Skip to main content
. Author manuscript; available in PMC: 2022 Apr 1.
Published in final edited form as: J Eval Clin Pract. 2020 Jun 10;27(2):356–364. doi: 10.1111/jep.13428

Table 1.

Illustration of a 2×2 table, event status modifications, and three artificial datasets

Treatment Event Non-event Sample size
Illustrative dataset:
Arm 0 e0 n0 − e0 n0
Arm 1 e1 n1 − e1 n1
Illustrative dataset with event status modifications:
Arm 0 e0 + f0 n0 − e0 − f0 n0
Arm 1 e1 + f1 n1 − e1 – f1 n1
Artificial dataset 1:
Arm 0 10 90 100
Arm 1 30 70 100
Artificial dataset 2:
Arm 0 25 75 100
Arm 1 30 70 100
Artificial dataset 3:
Arm 0 15 85 100
Arm 1 30 70 100

Note: n0 and n1 are sample sizes in treatment arms 0 and 1; e0 and e1 are event counts in the two arms in the original trial; and f0 and f1 are the numbers of modified events in the two arms for deriving the fragility index.