There is now a fairly large number of trials on guided ICBT for depression that have been summarized in systematic reviews and meta-analyses. Many reviews do not distinguish between computerized treatments in general and Internet-delivered treatments, but increasingly studies on ICBT dominate reviews as few new studies on computerized treatments appear (e.g., delivered via CD-ROM in a clinic, like the Beating the Blues program). First, we will comment on the overall effects as presented in meta-analytic reviews. Then we will cover long-term effects of ICBT for depression and after that comparisons against face-to-face CBT. Finally, effectiveness data will be mentioned.

One early meta-analytic review by two of us included both computerized interventions and ICBT, but there were only two studies out of 12 that did not involve the Internet (Andersson and Cuijpers 2009). We found an average effect size vs. control groups at posttest of d = 0.41, but it was also clear that this estimate was moderated by a significant difference between guided (d = 0.61) and unguided (d = 0.25) treatments. Subsequent reviews have included more studies but have largely found the same results (Richards and Richardson 2012; Johansson and Andersson 2012), with a linear association between support and outcome. Even the effect size of d = 0.61 might be higher when considering subsequent studies. Johansson and Andersson (2012) found that totally unguided treatments had a between-group effect d = 0.21, whereas studies with contact before treatment only had an effect of d = 0.44. When there was contact with a person/staff contact during treatment, the effect was d = 0.58, and finally when there was contact both before and during the intervention, the effect was d = 0.76. This latter effect size is in line with what is found for face-to-face psychotherapies in general (Cuijpers et al. 2011). A more recent update showed that only in the period between January 2013 to September 2014, as many as 11 controlled trials on Internet treatment for depression have been published only on guided treatments (Andersson et al. 2014a).

The literature in this area is expanding rapidly. From the early controlled trials on, e.g., unguided (Clarke et al. 2002; Christensen et al. 2004) and guided ICBT (Andersson et al. 2005), a considerable number of controlled trials have followed. For example, guided ICBT for depression has been found to work in controlled trials from Australia (Perini et al. 2009), Switzerland (Berger et al. 2011), Germany (Wagner et al. 2014), the Netherlands (Warmerdam et al. 2008; Ruwaard et al. 2009), and the United States (Mohr et al. 2013). Several formats of ICBT have been tested as well such as e-mail therapy (Vernmark et al. 2010), acceptance-oriented CBT (Carlbring et al. 2013), but also trials on smartphone-delivered CBT together with Internet support (Ly et al. 2014). Another example is a study from Japan in which an intervention incorporating manga pictures for mild depression was tested (Imamura et al. 2014). As mentioned earlier, there are also studies on different subgroups, for example, persons with diabetes and depression (van Bastelaar et al. 2011) and postpartum depression (O’Mahen et al. 2014), adolescents (Saulsberry et al. 2013), and older adults (Titov et al. 2015a), and programs have been translated and adapted to for other non-Western languages such as Chinese (Choi et al. 2012). While most studies have focused on mild to moderate depression, with an increasing number using validated instruments to diagnose depression, there is at least one example of a relapse prevention program from Sweden showing promising results (Holländare et al. 2011), also at a 2-year follow-up (Holländare et al. 2013).

Earlier in this chapter, transdiagnostic and tailored ICBT were mentioned. One advantage with transdiagnostic and tailored approaches is the possibility to target comorbid problems that are known to exist alongside depression, anxiety being one example (Andersson and Titov 2014). In one controlled trial, tailored ICBT was found to be more effective for the more severe clients in the trial when compared against standard ICBT (Johansson et al. 2012b). More research is needed to confirm this observation as different depression treatments usually tend to lead to equivalent outcomes. For example, the previously mentioned psychodynamic Internet treatment yielded large treatment effects in one trial (Johansson et al. 2012a). An interesting way to study differences between treatments is to allow clients to choose treatment form, which was done in one trial showing largely equivalent findings but a preference for ICBT over the psychodynamic Internet treatment (Johansson et al. 2013).

Some studies in the ICBT field have included longer-term follow-ups, often included in the original publication. For example, in one controlled trial where ICBT was compared against face-to-face group treatment, data from a 3-year follow-up were included suggesting sustained treatment effects and no differences between the two formats (Andersson et al. 2013b). In another study, 3.5-year follow-up data were presented showing maintained effects (Andersson et al. 2013a), and other studies showing long-term effects up to 3 years posttreatment have been published (Ruwaard et al. 2009). A limitation of these studies however is that they have not documented the course of depression during the follow-up period.

With regard to the contrast between ICBT and face-to-face CBT, there are few direct comparisons, possibly because such trials are more costly and time consuming. In a recent review, Andersson and coworkers compiled the studies that had directly compared face to face and ICBT within the same trial (four trials) (Andersson et al. 2014a). The overall random effects effect size was a nonsignificant Hedges’ g =0.12 (95 % CI: −0.08 ~ 0.32) favoring guided ICBT over face-to-face therapy and with no signs of heterogeneity (I ²  = 00 %). This finding is in line with other reviews suggesting the guided self-help in general (Cuijpers et al. 2010), and ICBT (Andersson et al. 2014b) appears to be as effective as face-to-face therapy when directly compared in controlled trials.

It is probably the case that patients recruited via advertisement to research settings differ from patients seen in regular care, which calls for separate studies in these two settings. This is referred to as the difference between efficacy and effectiveness, with the latter being studies conducted in clinically representative settings with regular patients and clinicians (Hunsley and Lee 2007). An increasing number of studies on ICBT delivered in regular clinic have been published (Andersson and Hedman 2013), some of which are on depression. For example, effectiveness data have been published on the Dutch Interapy program (Ruwaard et al. 2012), the Swedish Internet psychiatry unit treatment (Hedman et al. 2014), and the Australian Sadness program (Williams and Andrews 2013). Data on the effectiveness of unguided treatments for depression have also been published (Leykin et al. 2014). In general, these studies indicate that the effects of ICBT interventions found in trials are comparable to those found in regular care.

In sum, the research to date clearly supports that guided ICBT is effective for depression and that the effects may be durable. It is possible that guided ICBT is as effective as face-to-face CBT, but it is important to remember that most studies have been on mild to moderate depression and not more severe forms. In addition, research suggests that guided ICBT is effective for different target groups, but most research has been on adults.

Inger had been depressed once in her 20s but is now a 45-year-old middle manager at a large company and has a family with two kids and a husband. She could never expect that she would be depressed again. During winter holiday, the family went skiing and she managed to fall and break her leg. In addition to staying away from work a few weeks, she did not get an expected promotion and finally her oldest son (16 years old) had problems at school and spent much time in front of the computer. All this eventually lead to her feeling depressed and increasingly more passive. For example, she ceased doing her regular exercise and avoided her friends. She only managed to focus on work and on her family and felt easily distracted and annoyed when things went against her. Family had noted her changed mood and the fact that she had problems with sleep. When she eventually went to the GP, she was recommended that she should seek help at the Internet psychiatry unit in Stockholm, Sweden. She lived in another city close by. To her surprise, she was offered a consultation with a psychiatrist within 2 weeks and then an interview with a clinical psychologist who described how guided ICBT would work if she decided to start treatment now. As she was not too severely depressed and had a university degree, she thought that this would be a good alternative for her as she did not need to take time off from work. She started with the treatment that lasted for 10 weeks with guidance from a psychologist at the unit (Erik) who supported her and gave feedback on homework assignments. Scheduling of activities was fairly easy for her but the most appreciated part of the treatment was to work with cognitive therapy techniques as she had always had tendencies toward negative thinking and found the registrations and homework assignments helpful. In addition, the advice given for sleep management was beneficial even if she fairly rapidly started to sleep better once she had control over her days and started to engage in rewarding activities (like getting back to exercise and seeing her friends). The last thing she did in the treatment was to do a relapse prevention plan. She was interviewed after the treatment and received feedback from the therapist. During the treatment, symptoms were monitored and she noted already after a few weeks that things were getting better. Afterward, she felt she had done a good job and that her therapist had supported her.

There is not much research on the cost-effectiveness of guided ICBT for depression, but there are some studies on unguided ICBT (de Graaf et al. 2008). However, Warmerdam and colleagues showed that both guided ICBT and problem solving therapy were cost-effective (Warmerdam et al. 2010), and results from a trial on real-time ICBT conducted in the United Kingdom showed similar findings (Hollinghurst et al. 2010). Moreover, cost data were reported in a trial on ICBT for older adults (Titov et al. 2015a). In a trial on depressive symptoms among employees, the intervention was partly cost-effective, but not for the employers (Geraedts et al. 2015). The literature is yet uncertain when it comes to the cost-effectiveness of guided ICBT for depression (Arnberg et al. 2014), and more studies are needed.

As mentioned previously in the chapter, effectiveness data are increasingly published and there are also reports from clinics like the MindSpot clinic in Australia with large series of patients (Titov et al. 2015b). It is hard to give an overview of how well guided ICBT for depression has been disseminated, in particular as not all who provide treatment publish research. Moreover, unguided programs like MoodGYM can be delivered by clinicians as an adjunct, and thus the treatment is blended with face-to-face services (Høifødt et al. 2013). In Sweden, guided ICBT for depression is offered in a few places (like the internet psychiatry unit at Karolinska University Hospital), but a national treatment platform has been developed which most likely will increase access to ICBT in Sweden. Moreover, in many places in the world, health care is insurance based, and programs from private companies like Deprexis in Germany are gradually being introduced as a treatment option (Meyer et al. 2015), and in the Netherlands, several programs are disseminated (e.g., Interapy). In the United Kingdom, Norway, and Sweden, tax-funded health care has been part of the implementation process. To conclude, the field of dissemination is highly dependent on legal and ethical considerations, but increasingly ICBT for depression is being incorporated into clinical guidelines.