Internet-based cognitive behavioural therapy (iCBT) interventions allow a wider access to evidence-based psychological treatments for people with anxiety or depressive disorders. Newby et al. (Reference Newby, Mackenzie, Williams, McIntyre, Watts, Wong and Andrews2013) recently published a two-part study including a comparison of iCBT versus wait-list control for mixed anxiety and depression. Using a wait-list control results in an inflated effect-size of treatment, especially in iCBT studies (So et al. Reference So, Yamaguchi, Hashimoto, Furukawa and McCrone2013). Moreover, such a control may represent a potential bias, as underlined by Andersson & Cuijpers (Reference Andersson and Cuijpers2009) in their meta-analysis of Internet-based treatments for depression. They found a significant difference between treatments with therapist support, compared with non-supported treatments (d = 0.61 v. d = 0.25, respectively); and recommended to further investigate the role of therapist factors in computerized treatments. Newby et al. (Reference Newby, Mackenzie, Williams, McIntyre, Watts, Wong and Andrews2013) provided regular contacts with their clinician to the treatment group participants until they completed one-third of iCBT; thereafter, this contact was made available, at patients' request. On the contrary, participants in the wait-list control group were not offered such contacts. Therefore, part of the observed difference of efficacy between groups may be attributed to this factor, thus introducing a systematic bias in comparison. Posternak & Zimmerman (Reference Posternak and Zimmerman2007) assessed the therapeutic impact of follow-up assessments on placebo response in antidepressant trials through meta-analysis. They found that clinical assessments incur a significant therapeutic effect for participants on placebo, representing up to 40% of the placebo response. Thus, minimal contact with a clinician may result in therapeutic effect that is confounded with the effect of iCBT as the wait-list control did not include such contact. It can be argued that contacts by email and phone are less effective than face-to-face contacts. But, according to Hassanian-Moghaddam et al. (Reference Hassanian-Moghaddam, Sarjami, Kolahi and Carter2011), even sending postcards to patients may result in their mental health improvement. It seems that in a treatment-as-usual (TAU) group, this factor is better controlled as contacts with a clinician are provided to this group, but nevertheless TAU results in a wide observed distribution of quantity and quality of clinician support. Newby et al. (Reference Newby, Mackenzie, Williams, McIntyre, Watts, Wong and Andrews2013) suggest that the quality of clinician support may modify adherence to iCBT. We, therefore, suggest that clinician support per se may contribute to the efficacy of the iCBT protocol, when compared with a simple wait-list control, in an unknown proportion.
Declaration of Interest
None.
