ARTICLE: PDF OnlyMeta-Analysis of Randomized Controlled Trials of Cranial Electrostimulation Efficacy in Treating Selected Psychological and Physiological ConditionsKLAWANSKY, SIDNEY M.D., Ph.D.; YEUNG, ALBERT M.D., Sc.D.; BERKEY, CATHERINE Ph.D; SHAH, NIRAV B.S.; PHAN, HAI B.S.; CHALMERS, THOMAS C. M.D.Author Information Department of Health Policy and Management, Harvard School of Public Health, Boston, Massachusetts. Send reprint requests to Dr. Klawansky at Technology Assessment Group, Department of Health Policy and Management, Harvard School of Public Health, Room LL-7, 677 Huntington Avenue, Boston, Massachusetts 02115. The Journal of Nervous and Mental Disease: July 1995 - Volume 183 - Issue 7 - p 478-484 Buy Abstract To clarify the diverse published results of cranial electrostimulation (CES) efficacy, we conducted an extensive literature review that identified 18 of the most carefully conducted randomized controlled trials of CES versus sham treatment. For the 14 trials that had sufficient data, we used the techniques of meta-analysis to pool the published results of treating each of four conditions: anxiety (eight trials), brain dysfunction (two trials), headache (two trials), and insomnia (two trials). Because studies utilized different outcome measures, we used an effect size method to normalize measures which we then pooled across studies within each condition. The meta-analysis of anxiety showed CES to be significantly more effective than sham treatment (p < .05). Pooling did not affect results that were individually positive (headache and pain under anesthesia) or negative (brain dysfunction and insomnia). Most studies failed to report all data necessary for meta-analysis. Moreover, in all but two trials, the therapist was not blinded and knew which patients were receiving CES or sham treatment. We strongly recommend that future trials of CES report complete data and incorporate therapist blinding to avoid possible bias. © Williams & Wilkins 1995. All Rights Reserved.