Computerised Cognitive Behaviour Therapy (CCBT) is an effective way of delivering Cognitive Behaviour Therapy (CBT) for people experiencing depression (Wright et al. 2019). Clinician-supported cCBT can be as effective as antidepressant medication and some other forms of treatment, but non-supported cCBT has smaller effects (Wright et al. 2019). A key benefit of clinician-supported cCBT is that it uses less clinician time than face to face CBT, but with comparable benefits (Carlbring et al. 2018). This means it can be a way to scale up provision of CBT, providing treatment for more patients per clinician.
Given the potential for clinician-supported cCBT to be scaled up and reach more people, it’s important to think about who it reaches, because when considering technology-based treatments, ‘digital inclusion’ is paramount. Digital inclusion is defined as a combination of a) the skills to use digital devices, b) access to the internet, and c) services being designed to meet all users’ needs (including users of assistive technology) (NHS Digital 2022). Digital exclusion can lead to inequalities in who’s able to benefit from tech-based healthcare. Within the UK, people who are more likely to be digitally excluded are also likely to be in one of the following groups: older; in lower income groups; without a job; in social housing; with disabilities; with lower educational attainment; living in rural areas (NHS Digital 2022).
This study looks at whether cCBT works amongst people recruited through Primary Care in the United States (US), including a high proportion of people who are economically deprived. It provides results of a randomised controlled trial of ‘Good Days Ahead’ (GDA), an online clinician-supported cCBT programme (see mindstreet.com). One of its key aims was to explore the effectiveness of cCBT, and to consider its suitability in US primary care settings of people with lower income, lower educational attainment and limited internet access.
The cCBT programme trialled (GDA) is a multimedia-based intervention that includes video and audio alongside text, and has already been subject to testing (Wright et al. 2002; Wright et al. 2005; Kim et al. 2014; Thase et al. 2018).
The cCBT intervention group received the intervention alongside TAU (cCBT plus TAU). Participants had access to:
- the cCBT programme (which has 9 lessons) over a period of 12 weeks
- Up to 12 sessions of telephone support with a mental health clinician (roughly 20 minutes per week, texts and email also permitted).
The control group received treatment as usual.
Participants were primary care patients aged 18+, recruited between 2016 and 2019 from clinical practices at the University of Louisville; all of the practices were in urban settings except for one rural.
The study loaned low-cost laptops with internet access to participants who requested them (9.7% of participants, or 17 out of 175), to enable inclusion of people without access to computers.
Participants were screened using a range of tests including the Wide Range Achievement Test (WRAT, for reading age). For the first 6 months of the study, a reading age of less than 14 years was used as an exclusion criteria, however this was removed when it led to the exclusion of nearly one sixth of potential participants who otherwise wished to participate. Other exclusion criteria included self-report of inability to read English, significant suicidal risk, and diagnosis of psychotic disorder or bipolar disorder.
The primary outcome measure was depression, secondary outcomes were negative self-statements, anxiety and quality of life.
Outcomes were measured at 12 weeks (completion of the intervention period); and 3 and 6 months after completion of the intervention period.
The study recruited 175 participants, 80 allocated to TAU and 95 allocated to intervention. 74.6% had less than a college education (for the UK context, this means not having education beyond secondary school or equivalent). Of the 82% that reported income, 61.5% reported annual income of less than $30,000 (circa £25,000 in UK at time of writing, late July 2022). Reading level was grade 11.7 (circa 17 years old). 10.9% had reading levels below ninth grade (14-15 years)
- Both groups experienced an improvement in depression but the cCBT group showed a larger improvement at all time points.
- The rates of response and remission were significantly higher in the cCBT group at all time points. For example, after 12 weeks: response rates were 58.4% for cCBT, and 33.1% for TAU; remission rates were 27.3% and 12.0% respectively.
- There were significant differences in effect sizes for all secondary outcomes at all time points, except anxiety at 6 months. At 12 weeks, patients in the cCBT group had lower negative self-statements, lower anxiety and higher quality of life
- The treatment completion rate for cCBT was 74.7%. Dropout rates were 22.1% for cCBT and 30.0% for TAU
- The number of cCBT modules completed was associated with greater symptomatic improvement
- Out of a maximum of 12 possible sessions, the mean number of sessions was 8.83. The mean amount of telephone support time was 17.4 minutes, with mean amount of time for emails and texts as 0.24 minutes and 0.90 minutes respectively (total mean time 18.54 minutes per session, and 163.7 minutes for the whole course of treatment).
The study shows that clinician-supported cCBT is effective at improving depression in the population studied, and effects were maintained at the 3-and 6-month follow ups. In addition, it has also been shown to be effective at improving anxiety, negative self-statements and quality of life. The authors conclude:
“Together, these results indicate that cCBT for depression has widespread favorable outcomes.” (p.8)
The unique contribution of these findings is inclusivity of people with lower educational attainment, reading proficiency, and lower incomes.
Strengths and limitations
The study contributes to a body of evidence that is already strong. The importance of these findings lies in the fact the population studied has not always been included in trials of cCBT, namely, it was inclusive of people more likely to be digitally excluded and from more economically deprived settings. The provision of loaned computers helped to overcome this. Whilst the authors point out there weren’t sufficient numbers to assess the relative effectiveness of cCBT for people with greater disadvantage, these are areas for further investigation. There are many ways in which digital exclusion can occur, alongside other forms of exclusion and disadvantage not included here. Whilst the findings are promising, there’s still much work around inclusivity and disadvantage to be done.
The results relate to the GDA programme, which includes video, audio and is interactive, suggesting it may be particularly engaging and accessible. This might have helped for people with a lower reading age, for example, with information presented via video and audio clips instead of text. As not all cCBT programmes are created equal, it’s worth being cautious about how applicable the results are for cCBT that may be less engaging and accessible.
The study was largely carried out in urban settings, with only one rural site. Loaning computers helped in this study, but it’s not clear how that would play out in rural settings with less infrastructure relating to internet connectivity. The authors point out that developments in technology (a GDA app for use on smartphones) means potential access may have increased since this study took place. This is positive, but routes to providing internet access (or support) will still be required where ownership or use of a smartphone may not be viable.
Implications for practice
The average amount of clinician time per participant was less than 3 hours, suggesting it brings the potential for access to psychological therapy in a relatively time-efficient manner. Clinician-supported cCBT can be effective in clinical populations with greater economic deprivation and less access to the internet. This was enabled by the loan of computers, and it’s important to think how services might achieve this locally. The study has implications for those providing care (including commissioners of services) who should consider issues of digital exclusion – and how to overcome it – as part of their offering.
Statement of interests
I have no competing interests relating to the study. I work at NIHR MindTech MedTech Co-operative, and do research relating to technologies for mental health.
Wright JH, Owen J, Eells TD, Antle B, Bishop LB, Girdler R, Harris LM, Wright RB, Wells MJ, Gopalraj R, Pendleton ME, Ali S. Effect of Computer-Assisted Cognitive Behavior Therapy vs Usual Care on Depression Among Adults in Primary Care: A Randomized Clinical Trial. JAMA Netw Open. 2022 Feb 1;5(2):e2146716. doi: 10.1001/jamanetworkopen.2021.46716. PMID: 35142833; PMCID: PMC8832170.
Carlbring, P., Andersson, G., Cuijpers, P., Riper, H. & Hedman-Lagerlof, 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, 1-18.
Good Days Ahead. Mindstreet. Accessed July 21, 2022. https://www.mindstreet.com
Kim DR, Hantsoo L, Thase ME, et al: Computer-assisted cognitive behavioral therapy for pregnant women with major depressive disorder. J Womens Health (Larchmt) 2014; 23:842–848
NHS Digital 2022 Available at: https://digital.nhs.uk/about-nhs-digital/our-work/digital-inclusion/what-digital-inclusion-is (accessed 1st August 2022)
Thase ME, Wright JH, Eells TD, et al. Improving the efficiency of psychotherapy for depression: computer assisted versus standard CBT. Am J Psychiatry. 2018;175(3):242-250. Doi: 10.1176/appi.ajp.2017.17010089
Wright JH,Wright AS, Salmon P, et al. Development and initial testing of a multimedia program for computer assisted cognitive therapy. Am J Psychother. 2002;56(1):76-86. doi:10.1176/appi.psychotherapy.2002.56.1.76
Wright JH, Wright AS, Albano AM, et al. Computer-assisted cognitive therapy for depression: maintaining efficacy while reducing therapist time. Am J Psychiatry. 2005;162(6):1158-1164. doi:10.1176/appi.ajp.162.6.1158