Archived

This content is available here for research, reference, and/or recordkeeping.

Author ORCID Identifier

https://orcid.org/0000-0003-4503-141X

Date Available

5-12-2026

Year of Publication

2026

Document Type

Doctoral Dissertation

Degree Name

Doctor of Philosophy (PhD)

College

Arts and Sciences

Department/School/Program

Psychology

Faculty

Shannon Sauer-Zavala

Abstract

The duration of psychotherapeutic treatment is a longstanding and clinically consequential question in psychotherapy research. Premature dropout from therapy is common, indicating some patients do not receive adequate benefit. On the other hand, some patients remain in treatment beyond the point at which they receive significant benefit, which may delay care for those on patient waitlists. Researchers have proposed two contemporary models of the relation between treatment duration and symptom improvement; 1) the dose-effect (DE) model, which posits that patients improve at similar rates following a negatively accelerated symptom curve, suggesting treatment should end when sessions no longer provide significant benefit, and 2) the good enough level (GEL) model, which asserts that patients improve at different rates and that treatment should end once symptoms or improvement have reached a “good enough” level, regardless of whether additional sessions lead to improvement. Prior research directly comparing these models in the same sample has largely favored the GEL model, suggesting ending treatment when patients reach a pre-determined level of symptom severity may be a promising approach to treatment discontinuation. However, these investigations have thus far omitted examination of potentially moderating variables of the relation between treatment duration and symptom change, and the patient perspective of treatment duration. Therefore, this mixed-methods study aimed to 1) directly compare DE and GEL models in a general outpatient sample, 2) investigate patient- and therapist-level variables as moderators, and 3) collect and analyze patient perspectives of psychotherapy duration and improvement using qualitative interviews to contextualize quantitative results. Quantitative participants were 55 outpatients (Mage=29.9, 58.2% female, 81.8% white) who completed routine outcome monitoring measures as part of usual treatment at a university training clinic. Qualitative participants were 25 of these patients (Mage=30.1, 60% female, 84% white) who attended at least three treatment sessions, began and ended treatment during the study window, and completed a qualitative interview. Quantitative results showed mixed support for the DE and GEL models. Neither depression nor anxiety symptom slopes were significantly related to symptom duration, in support of the DE model. However, visual inspection of symptom slopes suggested patients who attended fewer than 10 sessions improved more quickly than other patients, in support of the GEL model. Additionally, achievement of “good enough” improvement was significantly associated with steeper anxiety and depression symptom slopes, again in support of the GEL model. Qualitative responses aligned closely with the GEL model, in that patients often reported treatment duration was based in achievement of treatment goals, attainment of therapy skills, or the ability to utilize therapy skills outside of sessions. These mixed results and the discrepancy between quantitative and qualitative results may have occurred due to differences in defining what is considered “good enough” improvement between routine outcome measures and patient’s own definitions, highlighting the difficulty in quantifying patient progress and outcomes in psychotherapy. Overall, this study adds to the psychotherapy duration literature by providing mixed support for both DE and GEL models, suggesting that each may reflect different aspects of response patterns and their relation to treatment duration.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2026.278

Archival?

Archival

Download

Share

COinS