Data extraction and statistical errors: A quantitative critique of Gumley, Braehler, and Macbeth (2014).

Abstract

OBJECTIVE While oxytocin has been identified as having therapeutic properties for schizophrenia, the emerging evidence has been mixed which has resulted in meta-analytic reviews. We identified several errors in one such meta-analysis. Here, we highlight these errors, demonstrate the conclusions were incorrect, and state the importance of this report. METHODS We reproduced the methods of Gumley, Braehler, and Macbeth (), including: outcomes (positive, negative, and total symptoms, as well as general psychopathology) and meta-analytic estimates for fixed and random effect models. RESULTS Whereas Gumley, Braehler, and Macbeth () reported oxytocin had significant effects on three of four outcomes, we show that all effects were non-significant. CONCLUSIONS Based on these null results, we hope this report encourages a re-evaluation of intranasal oxytocin as a treatment for schizophrenia.

DOI: 10.1111/bjc.12130

Cite this paper

@article{Williams2017DataEA, title={Data extraction and statistical errors: A quantitative critique of Gumley, Braehler, and Macbeth (2014).}, author={Donald R. Williams and Paul-Christian B{\"{u}rkner}, journal={The British journal of clinical psychology}, year={2017}, volume={56 2}, pages={208-211} }