Cranial electrostimulation for headache: meta-analysis.

Calls attention to errors in the meta-analysis of cranial electrostimulation for headaches by S. Klawansky et al (see record [rid]1996-02475-001[/rid]). First, Figure 3 lists incorrect citations for studies on headache. Second, in the conclusions, 2 studies on headache are listed as 'independently positive' but are included among a list of studies described as independently negative. Finally, Klawansky et al calculated effect size by calculating the posttreatment mean minus the pretreatment mean score, yet claim that none of the studies presented mean and standard errors of patient change. These errors cloud the conclusions of the meta-analysis.