H Williams1, R Gwyn, A Smith, A Dramis, J Lewis. 1. Trauma and Orthopaedic Department, University Hospital of Wales, 2 Canada Road, Cardiff, CF14 3BW, UK, huwlmwilliams@gmail.com.
Abstract
PURPOSE: With restructuring within the NHS, there is increased public and media interest in surgical outcomes. The Nottingham Hip Fracture Score (NHFS) is a well-validated tool in predicting 30-day mortality in hip fractures. VLAD provides a visual plot in real time of the difference between the cumulative expected mortality and the actual death occurring. Survivors are incorporated as a positive value equal to 1 minus the probability of survival and deaths as a negative value equal to the probability of survival. Downward deflections indicate mortality and potentially suboptimal care. METHODS: We prospectively included every hip fracture admitted to UHW that underwent surgery from January-August 2014. NHFS was then calculated and predicted survival identified. A VLAD plot was then produced comparing the predicted with the actual 30-day mortality. RESULTS: Two hundred and seventy-seven patients have completed the 30-day follow-up, and initial results showed that the actual 30-day mortality (7.2 %) was much lower than that predicted by the NHFS (8.0 %). This was reflected by a positive trend on the VLAD plot. CONCLUSION: Variable life-adjusted display provides an easy-to-use graphical representation of risk-adjusted survival over time and can act as an "early warning" system to identify trends in mortality for hip fractures.
PURPOSE: With restructuring within the NHS, there is increased public and media interest in surgical outcomes. The Nottingham Hip Fracture Score (NHFS) is a well-validated tool in predicting 30-day mortality in hip fractures. VLAD provides a visual plot in real time of the difference between the cumulative expected mortality and the actual death occurring. Survivors are incorporated as a positive value equal to 1 minus the probability of survival and deaths as a negative value equal to the probability of survival. Downward deflections indicate mortality and potentially suboptimal care. METHODS: We prospectively included every hip fracture admitted to UHW that underwent surgery from January-August 2014. NHFS was then calculated and predicted survival identified. A VLAD plot was then produced comparing the predicted with the actual 30-day mortality. RESULTS: Two hundred and seventy-seven patients have completed the 30-day follow-up, and initial results showed that the actual 30-day mortality (7.2 %) was much lower than that predicted by the NHFS (8.0 %). This was reflected by a positive trend on the VLAD plot. CONCLUSION: Variable life-adjusted display provides an easy-to-use graphical representation of risk-adjusted survival over time and can act as an "early warning" system to identify trends in mortality for hip fractures.
Authors: R A Lawrance; M F Dorsch; R J Sapsford; A F Mackintosh; D C Greenwood; B M Jackson; C Morrell; M B Robinson; A S Hall Journal: BMJ Date: 2001-08-11
Authors: Christina Pagel; Martin Utley; Sonya Crowe; Thomas Witter; David Anderson; Ray Samson; Andrew McLean; Victoria Banks; Victor Tsang; Katherine Brown Journal: Heart Date: 2013-04-05 Impact factor: 5.994