The complexity of treatment failure – prevalence and predictors of dropout and non-response in psychological treatment for traumatized populations

Semmlinger, Verena (2024): The complexity of treatment failure – prevalence and predictors of dropout and non-response in psychological treatment for traumatized populations. Dissertation, LMU München: Fakultät für Psychologie und Pädagogik

Abstract

The efficacy of psychological interventions in the treatment of traumatized patients has been widely demonstrated (Martin et al., 2021) and the current evidence shows promising results for specific populations, such as refugees and asylum seekers (e.g., Thompson et al., 2018). However, there is strong evidence that a substantial proportion of patients does not benefit sufficiently from treatment or discontinues treatment prematurely (Schottenbauer et al., 2008; Varker et al., 2021). Treatment failure is a complex construct that can be viewed as an umbrella term encompassing several aspects, such as dropout and non-response (Oasi & Werbart, 2020). The consequences of treatment failure are far-reaching and include negative effects on the patient, the therapist, society, and the healthcare system in general (e.g., Ogrodniczuk et al., 2005; Smith-Apeldoorn et al., 2019; Swift et al., 2012). However, to date there is a significant lack of research on dropout and non-response in the treatment of traumatized patients. In order to gain an in-depth understanding of both aspects, which can later form the basis for deriving preventive measures, it is important to examine the prevalence and identify baseline predictors. Therefore, the overarching aim of this thesis was to fill this gap by providing new evidence on the prevalence and predictors of dropout and non-response in the treatment of traumatized populations. In particular, this thesis covers three publications designed to investigate the prevalence and predictors of dropout in understudied areas, namely the treatment of refugees and treatment of PTSD patients in naturalistic settings. The fourth publication focused on non-response, aiming to investigate its prevalence and predictors in PTSD treatment. Publication I and Publication II were the first to provide comprehensive evidence on the prevalence and predictors of dropout in the treatment of refugees and asylum seekers. In the absence of previous knowledge, Publication I was designed as a review, synthesizing refugee-specific findings and additionally reviewing existing evidence on treatment dropout in general and applying the findings to the refugee population. Further, we reviewed the current evidence on measures to prevent dropout. The review revealed a significant range of reported dropout rates, varying from 0% to 64.7%. Additionally, the review emphasized the importance of predictors specific to refugees, such as high initial impairment, differing perceptions of mental health, deviating expectations of psychological treatment, and external treatment barriers. To prevent dropout, it is crucial to prioritize the promotion of cultural competencies, cultural adaptation of treatment, and preparation for treatment. Based on the findings of the review, Publication II aimed to provide the first comprehensive evidence on the prevalence and predictors of dropout in psychological or psychosocial interventions for refugees and asylum seekers. The meta-analytic results of 28 eligible randomized controlled trials (RCTs), with 39 active treatment conditions, and 2,691 participants, revealed a weighted average dropout rate of 19.14%. Dropout was less frequent in the treatment condition compared to the control condition (OR = 0.52). The results revealed no significant predictor of dropout, except the country in which the study was conducted, but showed a potential influence of refugee-specific variables on dropout. Overall, the findings suggest that the dropout rate is comparable to those reported in non-refugee populations. Future research should focus on refugee-specific variables, such as duration of stay in the country of resettlement and asylum status, rather than applying predictors of dropout from Western samples directly to the refugee population. Publication III examined the dropout rates and predictors of dropout in PTSD treatment in a naturalistic setting. Of the 195 adults diagnosed with PTSD included in the study, 15.38% discontinued trauma-focused cognitive behavioral therapy prematurely, which was provided in three specialized outpatient centers. Dropout rates were higher in younger patients, and lower in patients who lived with their parents compared to living alone. Results showed that the dropout rate found in naturalistic settings was comparable to dropout rates found in RCT studies. Although routinely assessed baseline patient variables were associated with dropout, the results on prediction performance indicate that the overall model, comprising different pretreatment variables, could not predict dropout to a practically useful level. Publication IV was the first study to provide comprehensive evidence on the prevalence and predictors of non-response to first-line guideline-recommended psychological treatments for PTSD. The meta-analysis employed a methodology similar to Publication II, and meta-analyzed 86 studies, with 117 active treatment conditions, and 7,894 patients. The weighted average non-response rate was 39.23%, and non-response was less frequent in the treatment condition compared to the control condition (OR = 0.22). Higher non-response rates were found to be associated with male gender, older age, and with being a refugee or veteran. Further, higher PTSD symptom severity and the presence of comorbid depressive disorder or higher depressive symptoms was associated with non-response. Treatment type and treatment format were identified as significant treatment-related predictors, with lowest non-response rates in treatments combining prolonged exposure (PE) and cognitive therapy (CT), and in a combination of individual and group therapy. Finally, non-response was significantly higher in studies reporting intention-to-treat (ITT) analysis compared to per-protocol (PP). The findings indicate that treatment modifications should be considered for specific subgroups of PTSD patients characterized by one or more of the identified baseline predictors. In conclusion, this thesis addresses the lack of research on treatment failure in traumatized populations. The integrated findings of the four publications provide comprehensive knowledge on the prevalence and predictors of dropout and non-response in the treatment of PTSD in general, in specific subpopulations of traumatized patients, and in specific treatment settings. Future research should focus on a wider range of specific predictors and examine underlying mechanisms and process variables beyond pretreatment predictors. In clinical practice, the findings have implications for the derivation of measures to prevent and reduce dropout and non-response in the treatment of traumatized populations.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Keywords: Dropout, Non-Response, Refugees, Posttraumatic Stess Disorder, Meta-analysis
• Themengebiete: 100 Philosophie und Psychologie ; 100 Philosophie und Psychologie > 150 Psychologie
• Fakultäten: Fakultät für Psychologie und Pädagogik
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 4. November 2024
• 1. Bericht­erstatter:in: Ehring, Thomas