Comment on "Inpatient mortality following hip fracture in the United States: an updated analysis of over one million cases".

This commentary examines the statistical and methodological limitations of a recent national analysis of inpatient hip fracture mortality using the Epic Cosmos database. Four concerns are raised. First, the reported Spearman correlation between a four-category Social Vulnerability Index and mortality lacked statistical power because it was calculated across only four aggregated data points, and collapsing a continuous index into ordinal bins discarded patient-level variation needed to detect a dose-response relationship. Second, do-not-resuscitate status, documented in over one-third of patients, was omitted from the analysis of racial mortality differences, despite its direct influence on in-hospital mortality timing and its potential to confound comparisons unrelated to surgical risk. Third, the regional analysis reported the distribution of deaths across regions rather than region-specific mortality rates, conflating hospital volume with mortality risk. These limitations should be addressed before the reported demographic and regional patterns are used to guide resource allocation or clinical practice.