Patient demographics and psychological functioning as predictors of unilateral termination of psychodynamic therapy

Abstract

Approximately one in five patients drops out of treatment before its completion. Little is known about consistent predictors of dropout, and most studies focus on patients’ demographic characteristics. A mass of information is collected daily at intake in clinical practice. Based on psychodynamic theoretical conceptualizations and accumulative clinical experience, this information may help predict dropout, and thereby expand the empirically based predictors of dropout. Objective: The present study aims at bridging between scientific research and clinical practice by investigating potential predictors of unilateral termination collected at intake, before therapy, in addition to predictors already identified in the literature. Method: The study was based on data from 413 patients from a university consulting center. Each patient completed a pre-intake questionnaire collecting demographic information, and underwent an interview conducted by a professional intaker. Results: Results indicate that the consistent predictors described in the literature, education, and age, were related to unilateral termination rates. Additionally, lower intrapsychic functionality, as evaluated by the intakers, was also found to contribute uniquely to higher unilateral termination rates. Conclusion: This finding attests to the unique value of professional evaluations of patients’ intrapsychic functionality, frequently conducted in clinical practice, to detect patients at risk of unilateral termination of treatment.

Keywords:

• unilateral termination
• dropout
• psychodynamic treatment
• intrapsychic functionality
• dropout predictors

Acknowledgements

We would like to thank the Counseling Center at the University of Haifa, headed by Dr Haim Kaplan, for their generosity in sharing their data with us and especially Dr. Polina Viksman for her assistance.

Notes

1 The majority of participants were evenly distributed between 13 and 17 years of education. Mean years of education was 14.9, with an SD of 1.59, a median of 15, and a skewness value of .65.

2 We also conducted an exploratory analysis to examine the potential effect of gender match between intakers and patients. There were 277 (67.1%) matched gender cases between intaker and patients. In a logistic regression analysis, the effect of gender matching on unilateral termination was not found to be significant (B = 0.243; SD = 0.220; OR = 1.275; 95% CI = 0.83–1.96; p = .26).

3 Although the ability of age to predict unilateral termination was significant only in the fully imputed data and not in the complete sample, the ORs for age in predicting unilateral termination in the two data sets were the same (1.05), therefore the differences between the two samples can be attributed to sample size.