Abstract The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines aim to improve the reporting of observational studies, including cohort, case-control, and cross-sectional designs, relevant to assessing associations, risks, and outcomes in real-world clinical settings. They are conceived to optimize the meaningfulness of epidemiological and clinical studies, aligning them with the Aristotelian “mesotes” curve, namely, the principle of achieving balance and avoiding extremes to best reflect the most truthful approximation of reality. This commentary addresses situations where strict adherence to STROBE guidelines may be impossible or inappropriate, potentially distorting results, and shows how statistical tools can mitigate these issues. Specific challenges arise in real-life randomization scenarios, situations lacking placebo arms, and in correcting for multiple comparisons. We discuss these challenges, examine the role of Bonferroni and similar corrections, and propose alternative approaches such as false discovery rate (FDR) and Bayesian hierarchical models. We illustrate these points with examples from the literature and a simulation study evaluating p-value adjustments in multiple hypothesis testing. This work provides a framework for researchers to navigate STROBE guidelines thoughtfully, ensuring that observational studies are both rigorous and relevant.
Chrousos et al. (2026) studied this question.