Normal or Not? Acetabular Morphology Is Not a Binary Classification

Orthopaedic surgical decision making is a combination of clinical intuition, radiological measurements, and referenced standards. As clinicians and researchers, we consider hip pathology as a mechanical problem described in geometric and statistical language. For a clinical measurement to be useful, it must be easy to perform, reproducible, and demonstrably associated with the risk of the condition it seeks to diagnose or prevent. Using acetabular dysplasia as an analytic lens, this review discusses what our radiographs actually tell us about hip morphology risk, by separating population reference ("normative") intervals from outcome-anchored decision limits and by tracing the evidentiary lineage of the field's workhorse measurements. We revisit the origins, current use, and statistical power of the lateral center-edge angle, acetabular index/Tönnis angle, femoral head extrusion index, and Graf ultrasonography classification and synthesize what is known about their reproducibility and what clinical decisions can appropriately be made. We examine how nomenclature drift fuels routine miscommunication and show that superficially similar measurements are often tied to reference datasets they were neither derived from nor validated against. The current, normal or not, phenotypic model of risk allocation is considered, and the alternative of a continuous, dose-response relationship is proposed. The aim of this narrative review is to prompt clinicians and researchers to consider has our use of legacy morphological risk models actually curtailed osteoarthritis progression in conditions such as hip dysplasia and can we continue to depend on them? Or do models grounded more in lineage than validation now warrant a fundamental reconstruction.