Immortal time bias: a frequently unrecognized threat to validity in the evaluation of postoperative radiotherapy

To evaluate the influence of immortal time bias on observational cohort studies of postoperative radiotherapy (PORT) and the effectiveness of sequential landmark analysis to account for this bias.

First, we reviewed previous studies of the Surveillance, Epidemiology, and End Results (SEER) database to determine how frequently this bias was considered. Second, we used SEER to select three tumor types (glioblastoma multiforme, Stage IA-IVM0 gastric adenocarcinoma, and Stage II-III rectal carcinoma) for which prospective trials demonstrated an improvement in survival associated with PORT. For each tumor type, we calculated conditional survivals and adjusted hazard ratios of PORT vs. postoperative observation cohorts while restricting the sample at sequential monthly landmarks.

Sixty-two percent of previous SEER publications evaluating PORT failed to use a landmark analysis. As expected, delivery of PORT for all three tumor types was associated with improved survival, with the largest associated benefit favoring PORT when all patients were included regardless of survival. Preselecting a cohort with a longer minimum survival sequentially diminished the apparent benefit of PORT.

Although the majority of previous SEER articles do not correct for it, immortal time bias leads to altered estimates of PORT effectiveness, which are very sensitive to landmark selection. We suggest the routine use of sequential landmark analysis to account for this bias.