Hao Liu
Expansion outcomes were consistently more favourable than those observed in the IPS-AD study. Factors that may have contributed to improved outcomes included implementation and delivery enhancements, broadened eligibility criteria, and favourable environmental and economic conditions.
PURPOSE: To examine the statistical robustness and interpretive framing of a recent retrospective cohort study examining perioperative RAAS inhibitor use and outcomes after total shoulder arthroplasty.
METHODS: We evaluated the multiplicity of reported outcomes, sample-size implications for statistical significance, absolute risk differences, and the transparency of propensity score matching covariates against accepted observational-study principles.
RESULTS: The study reported at least 14 separate outcomes across 90-day and 1-year timepoints in two propensity-matched comparisons (ARB vs control, ACEi vs control), with no correction for multiple comparisons. With cohort sizes of 27,938-30,602 per group, absolute risk differences of less than 0.5% may reach conventional statistical significance, yet the abstract reports only directional findings ("higher rates") without absolute risk differences or confidence intervals for individual outcomes. The propensity score matching covariates are not specified in the abstract, and the title and abstract emphasize "higher complications and revision rates" while the conclusion acknowledges that findings should be interpreted as associative rather than causal.
CONCLUSION: Clarification of multiplicity correction, absolute risk differences, matching covariates, and consistency between the title/abstract framing and the conclusion's causal caveat would strengthen the interpretability and clinical utility of this study.