Nestoriuc, Rief, & Martin (2008)* conducted a meta-analysis of biofeedback applications for tension-type headache (TTH). A meta-analysis combines data from published studies that meet inclusion criteria to increase statistical power and the ability to detect a statistically meaningful treatment effect, often expressed as an “effect size.” The researchers provide an overview of TTH and report that episodic (versus chronic) TTH is the most prevalent type of headache in adults, adolescents, and children with a 1 year prevalence rate of 38%. The researchers further report that pharmacotherapy (anti-depressants and analgesics) is the most common form of treatment, but that medications are only moderately effective and can have significant side effects. Thus, behavioral interventions, including biofeedback, may represent a promising alternative or complimentary treatment if demonstrated to be effective.
One important distinction must first be made. EEG biofeedback, better known as neurofeedback, has been frequently discussed at BMED Report. The current TTH study mostly involves peripheral biofeedback. The basic principles of peripheral biofeedback are very similar to neurofeedback except that instead of the patient monitoring his or her EEG (in neurofeedback), the patient receives feedback and reinforcement of their body signals. Biofeedback can include a number of different measures, such as heart rate variability (HRV), skin temperature, electromyography (EMG; muscle electricity), and electrodermal response (sweat). These signals are indirect measures of autonomic nervous system activity and/or muscular activity. Similar to neurofeedback, the biofeedback therapist encourages certain desired states with visual and auditory reinforcements.
53 studies involving 1532 patients, ages 10.3 to 66.7, with TTH were identified for inclusion in this meta-analysis. EMG biofeedback was the most frequently used modality across all studies; however, studies using temperature (2), electrodermal (1), and EEG (1) biofeedback, as well as combined relaxation (9) were also included. Please review the original article for a more comprehensive overview of inclusion criteria and descriptions of each study.
An overall medium-to-large effect size of .73 was found across all 53 studies for pre/post comparisons of symptom improvement (or lack thereof) following biofeedback treatment. Statistically significant medium-to-large effect sizes were reported for headache reduction (frequency, intensity, duration, and headache index). Interestingly, patient anxiety, depression, and self-efficacy also significantly improved across 19 studies. Biofeedback treatment compared no-treatment or placebo control groups again produced statistically significant improvements with medium-to-large effect sizes (.81, .50 respectively). Biofeedback also improved patient outcome over and above relaxation modalities as evidenced a by small, but statistically significant, effect size (.20). Combined EMG biofeedback and relaxation therapy were the most effective treatment for TTH. Children and adolescents had the best results, though adults did very well too. Adults with chronic (long term) TTH actually fared better than those with acute (short-term) TTH. Geriatric patients did not realize significant treatment effects, but only 2 groups comprising a total of 12 participants were included in the analysis. Importantly, treatment gains were found to persist for years at follow up, with an average follow up period of 15 months.
Unfortunately, only a handful of studies were located to allow for adequate comparison of biofeedback to pharmacotherapy and traditional behavioral/talk therapies; however, researchers noted that in 2 of the 3 studies biofeedback treatment was favored over pharmacotherapy with large effect sizes.
Remember that a meta-analysis is not a “cause and effect” experiment. Nonetheless, it does provide a statistically valid and reliable way to make standardized comparisons across studies (subject to statistician error and bias of course), and provides reasonable estimates of a treatment effectiveness in a number of different settings with different populations.
In summary, the results of this meta-analysis suggest that peripheral biofeedback may represent an efficacious treatment of TTH for many people.
*Nestoriuc, Y., Rief, W., Martin, A. (2008). Meta-analysis of biofeedback for tension-type headache: Efficacy, specificity, and treatment moderators. Journal of Consulting and Clinical Psychology, 76(3), 379-396.