To the Editor
We thank Guaraldi et al. [1] for the opportunity to clarify specific methodological issues that were identified in our review [2]. We agree that the magnitude of immortal bias may be small if the time span between the start of follow up and the treatment initiation is very short. However, this bias has already created many flawed publications in many epidemiological areas, so it cannot be ignored. We also highlight that immortal time is a time-dependent bias that may refer to other non-fatal outcomes under interest, such as discharge alive or initiation of mechanical ventilation [3]. In addition, we would like to remark that the quantification of the magnitude of the biases was beyond the scope of our review.
Competing risk events can occur in both randomized trials and observational studies and competing risk analysis should be performed irrespective of the primary study outcome [4]. In the review we pointed out that there are two main approaches for competing risks and the cause-specific hazard model is considered as an appropriate method for aetiological research [5].
Regarding time-varying confounding, the authors [1] considered glucocorticoids during follow up and therefore potentially later than the initiation of tocilizumab. However, time-varying confounding is evoked by covariates that influence the decision of administrating tocilizumab, so confounders are measured before the potential administration.
Finally, regarding the validity assessment of effect estimates obtained from studies with different design, we fully agree that well-designed observational studies with accurate results might reflect findings from randomized trials and should complement the clinicians' knowledge and support clinical decision-making.
Author contributions
MW, MvC and OM contributed to the conceptualization of the letter, writing of the original draft and reviewing of the letter.
Transparency declaration
The authors have no conflicts of interest to declare. No funding was received for this letter.
Editor: L. Leibovici
References
- 1.Guaraldi G., Meschiari M., Cozzi-Lepri A., Milic J., Tonelli R., Menozzi M., et al. Tocilizumab in patients with severe COVID-19: a retrospective cohort study. Lancet Rheumatol. 2020;2:e474–e484. doi: 10.1016/S2665-9913(20)30173-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Martinuka O., Cube M von, Wolkewitz M. Methodological evaluation of bias in observational COVID-19 studies on drug effectiveness. Clin Microbiol Infect. 2021;27:949–957. doi: 10.1016/j.cmi.2021.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Renoux C., Azoulay L., Suissa S. Biases in evaluating the safety and effectiveness of drugs for COVID-19: designing real-world evidence studies. Am J Epidemiol. 2021 doi: 10.1093/aje/kwab028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Austin P.C., Fine J.P. Accounting for competing risks in randomized controlled trials: a review and recommendations for improvement. Stat Med. 2017;36:1203–1209. doi: 10.1002/sim.7215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Latouche A., Allignol A., Beyersmann J., Labopin M., Fine J.P. A competing risks analysis should report results on all cause-specific hazards and cumulative incidence functions. J Clin Epidemiol. 2013;66:648–653. doi: 10.1016/j.jclinepi.2012.09.017. [DOI] [PubMed] [Google Scholar]