Measuring the success of Medical Emergency Teams: potentially preventable deaths versus total cardiac arrest deaths. A single centre observational study.

Background: Rapid response systems have been implemented with the aim of preventing patient deterioration, in-hospital cardiac arrests (IHCA) and related deaths. Not all 'unexpected deaths' are preventable, thus compromising the use of unexpected deaths as an outcome measure.

Aims: To assess temporal trends in potentially preventable deaths as a subset of total unexpected death rates over a 4-year period.

Methods: A single centre, cohort study of all unexpected deaths between 1 January 2010 and 31 December 2013. Unexpected deaths were identified from the rapid response systems database and patients' case histories were reviewed to reclassify the deaths into one of three categories: potentially preventable: if earlier MET activation may have prevented death; missed not for resuscitation opportunity; and not preventable. Total bed days were obtained from the hospital's patient administration system.

Results: The rate of potentially preventable deaths decreased from 5.3 to 0.7 per 100 000 bed days (incident rate ratio (IRR) 0.53 (95% CI 0.31-0.90), P = 0.02). The rate of total unexpected deaths was unchanged (IRR 0.96 (0.80-1.16), P = 0.70), as were the rates of non-preventable deaths (IRR 1.06 (0.78-1.42), P = 0.72) and missed NFR deaths (IRR 1.1 (0.83-1.42), P = 0.56).

Conclusion: The rate of potentially preventable deaths has decreased by 47% per year over a 4-year period without any change in the overall rate of unexpected deaths. Distinguishing between potentially preventable deaths in contrast to total unexpected deaths enables more targeted evaluation of rapid response systems.