To the Editor: Although the direct meta-analysis methods that Dr Anothaisintawee and colleagues used in their review of the management of chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS)1 are mostly appropriate, we are concerned about the alternation between a random-effects and a fixed-effects model. In meta-analysis theory, when the heterogeneity of the studies is sufficiently low, the 2 models should produce the same result.2 However, if there is a significant difference between the results using fixed- and random-effects models, then a random-effects model should be used because it is more conservative. The tests for heterogeneity are notoriously insensitive.2
Jackson JL, Cohen JM, Dimitrakoff J. Treatment of Chronic Prostatitis/Chronic Pelvic Pain Syndrome. JAMA. 2011;305(13):1298–1299. doi:10.1001/jama.2011.399
Artificial Intelligence Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.