Padua Inventory – Washington State University Revision (PI-WSUR)

Abstract

The Padua Inventory – Washington State University Revision (PI-WSUR; Burns et al., 1996) is an extensively validated self-report psychometric instrument designed to assess the severity, phenotypic phenomenology, and multidimensional topography of obsessive-compulsive disorder (OCD) symptoms. Developed to remedy psychometric shortcomings present in Ezio Sanavio’s original 60-item Padua Inventory (Sanavio, 1988)—most notably its substantial item overlap with generalized trait anxiety, depressive rumination, and non-specific worry—the PI-WSUR refines the inventory to 39 items distributed across five distinct, empirically derived subscales: (1) Contamination Obsessions and Washing Compulsions, (2) Dressing/Grooming Compulsions, (3) Checking Compulsions, (4) Obsessional Thoughts of Harm to Self/Others, and (5) Obsessional Impulses to Harm Self/Others.

Each item is evaluated on a 5-point Likert-type scale ranging from 0 (Not at all) to 4 (Very much), reflecting the degree of subjective disturbance generated by specific ego-dystonic mental intrusions or compulsive behavioral sequences. Extensive psychometric evaluations demonstrate that the PI-WSUR possesses exceptional internal consistency (subscale Cronbach’s alphas typically ranging from .77 to .89; total scale α ≥ .92) and robust test-retest reliability across clinical and non-clinical populations. Crucially, confirmatory factor analysis (CFA) validates that the removal of 21 confound-laden items effectively decouples genuine obsessions and compulsions from generalized worry, establishing superior discriminant validity against measures such as the Penn State Worry Questionnaire (PSWQ) and the Beck Depression Inventory (BDI). Today, the PI-WSUR is recognized internationally as a gold-standard instrument for dimensional OCD research, experimental psychopathology, and clinical outcome monitoring.

Keywords

Padua Inventory, PI-WSUR, Obsessive-Compulsive Disorder, Psychometrics, Factor Analysis, Obsessional Intrusions, Compulsions, Discriminant Validity, Assessment, Exposure and Response Prevention

Authors

The revision of the Padua Inventory was spearheaded by clinical psychometricians at the Department of Psychology, Washington State University (Pullman, Washington, USA):

• G. Leonard Burns, Ph.D. — Professor Emeritus of Psychology, Department of Psychology, Washington State University, Pullman, WA. Dr. Burns is an internationally recognized expert in clinical child and adult psychometrics, measurement invariance, and the structural taxonomy of psychopathology.
• Sherry L. Keortge, M.S. — Clinical Research Associate, Department of Psychology, Washington State University.
• Gregory M. Formea, Ph.D. — Clinical Psychologist and Psychometrician, Department of Psychology, Washington State University.
• L. G. Sternberger, Ph.D. — Clinical Psychology Faculty and Researcher, collaborating within the psychopathology measurement laboratory at Washington State University.

The revision built directly upon the seminal foundational work of Ezio Sanavio, Ph.D. (University of Padua, Italy), who constructed the original 60-item Padua Inventory in 1988 to bridge European and Anglo-American operationalizations of obsessive-compulsive phenomena.

Purpose

The primary clinical and empirical purpose of the Padua Inventory – Washington State University Revision (PI-WSUR) is to provide a pure, dimensionally precise, and psychometrically uncontaminated self-report measure of obsessive-compulsive symptomatology. While the original 60-item Padua Inventory (Sanavio, 1988) represented a significant advance over earlier tools like the Maudsley Obsessional-Compulsive Inventory (MOCI; Hodgson & Rachman, 1977), empirical research rapidly uncovered a critical flaw in its construct validity: substantial criterion contamination by general affective distress, depressive rumination, and generalized anxiety worry.

Investigations conducted in the early 1990s revealed that several items on Sanavio’s original instrument—such as general concerns over making wrong decisions, persistent self-doubt regarding minor life details, and worries about hypothetical future misfortunes—correlated just as strongly with the Penn State Worry Questionnaire (PSWQ; Meyer et al., 1990) as they did with structured interviews for OCD such as the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). Consequently, clinicians and researchers frequently misattributed general anxiety disorder (GAD) symptomatology to obsessive-compulsive neurosis, resulting in inflated false-positive rates and blurred diagnostic boundaries.

Burns, Keortge, Formea, and Sternberger (1996) undertook a rigorous psychometric overhaul designed to:

• Eliminate items that shared higher empirical variance with generalized worry, neurotic perfectionism, or dysphoria than with core obsessive-compulsive phenotypes.
• Disentangle covert mental rituals and obsessional impulses from generalized anxious rumination.
• Delineate distinct symptom dimensions (e.g., contamination, checking, grooming, aggressive intrusions) to facilitate phenotypic subgrouping in clinical genetics, neuroimaging, and cognitive-behavioral trials.
• Establish a reliable metric for tracking symptom reduction during pharmacological regimens (e.g., SSRIs) and evidence-based psychotherapy, specifically Exposure and Response Prevention (ERP).

Psychological Construct

The PI-WSUR conceptualizes Obsessive-Compulsive Disorder not as a monolithic, unitary entity, but as a multi-dimensional spectrum comprising distinct yet inter-correlated behavioral, emotional, and cognitive phenomena. The 39 items of the scale are structured into five discrete psychometric dimensions:

1. Contamination Obsessions and Washing Compulsions (Items 1–10)

This subscale captures obsessional fears regarding contact with viral or bacterial pathogens, bodily secretions (e.g., saliva, perspiration, urine), public contaminants, or substances touched by unfamiliar individuals, coupled with repetitive, ritualized cleansing acts. Sample manifestations include washing hands substantially longer and more frequently than hygiene dictates (Item 7) or feeling compelled to cleanse oneself immediately after incidental contact with perceived environmental vectors (Items 8, 9, 10). The construct reflects both autonomic revulsion (the disgust-avoidance axis) and catastrophic cognitive threat appraisals regarding disease transmission.

2. Dressing/Grooming Compulsions (Items 11–13)

Comprising a focused 3-item triad, this factor isolates rigid, stereotyped behavioral sequences executed during self-care, clothing management, and bedtime routines. Individuals scoring high on this dimension experience an intense subjective need for symmetry, motoric precision, and “just-right” perceptual sensations (Item 11: “I feel obliged to follow a particular order in dressing, undressing and washing myself”; Item 13: “Before going to bed I have to hang up or fold my clothes in a special way”). If interrupted, the behavioral routine must typically be restarted from the beginning to resolve profound psychic unease.

3. Checking Compulsions (Items 14–23)

This dimension operationalizes perseverative verification behaviors driven by pathologically elevated inflated responsibility and pervasive cognitive self-distrust. Manifestations include repetitive checking of security measures such as locks, windows, and drawers (Item 17), physical utilities such as gas taps and electrical switches (Item 16), and administrative details such as financial documents and outgoing letters (Items 18, 20, 21). Crucially, the subscale also taps meta-cognitive memory distrust (Item 22: “Sometimes I am not sure I have done things which in fact I know I have done”) and perseverative reading rituals stemming from the fear of informational omission (Item 23).

4. Obsessional Thoughts of Harm to Self/Others (Items 24–30)

This factor measures intrusive, non-delusional, ego-dystonic mental representations concerning catastrophic disasters, accidental injuries, or covert culpability. In contrast to normal worry—which concerns realistic, real-life difficulties—these obsessions involve irrational, horrific outcomes for which the individual irrationally assumes moral liability. Examples include persistent, agonizing worry about having inadvertently harmed someone without knowing it (Item 25), magical-thinking guilt attributing distant catastrophes to oneself (Item 26), and phobic distress upon visual exposure to sharp or pointed objects (Item 28).

5. Obsessional Impulses to Harm Self/Others (Items 31–39)

This dimension isolates ego-dystonic urges, visceral sensations, or terrifying mental impulses to commit catastrophic, violent, taboo, or antisocial acts. Importantly, individuals diagnosed with OCD almost never act upon these impulses; rather, the impulse itself triggers extreme horror, hypervigilance, and moral distress. Representative items include the sudden impulse to leap from high elevations or under incoming trains (Items 31, 32), sudden urges to steer a vehicle into obstacles or bystanders (Item 34), impulses to disrobe inappropriately in public spaces (Item 33), and taboo urges to inflict harm upon vulnerable populations such as children or animals (Item 39).

Theoretical Framework

The PI-WSUR is anchored theoretically in modern Cognitive-Behavioral Models of OCD, synthesized with experimental psychopathology paradigms advanced by S. Rachman (1997, 1998), Paul Salkovskis (1985, 1989), and David A. Clark (2004). This framework rests upon several foundational tenets:

The Ubiquity of Intrusive Thoughts

A seminal premise of cognitive theory is that intrusive, taboo, and bizarre thoughts, images, and impulses occur naturally and universally across the non-clinical population (Rachman & de Silva, 1978; Purdon & Clark, 1993). Intrusive thoughts become clinical obsessions not because of their mere occurrence, but because of the idiosyncratic, dysfunctional cognitive appraisals the individual assigns to them.

Dysfunctional Belief Domains and Inflated Responsibility

According to Salkovskis’ cognitive formulation (1985), individuals vulnerable to OCD evaluate benign intrusive thoughts as signs of impending personal catastrophe or unpardonable moral culpability. Central to this process are specific maladaptive meta-cognitive beliefs, as cataloged by the Obsessive Compulsive Cognitions Working Group (OCCWG, 1997):

• Inflated Responsibility: The belief that one holds total power to cause or prevent catastrophic harm to oneself or others.
• Thought-Action Fusion (TAF): The cognitive bias wherein having a taboo thought is morally equivalent to performing the action (Moral TAF), or increases the physical likelihood that the event will manifest in reality (Likelihood TAF).
• Overestimation of Threat: A systemic cognitive bias that magnifies the probability and severity of dangerous outcomes.
• Intolerance of Uncertainty: The catastrophic appraisal of ambiguous or indeterminate scenarios.

The Neutralization Cycle and Negative Reinforcement

Building upon Mowrer’s two-factor theory, compulsions (e.g., washing, checking, mental counterspeech, symmetry routines) are conceptualized as active avoidance behaviors. Because completing a compulsive ritual temporarily terminates acute subjective anxiety and distress, the behavior is maintained through potent negative reinforcement. However, this neutralization prevents the disconfirmation of catastrophic beliefs, perpetuating the vicious cycle of chronic psychopathology.

By segregating pure intrusive content (Subscales 4 and 5) from overt neutralizing behaviors (Subscales 1, 2, and 3), the PI-WSUR operationalizes this exact theoretical interaction between obsessional threat activation and compulsive safety behaviors.

Validity

The psychometric validity of the PI-WSUR has been demonstrated through extensive multi-method testing across non-clinical, student, and clinical psychiatric samples globally:

Construct and Structural Validity

The factorial construct validity of the PI-WSUR was originally established by Burns et al. (1996) utilizing exploratory factor analysis with oblique rotation, demonstrating that a 5-factor solution accounting for approximately 48.6% of total variance provided the most parsimonious and clinically meaningful representation of OCD symptoms. Subsequent independent confirmatory factor analytic studies (e.g., Burns et al., 1996; Swinbourne, 2008; Anisi et al., 2011) have consistently replicated this 5-factor topology across cross-cultural demographics, yielding superior fit indices compared to single-factor or hierarchical general-neurosis structures.

Convergent Validity

The PI-WSUR demonstrates strong, statistically significant convergent validity when evaluated against established collateral OCD assessment tools:

• Correlations with the Maudsley Obsessional-Compulsive Inventory (MOCI) range from r = .60 to .74 for corresponding subscales (e.g., Checking with MOCI Checking; Washing with MOCI Cleaning).
• Correlations with the Yale-Brown Obsessive Compulsive Scale (Y-BOCS) clinician-administered total severity score range from r = .52 to .68 in primary OCD clinical samples.
• Correlations with the Obsessive-Compulsive Inventory-Revised (OCI-R) total and dimension-specific scales regularly exceed r = .70.

Discriminant Validity

The primary psychometric justification for creating the PI-WSUR was the purification of discriminant boundaries. Burns et al. (1996) demonstrated that while Sanavio’s original 60-item scale correlated substantially with the Penn State Worry Questionnaire (PSWQ) (r ≥ .55), the PI-WSUR subscales demonstrated dramatically attenuated associations with the PSWQ (correlations dropping into the .18 to .34 range). Similarly, correlations between the PI-WSUR checking and contamination subscales and measures of depression (such as the Beck Depression Inventory; BDI-II) remain low-to-moderate (r ≈ .20–.35), verifying that the instrument captures core OCD pathology rather than pervasive negative affectivity or demoralization.

Criterion and Predictive Validity

Receiver operating characteristic (ROC) curves confirm that the PI-WSUR accurately discriminates patients formally diagnosed with OCD (DSM-IV/DSM-5 criteria) from patients diagnosed with other anxiety disorders (e.g., Generalized Anxiety Disorder, Social Anxiety Disorder, Panic Disorder) and healthy control participants. Furthermore, longitudinal outcome studies show that PI-WSUR subscale scores drop substantially and selectively following successful course completion of cognitive-behavioral therapy with Exposure and Response Prevention (ERP).

Reliability

The PI-WSUR exhibits exceptional reliability across various psychometric modalities:

Internal Consistency

Across the development and validation samples documented by Burns et al. (1996), Cronbach’s alpha coefficients for the individual subscales consistently exceed established psychometric thresholds for clinical measurement:

• Contamination Obsessions & Washing Compulsions: α = .88 to .90
• Dressing/Grooming Compulsions: α = .77 to .82 (notably high given this scale consists of only 3 items)
• Checking Compulsions: α = .87 to .89
• Obsessional Thoughts of Harm: α = .81 to .85
• Obsessional Impulses of Harm: α = .82 to .86
• Total Scale (39 items): α = .92 to .95

Test-Retest Stability

Temporal stability over time has been verified across diverse intervals. Burns et al. (1996) reported 4-week test-retest reliability coefficients ranging from r = .76 to .86 across individual subscales, with total score stability measured at r = .87. Longer intervals spanning 3 to 6 months in stable non-treatment clinical samples have documented retest correlations exceeding .72, indicating that the scale accurately taps durable behavioral traits and chronic symptoms rather than transient situational fluctuations.

Factor Analysis

The factorial composition of the PI-WSUR is one of its most heavily scrutinized and empirically validated attributes. In the initial derivation study, Burns et al. (1996) subjected the 60 items of the original Padua Inventory to exploratory factor analysis (EFA) using principal axis factoring with promax (oblique) rotation on a sample of 784 undergraduate participants, with parallel validation in a replication cohort (N = 786).

Identification of Confounds and Item Deletion

A rigorous empirical elimination criterion was applied. Items were excised if they:

1. Failed to achieve a primary factor loading ≥ .40 on a definitive clinical OCD factor.
2. Demonstrated cross-loadings ≥ .30 across multiple discordant factors.
3. Loaded primarily onto an amorphous “generalized worry/distress” factor that correlated preferentially with the Penn State Worry Questionnaire.

Through this systematic methodology, 21 items were eliminated. The remaining 39 items coalesced cleanly into the 5-factor structure described below:

Confirmatory Factor Analytic (CFA) Fit Indices

Subsequent Confirmatory Factor Analyses (CFA) testing the 5-factor model across non-clinical, student, and diagnosed OCD cohorts have substantiated excellent goodness-of-fit parameters:

• Comparative Fit Index (CFI): .92 – .96
• Tucker-Lewis Index (TLI): .91 – .95
• Root Mean Square Error of Approximation (RMSEA): .042 – .056 (with 90% confidence intervals bounded below .06)
• Standardized Root Mean Square Residual (SRMR): .045 – .052

Alternative models—such as a single undifferentiated higher-order factor or a simple two-factor model (obsessions vs. compulsions)—exhibit significantly degraded model fit (Δχ² p < .001), underscoring that the phenotypic structure of OCD is fundamentally multidimensional.

Instrument / Measurement Tool

The PI-WSUR is structured as a self-administered pencil-and-paper or computerized questionnaire. Key administrative parameters include:

• Test Type: Multidimensional self-report rating scale.
• Target Population: Adolescents and adults (ages 16 and older). Validated for both clinical psychiatric cohorts and subclinical research populations.
• Item Count: 39 discrete statements.
• Estimated Completion Time: 10 to 15 minutes.
• Response Format: 5-point Likert-type scale reflecting the degree of subjective disturbance created by the thought or behavior:
  • 0 = Not at all
  • 1 = A little
  • 2 = Quite a lot
  • 3 = A lot
  • 4 = Very much
• Scoring and Metrics:
  • Contamination / Washing Subscale: Sum of items 1–10 (Score range: 0–40).
  • Dressing / Grooming Subscale: Sum of items 11–13 (Score range: 0–12).
  • Checking Compulsions Subscale: Sum of items 14–23 (Score range: 0–40).
  • Thoughts of Harm Subscale: Sum of items 24–30 (Score range: 0–28).
  • Impulses of Harm Subscale: Sum of items 31–39 (Score range: 0–36).
  • Total PI-WSUR Score: Sum across all 39 items (Score range: 0–156).
• Clinical Interpretation Guidelines:
  • Non-clinical mean total scores typically cluster between 15.0 and 28.0 (SD ≈ 15–18).
  • Clinical OCD mean total scores typically range between 55.0 and 78.0 (SD ≈ 22–26).
  • Elevations on specific subscales can reveal clinical phenotypes (e.g., checking-predominant, contamination-predominant, or pure mental/impulsive intrusions), informing tailored cognitive-behavioral behavioral hierarchies and exposure tasks.

Permissions & Fee and Test Year

The Padua Inventory – Washington State University Revision was formally published in 1996 in the journal Behaviour Research and Therapy following initial distribution via academic monograph in 1995 by Dr. G. Leonard Burns at Washington State University (Pullman, WA).

Licensing and Accessibility: The PI-WSUR is placed in the public domain for non-commercial research, academic, and clinical healthcare applications. No licensing royalty or administrative fee is required to administer, score, or duplicate the scale for clinical practice or scientific investigation, provided appropriate academic attribution is rendered to the authors (Burns et al., 1996) and the original developer (Sanavio, 1988). Commercial reproduction, software bundling, or incorporation into proprietary electronic medical record systems requires explicit prior permission from the primary author or copyright stakeholders.

References

• Anisi, J., Ghofrani, F., & Rahnejat, A. M. (2011). Psychometric properties of the Padua Inventory-Washington State University Revision (PI-WSUR) in Iranian military personnel. Journal of Military Medicine, 13(2), 79-84. http://www.sid.ir/fa/VEWSSID/J_pdf/68113894501.pdf
• Burns, G. L. (1995). Padua Inventory-Washington State University Revision. Pullman, WA: Author.
• Burns, G. L., Keortge, S., Formea, G. M., & Sternberger, L. G. (1996). Revision of the Padua Inventory of obsessive compulsive disorder symptoms: Distinctions between worry, obsessions, and compulsions. Behaviour Research and Therapy, 34(2), 163-173. https://doi.org/10.1016/0005-7967(95)00035-6
• Clark, D. A. (2004). Cognitive-behavioral therapy for OCD. Guilford Press.
• Hodgson, R. J., & Rachman, S. (1977). Obsessional-compulsive complaints. Behaviour Research and Therapy, 15(5), 389-395. https://doi.org/10.1016/0005-7967(77)90042-0
• Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 28(6), 487-495. https://doi.org/10.1016/0005-7967(90)90135-6
• Obsessive Compulsive Cognitions Working Group (1997). Cognitive assessment of obsessive-compulsive disorder. Behaviour Research and Therapy, 35(7), 667-681. https://doi.org/10.1016/S0005-7967(97)00017-X
• Purdon, C., & Clark, D. A. (1993). Obsessional intrusive thoughts in nonclinical subjects. Part I. Content and relation with depressive, anxious and obsessional symptoms. Behaviour Research and Therapy, 31(8), 713-720. https://doi.org/10.1016/0005-7967(93)90001-B
• Rachman, S. (1997). A cognitive theory of obsessions. Behaviour Research and Therapy, 35(9), 793-802. https://doi.org/10.1016/S0005-7967(97)00040-5
• Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour Research and Therapy, 16(4), 233-248. https://doi.org/10.1016/0005-7967(78)90022-0
• Salkovskis, P. M. (1985). Obsessional-compulsive problems: A cognitive-behavioural analysis. Behaviour Research and Therapy, 23(5), 571-583. https://doi.org/10.1016/0005-7967(85)90105-6
• Sanavio, E. (1988). Obsessions and compulsions: The Padua Inventory. Behaviour Research and Therapy, 26(2), 169-177. https://doi.org/10.1016/0005-7967(88)90116-7
• Swinbourne, J. (2008). The co-morbidity of eating disorders and anxiety disorders (Doctoral dissertation, University of Sydney). http://ses.library.usyd.edu.au/bitstream/2123/4026/1/j-swinbourne-thesis.pdf

Items of the Scale