An evaluation of therapeutic alliance and outcome in an internet chat therapy service

Although the Internet has increasingly been the focus of research over the past decade, there have been relatively few studies about how the full variety of Internet communication tools can be used for the purpose of delivering psychological services. Much of the recent emphasis has been on web-based self-guided psychological interventions, where interactions with a psychologist are minimal (Amstadter, Broman-Fulks, Zinzow, Ruggiero, & Cercone, 2009; Spek, Cuijpers, Nyklicek, Riper, Keyzer, & Pop, 2007). A limited number of studies have investigated the processes and outcomes of psychological interventions applied over Internet chat communication (Cook & Doyle, 2002; King, Bambling, Reid, & Thomas, 2006a; Mallen, Day, & Green, 2003; Rassau & Arco, 2003); however, there has been a general tendency to avoid comparing these forms of intervention with face-to-face therapy (Anthony, 2000a). This has had the unfortunate consequence of placing the existing research beyond the reach of evidence-based practice, where various forms of intervention are compared and contrasted. The main goal of this research is to make a direct comparison of psychological processes and therapeutic outcomes when the same group of therapists deliver psychological interventions to clients over both Internet chat and face-to-face therapy.A mixed quantitative and qualitative approach was utilised to integrate findings from outcome measures with the subjective report of clients and therapists who undertook Internet chat therapy. A team of 20 therapists provided psychological services to clients who self-selected either face-to-face or Internet chat therapy. Both therapists and clients completed measures for symptom severity and the therapeutic alliance at the first and third session. Therapeutic alliance was measured using the client, therapist and observer rated versions of the CALPAS, while symptom severity was measured using the BSI and SCL-90 Analogue. The final sample consisted of 17 matched pairs of Internet and face-to-face therapy cases, with 3 additional cases where the therapist was only able to obtain an Internet case for the data. The hypotheses of this study predicted that symptom severity would decrease and that the therapeutic alliance would increase over 3 sessions in both treatment modalities. It was also hypothesised that face-to-face therapy would outperform Internet chat therapy on each measure.Results of ANOVA analyses supported all hypotheses related to improvement over the first 3 sessions of treatment, with the exception of therapist-rated symptom severity. There was a strong main effect for client-rated alliance, increasing significantly at the same rate in both treatment groups: F(1,35) = 23.021, p < .001, partial 2 = .397, Cohen’s d = 1.15. There was also a strong main effect for client-rated symptom severity across both groups over the first 3 sessions of treatment: F(1,35) = 15.191, p < .001, partial 2 = .303, Cohen’s d = .92. Results for ANOVA analyses comparing treatment modalities did not identify statistically significant differences, with the exception of significantly higher alliances rated by clients receiving Internet chat therapy: F(1,35) = 6.972, p = .012, partial 2 = .166, Cohen’s d = .76. In addition to statistically significant change, an analysis for clinically significant change was also undertaken (Jacobson & Traux, 1991). Results of this analysis showed that there were only minor differences between groups at both the first and third session. In the Internet chat therapy group 10% of cases were ‘improved’ and 5% were classed as ‘recovered’. In the face-to-face therapy group, 11.8% were categorised as ‘improved’ and 5.9% classed as being ‘recovered. In summary, there were relatively few measurable differences between these modes of service delivery.Qualitative data generated from interviews at the conclusion of Internet chat therapy showed the importance of therapeutic distance for people who select this form of treatment. Clients described how the distant way of communicating to a psychologist over Internet chat led to a more personal experience for them. Therapists described the challenge of understanding the emotions of their client when relying on the written word alone. Clients and therapists also described the formation of mental images of the other party and the role this had in terms of constructing a sense of interaction with the other person. The formation of mental images had an influence on the quality of the working alliance that developed, with both clients and therapists being generally satisfied with the therapeutic relationship overall.Despite the widespread public use of Internet chat, this popular form of communication has received very little attention from researchers. The results of this study are promising in that they demonstrate that positive clinical outcomes are associated with client interactions with psychologists offering services over this modality. With meta-analytic reviews showing that larger treatment effects are associated with web-based interventions where there is greater therapist involvement (Barak, Hen, Boniel-Nissim, & Shapira, 2008; Spek et al., 2007), the present study raises questions about whether Internet chat could be utilised more broadly as a mode of service delivery. This study provides a detailed first glimpse at how real-time written communication over the Internet could be used for psychotherapeutic purposes