RESUMEN
INTRODUCTION: Proximal femoral fracture is common with a high mortality (7% mortality at 30 days). Accurate determination of mortality risk allows better consenting, clinical management and expectation management. Our study aim was to develop a prognostic tool to predict 30-day mortality after proximal femoral fracture, among patients treated within a dedicated hip fracture unit. MATERIALS AND METHODS: We collected data from our hospital concerning 2210 patients with 2287 proximal femoral fractures. The clinical parameters of 97 patients who died within 30 days of surgery were analysed. We used logistic regression to determine if the parameters' relationship with 30-day mortality was statistically significant or not. The statistically significant parameters were used to create a prognostic model for predicting 30-day mortality. RESULTS: The 5 independent predictors of 30-day mortality were gender, age, admission source, preoperative Abbreviated Mental Test Score (AMTS) and American Society of Anesthesiologists Score (ASA). The highest risk was for males >85 years, admitted from institutional care, with low preoperative mental test score and high ASA grade. Using these predictors, we formulated the G4A score. The Hosmer-Lemeshow 'goodness of fit' test showed good concordance between observed and predicted mortality rates. CONCLUSIONS: We recommend the use of the G4A score to predict 30-day mortality after surgery for proximal femoral fracture, particularly within dedicated hip fracture units. Further research is needed to establish whether the findings of this study are applicable on a national scale.