The mean bias was approximately zero for the Cox model across all of the scenarios. However, the ORs from the logistic regression model tended to be larger than the underlying HR, and this difference increased as the underlying HR and the length of follow-up increased (Table 2 and Supplementary Table S5). In contrast to the cohort simulations, the HRs were not consistently more precise than the ORs. The mean SEs were smaller for the logistic regression model than the Cox model in the survey and random follow-up models but were slightly larger for the complete follow-up model (Table 2 and Supplementary Table S5). This increased precision of the logistic regression model relative to the Cox model in case-cohort studies (compared with cohort studies) occurs because robust SEs were not required for the logistic regression model but they were for the Cox model to account for the sampling process of this study design. The HRs also became less precise as the average length of follow-up time decreased. For instance, the mean SE of the logarithm of the estimated HR for the
