Language Use as a Progress Monitoring Marker: An Exploratory Study of Change in Psychotherapy
MetadataShow full item record
Use of progress monitoring and outcome assessment (PMOA) tools has been shown to have a significant, positive effect on rate of change in therapy (e.g. De Jong et al., 2014), and in reducing treatment failure (e.g. Lambert & Hawkins, 2001) and is increasingly more frequently required of clinicians. The ability to use language is a uniquely human quality and the patterns of language use that individuals use has been tied to various psychologically-relevant variables. Automated linguistic analysis may represent one path forward to add the clinical utility provided by PMOA measures without drastically increasing time demands or suffering from the measurement reactivity other tools may induce. The present study analyzed the transcripts of 24 psychotherapy sessions, one early in the treatment process and one late in the course of treatment, split between six good and six poor outcome cases to explore the utility of Linguistic Inquiry and Word Count (LIWC; Pennebaker, Booth, Boyd, & Francis, 2007) for PMOA; specifically, examining its ability to predict treatment outcome and to track change over time. Results indicated significant differences in language use patterns between good and poor outcome cases but failed to find support for the hypothesis that language use patterns would change from time one to time two during the course of therapy. Logistic regression models were able to predict good vs. poor treatment outcome in between 70 – 81.7% of transcripts.