The therapeutic alliance is seen as an important dimension in any type of psychotherapy. But patient, therapist, or observers can have different views on the therapeutic alliance. The question is which perspective best represents the therapeutic alliance, and what are the differences between these alternative views. In the present study, the therapist-patient alliance (TPA, the view of the therapist), patient-therapist alliance (PTA, the view of the patient), and mutual therapeutic alliance (MTA, the view of an observer) were measured simultaneously in cognitive behavior therapy of patients suffering from generalized anxiety disorder. Additionally, the concordance between patient and therapist ratings (TPC) was calculated. Cognitive behavior therapists attained high positive scores in all perspectives for all dimensions of the therapeutic alliance, such as empathy, cooperation, transparency, focusing, and assurance of progress. Correlations were consistently higher for ratings between therapist and patient than between observer and patient. A relation with outcome (Hamilton Anxiety Scale) was only found for observer ratings. It was concluded that cognitive behavior therapists can achieve good alliances with their patients. Different perspectives on the therapeutic alliance should be distinguished and taken into account separately in studies on the therapeutic process and outcome.
Keywords: therapeutic alliance; therapist-patient relationship; generalized anxiety disorder; cognitive behavior therapy
The relationship between the therapist and patient is one of the primary areas of psychotherapy research. The therapeutic alliance has been investigated in various psychotherapeutic orientations and settings (Barber, 2000; Castonguay, 1996; Cottraux et al., 1995; Frieswyk et al., 1986; Gaston, 1991, 1998; Hartley & Strupp, 1983; Hintikka, Laukkanen, Marttunen, & Lehtonen, 2006; Hogduin, De Haan, & Schaap, 1989; Keijsers, Scraap, Hoogduin, & Lainmors, 1995; Krupnick, 1996; Loeb et al., 2005; Luborsky, McLellan, Woody, O'Brien, & Auerbach, 1985; Malik, Bentler, Alimohamed, Gallagher-Thompson, & Thompson, 2003; Newman & Strauss, 2003; Orlinsky, Grawe, & Parks, 1994; Vogel, Hansen, Stiles, & Gotestam, 2006). More recent studies have examined the working alliance even in online psychotherapy (Knaev-elsrud & Maercker, 2006).
The scientific literature suggests that the quality of the therapeutic alliance is positively related to treatment outcome and can even lead to therapeutic changes by itself. In a meta-analysis of 25 studies, Horvath and Symonds (1991) found a moderate overall effect size of .26 for the impact of the quality of the therapeutic alliance on treatment efficacy. Similarly, Martin, Garske, and Davis (2000) reported an effect size of .22 on the basis of 79 studies. However, there are also studies that did not find significant correlations between the quality of the therapeutic alliance and treatment outcome (DeRubeis, 1990; Feeley, DeRubeis, & Gelfand, 1999). This may be due to discrepant concepts of the therapeutic alliance (Dill-Standiford, Stiles, & Rorer, 1988; Horvath, 2000; Wolfe & Goldfried, 1988; Zimmer, 1983).
Different psychotherapeutic schools have different concepts of the therapeutic alliance. In psychoanalysis, transference and countertransference are held to be indispensable treatment factors (Freud, 1958), and in client-centered psychotherapy the therapeutic alliance is seen as the primary treatment element (Rogers, 1958; Schmidt-Traub, 2003; Truax & Carkhuff, 1967). Behavior therapy, in contrast, is sometimes suspected to neglect the therapeutic alliance and to be rather mechanical and less personal. There is empirical evidence that cognitive behavior therapists are more active (Greenwald, Kornblith, Hersen, Bellack, & Himmelhoch, 1981; Hardy & Shapiro, 1985; Sloane, Staples, Cristol, Yorkston, & Whipple, 1975; Stiles, Shapiro, & Firth Cozens, 1988) and more directive (Brunik & Schroeder, 1979) than other psychotherapists, but they, nevertheless, do also reach high levels of emotional support, empathy, and unconditional positive acceptance towards the patient (Keijsers, Scraap, & Hoogduin, 2000; Sloane et al. …