Efficacy of Expectancy Focused Exposure Therapy for Anxiety Disorders

Anxiety disorders are highly prevalent worldwide and are a major source of disability and reduced quality of life. They involve excessive fear and anxiety, physiological activation, hypervigilance, and avoidance behaviors that interfere with social and occupational functioning. They include social anxiety disorder, specific phobia, generalized anxiety disorder, and panic disorder with or without agoraphobia. In Chile, anxiety disorders are a major public health concern, with estimated lifetime prevalence of approximately 16.2% in adults and 6-month prevalence of 7.9%. They are among the leading causes of years lived with disability, particularly among women, and recent reports suggest that approximately one in four Chileans presents anxiety symptoms. Chilean clinical guidelines recommend cognitive behavioral therapy (CBT) among the primary treatments.

CBT is a first-line treatment for anxiety disorders, and exposure therapy is its central component. Exposure involves systematic confrontation with feared stimuli or situations in the absence of the anticipated aversive outcome. Although effective, symptom reduction during treatment does not necessarily predict long-term maintenance, and return of fear remains common.

Experimental psychopathology has emphasized learning mechanisms underlying exposure. Extinction occurs when the feared stimulus is repeatedly encountered without the expected outcome. Rather than erasing the original fear association, extinction produces inhibitory learning that competes with the original fear memory. Because the original association remains intact, fear can reappear through renewal, spontaneous recovery, reinstatement, or rapid reacquisition.

Contemporary models propose that therapeutic change is better understood as inhibitory learning than habituation. A key mechanism is expectancy violation: the discrepancy between what individuals expect to occur during exposure and what actually occurs. Expectancy-focused approaches emphasize learning that feared outcomes are less likely, less intense, or less catastrophic than anticipated.

Randomized clinical evidence evaluating optimized exposure remains limited. Recent evidence suggests that inhibitory learning and expectancy violation may improve treatment response, but short follow-up periods may be insufficient to evaluate relapse or recurrence. This trial evaluates whether exposure focused on expectancy violation and inhibitory learning produces more durable improvements than exposure focused primarily on anxiety reduction, with outcomes assessed during treatment and at 6, 12, and 18 months.

Study Design This is a superiority randomized controlled clinical trial using an open-cohort stepped-wedge design with sequential recruitment. Participants will be enrolled in successive cohorts. At each step, newly enrolled participants will be randomly assigned to begin treatment immediately or complete a brief waitlist before treatment. Participants assigned to waitlist will begin treatment at the subsequent step.

When treatment begins, participants will be randomly assigned to Expectancy Focused Exposure (EFE) or Anxiety Focused Exposure (AFE). Each participant will receive only one intervention. Both conditions will be manualized and will follow identical assessment schedules. Randomization will be stratified by principal diagnosis and gender. Participants, outcome evaluators, and statistical analysts will remain blinded to treatment allocation. Therapists will not be informed of the study hypothesis.

Participants Participants will be adults aged 18 to 70 with a principal diagnosis of social anxiety disorder, generalized anxiety disorder, specific phobia, or panic disorder with or without agoraphobia, established through a structured diagnostic interview. Individuals taking psychotropic medication may participate if dosage has been stable for at least four weeks before enrollment. Exclusion criteria are current psychotic disorder or bipolar disorder, high suicide risk, severe active substance use disorder, concurrent disorder-specific psychotherapy, and medical conditions that contraindicate exposure.

Recruitment and Screening Participants will be recruited through social media, university channels, and referrals from university-affiliated mental health services. Interested individuals will complete screening to identify potential anxiety-related problems. Those meeting preliminary eligibility criteria will complete a structured diagnostic interview to confirm diagnosis and eligibility.

Therapist Recruitment and Training Therapists will be clinicians with formal CBT training and at least one year of postgraduate or post-degree CBT training. Selection and training will include evaluation of professional background and random assignment to one protocol through workshops. Training will include instruction, demonstrations, and supervised role-playing. Therapists who fail to meet performance or attendance criteria will be excluded. Therapists will be blinded to the alternative condition and will receive weekly supervision.

Therapist adherence will be monitored to ensure fidelity and differentiation between models. All sessions will be audio- or video-recorded, and approximately 30% of sessions per condition will be evaluated using standardized adherence checklists. EFE indicators include threat expectancies, associative mapping, expectancy-violation exposures, and post-exposure learning. AFE indicators include use of a graduated hierarchy, repeated confrontation with feared stimuli, reduction of safety behaviors, and anxiety monitoring.

Treatment Structure Both treatments consist of approximately ten weekly individual psychotherapy sessions. Session 1 includes orientation, psychoeducation, and introduction to the treatment model. Session 2 focuses on individualized conceptualization and construction of an associative map or exposure hierarchy. Sessions 3 to 9 involve exposure according to the assigned condition. Session 10 includes consolidation, review of learning, and relapse prevention planning.

Although both conditions involve repeated exposure, their strategies differ. EFE identifies and tests explicit threat predictions associated with feared situations. Exposures are designed to maximize expectancy violation and strengthen inhibitory learning. AFE follows a graduated exposure framework, with exercises organized along an anxiety hierarchy and the goal of reducing anxiety responses through repeated confrontation with feared stimuli.

Between-Session Exposure Practice Participants in both conditions will complete between-session exposure practices to consolidate learning. Daily practice will be encouraged, with at least three times per week considered acceptable. Participants will repeat in-session tasks whenever possible or design functionally equivalent tasks with the therapist. In EFE, practices are behavioral experiments aimed at testing threat expectancies and reinforcing inhibitory learning; variability across contexts will be encouraged to promote generalization. In AFE, practices follow structured replication of in-session exposures, emphasizing repeated confrontation with feared stimuli and anxiety monitoring with standardized logs.

Exposure Monitoring Exposure exercises will be recorded using structured logs. In AFE, participants will record duration, hierarchy level, and subjective units of anxiety (USA) at the beginning, peak, and end of exposure. In EFE, logs will focus on threat expectancy using the Threat Expectancy Index (IEA), including initial IEA, peak IEA, whether the feared outcome occurred, what was learned, and adjusted IEA at the end. Participants in both conditions will also report relief, hope, joy, and peak anxiety.

Sample Size Sample size estimation was based on the primary comparison between EFE and AFE using repeated-measures assumptions. Because the estimate does not accommodate the full complexity of the open-cohort stepped-wedge delayed-start design, including time-varying treatment exposure and therapist-level clustering, it is an initial conservative approximation. The estimation assumed 14 measurement occasions: 11 before, during, and immediately after treatment, plus 6, 12, and 18 month follow-ups. The minimum required sample was estimated at 66 participants per arm (N = 132). With expected attrition of approximately 35%, the recruitment target was increased to 203 participants.

Outcome Measures and Assessment Schedule Clinical outcomes will be assessed using general self-report measures and disorder-specific symptom scales. Weekly assessments include general psychological distress and anxiety symptoms. Disorder-specific measures will be administered according to principal diagnosis. All scales have been validated for use in Chile. Assessments will occur at baseline, weekly throughout treatment, immediately after treatment, and at 6, 12, and 18 months after treatment. Additional pre- and post-treatment measures will assess quality of life, state-trait anxiety, depression, treatment satisfaction, overall anxiety severity and impairment, and alcohol, smoking, and substance involvement.

Statistical Analysis Plan Longitudinal changes in anxiety symptoms and psychological distress will be analyzed using multilevel mixed-effects models appropriate for repeated-measures data within a stepped-wedge delayed-start design. Models will account for observations nested within participants and participants nested within therapists. Therapist-level clustering will be modeled using random effects. Fixed effects will include time, treatment condition, and treatment exposure status. Additional covariates will include step and baseline symptom severity when appropriate.

Primary analyses will evaluate differences in symptom trajectories between EFE and AFE from baseline through post-treatment and follow-ups. The design also allows secondary comparisons of symptom change during waitlist and after treatment initiation, helping distinguish treatment effects from natural symptom fluctuations.

Secondary analyses will evaluate clinically significant change and relapse or recurrence across follow-ups. When Chilean reliable change indices are unavailable, alternative estimates will be used, including internal consistency-based methods and the Standardized Individual Difference. Mediation analyses will examine whether symptom reduction is explained by changes in anxiety during AFE and changes in expectancies during EFE. Moderation analyses will examine whether treatment effects vary by sex, age, socioeconomic status, and symptom severity.

All analyses will follow an intention-to-treat approach. Missing data will be handled using maximum likelihood estimation within mixed-effects models. Sensitivity analyses will examine robustness under alternative model specifications and the impact of attrition during long-term follow-up.

Ethical Considerations The protocol has been approved by the Ethics Committee of the Faculty of Social Sciences at the University of Chile. Participant confidentiality will be maintained. Identifying information will be stored separately from research data, and electronic data will be stored on secure password-protected servers. Participants may withdraw at any time without penalty or loss of access to clinical services.

Expected Contribution By assessing outcomes through 18 months, this trial tests whether inhibitory learning-based exposure improves durability of therapeutic gains.