adult Separation Anxiety Questionnaire (ASA-27) is a 27-item self-report psychometric instrument designed to evaluate the presence and severity of separation anxiety symptoms experienced during adulthood (specifically individuals aged 18 years and older). Developed by clinical psychologists and psychiatrists Vijaya Manicavasagar, Derrick Silove, Rosalind Wagner, and Julia Drobny at the University of New South Wales, the ASA-27 was constructed to adapt diagnostic formulations of childhood separation anxiety disorder from the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) into developmentally appropriate phenomenological expressions in adult life. Each item is rated on a 4-point Likert scale ranging from 0 (“This has never happened”) to 3 (“This happens very often”), yielding an aggregate dimensional severity score ranging between 0 and 81.

Psychometric evaluations across non-clinical, community, and clinical cohorts have demonstrated exceptional internal consistency (Cronbach’s alpha coefficients consistently ranging from .90 to .95) and robust test-retest reliability over multiple-week intervals ($r = .86$ to $.91$). Factor analytic studies demonstrate strong evidence for both a primary general factor capturing separation-induced distress and distinct multidimensional sub-domains, including catastrophic fears of harm to attachment figures, separation-induced panic/somatic manifestations, proximity maintenance behaviors, and sleep-related separation distress. The ASA-27 possesses notable convergent validity with measures of generalized anxiety, panic disorder, and anxious adult attachment, while successfully establishing discriminant validity from agoraphobia and dependent personality features. The instrument played a pivotal empirical role in establishing Adult Separation Anxiety Disorder (ASAD) as an independent diagnostic category in the transition to the DSM-5.

2. Keywords

Adult Separation Anxiety Questionnaire, ASA-27, Adult Separation Anxiety Disorder, attachment theory, psychometrics, anxiety disorders, separation panic, interpersonal distress, scale validation, clinical assessment

3. Authors

The ASA-27 was designed and validated by an influential research group specializing in anxiety disorders and phenomenology at the School of Psychiatry, University of New South Wales (UNSW), and the Centre for Population Mental Health, Sydney, Australia:

• Vijaya Manicavasagar, Ph.D. — Professor and Consultant Clinical Psychologist, School of Psychiatry, University of New South Wales, and the Black Dog Institute, Prince of Wales Hospital, Randwick, NSW, Australia. Recognized internationally for her pioneering phenomenological and empirical research on separation anxiety in adulthood.
• Derrick Silove, MD, FRANZCP — Scientia Professor of Psychiatry, School of Psychiatry, University of New South Wales; Director of the Psychiatry Research and Teaching Unit, South Western Sydney Local Health District, Sydney, Australia. Renowned for work on anxiety classification, complex trauma, and transcultural psychopathology.
• Rosalind Wagner, M.Clin.Psych. — Clinical Psychologist and Psychometrician, School of Psychiatry, University of New South Wales, Sydney, Australia.
• Julia Drobny, Ph.D. — Clinical Psychologist and Research Fellow, School of Psychiatry, University of New South Wales, Sydney, Australia.
• Collaborating Contributor (Earlier Phenomenological Formulations): John M. Curtis, MBBS, FRANZCP — Specialist Psychiatrist, St George Hospital and UNSW Faculty of Medicine.

4. Purpose

The primary purpose of the Adult Separation Anxiety Questionnaire (ASA-27) is to provide an empirically validated, sensitive, and standardized self-report metric for quantifying separation anxiety symptoms that manifest after the age of 18. Historically, separation anxiety disorder was relegated strictly to infancy, childhood, or adolescence within psychiatric taxonomies such as the DSM-III-R and DSM-IV. Clinicians routinely misattributed adult manifestations of excessive distress regarding physical or emotional distance from primary attachment figures (such as spouses, romantic partners, adult children, or elderly parents) to other nosological entities, most notably panic disorder with agoraphobia, generalized anxiety disorder (GAD), dependent personality disorder (DPD), or borderline personality disorder (BPD).

The ASA-27 was specifically formulated to resolve this clinical blind spot. By translating the developmental criteria of childhood separation anxiety into adult-specific contexts, the instrument assesses the emotional, cognitive, behavioral, and physiological sequelae that emerge when an adult faces actual or threatened separation from a primary attachment figure. Applications of the instrument span several critical domains:

• Clinical Diagnosis and Differential Assessment: The ASA-27 aids psychiatrists and clinical psychologists in distinguishing between agoraphobic avoidance (which is primarily driven by fears of experiencing panic attacks or physical collapse in places where escape might be difficult) and separation-based avoidance (which is driven exclusively by fear of being distant from an attachment figure or catastrophic fears regarding the safety of that figure).
• Treatment Planning and Monitoring: Because separation anxiety does not always respond optimally to traditional cognitive-behavioral protocols formulated for generalized anxiety or agoraphobia, administering the ASA-27 allows clinicians to identify specific attachment-related triggers, catastrophic cognitions, and safety behaviors (such as excessive telephone monitoring or carrying transitional comfort objects). It provides a baseline dimensional metric against which pharmacotherapeutic or psychotherapeutic progress can be longitudinally tracked.
• Epidemiological and Phenomenological Research: The scale enables researchers to investigate the developmental trajectory of separation anxiety across the lifespan, delineating individuals who suffer from adult-onset separation anxiety from those who experience a persistent continuation of childhood-onset separation anxiety.
• Interpersonal and Couples Counseling: In marriage and relationship clinics, the ASA-27 serves as an objective assessment to explore relationship strain, suffocating proximity-seeking behaviors, and maladaptive conflict arising from one partner’s profound fears of abandonment or illness befalling the other.

5. Psychological Construct

The construct measured by the ASA-27 is Adult Separation Anxiety Disorder (ASAD), characterized by developmentally inappropriate, excessive, and functionally impairing fear or distress concerning separation from individuals to whom the individual is attached. The construct is inherently multifaceted, integrating affective distress, cognitive catastrophe, behavioral proximity seeking, and vegetative-autonomic arousal. The scale systematically operationalizes five interrelated core dimensions:

1. Separation-Induced Panic and Somatic Hyperarousal

Adults with elevated separation anxiety frequently experience acute autonomic discharges when anticipating or experiencing physical separation. This domain measures symptoms analogous to panic attacks—such as tachycardia, sweating, tremulousness, dyspnea, headaches, and gastrointestinal distress (e.g., nausea or stomach aches)—that are triggered specifically by departing from home or leaving an attachment figure (e.g., Items 9, 21, and 24). Unlike unconditional panic disorder, these somatic surges occur almost exclusively in the context of relational detachment or geographical separation.

2. Catastrophic Apprehensions Regarding Harm to Attachment Figures

This cognitive dimension assesses persistent, intrusive worries that catastrophic events, such as motor vehicle accidents, severe medical crises, violence, or sudden death, will befall primary attachment figures when they are out of contact (e.g., Item 16). The construct encompasses not only fears of physical harm to others, but also fears of emotional abandonment, relationship termination, or the perceived psychological inability to survive independently without the attachment figure (e.g., Items 18 and 23).

3. Proximity Maintenance and Safety-Seeking Behaviors

This behavioral dimension captures the continuous efforts made by the individual to preserve sensory, geographical, or communicative contact with attachment figures. Operationalized behaviors include excessive or compulsive telephone contact (Item 22), excessive talking designed to keep the attachment figure physically and emotionally engaged (Items 10 and 26), avoiding being home alone when significant others are away (Item 20), and carrying comfort or security items (e.g., transitional objects in a purse or wallet; Item 3).

4. Distress upon Departure and Anticipatory Disruption

This affective-cognitive dimension quantifies severe distress, emotional dysregulation, and functional avoidance prior to or during travel, commuting, or standard departures from home (e.g., Items 2, 4, 6, 14, and 25). It also evaluates intolerance toward routine disruptions that impede or threaten regular relational contact (Items 7 and 17).

5. Sleep-Related Separation Disturbances

Nighttime represents an acute and protracted period of sensory and geographical detachment. This dimension captures sleep-related vulnerabilities, including difficulty falling asleep unless an attachment figure is physically present or vocal background noise (such as radio or television) is audible (Items 12 and 13), needing lights to remain illuminated to diminish feelings of vulnerability (Item 19), and experiencing recurring nightmares involving separation, abandonment, or distance from home (Items 5 and 15).

6. Theoretical Framework

The ASA-27 is grounded in John Bowlby’s Attachment Theory (1969, 1973, 1980), integrated with contemporary models of evolutionary psychopathology and adult developmental psychology. Bowlby postulated that the human attachment behavioral system evolved as an innate homeostatic mechanism designed to ensure survival by maintaining physical proximity between a vulnerable infant and a primary caregiver (the attachment figure). When proximity is preserved, the individual experiences security; when the attachment figure becomes inaccessible or unresponsive, the behavioral system activates a predictable sequence of protest, despair, and detachment accompanied by intense distress and physiological arousal.

Bowlby explicitly argued that although the attachment behavioral system is most obvious during infancy and childhood, it remains active and continually operates throughout the entire human lifespan. In adulthood, romantic partners, spouses, close family members, and occasionally adult offspring become primary “attachment figures” who serve as both a *safe haven* during times of acute distress and a *secure base* from which the adult explores the occupational and social environment. In normal adult development, internal working models of the self and others become sufficiently stable to permit prolonged separations without debilitating anxiety.

In individuals with pathological adult separation anxiety, however, the attachment system remains hyper-activated or becomes dysregulated due to genetic predispositions, neurobiological vulnerabilities (e.g., heightened amygdaloid sensitivity or noradrenergic dysregulation), and early developmental experiences involving inconsistent caregiving, parental overprotection, unexpected childhood bereavement, or traumatic attachment ruptures. Consequently, adult separation anxiety reflects a catastrophic failure of the individual’s internal working models to maintain a mental representation of safety during physical absence, leading to cognitive hypervigilance, behavioral clinging, and profound panic.

Furthermore, Donald Klein’s neurobiological “suffocation alarm” hypothesis of panic disorder provides supplementary theoretical backing for the ASA-27. Klein theorized that recurrent panic attacks stem from an evolutionary hypersensitivity to suffocation or separation distress. The ASA-27 bridges Bowlby’s socio-emotional construct and Klein’s physiological model by measuring how emotional detachment directly provokes acute autonomic and somatic dysregulation.

7. Validity

The psychometric validity of the ASA-27 has been empirically confirmed through an extensive array of validation investigations involving non-clinical university cohorts, community-dwelling adults, and clinical psychiatric outpatients diagnosed with anxiety and mood disorders.

Construct and Structural Validity

In the seminal validation study by Manicavasagar et al. (2003), construct validity was confirmed by demonstrating that the ASA-27 reliably discriminated between adult clinical cohorts with clinically confirmed separation anxiety and healthy control samples, as well as between patients with separation anxiety and those presenting with other established anxiety disorders such as pure panic disorder or social phobia. Individuals diagnosed with adult separation anxiety scored significantly higher ($p < .001$) on the overall scale than patients with other psychiatric diagnoses, demonstrating that the scale accurately captures a construct that is phenomenologically distinct from general distress.

Convergent Validity

Convergent validity is supported by strong, statistically significant correlations between the ASA-27 total score and established clinical measures assessing overlapping emotional constructs:

• Panic Disorder Severity Scale (PDSS): Moderate to high correlations ($r = .52$ to $.64$), reflecting shared autonomic hyperarousal while maintaining independence based on trigger specificity.
• Beck Anxiety Inventory (BAI): Robust positive correlations ($r = .58$ to $.68$), demonstrating that the ASA-27 captures clinically significant levels of pervasive physiological and cognitive anxiety.
• Adult Attachment Scales (e.g., Experiences in Close Relationships [ECR]): Significant positive correlations with Attachment Anxiety ($r = .61$ to $.72$), confirming that individuals who manifest high ASA-27 scores consistently exhibit anxious-preoccupied internal working models characterized by intense fear of rejection and hyper-activating proximity strategies.
• Adult Separation Anxiety Semistructured Clinical Interview (ASA-SCID): Exceptionally high concordance rates ($r = .78$ to $.84$), confirming that the self-report ASA-27 strongly mirrors clinician-rated diagnostic interviews.

Discriminant and Criterion Validity

Discriminant validity has been rigorously examined to refute the historic criticism that adult separation anxiety is merely a subtype of agoraphobia or dependent personality disorder. Factor analyses and multiple regression models show that while the ASA-27 correlates weakly to moderately with the Agoraphobic Cognitions Questionnaire (ACQ) and the Mobility Inventory for Agoraphobia (MIA) ($r = .30$ to $.42$), partial correlations controlling for general neuroticism reveal that the ASA-27 uniquely predicts separation-specific panic and avoidance of solitude, whereas agoraphobia instruments exclusively predict avoidance of public spaces, crowded venues, and enclosed environments regardless of interpersonal proximity. Furthermore, receiver operating characteristic (ROC) analyses have established optimal clinical cutoff scores (typically scores between 22 and 26) with sensitivity rates exceeding 85% and specificity rates above 80% for identifying individuals meeting formal clinical diagnostic thresholds.

8. Reliability

The ASA-27 demonstrates exceptional psychometric reliability across diverse demographic strata, cultural contexts, and clinical environments.

Internal Consistency

Across validation studies, the scale exhibits consistently high internal consistency. In the foundational study by Manicavasagar, Silove, Wagner, and Drobny (2003), the overall instrument yielded a Cronbach’s alpha ($\alpha$) of .95 in a clinical sample and .92 in a non-clinical sample. Subsequent international adaptations—including Italian, Turkish, Spanish, and Brazilian Portuguese translations—have documented alpha coefficients consistently ranging from $.90$ to $.94$. Corrected item-total correlations across the 27 items typically range from $.45$ to $.78$, indicating that every individual item makes a substantial and coherent contribution to the primary latent construct.

Test-Retest Reliability and Stability

Temporal stability has been verified across test-retest intervals ranging from two to four weeks in clinically stable outpatients and non-clinical participants. Test-retest reliability coefficients (Pearson’s $r$) are consistently reported between .86 and .91 ($p < .001$), indicating that the ASA-27 reliably reflects an enduring trait-like psychopathological vulnerability while remaining sufficiently sensitive to demonstrate symptom reduction following therapeutic interventions (such as targeted cognitive behavioral therapy or pharmacotherapy).

Standard Error of Measurement

The Standard Error of Measurement (SEM) for the ASA-27 total score is reliably estimated around 3.2 to 3.8 points. This relatively narrow error margin provides clinicians with high confidence when calculating the Minimal Detectable Change (MDC) or evaluating clinically significant change indices during treatment trials.

9. Factor Analysis

Extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have examined the structural dimensionality of the ASA-27. Early exploratory analyses (using principal component and principal axis factoring with varimax and oblimin rotations) consistently indicated the presence of one dominant general factor accounting for the majority of common variance (often between 38% and 48% of the variance), affirming that summing the 27 items to create a unified unidimensional index of adult separation anxiety is psychometrically sound.

However, multi-factor CFA solutions frequently demonstrate superior model fit when evaluating sub-dimensions of clinical presentation. A well-established four-factor first-order model (often subsumed under a single higher-order Adult Separation Anxiety factor) demonstrates optimal goodness-of-fit indices across diverse samples:

• Factor 1: Separation Panic and Somatic Hyperarousal (Items 4, 6, 9, 21, 24). Item factor loadings range between .62 and .84. Captures acute autonomic panic sensations and anticipation of somatic collapse during departures.
• Factor 2: Catastrophic Worries Regarding Harm and Abandonment (Items 8, 14, 16, 18, 23, 25, 27). Item factor loadings range between .58 and .79. Represents the cognitive domain of intrusive fears that attachment figures will experience trauma, fatal accidents, or relational defection.
• Factor 3: Proximity Maintenance and Safety-Seeking Behaviors (Items 1, 2, 3, 7, 10, 11, 17, 20, 22, 26). Item factor loadings range between .48 and .74. Operationalizes direct behaviors designed to prevent separation, including excessive telephone calling, speech behaviors, and carrying comfort items.
• Factor 4: Nocturnal Separation and Sleep Vulnerability (Items 5, 12, 13, 15, 19). Item factor loadings range between .54 and .81. Delineates nighttime isolation vulnerabilities, requirements for background sensory cues (lights, TV, vocalizations), and separation nightmares.

In structural equation modeling evaluations of this higher-order model, standard fit indices frequently demonstrate strong fit: Comparative Fit Index ($\text{CFI}$) $\ge .93$, Tucker-Lewis Index ($\text{TLI}$) $\ge .92$, Root Mean Square Error of Approximation ($\text{RMSEA}$) $\le .055$ (90% CI [.048, .062]), and Standardized Root Mean Square Residual ($\text{SRMR}$) $\le .050$.

10. Instrument / Measurement Tool

• Instrument Name: Adult Separation Anxiety Questionnaire (ASA-27)
• Instrument Type: Self-report psychological questionnaire / psychometric rating scale
• Target Population: Adults aged 18 years and older
• Number of Items: 27 items
• Completion Time: Approximately 5 to 10 minutes
• Response Format: 4-point ordinal Likert-type scale:
  • 0 = This has never happened
  • 1 = This happens occasionally
  • 2 = This happens fairly often
  • 3 = This happens very often
• Scoring Procedure: Items are summed arithmetically to generate a continuous total score ranging from 0 to 81. All items are scored positively (no reverse-scored items).
• Clinical Cutoff Guidelines: While clinical thresholds may be adjusted depending on the specific demographic or clinical population, empirical investigations (e.g., Manicavasagar et al., 2003) suggest that a total score of 22 or higher reliably identifies adults with clinically significant separation anxiety warranting formal comprehensive clinical evaluation. A threshold of ≥ 26 indicates severe symptomatology typical of psychiatric outpatient samples.

11. Permissions & Fee and Test Year

The Adult Separation Anxiety Questionnaire (ASA-27) was originally developed in the late 1990s and published formally in its standardized 27-item format in 2003 by Vijaya Manicavasagar, Derrick Silove, Rosalind Wagner, and Julia Drobny in the journal Comprehensive Psychiatry. The authors and copyright holders grant permission for the instrument to be used free of charge for non-commercial, academic, scientific research, and clinical diagnostic purposes, provided appropriate scholarly attribution and standard academic citations are cited in resulting publications. Any commercial exploitation, digital redistribution behind a paywall, or inclusion in proprietary commercial software suites requires explicit prior written consent from the copyright holders and publishers.

12. References

• Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Basic Books.
• Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Basic Books.
• Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. Basic Books.
• Manicavasagar, V., Silove, D., & Curtis, J. (1997). Separation anxiety in adulthood: A phenomenological investigation. Comprehensive Psychiatry, 38(5), 274–282. https://doi.org/10.1016/S0010-440X(97)90060-6
• Manicavasagar, V., Silove, D., Wagner, R., & Drobny, J. (2003). A self-report questionnaire for measuring separation anxiety in adulthood. Comprehensive Psychiatry, 44(2), 146–153. https://doi.org/10.1053/comp.2003.50024
• Milrod, B., Markowitz, J. C., Shear, M. K., & Skodol, A. E. (2014). Adult separation anxiety: An unrecognized disorder? Harvard Review of Psychiatry, 22(4), 226–236. https://doi.org/10.1097/HRP.0000000000000010
• Silove, D., Marnane, C., Wagner, R., & Manicavasagar, V. (2010). The epidemiology and clinical correlates of adult separation anxiety disorder. Psychiatric Clinics of North America, 33(3), 573–587. https://doi.org/10.1016/j.psc.2010.04.011
• Silove, D., Momartin, S., Marnane, C., Abdalla-Steinkellner, L., & Manicavasagar, V. (2007). Adult separation anxiety in clinical and nonclinical populations: Validating the Adult Separation Anxiety Questionnaire (ASA-27). The Journal of Nervous and Mental Disease, 195(6), 534–537. https://doi.org/10.1097/NMD.0b013e318064a7dc

13. Items of the Scale (Questionnaire)