Can CBT impact MS-related pain?After seeing the beneficial effects of CBT for chronic pain in the general population, Dr. Kerns and colleagues set out to assess whether the intervention could improve pain outcomes in patients with MS, randomly assigning 20 patients with moderate MS-related pain to receive 12 weekly sessions of either CBT or MS-related education in addition to standard care with follow-up at 15 weeks. Overall, the participants were mostly white (75%), male (60%) patients with relapsing remitting MS (70%). The cohort had a mean age of 52.6 years and a disease duration of 13.25 years. Nearly all participants reported pain in their legs or feet (95%) or their lower back (80%). Pain was reported in a mean of 4.25 different locations.
Analyses were approached as intention to treat. Differences between the two treatment groups’ demographic and disease-related characteristics were assessed using T tests for continuous data and chi-square tests for categorical data. Changes on the outcome measures were evaluated using a 2 (CBT/standard care and education/standard care) × 2 (before and after treatment) factorial design. Treatment credibility, treatment satisfaction, and behavioral goal accomplishment between the two conditions were compared using T tests.
The investigators examined participants’ pain severity, pain interference, depressive symptoms, and treatment satisfaction at 15 weeks, compared with baseline. At baseline, there was no difference in composite pain scores between groups (t18, –0.30; P = .767), but a significantly lower severity of pain in the CBT group (t18, –3.42; P = .003). Participants in the CBT group also had lower depressive symptom severity scores at baseline, compared with the educational group (t18, –2.04; P = .056).
Time carried a significant effect for pain severity (F1,18, 4.61; P = .046) and for pain interference (F1,18, 4.63; P = .045) but there was not a statistically significant effect for time × treatment interaction for either pain severity (F1,18, 0.61; P = .444) or pain interference (F1,18, 0.06; P = .813) at 15 weeks. In the CBT group, there was a significantly different mean behavioral goal accomplishment rating (1.27; 95% confidence interval, 0.87-1.66) from zero, compared with the educational group (0.65; 95% CI, –0.35 to 1.65), but there were no significant differences between the CBT and educational groups (t9,37, 1.36; P = .207).
Regarding depressive symptom change, the CBT group had a lower but not statistically significant difference in symptom severity, compared with the educational group (t18, –2.04; P = .056). Time also carried a significant effect for depression symptom severity, but the time × treatment effect was not statistically significant (F1,18, 1.24; P = .280). Participants in the CBT group rated their treatment credibility and expectations for improving higher than the educational group (t15, 1.52; P = .150), but there was no significant difference between groups (t16, –0.04; P = .967), with high total satisfaction in both CBT and educational groups.
The investigators attributed the underpowered nature of the study to the low number of patients who were ultimately enrolled, and said that a low engagement rate, inability of a neurologist to confirm an MS diagnosis, and failure to meet the study’s pain thresholds contributed.
Top image credit: Photographee.eu/Shutterstock
Image credit: ESB Professional/Shutterstock