Improving the Signal - An analysis into the afferent arm of deterioration

Abstract

Aim: There were two aims to this portfolio: 1) to test the sensitivity of a track and trigger scoring system at identifying early deterioration: 2) to identify useful design features of current New Zealand observation charts that may assist with recognising clinical deterioration. Background: Signs of life-threatening physiological dysfunction may be missed, misinterpreted or mismanaged by doctors and nurses with varying levels of experience, potentially resulting in catastrophic consequences. Track and trigger scoring systems have evolved to allow for increased recognition and response to deterioration. These scoring systems aid recognition of a patient's deteriorating condition at the bedside based on vital signs routinely recorded on the observation chart. Failure to recognise patients who are deteriorating can be the result of failure to assess and record vital signs, failure to recognise changes in vital signs or failure to take appropriate actions to such changes. The design of the observation chart itself may be a key factor in these failures. Methods: An audit was conducted that applied a track and trigger scoring system to the pre-critical event observations on patients who triggered a cardiac arrest or medical emergency team (MET) call at Auckland City Hospital. The study period was between April to September 2009 for MET and for the entire 12 months of 2009 for cardiac arrest calls. The second aim of this portfolio was achieved by conducting a heuristic analysis of all current observation charts from within New Zealand that use a track and trigger scoring system. Results: Complete sets of observations were not recorded in almost 50% of patients. Incomplete sets could prevent the recognition of deterioration and appropriate response that may prevent further deterioration, intensive care unit admission or death. Despite incomplete vital sign collection the majority of patients still showed early signs of deterioration prior to a critical event suggesting that MET calls were indicated much earlier than when they were actually called, in many cases up to eight hours earlier. There was considerable variation in the observation charts. Many of the charts had aspects of good design such as layout and grammar but the more important features such as the use of colour to help identify abnormal observations were less evident in the forms. Only two charts incorporated all optimal design features to a high standard. There was considerable room for improvement. Conclusions: A track and trigger scoring system can identify up to 75% of patients who need increased monitoring as well as those requiring non-urgent or urgent review up to eight hours earlier. A new approach should be taken when designing observation charts for the future. This better design may help improve the recognition and response to patients who exhibit signs of clinical deterioration.