Improving psychotherapy effectiveness by addressing the problem of premature termination: Introduction to a special section

Abstract

Premature termination appears to be a consistent problem in psychotherapy, showing up across client, treatment, and therapist types. As psychotherapy researchers and practitioners, it is important that we gain a better understanding of this negative psychotherapy event and identify methods for reducing its occurrence. This article introduces a special section on premature termination in psychotherapy. In addition to briefly introducing the articles contained in the special section, this article offers suggestions for future research on the topic of premature termination.

Keywords:

• attrition
• dropout
• premature termination
• psychotherapy

Clinical or methodological significance of this article: This article reviews recent research findings on premature termination from psychotherapy. Future directions for research on client dropout are discussed.

Even though there is clear evidence that psychotherapy is an effective intervention for psychological problems, there is also clear evidence that psychotherapy, as currently practiced, does not work for everyone (McAleavey et al., 2017; Werbart, Below, Brun, & Gunnarsdottir, 2015). One of the most significant impediments to the effectiveness of psychotherapy is the problem of premature termination. Although dropout rates in psychotherapy have been found to be lower than rates of premature termination for other mental health interventions (e.g., pharmacotherapy; see Swift, Greenberg, Tompkins, & Parkin, 2017), in psychotherapy approximately one in every five clients still chooses to end treatment prior to its completion (Swift & Greenberg, 2012). Given the deleterious impacts that premature termination can have on clients, therapists, and others, it is important that we seek to gain a better understanding of this negative therapy event and seek to identify methods for decreasing its occurrence. This special section includes three articles that seek to meet that goal.

In an investigation of dropout rates in adolescents receiving treatment for depression, O’Keeffe and colleagues (2017) tested several potential predictors of premature termination using data from a multi-site randomized controlled trial. Their results indicated that older age, higher levels of antisocial behavior, lower levels of child verbal intelligence, and lower levels of the therapeutic alliance were all significantly associated with higher rates of premature termination; however, gender, ethnicity, depression and other symptom severity, and parental well-being and parenting styles were not. One of the most interesting findings from this study was that the ratio of sessions missed after session number 4 was a more accurate predictor of eventual dropout than the ratio after sessions 1 through 3 or sessions 5 and above.

In the second study, Rubin, Dolev, and Zilcha-Mano (2016) used data from a sample of clients receiving psychodynamic psychotherapy at a university counseling center in Northern Israel to also test several predictors of premature termination. In addition to testing several client demographic variables, in this study they tested three client psychological functioning variables that they believed had high theoretical support for predicting dropout—psychological distress level, intrapsychic functionality, and global functionality. Each of these variables was assessed by the intake therapist at the time of the initial session. They found that younger clients and those with higher intrapsychic functionality levels were less likely to prematurely terminate from psychotherapy; however, client diagnosis, gender, marital status, educational level, and the other two psychological functioning variables were not associated with dropout.

The third article in this special section (Gulum, Soygut, & Safran, 2016) describes a mixed-methods study conducted in Turkey that examined the relationship between alliance ruptures and premature termination. Specifically, researchers in this study compared the content of 16 sessions that contained alliance ruptures and were followed by continued client attendance to 16 sessions that contained alliance ruptures, but were followed by client dropout. They found that temporary rupture sessions differed from the pre-dropout sessions in the amount and variety of the session content, the type of ruptures that occurred, and the frequency of positive therapist behaviors.

Taken together, the three articles included in this special section have important clinical and research implications. Both O’Keeffe, Martin, Goodyer, Wilkinson, and Midgley (2017) and Rubin et al. (2016) found that age was related to premature termination with their samples; however, in a direction that is opposite to what has been seen in the existing literature (Swift & Greenberg, 2012). Their samples did include younger clients (adolescents and college students) and it is possible that when considering all client age groups, an inverse U relationship between age and risk of premature termination is present. Importantly, both studies were able to identify client psychological factors that are associated with premature termination that have received little attention in the existing literature—antisocial behavior, verbal intelligence, and intrapsychic functioning. In addition to identifying predictors of premature termination, the articles in this special section provide some preliminary insight into clients’ timing and reasons from premature termination. Specifically, O’Keeffe and colleagues and Gulum and colleagues (2016) results illustrate the importance of the therapeutic alliance and the therapist’s ability to repair alliance ruptures when they occur. The findings from these studies are particularly noteworthy given that the data came from diverse groups of clients, were drawn from multiple clinics across different regions of the world, and were obtained during the courses of very different types of psychotherapies.

Although these three papers do add to our understanding of premature termination in psychotherapy, there are still significant gaps in the research on dropout that need to be addressed. First, it is important that future studies on premature termination use a consistent and valid method for measuring the construct. This issue has been discussed previously (Swift, Callahan, & Levine, 2009); however, many researchers still use operationalizations based solely on treatment duration or protocol completion, which are far from ideal. Second, we need more research seeking to identify the reason or cause behind clients’ choices to prematurely terminate. Several studies have tested the relationship between process variables and premature termination. Others have asked prematurely terminating clients to qualitatively share their reasons for prematurely terminating. What is needed are more studies similar to Gulum et al.’s (2016) that qualitatively and quantitatively examine what is occurring in the sessions immediately prior to the unilateral termination. A micro-process approach may be particularly useful for this area of study (Pascual-Leone, 2018; Swift, Tompkins, & Parkin, 2017). Third, research is needed to further examine why some strategies that have been proven to be effective for reducing premature termination (e.g., outcome monitoring) are not always used in practice. Last, further effort is needed to develop and test new methods for reducing premature termination in psychotherapy. Research into treatment techniques that produce positive outcomes will also likely have the benefit of reducing premature termination; however, dedicated research on strategies designed specifically for treatment dropout is needed. These strategies should include methods that can be easily transferred to practice settings.