Erratum: Rilonacept for colchicine-resistant or -intolerant familial mediterranean fever ((Annals of Internal Medicine (2012) 157 533-541))

摘要

A recent article (1) had several errors in the results that were caused by errors in the published WinBUGS code. Page 533, Results section of abstract, left column, last 3 lines: ". . . the rilonacept-placebo attack risk ratio was 0.45 [not 0.59] (SD, 0.13 [not 0.12]) (equal-tail 95% credible interval, 0.26 [not 0.39] to 0.77 [not 0.85])." Page 535, right column, Statistical Analysis section, 3rd paragraph, lines 5 to 9: "We developed models based on noninfor-mative (log-normal distribution mean, 0 [SD, 10]), enthusiastic (mean, ln(0.5) [not 0.5] [SD, 0.125]), and skeptical (mean, 0 [SD, 0.125]) priors for the attack rate risk ratio (RR) (rilonacept-placebo)."