RESUMO
For the analysis of time-to-event data, frequently used methods such as the log-rank test or the Cox proportional hazards model are based on the proportional hazards assumption, which is often debatable. Although a wide range of parametric and non-parametric methods for non-proportional hazards has been proposed, there is no consensus on the best approaches. To close this gap, we conducted a systematic literature search to identify statistical methods and software appropriate under non-proportional hazard. Our literature search identified 907 abstracts, out of which we included 211 articles, mostly methodological ones. Review articles and applications were less frequently identified. The articles discuss effect measures, effect estimation and regression approaches, hypothesis tests, and sample size calculation approaches, which are often tailored to specific non-proportional hazard situations. Using a unified notation, we provide an overview of methods available. Furthermore, we derive some guidance from the identified articles.
RESUMO
Dropping out of psychotherapeutic treatment (i.e., the patient ending treatment unilaterally) poses a problem for patients, therapists, and the health care sector. Previous research showed that changes in symptom severity and general change mechanisms (GCMs), such as interpersonal experiences, intrapersonal experiences, and problem actuation, might be related to drop-out. We investigated the relationship of these predictors and drop-out in a sample of 724 patients (21.1% drop-out) receiving cognitive-behavioral therapy in routine care from a German outpatient clinic. Survival analysis was used to account for the longitudinal nature of the data created by routine outcome monitoring and to deal with the time varying predictors, GCMs, and changes in symptom severity. As outcome, we predicted the risk of dropping out. Results showed that patient- and therapist-rated interpersonal experiences, which include alliance, significantly predicted the risk for drop-out. Contrary to previous research, intrapersonal experiences and symptom severity change did not predict drop-out. Overall, GCMs and symptom severity change accounted for 3.8% of explained variance in the outcome. These results entail that it is important to monitor interpersonal experiences over the course of treatment to identify patients at risk for drop-out. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
