Prognostic scoresBMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g1402 (Published 07 February 2014) Cite this as: BMJ 2014;348:g1402
The author of this Endgames article, Philip Sedgwick, would like to correct some errors in both the scenario and the answers section (BMJ 2014;348:g282, doi:10.1136/bmj.g282). In the third paragraph of the scenario section, the fourth sentence reads that “For each child a prognostic score for predicting immediate, early, and late death was derived by summing the total number of clinical indicators present.” This article’s scenario was taken from a research paper (cited as reference 1); however, the above description of how a prognostic score for each child was calculated is wrong. In fact, in the referenced study, the researchers describe calculating prognostic scores for each child using a multivariate analysis, which gave greater importance (weight) to some clinical indicators because of their predictive ability. Consequently, statement b (“Each clinical indicator was assumed to have equal weight in predicting mortality”) is false, and not “true” as was published in the first paragraph of the answers section. Furthermore, the reader would not be able to discern this answer from the information given in the article’s scenario section, and so the author Philip Sedgwick would suggest that statement b is ignored when attempting the questions.
The answers section also contains an explanation of statement b in the fifth paragraph—the first sentence of which begins with “For each child in the validation cohort . . .” For the reasons described above, the reader is also advised to ignore the reference to statement b in this paragraph of the answers section. In order to read correctly, this entire paragraph should be rephrased to say: “For each child in the validation cohort, the prognostic models yielded scores on admission to hospital that predicted the risk of immediate, early, and late death. These scores were calculated for each child using a multivariate analysis, which gave greater importance (weight) to some clinical indicators because of their predictive ability.”
Cite this as: BMJ 2014;348:g1402