Date of Award:
Doctor of Philosophy (PhD)
Kevin S. Masters
Several systems have been designed to monitor psychotherapy outcome, in which feedback is generated based on how a client's rate of progress compares to an expected level of progress. Clients who progress at a much lesser rate than the average client are referred to as signal-alarm cases. Recent studies have shown that providing feedback to therapists based on comparing their clients' progress to a set of rational, clinically derived algorithms has enhanced outcomes for clients predicted to show poor treatment outcomes. Should another method of predicting psychotherapy outcome emerge as more accurate than the rational method, this method would likely be more useful than the rational method in enhancing psychotherapy outcomes. The present study compared the rational algorithms to those generated by an empirical prediction method generated through hierarchical linear modeling. The sample consisted of299 clients seen at a university counseling center and a psychology training clinic. The empirical method was significantly more accurate in predicting outcome than was the rational method. Clients predicted to show poor treatment outcome by the empirical method showed, on average, very little positive change. There was no difference between the methods in the ability to accurately forecast reliable worsening during treatment. The rational method resulted in a high percentage of false alarms, that is, clients who were predicted to show poor treatment response but in fact showed a positive treatment outcome. The empirical method generated significantly fewer false alarms than did the rational method. The empirical method was generally accurate in its predictions of treatment success, whereas the rational method was somewhat less accurate in predicting positive outcomes. Suggestions for future research in psychotherapy quality management are discussed.
Spielmans, Glen I., "A Comparison of Rational Versus Empirical Methods in the Prediction of Psychotherapy Outcome" (2004). All Graduate Theses and Dissertations. 6216.