The therapeutic alliance construct refers to the collaborative aspect of the relationship between therapist and client(s) in the context of psychotherapy. The conceptualization of the alliance that has been most prominent in individual psychotherapy, proposed by E.S. Bordin in 1979, includes three components: (1) an agreement between therapist and client about the goals of treatment, (2) an agreement about the therapy tasks needed to accomplish those goals, and (3) the emotional bond developed between therapist and client that allows the client to make therapeutic progress. The therapeutic alliance is undoubtedly a highly relevant construct in psychotherapy: In process-outcome research in individual psychotherapy, the most important finding that has emerged from a considerable number of studies is that the alliance assessed early in treatment predicts ultimate therapeutic success across a variety of clinical issues and treatment modalities.

Although family therapy models tend not to formally incorporate the therapeutic alliance construct in their approach, most models rely on changing relational patterns within the family and consider the therapeutic relationship as a central tool for doing so. The acknowledged complexity of the therapist-family relationship as well as within-family relationships in conjoint family treatment suggests that the therapeutic alliance is a key construct for understanding the process and outcome of family therapy. In fact, although the therapeutic alliance is much less well studied in family therapy than in individual therapy, some recent investigations have found positive associations between therapeutic alliance and treatment retention or treatment success.

The SOFTA-o is an observational instrument for investigating the process of family therapy and guiding the work of clinicians, trainers and supervisors. The difference between the SOFTA-o and self-report measures of the alliance in family therapy, like the FTAS (Pinsof & Catherall, 1986) or WAI-couples (Symonds & Horvath, 2004,) is that the SOFTA-o is an observer rating scale based on the clients’ behavioral manifestations of therapeutic alliance, as well as on the therapists’ observable contributions to the alliance.

The SOFTA-o is based on a set of validated behavioral indicators representing positive or negative expressions of each of the four dimensions composing the SOFTA model. Using those behavioral indicators, client (and therapist) behaviors contributing to strong and weak alliances are noted. Then, a set of rules on the frequency, valence, intensity, and clinical meaningfulness of the noted indicators are used to make a rating, from -3 (extremely problematic) to +3 (extremely strong), on the 4 dimensions of the SOFTA model: Engagement in the Therapeutic Process, Emotional Connection with the Therapist, Safety within the Therapeutic System, and Shared Sense of Purpose within the Family.

Good interrater reliabilities can typically be achieved in 10-15 hours of training. In selecting raters, the investigator need not choose individuals who are highly experienced clinically. Although clinical experience can facilitate the rating task, SOFTA-o does not require a great deal of clinical sophistication, and even students with little or no clinical experience can achieve good reliabilities. Good social skills are necessary, since judgments need to be made on the meaningfulness of the SOFTA-o items in context. We also recommend using both male and female raters if at all possible, inasmuch as women and men tend to see interpersonal behavior differently. To achieve good interrater reliability, we recommend selecting 6-8 practice videotapes that are representative of the investigator’s final data set. Ideally, these training tapes should have few clients and good sound and video quality, as well as examples of both negative and positive Engagement, Emotional Connection, Safety, and Shared Purpose. The difficulty level should be increased gradually as raters become fluent with the task.

The e-SOFTA is a software application that allows users to rate the quality of the therapeutic alliance in conjoint couple and family therapy sessions. The main advantage of this software version over the paper-and-pencil instrument is that e-SOFTA provides a direct comparison of two or more users’ ratings (e.g., supervisor and trainee or several research assistants). In addition, qualitative impressions can be recorded and compared across users. That is, built into the software program is the ability to directly compare two users’ qualitative impressions and alliance ratings of the same session. In the context of research, a therapy session can first be rated by the SOFTA-o trainer. Next, the individuals being trained can rate the same session, using the time-stamped items and qualitative comments by the trainer to understand differences between their own results and those of the trainer. In the context of supervision, a supervisee can rate his or her own sessions after (or before) the supervisor has independently done so. The program is available free from this website and includes a series brief training vignettes (8 client, 8 therapist) in both Spanish and English.

Subsequent to the development of the SOFTA-o, this self-report measure (the SOFTA-s) was created, with versions for both clients and therapists. This is a 16-item questionnaire with 4 subscales, corresponding to the 4 SOFTA dimensions. The SOFTA-s questionnaires have two subscales that can be considered common to all formats of therapy (individual, group, and couple/family therapy (Engagement in the Therapeutic Process and Emotional Connection to the Therapist) and two subscales that reflect elements unique to couple and family treatment (Safety within the Therapeutic System and Shared Sense of Purpose within the Family).

Because the SOFTA project evolved from a collaboration of Spanish (Escudero) and United States (Friedlander and Heatherington) family therapy researchers, nurtured by a Spain-U.S. exchange program in family interventions and counseling psychology for graduate students. As a product of this collaboration, the SOFTA model and instruments were created simultaneously in English and Spanish (called Sistema de Observación de la Alianza en Terapia Familiar). For example, the SOFTA-o was originally written in English and translated into Spanish, whereas the opposite process was undertaken for the SOFTA-s, but in each step of their development process, the items, operational definitions, and training manual were translated and back-translated to ensure accuracy. Similarly, reliability and validation studies were conducted in both North America and Spain, obtaining highly congruent results. Additionally a French version of the SOFTA-s is available on this website.

The SOFTA-o is valuable for the training and supervision of novice family therapists, as well as for the analysis of difficult cases seen by expert therapists. Using the behavioral descriptors within each dimension, a therapist or supervisor can analyze the strength of the therapeutic alliance in one session or across sessions. The analysis can be focused on one or more alliance dimensions, on one member of the family, or on the family as a group. With the e-SOFTA (the software application), the observer can create a file with qualitative notes, automatically arranged in chronological order by the program to accompany each behavioral descriptor as it is detected and marked. In this way, the SOFTA-o can be a useful aid in training, supervision, and self-supervision.