AOTM: Surviving Sepsis Campaign: association between performance metrics and outcomes in a 7.5-year
Levy MM, Rhodes A, Phillips GS, Townsend SR, Schorr CA, Beale R, Osborn T, Lemeshow S, Chiche JD, Artigas A, Dellinger RP
Intensive Care Medicine 2014, 40 (11): 1623-33
The Surviving Sepsis Campaign (SSC) was established in 2002 as a major international initiative to improve both the awareness about and the management of sepsis, particularly in its severest forms. The intention was to deliver this program through the development of evidence-based guidelines for the management of severe sepsis and septic shock. It very quickly became apparent that for the program to succeed then achieving clinical behavior change during its implementation was of equal importance to the quality of the guidelines themselves. The mechanism constructed to deliver this sustained change in clinical practice was through the adoption of clinical care bundles.
In the 2010 published results covering 15,022 patients no difference was found in the mortality rates throughout the two year study period. This paper presents the results after 7.5 years of running the program across 218 sites in Europe (36.2% of collaborating centres), the United States (49.1% of collaborating centres) and South America (14.7% of collaborating centres). The primary measurements in the study were the mortality rates and compliance with the care bundle. Compliance was defined as evidence that all bundle elements were achieved within 6 hours for the resuscitation bundle and 24 hours for the management bundle. Low compliance for the resuscitation and management bundles were defined as <15% and <20% respectively, with high compliance been greater than this.
The results found high compliance in 46.8% of sites for the resuscitation bundle and 47.2% of sites for the management bundle. For both bundles, the highest compliance was in the United States. There was a relationship between duration of participation in the program and compliance achievement. Sites with <2 years’ duration of participation had low compliance for the resuscitation (67.2 %) and management (64.3 %) bundles. The majority of high compliance sites had >2 years in the program (p <0.001 and p = 0.010, respectively).
A relationship was also found between the level of compliance and the observed overall mortality rates. In considering the differences between high and low compliance sites for the management bundle mortality was not significantly different. However overall mortality was lower in high versus low resuscitation bundle compliance sites (29.0% versus 38.6 %; p <0.001). It was found that compliance with the resuscitation bundle increased quarter on quarter with participation in the study and that although greatest in the first two years this continued to be observed thereafter. The hospital mortality rate was measured to drop 0.7% per quarter of participation in the program and again this was a sustained progression of impact. A secondary endpoint supporting the presence of this improvement was the finding that hospital and intensive care unit length of stay decreased by 4% for every 10% increase in the site resuscitation bundle compliance
Recognising that mortality in severe sepsis has been reducing over time mathematical modelling was used to assess whether the observed impact from the SSC was independent of such an effect and this was confirmed to be the case.
Although this is the largest presented prospective study of severe sepsis patients it is nevertheless important to recognise its limitations. Firstly, it is not a randomised, controlled trial so no direct causation can be claimed. A second major limitation is the lack of a rigourous quality control approach to data entry and particularly to the lack of availability of robust baseline mortality data at the participating sites.
Notwithstanding these limitations this paper provides conclusive support for the impact of the SSC by demonstrating that high performance in terms of bundle compliance relates to improved mortality outcomes in patients suffering from severe sepsis. It also provides quantification of the quantum of this effect with an observed 25% relative risk reduction in the mortality rate. The link between clinical behaviour and patient outcomes is clearly illuminated by the performance metrics used and is thus shown to be integral to achieving improvement in the quality of care.
Levy MM, Rhodes A, Phillips GS, Townsend SR, Schorr CA, Beale R, Osborn T, Lemeshow S, Chiche JD, Artigas A, Dellinger RP. Surviving Sepsis Campaign: association between performance metrics and outcomes in a 7.5-year study