Chronic severe pain, previously thought to be a uniquely adult condition, affects approximately 8% of children and has the potential to impact on all aspects of a child’s life.
Cognitive Behavioural Therapy (CBT) is a therapeutic intervention that draws on a range of cognitive, behavioural and emotional strategies and has been shown to be effective for reducing chronic pain in children and adolescents (1).
Despite the importance of alleviating the suffering of this population there is a lack of access to CBT in the UK child population (2). One solution is to develop computerized versions of CBT (cCBT).
Velleman and colleagues reviewed the effectiveness of cCBT for relieving chronic pain in children and adolescents.
The reviewers searched Pubmed (including Medline), Embase and Psychinfo to identify studies that met the following criteria:
- Participants up to 18 years of age
- Experienced continuous or recurrent pain
- Original research published in a peer reviewed journal
- Evaluated a cCBT intervention using standardized measures
- Written in English
Out of 2,730 articles that were identified by the search, 4 were ultimately included. Three trials compared cCBT intervention with a waiting list control group and one compared cCBT with a computerized educational programme. All the studies were Randomised Controlled Trials (RCTs).
The four studies are summarised below:
|Connelly 2006||37 children aged 7–12||Greater reduction in pain and headache related disability in cCBT group|
|Hicks 2006||47 children aged 9–16||Greater reduction in pain in cCBT group but no difference in quality of life|
|Palermo 2009||48 children aged 11-17||Greater reduction in pain, pain intensity and activity limitations in cCBT group|
|Trautmann 2008||47 children aged 9–16||Greater reduction in pain and pain catastrophizing in cCBT group|
All studies found that cCBT was more effective than the control in reducing pain intensity and these gains were maintained at follow-up (3 and 6 months later).
The difference between cCBT and the control were statistically significant, meaning that the difference is very unlikely to be due to chance. We can be confident that the findings weren’t a fluke but statistical significance doesn’t tell us about the size of the difference.
Effect sizes can tell us how much of a difference the intervention made overall. In this case, the effect sizes for individual studies ranged from small to medium. Therefore, cCBT made a difference but not a big difference. In addition, one study found that there was no difference in reported quality of life between cCBT and the control group at any of the time points.
The authors also measured the clinical significance of cCBT. Clinical significance refers to the importance of practical and experiential change which affects the quality of day to day life. These studies based their clinical significance on a 50% or more reduction in frequency, duration and intensity of pain between the start and end of the study. They found that those in the cCBT group were between 4 and 10 times more likely to achieve this level of improvement than those who did not. So even though the effect size metrics would be regarded as small to medium, computerized CBT was a strong predictor of clinically important improvements in pain and pain related disability.
Finally, the studies reported on the acceptability of the intervention, meaning how well the intervention was liked and whether it was perceived as helpful. Overall, these studies reported good acceptability of cCBT for children. They liked being able to do ‘treatment from home’, being ‘better able to manage pain’ and having ‘flexibility’ in when to use the intervention. Children didn’t think ‘having parents participate with treatment’ was as important. Parents thought that contact with therapists was important. Lack of contact with a therapist may be an issue for some participants. A more recent study (3) found that paediatricians would prefer face to face interventions to computerised interventions for children’s chronic pain too.
Strengths and limitations
The authors have done a good job in offering a thorough and appropriate review of the limited evidence base for cCBT for chronic pain in children. However, they themselves acknowledged that the searches may have been more comprehensive if they had been able to include grey literature and non-English language studies.
Grey literature is anything not published in peer-reviewed journals, for example government reports or reports by charities who directly fund evaluation of interventions. By looking at only published sources, as this review did, it may suffer from a ‘publication bias’ where certain types of study may be more likely to be published than others, such as those finding positive results.
It is clear that all studies are not created equal!
Studies varied in design and outcome measures but all were limited as they were small scale pilot studies, did not compare cCBT to face-to-face CBT and had no follow ups after 6 months. Crucially, they included some level of therapist contact either via a webpage, email or telephone. This interaction with a therapist means that the cCBT was not solely computerised. There was an element of a therapeutic relationship. The only study that had no therapist contact was the study that found no effect for quality of life. However, with only four studies very little can be concluded from this.
The limited evidence base and variation in study design means that further research is required (this time using standardised design please) to fully measure the success of cCBT, compared to both no intervention and to face to face CBT therapy, no therapist contact and some therapist contact. The following are areas which may warrant further investigation:
- Appropriate age range for the intervention and the level of parent involvement required
- Level of pain severity with which cCBT has the most success
- Optimisation of the duration and frequency of the programme
- Quantification of the possible reduction in cost to the service
- Length of waiting time for therapy
- Uptake and completion of the programme
The big picture!
The authors report that these results provide preliminary evidence that cCBT is useful in the treatment of pain in children and adolescents. They conclude that it is a cost-effective and accessible intervention because it uses less therapist time and is available in the child’s home. They also report that it can be seen as helpful and likeable by children and parents using it.
However… whilst this conclusion on face value seems very exciting and may tempt us into rushing to establish cCBT within our practice areas post haste; we do not know whether the reduction in pain intensity can be attributed to cCBT without first comparing it to face-to-face CBT and cCBT controlling for the amount of therapist contact. This is particularly relevant given that parents felt that therapist contact was important.
So can computers replace therapists? This elf’s answer is a tentative maybe.
The results show that cCBt can be effective in reducing chronic pain for children and adolescents, but such interventions may still require frequent therapist contact. We still don’t know how much therapist contact is needed to maximise effectiveness and whether cCBT can be as effective as face-to-face CBT. Such research is important, as cCBt has the potential to improve accessibility of CBT and increase therapists caseload capacity.
Velleman, S, Stallard, P, Richardson, T. (2010). A review and meta-analysis of computerized cognitive behaviour therapy for the treatment of pain in children and adolescents. Child: care, health and development, 36, 4, 465-472.
1. Eccleston, C., Morley, S., Williams, A., Yorke, L. & Mastroyannopoulou, K. (2002) Systematic review of randomised controlled trials of psychological therapy for chronic pain in children and adolescent, with a subset meta-analysis of pain relief. Pain, 99, 157–165.
2. Stallard, P., Udwin, O., Goddard, M. & Hibbert, S. (2007) The availability of Cognitive Behaviour Therapy within specialist child and adolescent mental health services (CAMHS): a national survey. Behavioural and Cognitive Psychotherapy, 35, 501–505.