Add on iCBT: weak evidence of modest benefits in depression and anxiety

The Mental Health Foundation reports that 1 in 6 people likely experienced mental health symptoms in the last week, and 1 in 4 people may experience a mental health disorder at some point in their lives. However, news articles regularly remind us that waiting lists prevent many from accessing the care that they need.

Psychological treatment delivered via the internet, including the use of internet-delivered cognitive behavioural therapy (iCBT), could improve accessibility to treatment.

With outcomes that are comparable to face-to-face therapies (Andrews et al., 2018), meta-analyses have found that the use of iCBT in comparison to waitlists may lead to reductions in anxiety and depression (Andrews et al., 2018) and symptoms of Post-Traumatic Stress Disorder (PTSD; Carlbring et al., 2018). iCBT can be as effective as face-to-face therapy (Andrews et al., 2018) and more effective than control conditions (Carlbring et al., 2018). Similar findings advocating for the effectiveness of iCBT have also been found in the context of Obsessive-Compulsive Disorder (OCD) and social anxiety (Andersson et al., 2015; Titov et al., 2009).

However, it is unclear as to what the advantages of iCBT may be in addition to treatment as usual (TAU), meaning an up-to-date examination is required to inform clinical decision-making. The review by Jonsson et al. (2023) discussed in this blog examined the advantages and risks of internet-delivered psychological treatments delivered as add-ons to TAU for a variety of presentations across the lifespan.

iCBT can be effective in reducing symptoms of depression, anxiety, OCD and PTSD – could its accessibility help reach more people?

Methods

14 databases were searched for studies that included:

• Participants of all ages, diagnosed with either depression, anxiety, OCD and related disorders, or trauma- and stress-related disorders;
• Self-guided, internet-delivered CBT (including mindfulness- and acceptance-based treatments), psychodynamic psychotherapy, or interpersonal psychotherapy as an add-on to TAU;
• TAU as the control;
• Randomised controlled trial (RCT) design;
• Reported outcomes such as clinical variables, quality of life, functioning, adverse effects, healthcare consumption, and adherence;
• Published in a peer-reviewed journal in English, Swedish, Norwegian, or Danish.

All identified articles were independently screened by two authors at title/abstract and full-text stage. Data from the included studies was extracted by one author, with a second author checking for integrity.

A range of meta-analyses were carried out using random effects models. If different scales were used, the authors calculated effect sizes into standardised mean difference (SMD; Hedges’ g); conversely, if the same scale was used, both SMD and the scale’s original effect size metric were presented. Binary outcomes (i.e., yes/no) were reported as either odds ratios or risk differences.

Risk of bias was assessed through the Cochrane risk-of-bias tool for randomised trials (Higgins et al., 2011).  11 of 15 studies were assessed as having a high risk of bias, and 4 were assessed as having some concerns.

Certainty of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) tool (Balshem et al., 2011), and was found to be very low.

Results

The review included 16 articles on 15 trials. Out of these articles, 9 included participants with depressive disorders, 4 included participants with anxiety or depression, and 2 included participants with PTSD or ‘unspecified trauma’. Sample sizes ranged from 43 to 647 participants (N = 3,339), with most participants being female in their 30’s or 40’s.

All add-on internet-delivered interventions were CBT-based, with some including mindfulness-based practices. The length of interventions varied, with programs ranging from 5 to 24 sessions over a period of 5 to 26 weeks. TAU usually consisted of pharmacological or psychological treatments.

Results from the meta-analysis found:

• From 12 trials, a small overall effect size in favour of iCBT as an add-on in studies of depressive disorders at post-treatment (g = -0.23, 95% CI -0.37 to -0.09, p = .001). The findings were statistically significant at both short-term follow-up (<6 months; g = -0.19, 95% CI -0.33 to -0.05, p = .007) and long-term follow-up (≥6 months; g = -0.22, 95% CI -0.39 to -0.04, p = .01).
• From 3 trials, a small overall effect in favour of iCBT as an add-on in studies of anxiety disorders at post-treatment (g = -0.28, 95% CI -0.44 to -0.11, p < .001). The findings were significant at short-term follow-up (<6 months; g = -0.25, 95% CI -0.49 to -0.01, p = .04) but not at long-term follow-up (≥6 months; g = -0.09, 95% CI -0.29 to 0.11, p = .37).
• From 3 trials, no significant effects for iCBT as an add-on to TAU compared to TAU for PTSD symptoms at post-treatment, 1-month follow-up, or 3-month follow-up.
• In terms of treatment response, 3 trials showed that iCBT as an add-on had a better response rate for depression compared to TAU (OR = 2.46, 95% CI 1.31 to 4.64, p = .005)
• 4 trials showed that iCBT as an add-on compared to TAU had an increased remission rate for adults with depression at post-treatment (OR = 1.70, 95% CI 1.19 to 2.42, p = .003), which was maintained at short-term follow-up (<6 months; OR = 2.18, 95% CI 1.55 to 3.06, p < .001).
• The effects of iCBT as an add-on on quality of life in depression were small, with positive post-treatment results (g = 0.15, 95% CI 0.04 to 0.26, p = .006)but inconsistent results at short-term (<6 months; g = 0.13, 95% CI -0.14 to 0.39 p = .34) and long-term follow-ups (≥6 months; g = 0.14, 95% CI -0.06 to 0.35, p = .18).
• 2 trials examined functioning in iCBT as an add-on for depression treatment, but no significant differences were found.

Meta-analyses found small but significant effects for iCBT as an add-on to treatment for depression and anxiety at post-treatment, but not for PTSD at any timepoints.

Conclusions

The authors concluded that:

This systematic review suggests that iCBT as an add-on to usual care for mild to moderate depression in adults may have a small incremental effect, at least in the short-term perspective.

This effect is comparable to TAU for adult depression.

The authors also found a small effect for anxiety.  However, the small number of trials investigating iCBT for anxiety and PTSD, the lack of trials for iCBT for OCD, and the lack of studies on children, adolescents, and the elderly was raised as a concern. It was highlighted that this will likely improve in the future, with the rise of studies using TAU as a comparator for internet-delivered therapy.

We are only just starting to understand how digital technology can improve effectiveness and access to mental health care – what could our understanding look like in 5 years time?

Strengths and limitations

This study addressed a research question which is timely and relevant to mental health practice across a variety of settings. The methodology was robust and reliable, with the review incorporating only peer-reviewed RCTs and following standards of good practice.

Only studies using validated measures were included, contributing to interpretability of the measured concepts and comparability with other studies using the same measures. The inclusion of articles in four different languages will have contributed to the scope, as many systematic reviews only include English-language articles. Sensitivity analyses were carried out where applicable, taking into account age groups, type of treatment, models of delivery, degree of support, design and risk of bias.

However, there were some limitations, such as the review not taking into consideration potential adverse effects of iCBT. Furthermore, comparability, interpretation and generalisation of results is limited because of characteristics of the included samples and study designs. For example:

• Included studies were conducted mainly in primary (health) care settings, limiting the applicability of these results to more intensive types of care and likely more complex symptomatology.
• Sample sizes of the studies varied widely (n = 43 to n = 647), with a restricted age range (mid-30s to mid-40s) and an over-representation of the female gender in most symptom categories (except PTSD).
• A high percentage of participants were educated to a higher-education level (30% to 100% across samples).
• Treatment duration varied considerably (from 5 to 24 modules), with not all studies providing information on adherence to iCBT. Additionally, rates of completion for all modules were inconsistent across studies (13% to 86%).
• TAU could consist of pharmacological or psychological treatment, with large differences in rates of participants on medication across studies.

It is unclear how these differences may have affected outcomes. However, it is also important to keep in mind that iCBT may look different across different mental healthcare providers. There are also currently no requirements for standardisation of treatments, so comparability of different studies using iCBT may in general have its limits.

iCBT may be an effective add-on to existing treatments, but more evidence is needed as the limitations of the studies included in this review prevent generalisability to all service users.

Implications for practice

The review showed that robust evidence for the effectiveness of iCBT as an add-on to TAU is still limited, with results not necessarily generalisable to people from different backgrounds based on age, gender, education, and mental healthcare settings. Clearly more research is needed, as remotely delivered, self-guided iCBT has the promise to increase accessibility and be more cost-effective.

Despite the lack of evidence for iCBT in certain groups, clinical practitioners working with online self-help modules, like the authors of this blog, can provide anecdotal evidence for its engagement and efficacy.

Although stand-alone iCBT has substantial research support, many clients may still want to see a therapist. For those individuals, ‘blended treatments’ (such as offered by Mindler UK, who employ both authors) that combine remotely delivered therapy with self-guided iCBT may be a good option.

Our experience is that clients enjoy the flexibility and choice of iCBT as an add-on to therapy, with the use of an app meaning that all resources are stored in one place, making accessing and checking the materials easier. As evidence suggests that clients may only retain about 15% of the content from their therapy session (Huppert et al., 2001) and may only recall an average of 65% of what was spoken about in therapy (Boswell et al., 2007), it is important that they can easily revisit material.

However, it is important to note that iCBT may not be suitable for individuals that are not comfortable using technology, either because of lack of familiarity, lack of access or because they prefer face-to-face therapy (Kaltenthaler et al. 2006). In addition, concerns around anonymisation and data safety can be deterrents to online treatments, whether it is iCBT or remotely delivered therapy sessions. Even though clinicians are bound to adhere to the same data protection policies online as they would in person, it is important to be aware of data protection concerns. We cannot emphasise enough that client data needs to be kept secure to prevent a breach of confidentiality and subsequently trust.

*With thanks to Dr Victoria Kinnear for her valuable contributions to this section.

Data privacy and confidentiality should be guaranteed at all times to make sure clients can feel safe.

Statement of interest

Both authors of this blog are psychologists that work remotely for an online therapy platform called Mindler UK, whose primary aim when it was founded was to improve accessibility to treatment by using iCBT and reducing session times. Mindler UK is part of Mindler, with offices in Sweden, Denmark, and the Netherlands.

Links

Primary paper

Jonsson, U., Linton, S. J., Ybrandt, H., Ringborg, A., Leander, L., Moberg, K., Hultcrantz, M., & Arnberg, F. K. (2023). Internet-delivered psychological treatment as an add-on to treatment as usual for common mental disorders: A systematic review with meta-analysis of randomized trials. Journal of Affective Disorders, 322, 221–234.

Other references

Andersson, E., Hedman, E., Enander, J., Radu Djurfeldt, D., Ljótsson, B., Andersson, G., & Rück, C. (2015). Internet-based cognitive behavior therapy for obsessive–compulsive disorder: A randomized controlled trial. Psychotherapy and Psychosomatics, 84(6), 342-352.

Andrews, G., Basu, A., Cuijpers, P., Craske, M. G., McEvoy, P., English, C. L., & Newby, J. M. (2018). Computer therapy for the anxiety and depression disorders is effective, acceptable and practical health care: an updated meta-analysis. Journal of Anxiety Disorders, 55, 70-78.

Balshem, H., Helfand, M., Schünemann, H. J., Oxman, A. D., Kunz, R., Brozek, J., Vist, G. E., Falck-Ytter, Y., Meerpohl, J., Norris, S., & Guyatt, G. H. (2011). GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology, 64(4), 401–406.

Boswell, J. F., Castonguay, L. G., & Wasserman, R. H. (2010). Effects of psychotherapy training and intervention use on session outcome. Journal of Consulting and Clinical Psychology, 78(5), 717.

Carlbring, P., Andersson, G., Cuijpers, P., Riper, H., & Hedman-Lagerlöf, E. (2018). Internet-based vs. face-to-face cognitive behavior therapy for psychiatric and somatic disorders: An updated systematic review and meta-analysis. Cognitive Behaviour Therapy, 47(1), 1-18.

Higgins, J. P., Altman, D. G., Gøtzsche, P. C., Jüni, P., Moher, D., Oxman, A. D., Savovic, J., Schulz, K. F., Weeks, L., Sterne, J. A., Cochrane Bias Methods Group, & Cochrane Statistical Methods Group (2011). The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ, 343, d5928.

Huppert, J. D., Bufka, L. F., Barlow, D. H., Gorman, J. M., & Shear, M. K. (2001). Therapists, therapist variables, and cognitive-behavioral therapy outcome in a multicenter trial for panic disorder. Journal of Consulting and Clinical Psychology, 69(5), 747-755.

Kaltenthaler, E., Shackley, P., Stevens, K., Beverley, C., Parry, G., & Chilcott, J. (2006). Computerised cognitive behaviour therapy for depression and anxiety update: a systematic review and economic evaluation. Health Technology Assessment, 10(33).

Titov, N., Andrews, G., Schwencke, G., Robinson, E., Peters, L., Spence, J., & Choi, I. (2009). Randomized controlled trial of internet cognitive behavioural treatment for social phobia with and without motivational enhancement strategies. Australian and New Zealand Journal of Psychiatry, 43(9), 829-835.

Photo credits