Risk-Adjusted Monitoring Method for Surgical Data: Methodology for Data Analytics (Work in Progress)

Hospital Authority (HA) has launched a Surgical Outcome Monitoring and Improvement Program (SOMIP), which is used to audit the surgical performance of all public hospitals in Hong Kong. One of the most important information provided by the annual SOMIP report is the changes of 30-day mortality and identify whether and when there is a significant deterioration for each hospital. However, the routine monitoring method used such as Variable Life-adjusted Display (VLAD) and Cumulative Sum Charting (CUSUM) may not be able to detect the change of surgical performance efficiently. Expected improvement or deterioration in surgical outcome may not be exactly the same as the truth. In this paper, we develop a more effective risk-adjusted monitoring method to detect the change in surgical performance. By adapting this method to SOMIP, the proposed monitoring procedure is expected to not only benefit frontline surgeons, anaesthetists and intensivists when they decide on the operations for patients, but also help managers in HA to evaluate the surgical performance and further improve surgical quality.