Rethinking Research on Prediction and Prevention of Psychotherapy Dropout: A Mechanism-Oriented Approach

Abstract

Dropout is a ubiquitous psychotherapy outcome in clinical practice and treatment research alike, yet it remains a poorly understood problem. Contemporary dropout research is dominated by models of prediction that lack a strong theoretical foundation, often drawing on data from clinical trials that report on dropout in an inconsistent and incomplete fashion. In this article, we assert that dropout is a critical treatment outcome that is worthy of investigation as a mechanistic process. After briefly describing the scope of the dropout problem, we discuss the many factors that limit the field's present understanding of dropout. We then articulate and illustrate a transdiagnostic conceptual framework for examining psychotherapy dropout in contemporary research, concluding with recommendations for future research. With a more comprehensive understanding of the factors affecting retention, research efforts can shift toward investigating key processes underlying treatment dropout, thus, boosting prediction and informing strategies to mitigate dropout in clinical practice.

Divergent experiences of U.S. veterans who did and did not complete trauma-focused therapies for PTSD: A national qualitative study of treatment dropout

“…In line with recommendations by Cooper et al (2018) , researchers have examined potentially modifiable contributors of non-completion and mechanistic variables that may underlie PE and CPT dropout. For example, a prospective study of veterans initiating PE or CPT across four VA clinics found that veterans whose loved ones encouraged them to face their distress (i.e., exhibited social control) were twice as likely to finish treatment, even after accounting for demographic predictors and baseline illness severity ( Meis et al, 2019 ).…”

When adolescents stop psychological therapy: Rupture–repair in the therapeutic alliance and association with therapy ending.

“…Regardless of how dropout is defined, there is consensus in the literature that dropout presents a major challenge for clinicians (Leichsenring, Sarrar, & Steinert, 2019). Treatment dropout has been linked with dissatisfaction with treatment (O’Keeffe, Martin, Target, & Midgley, 2019), and those who drop out may not access the full benefits of treatment (Cooper, Kline, Baier, & Feeny, 2018), making dropout a substantial concern. Some studies have found dropout to be linked with poorer clinical outcomes in adults receiving therapy for depression (Saatsi, Hardy, & Cahill, 2007; Saxon, Firth, & Barkham, 2017), although there is not strong evidence for an association between dropout and clinical outcomes in adolescents with depression (O’Keeffe, Martin, Goodyer, et al, 2019).…”

The first session is the one that counts: An exploratory study of therapeutic alliance

“…This difficulty in finding precisely even whether the therapeutic alliance is a process common to the different therapeutic procedures, or whether it is rather a specific factor (Baier et al, 2020), demonstrates that we are still only at the beginning of scientific knowledge of the therapeutic process (Onken et al, 2014), perhaps still with basic analyses that are really additive, linear and less directed at identifying patterns of change (de Felice et al, 2019). However, as the therapeutic alliance is a necessary condition of therapy, predictive not only of the outcome of intervention (Norcross and Wampold, 2018;Flückiger et al, 2020a;Finsrud et al, 2022), but also of dropping out (Cooper et al, 2018), and the first sessions when the patient-therapist alliance is established and stabilized, are important (Flückiger et al, 2018(Flückiger et al, , 2020aDel Re et al, 2021), the objectives posed for this study were to analyze the relationship between the therapeutic alliance and perceived improvement during the first five sessions of the intervention, and check whether the therapeutic alliance remained stable during this period. For the first objective, and as a novelty in the analysis of the variables intervening in therapy, it should be mentioned that the STATIS procedure employed in this study clearly differentiated the contribution of each of the measurements, by clustering the results of the SRS questionnaire (therapeutic alliance), and the ORS questionnaire (satisfaction), in different bundles of vectors.…”

“…PCOMS seems to improve the results of cognitive behavioral treatment (Duncan and Reese, 2015;Waller and Turner, 2016), reducing the number of sessions required for sufficient or expected improvement (Janse et al, 2017), improves symptoms in fewer sessions (Janse et al, 2017;Koementas-de Vos et al, 2018;Janse et al, 2020) and reduces dropout rates (Janse et al, 2020;de Jong et al, 2021). As the first three-to-five sessions are considered essential in establishing the therapeutic alliance (Flückiger et al, 2018(Flückiger et al, , 2020b and its counterpart, therapeutic dropout (Roos and Werbart, 2013;Cooper et al, 2018), we proposed an exploratory analysis of what happens in the therapeutic alliance in a period of time limited to the first five sessions, regardless of the changes that could take place during the rest of the therapeutic process. Thus, from a state perspective, which enables the changes in perception of the alliance and well-being/symptoms during the treatment to be analyzed (Zilcha-Mano et al, 2016), this study posed the following objectives: (1) analyze the relationship between therapeutic alliance and patient perceived improvement during the first five sessions of therapy, and (2) check whether the therapeutic alliance remains stable over time or varies, as well as its relationship with perceived improvement.…”