Anxiety disorders, such as generalised anxiety disorder (GAD) and social anxiety disorder, are the most prevalent mental illnesses worldwide (Bandelow & Michaelis, 2015). However, approximately 80% of individuals with anxiety do not receive sufficient treatment (Kasteenpohja et al., 2016). Lack of qualified mental health practitioners and time restraints are some of the reasons that people are not receiving adequate treatment (Stefanopoulou et al., 2019).
We clearly need scaleable interventions that can help overcome these barriers. Digitally delivered interventions show promising potential to do just that, as they provide, for example, convenience, patient privacy and self-guidance (Stefanopoulou et al., 2019). Considering this, a recent study by Darin Pauley and colleagues (2021) examined the effectiveness of digital interventions across all anxiety disorders.
The following databases were used to identify relevant studies: PubMed, EMBASE, PsycINFO, and The Cochrane Central Register of Controlled Trials.
Inclusion criteria for experimental studies:
- aged 18 or older,
- diagnosis of any anxiety disorder based on DSM5,
- use of a guided or unguided digital intervention performed without requiring a therapist to be physically present or the prerequisite to participate outside of a personal environment of choice,
- randomised controlled trial design,
- use of an inactive control comparison group such as wait-list control,
- published in a peer-reviewed journal.
The quality and the risk of bias of studies were assessed using the Risk of Bias 2 tool. Data were analysed using a meta-analysis approach based on a random-effects pooling model, which assessed the effectiveness of digitally delivered interventions on anxiety symptoms.
47 studies were identified, resulting in 4,958 participants (2,808 treatment group and 2,150 control group) and 53 comparisons quantified in analysis.
The following anxiety disorders were represented in the sample:
- Social anxiety disorder (38% of comparisons)
- Panic disorder/agoraphobia (28% of comparisons)
- Generalised anxiety disorder (17% of comparisons)
- Mixed anxiety (17% of comparisons).
A waiting list control was employed in all studies. Cognitive-behavioural therapy was the most common digital intervention used in 90% of comparisons, and a guided treatment was offered in 79% of comparisons. Most studies were recruited from a community setting (92% of comparisons) and were largely conducted in Europe (70% of comparisons). The mean number of completed treatment sessions was 5.8, while treatment adherence was 55.9%.
Risk of bias
The risk of bias was rated as ‘low’ in 11% of study comparisons, ‘some concern’ in 64% of study comparisons, and ‘high’ in 25% of study comparisons. Missing outcome data and measurement of outcomes were the main areas of concern.
- The findings for the meta-analysis demonstrated that digital interventions have a large positive impact on anxiety symptoms, g = 0.80 [95% CI: 0.68 to 0.93]. Furthermore, confidence in this finding is enhanced as results showed no publication bias present.
- Moderate to large pooled effect sizes favoring digital interventions were found for subgroups:
- generalised anxiety disorder (g = 0.62)
- mixed anxiety samples (g = 0.68)
- panic disorder with or without agoraphobia (g = 1.08)
- social anxiety disorder (g = 0.76)
- No significant differences were found between guided (g = 0.84) and unguided (g = 0.64) treatment support, the mean number of treatment sessions, and treatment adherence.
This recent systematic review and meta-analysis suggests that digital interventions may be an effective treatment for anxiety disorders compared to wait-list controls. Moreover, digital interventions could be effectively facilitated by clinicians, non-clinicians, or self-administered as no difference was found between guided and unguided support. Digitally delivered interventions have the potential to address access-to-treatment challenges as they offer convenience, patient privacy and reduce the overall cost of care provision.
Strengths and limitations
The present meta-analysis quantified 53 comparisons, enhancing the conclusions’ power and precision. Also, a search strategy was adopted, which allowed the identification of studies investigating various anxiety disorders. This provides a comprehensive understanding of how effective digitally delivered interventions are for all anxiety disorders compared to other studies that have looked at specific mental illnesses (i.e., GAD). Furthermore, impressively, no publication bias was detected; suggesting that included studies are a representative sample of the available evidence on treatment effectiveness of digital interventions.
However, there are some limitations in the meta-analysis. All included studies compared digital interventions to wait-list controls. Research showed that participants in wait-list controls tend to improve less than other control conditions (Guidi et al., 2018), which could have consequently inflated the treatment effects of digital interventions. On a similar note, most studies provided minimal information on the nature of the wait-list condition, e.g., whether usual care was continued. These limit the generalisability of the findings for comparisons other than wait-list controls. The generalisability is further questioned as the samples were recruited largely from community settings and 96% were either European or Oceanian. Therefore, the applicability of these findings to clinical settings and other continents should be approached with caution.
Furthermore, the meta-analysis only included studies with adult samples. Anxiety disorders are highly prevalent among adolescents and young people, with 20% of young people experiencing these mental illnesses (Beesdo et al., 2009). Symptoms of anxiety largely interfere with school performance, the ability to form and maintain relationships and cognitive function, all of which are important attributes to carry into adulthood (Chartier et al., 2016). Therefore, future research should prioritise seeing the impact of digitally delivered interventions for young people, especially considering the amount adolescents use technological devices (Hagell, 2012). Furthermore, it would have been interesting to know the effect sizes of the relationship between digitally delivered interventions and anxiety symptoms across subgroups of adults. For instance, it is not clear how effective these interventions would be for older adults. Considering this population’s unique needs and preferences regarding technology (Andrews et al., 2019), a better understanding of this is necessary.
Implications for practice
The findings have important clinical implications. Considering the study demonstrates the effectiveness of digital interventions in treating anxiety disorders, it shows the need to adopt this treatment modality across health care settings. This will likely have a positive impact on both patients and the government. Regarding the latter, the National Director of Patients and Information stated that technological interventions could save the NHS approximately £10 billion (Castle-Clarke, 2015). Furthermore, it could be helpful for clients as it facilitates convenience, patient privacy and self-guidance (Stefanopoulou et al., 2019). Furthermore, this is particularly relevant during the current pandemic, enabling people to be provided with support without requiring in-person treatment.
Statement of interests
Pauley, D., Cuijpers, P., Papola, D., Miguel, C., & Karyotaki, E. (2021). Two decades of digital interventions for anxiety disorders: a systematic review and meta-analysis of treatment effectiveness. Psychological Medicine, 1-13.
Andrews, J. A., Brown, L. J., Hawley, M. S., & Astell, A. J. (2019). Older adults’ perspectives on using digital technology to maintain good mental health: interactive group study. Journal of Medical Internet Research, 21(2), e11694.
Bandelow, B., & Michaelis, S. (2015). Epidemiology of anxiety disorders in the 21st century. Dialogues in Clinical Neuroscience, 17(3), 327.
Beesdo, K., Knappe, S., & Pine, D. S. (2009). Anxiety and anxiety disorders in children and adolescents: developmental issues and implications for DSM-V. Psychiatric Clinics, 32(3), 483-524.
Castle-Clarke, S. (2015, June 30). Will technology sabe the NHS £10 billion? [Blog post]. Retrieved from https://www.nuffieldtrust.org.uk/news-item/will-technology-save-the-nhs-10-billion
Chartier M, Brownell M, MacWilliam L, Valdivia J, Nie Y, Ekuma O, Burchill C, Hu M, Rajotte L, Kulbaba C. The Mental Health of Manitoba’s Children. Winnipeg, MB. Manitoba Centre for Health Policy, Fall 2016.
Guidi, J., Brakemeier, E. L., Bockting, C. L., Cosci, F., Cuijpers, P., Jarrett, R. B., … & Fava, G. A. (2018). Methodological recommendations for trials of psychological interventions. Psychotherapy and psychosomatics, 87(5), 276-284.
Hagell, A. (2012). Health implications of new technology. Association for Young People’s Health. https://www.youngpeopleshealth.org.uk/wp-content/uploads/2015/07/New-technology.pdf
Kasteenpohja, T., Marttunen, M., Aalto-Setälä, T., Perälä, J., Saarni, S. I., & Suvisaari, J. (2016). Treatment adequacy of anxiety disorders among young adults in Finland. BMC Psychiatry, 16(1), 1-13.
Stefanopoulou, E., Lewis, D., Taylor, M., Broscombe, J., & Larkin, J. (2019). Digitally delivered psychological interventions for anxiety disorders: a comprehensive review. Psychiatric Quarterly, 90(1), 197-215.