This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
To the Editor.—
I read with interest the article by Leppik et al entitled "Double-blind Study of Lorazepam and Diazepam in Status Epilepticus" (1983;249:1452). As a statistician, what interested me most was not the basis of the article but rather the interpretation of the statistical analysis. In particular, the authors commented: "For all other seizure types, lorazepam was effective more frequently. Although the difference is not statistically significant, 32% of the episodes, exclusive of the generalized type, were not controlled by diazepam, while only 12% of these seizure types were not controlled by lorazepam." Unfortunately, this type of comment is still all too often found in the medical literature. The problem here is that the difference looks so large that, regardless of the lack of statistical significance, it must be noteworthy. We must keep in mind that when a statistical test result is nonsignificant, the implication is that the difference
Lee ML. Statistical Indiscretion. JAMA. 1983;250(18):2470. doi:10.1001/jama.1983.03340180032013
Browse and subscribe to JAMA Network podcasts!
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: