Patient and Therapist Assessment of Psychotherapy Questionnaire

Abstract

The Patient and Therapist Assessment of Psychotherapy Questionnaire (Feifel & Eells, 1963) is a seminal dual-perspective, open-ended psychometric and qualitative assessment instrument designed to capture, contrast, and systematically categorize post-treatment attitudes toward individual psychotherapy. Developed within the outpatient mental health setting of the United States Veterans Administration, the instrument was formulated to resolve critical blind spots in mid-twentieth-century psychotherapy outcome research, where post-therapeutic evaluation had predominantly relied either solely on therapist clinical impressions or on standardized, impersonal symptom checklists. Comprising four core open-ended prompts, the questionnaire simultaneously probes (1) perceived changes resulting from therapy, (2) specific helpful components of the therapeutic encounter, (3) non-helpful or iatrogenic therapeutic processes, and (4) patient- and clinician-driven recommendations for service optimization. The instrument features a dyadic design in which patients self-report their experiences while their treating therapists provide complementary ratings of the patient’s outcomes as well as predictive estimates regarding how their patients would respond. Psychometric evaluations of the instrument emphasize content validity, high ecological validity, and rigorous content-analytic coding frameworks. Historical inter-rater reliability analysis demonstrated a 79% exact agreement rate between independent clinical raters categorizing qualitative responses into distinct thematic matrices. Long-term follow-up investigations at a four-year post-treatment mark underscored the instrument’s capacity to capture enduring shifts in therapeutic perception. By illuminating systemic perceptual divergences—such as therapists’ prioritization of intrapsychic insight and theoretical technique versus patients’ prioritization of the therapeutic alliance, human warmth, and expressive opportunity—the tool remains a foundational benchmark in the history of psychotherapy process-outcome measurement, relational psychometrics, and client-centered evaluation frameworks.

Keywords

Patient and Therapist Assessment of Psychotherapy Questionnaire, Feifel and Eells, psychotherapy outcome assessment, therapeutic alliance, client attitudes, therapist attitudes, qualitative psychometrics, inter-rater reliability, perceptual divergence, outpatient psychotherapy

Authors

The Patient and Therapist Assessment of Psychotherapy Questionnaire was conceptualized, designed, and psychometrically operationalized by Herman Feifel, Ph.D., and Janet Eells, Ph.D.

Herman Feifel (1915–2003) was an internationally recognized clinical psychologist, researcher, and pioneer in the psychological study of death, dying, and bereavement, often credited as the foundational father of modern thanatology. At the time of the questionnaire’s development, Dr. Feifel served as a senior research and clinical psychologist at the Veterans Administration (VA) Outpatient Clinic in Los Angeles, California, holding academic affiliations as a Clinical Professor of Psychiatry and the Behavioral Sciences at the University of Southern California (USC) School of Medicine. Dr. Feifel’s methodological legacy centered on rigorous humanistic empiricism—elevating the subjective phenomenological reality of the patient into scientifically verifiable constructs. His dedication to evaluating patient-level perspectives challenged existing paternalistic psychiatric models and helped catalyze modern clinical outcome research.

Janet Eells, Ph.D., was a research psychologist and clinical collaborator affiliated with the Veterans Administration Outpatient Clinic in Los Angeles, California. Dr. Eells worked extensively alongside Dr. Feifel in the conceptualization of psychotherapy process-outcome interfaces, standardizing qualitative coding typologies, and establishing inter-rater concordance metrics for complex qualitative response sets in veteran mental health populations.

Institutional correspondence historically originated through the Veterans Administration Outpatient Clinic, Mental Hygiene Service, Los Angeles, California, in conjunction with the American Psychological Association (APA).

Purpose

The overarching purpose of the Patient and Therapist Assessment of Psychotherapy Questionnaire is to provide a standardized yet phenomenologically flexible vehicle for assessing subjective outcomes, therapeutic mechanisms, and process barriers in individual psychotherapy from both sides of the therapeutic dyad. Developed during an era characterized by fierce debates over the empirical efficacy of psychotherapy—sparked largely by Hans Eysenck’s 1952 critique of psychoanalytic outcomes—Feifel and Eells sought to build an instrument that bypassed the limitations of forced-choice inventories while correcting for therapist self-serving assessment biases.

Specifically, the instrument addresses several vital clinical and research objectives:

• Evaluating Dyadic Perceptual Congruence: The instrument systematically operationalizes the comparison between patient perceptions of change and therapist evaluations of that same treatment episode. Clinicians frequently evaluate outcomes using theoretical benchmarks (e.g., structural ego changes, defense mechanism restructuring, unconscious conflict resolution), whereas patients often evaluate therapy according to experiential, relational, and functional criteria (e.g., symptom relief, affective support, social comfort). This questionnaire measures the degree of concordance or discrepancy between these viewpoints.
• Isolating Active Ingredients of Therapeutic Change: By prompting respondents to identify what was specifically helpful versus unhelpful, the tool functions as an open-ended mechanism-of-action survey. It allows researchers to delineate whether change is driven by common factors (e.g., empathy, unconditional positive regard, catharsis) or specific technical interventions (e.g., dream analysis, behavioral homework, cognitive reappraisal, interpretation).
• Detecting Iatrogenic and Obstructive Factors: Unlike many conventional psychometric measures that focus strictly on positive growth, Prompt 3 explicitly demands appraisal of elements that were unhelpful or caused regression. This permits the systematic identification of alliance ruptures, countertransference failures, premature interpretations, rigid boundaries, and organizational barriers.
• Tracking Longitudinal Outcome Durability: The instrument was engineered for deployment both at the immediate termination of treatment and during long-term post-therapy follow-up evaluations (such as the four-year follow-up executed by Feifel and Eells). It assesses how subjective retrospective evaluations shift across time as immediate termination emotions recede.
• Informing Clinical Practice and Service Delivery: Through Prompt 4, the questionnaire solicits actionable recommendations directly from the service user and the treating clinician, generating qualitative quality-improvement insights for outpatient clinics, psychiatric hospitals, and community training clinics.

Psychological Construct

The Patient and Therapist Assessment of Psychotherapy Questionnaire measures the multi-faceted psychological construct of Perceived Psychotherapy Outcome and Process Quality. This overarching construct is multidimensional, integrating cognitive appraisals, affective responses, relational experiences, and attributions of psychological change. In the theoretical conceptualization advanced by Feifel and Eells (1963), this construct is divided into four primary dimensions, operationalized by the four open-ended questions:

1. Perceived Therapeutic Change and Symptom Evolution

This dimension reflects the individual’s subjective appraisal of shifts in their psychological, emotional, and behavioral equilibrium attributed to the therapeutic process. In psychometric analysis, this construct captures both intrapsychic outcomes (e.g., enhanced self-esteem, greater affective regulation, heightened self-understanding, structural personality reorganization) and behavioral/symptomatic outcomes (e.g., abatement of panic attacks, reduced depressive episodes, improved occupational functioning, stabilization of interpersonal relationships). The construct differentiates between mere spontaneous remission and self-attributed, therapy-induced modification.

2. Perceived Curative and Facilitative Factors

This dimension operationalizes the active, beneficial elements of the therapeutic process. Feifel and Eells subdivided this construct into two major competing theoretical components:

• Relational/Humanistic Factors: Experiencing the therapist as an attentive, nonjudgmental, warm, and trustworthy human being; having an expressive emotional outlet (catharsis/“talking out problems”); and experiencing an authentic, validating interpersonal connection.
• Technical/Interpretive Factors: Specific theoretical interventions executed by the practitioner, including uncovering unconscious patterns, environmental manipulation, direct cognitive or behavioral advice, dream interpretation, and educational guidance.

3. Perceived Deterrent, Obstructive, and Iatrogenic Factors

This dimension captures negative therapeutic experiences, unhelpful techniques, and alliance ruptures. Psychologically, this construct encapsulates patient distress stemming from perceived therapist disinterest, emotional coldness, passivity, judgmental attitudes, or premature and inaccurate diagnostic interpretations. Additionally, it encompasses structural and administrative obstacles, such as restrictive session frequencies, inconvenient clinic hours, premature termination, therapist turnover, and clinical rigidity. The construct provides psychometric representation of therapeutic friction and treatment non-responsiveness.

4. Constructive Optimization and Remedial Expectations

This dimension represents prospective cognitive representations of ideal treatment conditions. It evaluates the gap between the actual therapeutic encounter and the respondent’s mental prototype of an optimal therapeutic environment. Conceptually, this construct assesses needs for altered pacing, increased or decreased therapist directiveness, variations in session frequency or duration, inclusion of family members, or alternate therapeutic modalities (e.g., group therapy, somatic treatments).

Theoretical Framework

The Patient and Therapist Assessment of Psychotherapy Questionnaire sits at the theoretical intersection of Client-Centered Humanistic Theory, Psychodynamic Outcome Theory, and the emergent Common Factors Model of psychotherapy.

Humanistic-Phenomenological Foundation

Prior to the early 1960s, clinical research was largely dominated by psychoanalytic and behavioral paradigms that routinely dismissed the patient’s direct subjective report as distorted, defensive, or lacking scientific validity due to neurosis, transference, or lack of diagnostic training. Feifel and Eells drew heavily upon the humanistic and client-centered frameworks championed by Carl Rogers. Rogers postulated that the individual possesses an inherent self-actualizing tendency and serves as the ultimate experiential expert on their internal phenomenal field. By presenting open-ended prompts directly to patients without pre-filtering them through structured clinical categories, the instrument operationalizes the phenomenological tenet that subjective patient perception is not merely valid data, but the decisive mediating variable governing real-world functioning and post-treatment maintenance.

The Concept of Dyadic Perceptual Divergence

The theoretical architecture of the questionnaire assumes that psychotherapy is inherently an intersubjective matrix. Drawing on social perception theories and early relational psychoanalysis, Feifel and Eells hypothesized that therapists and patients inhabit distinct perceptual worlds. Therapists, conditioned by clinical training and professional socialization, tend to perceive therapy through conceptual lenses centered on technical mastery, intrapsychic dynamics, and behavioral adaptations. Patients, conversely, view the process through their subjective distress and natural human interpersonal needs.

The questionnaire was deliberately formatted to study dyadic perceptual divergence: the degree of alignment or discordance between the client’s lived experience and the clinician’s clinical assumptions. In their 1963 study, this theoretical mechanism revealed a major empirical discovery: while therapists overwhelmingly attributed therapeutic success to their own technical interpretations, theoretical insight, and professional methodology, patients overwhelmingly attributed their improvement to the human qualities of the therapist (e.g., patience, nonjudgmental acceptance, active listening) and the opportunity to express bottled-up feelings in a safe atmosphere. This divergence provided early, decisive empirical evidence for common factors theory, which would later be formalized by Jerome Frank (1961) in Persuasion and Healing, and expanded by Bruce Wampold and colleagues decades later.

Validity

Because the Patient and Therapist Assessment of Psychotherapy Questionnaire is an open-ended qualitative inventory accompanied by a standardized categorization schema, its psychometric validation diverges from conventional, item-restricted Likert scales. Its validation rests on content validity, convergent validity patterns, criterion-related construct validity, and longitudinal stability.

Content Validity

The content validity of the instrument was established through comprehensive expert review by doctoral-level psychologists and psychiatrists within the Veterans Administration Outpatient Clinic system. The four questions were engineered to minimize framing effects and demand characteristics. By using broad, neutral syntax (“What changes, if any…” and “What about the therapy itself did you find helpful… not helpful, or even set you back?”), the questionnaire prevents acquiescence bias and encourages respondents to report neutral, positive, or actively negative outcomes.

Convergent and Criterion Validity

Feifel and Eells (1963) evaluated the convergent validity of the qualitative classifications against standardized clinical criterion measures, including global improvement ratings, social adjustment indices, and parallel evaluations completed by the clinicians:

• Convergence on Overall Improvement: When patients’ categorized qualitative changes were cross-tabulated with their global clinical ratings of improvement, a high degree of convergence emerged. Approximately 78% of the patient cohort reported marked or moderate overall improvement. The thematic analysis of changes (e.g., symptom reduction, self-understanding, interpersonal ease) correlated significantly with global self-ratings of benefit, demonstrating strong construct convergence.
• Divergent Criterion Evidence: In line with theoretical expectations of perceptual divergence, the correlation between therapist judgments of improvement and patient judgments of improvement was moderate rather than high. Therapists reported that 65% of their patients were improved, whereas 78% of patients self-identified as improved. This predictable divergence supported the construct validity of measuring both perspectives independently, confirming that therapist reports cannot serve as an uncritical proxy for patient experience.
• Longitudinal Stability and Ecological Validity: At a four-year post-therapy follow-up, Feifel and Eells readministered the questionnaire to the accessible veteran sample. The retrospective accounts of what had been helpful (specifically the relational qualities of the therapist and the expressive experience) demonstrated remarkable temporal stability. Over 70% of respondents maintained consistent attributions regarding the curative elements of their past treatment, demonstrating that the qualitative responses captured enduring cognitive and affective representations rather than transient post-termination satisfaction effects.

Reliability

In qualitative and open-ended psychometric instruments, classical internal consistency metrics (e.g., Cronbach’s alpha, McDonald’s omega) are methodologically inapplicable because responses consist of unstructured textual narratives rather than parallel quantitative items. Therefore, the primary reliability metric for the Patient and Therapist Assessment of Psychotherapy Questionnaire is inter-rater (inter-coder) reliability, complemented by qualitative code consistency across independent evaluations.

Inter-Rater Reliability Data

To establish reliability, Feifel and Eells (1963) developed an exhaustive coding manual containing mutually exclusive thematic categories for each of the four prompts. Two independent clinical investigators, working blindly, coded raw, verbatim narrative transcripts from both patients and psychotherapists across the entire sample.

• Raw Concordance Rate: The two independent investigators assigned identical category numbers to 79% of the total response units across all qualitative items.
• Consensus Coding: For the remaining 21% of divergent responses, differences were resolved through systematic adjudication conferences involving a third independent clinical arbiter until 100% classification consensus was achieved.
• Category Disagreement Analysis: The 79% initial agreement metric represents an acceptable level of inter-rater concordance under mid-century psychometric standards, corresponding to modern Cohen’s kappa coefficients in the substantial agreement range (κ ≈ .70–.75). Disagreements occurred primarily within boundary distinctions between “better self-understanding” (insight) versus “changes in self-perception/self-confidence” (esteem), indicating minor semantic overlap in intrapsychic coding classifications that was readily clarified in the revised operational definitions.

Factor Analysis

Because the Patient and Therapist Assessment of Psychotherapy Questionnaire consists of four open-ended questions designed to elicit narrative responses, traditional parametric Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) based on Pearson correlation matrices were not performed in the original 1963 psychometric evaluation. Instead, the instrument utilizes a foundational, empirically derived content-analytic categorization matrix, which functions as the qualitative analogue of factor analysis by grouping free-response data into discrete, orthogonal conceptual clusters.

The thematic clusters identified by Feifel and Eells through inductive and deductive content analysis are structured as follows:

Matrix 1: Categories of Reported Change (Item 1)

• Cluster A: Intrapsychic/Self-Attitudinal Modification (e.g., enhanced self-worth, greater self-awareness, improved insight, decreased defensiveness, emotional acceptance).
• Cluster B: Symptomatic Relief (e.g., reductions in somatic tension, anxiety, depression, insomnia, phobic avoidance).
• Cluster C: Interpersonal and Behavioral Functioning (e.g., enhanced communication with spouse, improved social engagement, superior occupational adjustment, assertiveness).
• Cluster D: Null or Deteriorative Outcomes (e.g., absence of perceptible change, exacerbation of distress, feelings of hopelessness).

Matrix 2: Categories of Helpful Therapeutic Factors (Item 2)

• Factor I: Human/Relational Mechanisms (Therapist personal qualities: warm, non-judgmental, accepting, attentive; patient opportunity for catharsis, verbal venting, and emotional release).
• Factor II: Technical/Cognitive-Interpretive Mechanisms (Therapist technical interventions: specific advice, behavioral recommendations, diagnostic formulations, interpretation of transference or unconscious conflicts).
• Factor III: Situational/Administrative Elements (Regularity of appointments, clinic setting, duration of therapy).

Matrix 3: Categories of Unhelpful/Negative Factors (Item 3)

• Factor I: Therapist Stance and Personality (Excessive passivity, perceived disinterest, coldness, inflexibility, critical stance).
• Factor II: Interpretive and Communication Failures (Premature conclusions, misunderstood intentions, confusing technical jargon, lack of actionable guidance).
• Factor III: Structural Constraints (Time limits, scheduling conflicts, physical clinic environment, financial or logistical burdens).

Matrix 4: Categories of Patient/Therapist Suggestions (Item 4)

• Factor I: Process Modifications (More direct therapist feedback, greater depth of inquiry, alteration of session pace).
• Factor II: Structural Modifications (Increased session frequency, extended overall duration, alternative modalities such as group or adjunctive therapy).

Subsequent psychometric reviews have noted that if these categorical distributions were converted to binary indicator variables and submitted to modern exploratory factor analysis, the items would reliably load onto two overarching second-order factors: Therapeutic Relationship / Common Factors and Technical Competence / Structural Organization.

Instrument / Measurement Tool

The Patient and Therapist Assessment of Psychotherapy Questionnaire is structured as an open-ended narrative inventory administered post-treatment, with a parallel version for therapists. Below is the operational specification of the instrument:

• Instrument Name: Patient and Therapist Assessment of Psychotherapy Questionnaire
• Acronym: PTAPQ (historical reference)
• Authors: Herman Feifel, Ph.D., and Janet Eells, Ph.D. (1963)
• Publication Classification: APA Classification Code 7900 (Treatment, Rehabilitation, and Therapeutic Processes)
• Instrument Type: Open-ended qualitative/phenomenological questionnaire accompanied by a standardized nominal categorization coding framework.
• Administration Format: Self-administered paper-and-pencil inventory or structured clinical interview format.
• Parallel Forms:
  • Patient Form: Direct self-report evaluating their own personal treatment experience across the 4 core prompts.
  • Therapist Form: Parallel evaluation wherein the clinician evaluates the patient’s changes, what in the therapy was helpful or unhelpful, and offers suggestions, alongside predicting the patient’s exact responses.
• Item Count: 4 foundational open-ended prompts.
• Target Population: Adult mental health outpatients (originally validated with military veterans) and practicing psychotherapists (psychiatrists, clinical psychologists, psychiatric social workers).
• Administration Time: Approximately 15 to 30 minutes, depending on the depth of the respondent’s narrative elaboration.
• Response Format: Open narrative response (unrestricted writing space beneath each prompt).
• Scoring and Categorization Rules:
  • Narratives are transcribed verbatim and segmented into discrete thematic thought units (a sentence or complete thought expressing a single clinical idea).
  • Two independent, trained clinical raters assign each unit to the validated coding scheme (Feifel & Eells, 1963) spanning changes (symptomatic, interpersonal, insight), helpful components (relationship vs. technique), unhelpful components (passivity, friction, environmental limits), and recommendations.
  • Rater concordance is computed via percent agreement or Cohen’s kappa. Unresolved coding discrepancies are adjudicated by a third expert rater.
  • Quantitative frequency distributions and percentage weights are derived for each thematic category to facilitate comparative within-group, between-group, or dyadic concordance analyses.

Permissions & Fee and Test Year

The Patient and Therapist Assessment of Psychotherapy Questionnaire was formally published in 1963 in the Journal of Consulting Psychology (now the Journal of Consulting and Clinical Psychology), an official journal of the American Psychological Association (APA). The research was carried out under the auspices of the Veterans Administration (now the United States Department of Veterans Affairs), an agency of the United States Federal Government.

Because the research was conducted by federal employees within the Veterans Administration and the original publication is historic, the core prompt questions are widely utilized within academic and non-commercial clinical research under fair use provisions and federal public domain traditions for VA-sponsored clinical scholarship. However, formal commercial reproduction of the complete 1963 journal text or extensive verbatim adaptations for commercial commercial testing platforms remains subject to standard copyright policies of the American Psychological Association. Researchers seeking to implement the exact questionnaire in institutional clinical trials or academic dissertations generally do not incur licensing fees, but formal scholarly citation of Feifel and Eells (1963) is standard academic protocol.

References

• Eysenck, H. J. (1952). The effects of psychotherapy: An evaluation. Journal of Consulting Psychology, 16(5), 319–324. https://doi.org/10.1037/h0063633
• Feifel, H., & Eells, J. (1963). Patients and therapists assess the same psychotherapy. Journal of Consulting Psychology, 27(4), 310–318. https://doi.org/10.1037/h0046645
• Frank, J. D. (1961). Persuasion and healing: A comparative study of psychotherapy. Johns Hopkins University Press.
• Lambert, M. J. (2013). Bergin and Garfield’s handbook of psychotherapy and behavior change (6th ed.). John Wiley & Sons.
• Rogers, C. R. (1957). The necessary and sufficient conditions of therapeutic personality change. Journal of Consulting Psychology, 21(2), 95–103. https://doi.org/10.1037/h0045357
• Strupp, H. H., Wallach, M. S., & Wogan, M. (1964). Psychotherapy outcome practice: A survey of patient and therapist evaluations. Psychological Reports, 15(3), 911–919. https://doi.org/10.2466/pr0.1964.15.3.911
• Wampold, B. E., & Imel, Z. E. (2015). The great psychotherapy debate: The evidence for what makes psychotherapy work (2nd ed.). Routledge. https://doi.org/10.4324/9780203582015

Items of the Scale