Twenty-four studies were included in the review. Sample size ranged from 43 to 958 participants. Only four studies met all the combined quality assessment criteria specified by Chambless and Hollon and the Cochrane Collaboration. Most studies reported sample sizes larger than 25 participants per group. All studies clearly reported inclusion criteria and used reliable and well-validated measures. Five studies reported adequate allocation concealment. Most studies reported adequate blinding of assessors. Strategies to manage incomplete data were reported in 11 studies.
Chronic pain (nine studies): Three of eight studies using CBT reported statistically significant reductions in depression, anxiety, stress or distress compared with control, and two studies reported statistically significant reductions in catastrophising. Six out of nine studies reported statistically significant improvements in participants’ pain levels compared with control groups. One study found no difference between Internet CBT with or without telephone support on number of headaches.
Type 2 diabetes (five studies): There were no statistically significant improvements in distress outcomes between treatment and control groups. Three studies reported improvements in physical health measures.
Irritable bowel syndrome (four studies): Statistically significant improvements were reported for anxiety (two studies), quality of life (three studies), and depression (one study). Significant reductions were also reported in irritable bowel symptom severity (two studies), and abdominal pain (one study).
Tinnitus (two studies): One study reported treatment was superior to control groups in reducing tinnitus-distress, depression and annoyance. The follow-up study reported that benefits were maintained at 12 months follow-up. One study reported interventions were superior to control for tinnitus-specific physical outcomes, while one study found they were equivalent to the control group.
There were no significant differences between intervention and control groups for psychosocial outcomes for epilepsy (one study), fatigue (one study), or cancer (one study), however each of these reported some improvements in disease control outcomes.
Other results were reported.