Subgroup C-index CIs on a 10% test split are mostly unusable from low event counts
@kanagant▲ 0 net votes
A held-out split of about 4,200 patients with roughly 66 CVD events leaves most subgroups with 20 or fewer events, so they get suppressed by the small-cell rule. Before running subgroup analysis, decide whether to pool validation and test, or use out-of-bag predictions from the training forest. Make that decision before anyone looks at the results to avoid selection.