High Variability in Sepsis Guidelines in UK: Why Does It Matter?

Abstract

It is recommended that developers of Point Of Care Tests (POCTs) assess the care pathway of the patient population of interest in order to understand if the POCT fits within the pathway and has the potential to improve it. If the variation of the pathway across potential hospitals is large, then it is likely that the evaluation of effectiveness is harder and the route towards large-scale takes adoption longer. Evaluating care pathways can be a time-consuming activity when conducted through clinical audits or interviews with healthcare professionals. We have developed a more rapid methodology which extrapolates the care pathway from local hospital guidelines and assesses their variation. Sepsis kills 46,000 people per year in the UK with societal costs of up to £10 billion. Therefore, there is a clinical need for an optimized pathway. By applying our method in this field, we were able to assess the variation in current hospital guidelines for sepsis and infer the potential impact this may have on the evidence development on innovations in this applications. We obtained 15 local sepsis guidelines. Two independent reviewers extracted: use of the national early warning score (NEWS), signs and risk factors informing the decision to prescribe antibiotics, and the number of decisional steps up to this point. Considerable variation was observed in all the variables, which is likely to have an impact on future clinical and economic evaluations and adoption of POCT for the identification of patients with sepsis.

A new method for testing reproducibility in systematic reviews was developed, but needs more testing

“…Our result for data extraction was very satisfactory. Other studies also indicated that data extraction errors occur regularly leading to biased estimates [26][27][28], while this will usually not impact the conclusions of the SR [29,30]. One exception was the extraction on RoB ratings from the original publication.…”

The Potential Cost and Cost-Effectiveness Impact of Using a Machine Learning Algorithm for Early Detection of Sepsis in Intensive Care Units in Sweden

“…The results are sensitive to the sepsis incidence, which, as discussed by Bray et al, 16 depends, among other things, on how sepsis is diagnosed in the clinic. A broad definition of sepsis leads to a higher incidence, and a prediction algorithm would thus have a larger potential of reducing costs in situations where a broader definition is used.…”

“…In this study, we investigated the cost-saving potential of a machine learning algorithm such as NAVOY® Sepsis for early detection of sepsis in intensive care units (ICUs) in Sweden. Since the evaluation of costs and cost-effectiveness of novel diagnostic tools depend on multiple factors such as the definition of sepsis, the diagnostic tool currently used, and the sensitivity and specificity of the diagnostic tool, 16 we aimed at developing a health economic model that would enable us to simulate the cost-saving potential of a sepsis prediction algorithm in various contexts.…”

Destabilisation of T cell-dependent humoral immunity in sepsis

“…One-sixth of survivors experience post-sepsis syndrome with at least one cognitive, psychological, or physical impairment, and indeed are more prone to recurrent infection, renal failure, and cardiovascular episodes than matched patients hospitalised for other diagnoses [ 17 ]. As such, sepsis poses a significant medical and financial burden on healthcare services worldwide, with the National Health Service in the United Kingdom alone estimated to face annual costs of >£1 billion [ 19 ]. Although late-mortality and long-term symptoms following sepsis are well-studied, the causes of sequelae are poorly understood [ 20 ].…”