Recalculation of the numbers at risk - example classified as "consistent". Kaplan-Meier plot of a randomised trial comparing prednisolone and a control group . According to the information in the text of the publication, one patient was lost to follow-up in the prednisolone group and minimum follow-up was 120 days. At one time point beforehand (90 days), we read the survival probability from the curve (see vertical line). We recalculated the number of patients at risk by multiplying the survival probability with the number of randomised patients (number at risk: 17 in the prednisolone group vs. 6 in the control group). As the calculated and reported numbers matched (taking into account the one patient lost to follow-up in the prednisolone group), this example would be classified as "consistent".
Vervölgyi et al. BMC Medical Research Methodology 2011 11:130 doi:10.1186/1471-2288-11-130