Focus on Alternative and Complementary Therapies
www.pharmpress.com/fact
Focus Alternat Complement Ther©2005 Pharmaceutical Press
Focus Altern Complement Ther 2005; 10: 34–5
Regression-to-the-mean is a statistical phenomenon that occurs when measurements are repeatedly taken from a selected population. It is believed to be a major constituent of the placebo effect, often more important than patient expectations. We modelled the selection process in homoeopathic outcome studies in order to discriminate between regression-to-the-mean and treatment effects.
Reanalysis of recently published outcome data in homoeopathy (Steinsbekk A, Lüdtke R. Patients’ assessments of the effectiveness of homoeopathic care in Norway: a prospective observational multicentre outcome study. Homeopathy 2005; 94: 10–16. Becker-Witt C, Lüdtke R, Weber K, Willich SN. The effects of homoeopathic therapy on health-related quality of life. Focus Altern Complement Ther 2003; 8: 125) with a modified Mee-Chua algorithm (Mee RW, Chua TC. Regression-toward-the-mean and the paired sample t-test. Am Statistician 1991; 45: 39–42) and a newly developed approach for monotone increasing selection processes.
Effects of the homoeopathic treatment can be found for improvements in quality of life (SF-36 mental health: P < 0.01; SF-36 physical health, P < 0.001), general well-being (100 mm VAS, P < 0.01) and the major complaint’s impact on daily living (100 mm VAS, P < 0.1) even when the analysis accounts for strong regression-to-the-mean effects.
The success rates reported in homoeopathic outcome studies cannot solely be explained by regression-to-the-mean effects.
We thank A Steinsbekk for providing his data.
