User login
Early Recognition: The Rate-Limiting Step to Quality Care for Severe Sepsis Patients in the Emergency Department
From the Department of Medicine, University of Pennsylvania, and the Department of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, PA.
Abstract
- Objective: To detail strategies to improve sepsis recognition and the quality of care provided to the septic patient.
- Methods: Review of the literature.
- Results: Severe sepsis affects nearly 3 million individuals each year in the United States, and cost estimates for these hospitalizations exceed $24 billion. Effective management is predicated on timely recognition. In this review, we detail strategies to improve early identification of potentially septic patients as well as the quality of care provided to the septic patient in the emergency department (ED). The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients at risk for clinical deterioration.
- Conclusion: ED structures and processes can be used to increase adherence with sepsis management guidelines to improve patient outcomes.
Severe sepsis affects nearly 3 million individuals each year in the United States and cost estimates for these hospitalizations exceed $24 billion [1–3]. Sepsis is a life-threatening condition characterized by a suspected or identified infection accompanied by a vigorous host inflammatory response. In severe sepsis, end-organ dysfunction manifests in myriad forms, including altered mental status, acute kidney injury, liver dysfunction, pulmonary dysfunction, and hemodynamic compromise [4,5]. This protean presentation of a deadly condition makes identification and risk stratification both challenging and essential to improving patient outcomes. The majority of patients with severe sepsis will receive their initial care within an emergency department (ED) [6,7]. It is essential that emergency medicine providers have the means to appropriately identify patients presenting with severe sepsis in a timely manner—thus facilitating life-saving measures such as early intravenous fluid resuscitation and administration of timely and appropriate antimicrobials.
In this review, we detail strategies to improve sepsis recognition and the quality of care provided to the septic patient in the ED. The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients for clinical deterioration. Then, we review suggested ED structures and processes to increase adherence with sepsis-based guidelines to improve patient outcomes. Successful implementation is predicated on hospital administrative support towards the efforts given the time and resources required and strong and committed leadership across the health care system.
Epidemiology of Severe Sepsis
Estimates of annual cases of severe sepsis vary, ranging from 1 million to 3 million cases in the United States [1–3]. In-hospital mortality for this condition ranges from 14% to 30% [5]. The incidence of severe sepsis in the United States has been increasing at a rate of 13% annually, with an estimated cost of greater than $24 billion per year [1,2]. In 2 large cohorts of hospitalized patients, it was found that sepsis contributed to 1 in every 2 to 3 deaths following inpatient admission [8]. Coincident with these increased estimates, advances in the early identification and treatment of sepsis have led to decreasing mortality rates over the past decade [1,9].
Of importance to the ED clinician, an episode of sepsis has long-term effects on cognitive and physical function, quality-of-life, and survival [10,11]. Post-discharge, approximately one-quarter of sepsis survivors will be readmitted within 30 days [12–14]. In as many as half of these instances, another life-threatening infection is the cause for readmission, making the past medical history, including a detailed accounting of recent episodes of sepsis, an important part of the initial ED evaluation [12]. Furthermore, severe sepsis survivors spend a large proportion of their time following discharge within a health care facility, and will frequently present to the ED with an acute condition from such an environment. Important factors for predicting readmission after a sepsis hospitalization include patient age, severity of illness, hospital length of stay, and the need for intensive care during the initial hospitalization [12–14].
Principles of Effective Sepsis Management
The principles of effective sepsis management begin with early identification in the pre-hospital setting, at triage, or when a patient begins to decompensate in the hospital. After the point of initial recognition, core principles include risk stratification, timely and appropriate antimicrobial administration, initial intravenous fluid boluses and ongoing resuscitation guided by physical examination and objective resuscitation end-points [4,5]. These practices have been operationalized in the care bundles of the Surviving Sepsis Campaign Guidelines [4]. Within 3 hours, the resuscitation bundle includes measuring serum lactate to risk stratify patients, obtaining blood cultures, administering broad-spectrum antibiotics, and administering 30 mL/kg crystalloid in patients with hypotension or hyperlactatemia [4]. The 6-hour bundle expands upon these initial measures and includes additional management recommendations based on resuscitation end-points.
As effective management is predicated on timely recognition, an understanding of the impact of delayed recognition is essential to provide optimal care for the severe sepsis patient in the ED. Decades of research has revealed that certain markers predict adverse outcomes, including transition to septic shock and death, as do delayed processes of care. Importantly, while early quantitative resuscitation was demonstrated to improve outcomes in a meta-analysis, there was no demonstrable benefit when resuscitation was initiated late (> 24 hours) in the course in the ICU (odds ratio of death, 1.16 [95% confidence interval, 0.60–2.22]) [15].
Strategies To Improve Recognition
Pre-Hospital Environment
From EMS to ED Triage
Borrowing the principle “time equals tissue” from a variety of time sensitive conditions (eg, myocardial infarction management [“time equals muscle”] and stroke care [“time equals brain”]), clinicians and researchers have realized that expedited recognition of severe sepsis patients begins at the time of initial contact with the health care system. For severe sepsis patients, clinicians need to think “time equals organ function.” Given the frequency with which sepsis patients arrive to the ED via EMS, effective communication between EMS and ED providers could be leveraged to prepare the ED team to provide timely care for the sepsis patient via a “sepsis alert.” While confirmation of its applicability to sepsis care is required in the absence of a regionalized network of sepsis centers, the rationale is based on the experience of the effectiveness of trauma and stroke alert systems [20–22]. For patients not recognized as potentially being infected by EMS providers during transport, repeat vital signs during ED triage can be screened to identify patients exhibiting signs of the systemic inflammatory response syndrome (SIRS) [4,23]. The same principles of effective communication apply for patients being sent from medical clinics to the ED for evaluation and treatment of potential severe sepsis. For patients arriving independent of EMS, focused triage and initial vital signs are the starting point for identifying severe sepsis at the most proximal phase of entry into the health care system.
Vital Signs and SIRS Criteria in the ED
The Afferent Arm: Multimodal Screening Strategies
While institutional practice improvement initiatives to facilitate sepsis recognition and care should incorporate educational strategies, led by champions with expertise in sepsis, the complex presentation of sepsis requires multimodal approaches [29]. These multimodal approaches, beginning at the time of ED triage, should be designed to harness information technology to screen patients to improve severe sepsis recognition (the afferent arm) and to utilize structures and processes of care efficiently and effectively (the efferent arm) to guide severe sepsis management according to sepsis-care bundles espoused by guidelines (Figure) [4].
Operational processes to screen for sepsis in the ED will need to account for ED organizational flow (eg, average time from registration to triage, average time from triage to being seen by a physician, average length of stay in the ED, number of hospital beds) and hand-off practices (eg, care transition from ED team to floor or ICU team, or within ED at shift change). For ED organizations with shorter ED lengths of stay (eg, < 2 hours), screening practices at ED triage will serve as the focal point to identify cases of sepsis. Boarding, defined as caring for a patient in the ED pending transfer, is common, increasing as a result of ED closures [30,31], and associated with prolonged hospital length of stay and increased in-hospital mortality when ICU transfer is delayed [32]. Sepsis patients in particular appear to be a vulnerable group of patients. While many explanations exist to account for the relationship between delayed transfer and adverse outcomes, timely recognition and management of the septic patient could be compromised with prolonged boarding. To combat this potential effect, continual assessment during the entire ED stay may unmask an initially unclear presentation of sepsis.
One strategy to identify sepsis in ED organizations with prolonged ED lengths of stay is through the use of a track-and-trigger system, or early warning system. Traditionally, track-and-trigger systems were implemented on the hospital wards, as means to identify physiological deterioration in a timely manner to prevent clinical deterioration [33]. More recently, early warning systems have been used to identify patients with sepsis on the hospital wards and within EDs, as these systems rely on physiological parameters such as SIRS that are cardinal features of sepsis [34]. However, given the potential for alert fatigue, designing a system that operates with high accuracy is imperative.
Efforts are underway to redefine sepsis, using a simplified approach and readily available physiological variables, with the main goal of targeting those most at-risk of an adverse outcome during the hospitalization. Simultaneously, an understanding of the overt and more occult manifestations are essential to incorporate into the clinical decision-making and pattern recognition required to identify sepsis in a timely and accurate manner. In Table 2, the signs and symptoms that may serve as flags for severe sepsis are presented.
Mature early warning systems, designed to leverage the electronic medical record (EMR) by capturing vital signs, laboratory measures, (eg, elevated serum creatinine compared to a recent hospitalization) and symptoms (eg, altered mental status), are well-positioned to herald clinical deterioration (eg, cardiac arrest) with improved accuracy [35] and to be applied to sepsis specifically [34]. While sophisticated analytical strategies, such as machine learning, are being used to improve the test characteristics of these early warning systems, iterative, prospective chart review is an essential and complementary performance improvement step to refine the process. Further, chart review affords the opportunity to ensure compliance with sepsis care bundles.
Knowledge of the risk factors associated with development of sepsis is critical for the front-line emergency physician and nurse. Additionally, as many of these risk factors are associated with adverse outcomes, including unplanned ICU transfer and in-hospital mortality, which occur in as many as one out of 8 patients admitted directly to the ward, they have utility for early risk-stratification and triaging purposes in the ED. Advanced age and pre-existing comorbid conditions, particularly an oncologic diagnosis and/or chronic organ dysfunction, are major risk factors for sepsis and worse outcomes result in those who develop sepsis [2]. Further, illness severity, including an elevated serum lactate level, is associated with adverse outcomes. These factors can be incorporated into triage decisions and/or close monitoring for patients admitted to the general ward [36]. Conversely, because patients admitted to the ICU setting and subsequently stepped down through their hospitalization may experience better outcomes compared to patients admitted to the general ward who then require step-up to an ICU setting (37,38), attention to triage practices is critical.
These complementary strategies, which serve as the afferent arm of the system, summon health care providers to the bedside of a vulnerable patient. However, clinical effectiveness in the management of severe sepsis requires a robust, sophisticated, and mature efferent arm capable of delivering expert care to the now recognized septic patient.
Principles of Effective Management Post-Recognition
Risk Stratification
An elevated serum lactate level was initially described in pathological states in the mid 19th century by Johann Joseph Scherer [39] and has long been associated with increased mortality in hospitalized patients [40]. Lactate is a useful biomarker for risk stratification in a variety of patients arriving to the ED, particularly those who have been identified at high risk for sepsis. Jansen and colleagues examined the measurement of pre-hospital serum lactate at the time of paramedic on-scene assessment in a group of acutely ill patients [41]. Patients with point-of-care lactate levels of 3.5 mmol/L or greater were found to have an in-hospital mortality of 41% versus 12% for those with lactate levels less than 3.5 mmol/L. Within the population with an elevated lactate, patients with a systolic blood pressure greater than 100 mgHg experienced a mortality of nearly 30%, while it was greater than 50% in hypotensive patients with an elevated lactate, highlighting the value of both hemodynamic and serum lactate measures. Upon arrival to the ED, lactate measurements have a strong correlation with mortality. In one retrospective cohort, lactate level was linearly associated with mortality in a broad array of patients older than age 65 years [42]. An initial serum lactate level in the ED in the intermediate (2.0 – 3.9 mmol/L) or high range (≥ 4 mmol/L) has been associated with increased odds of death 2 to 5 times higher independent of organ dysfunction in severe sepsis specifically [43].
As the association between serum lactate levels and death is independent of organ dysfunction, serum lactate is a simple and reliable tool to both enhance detection and risk-stratify patients presenting to the ED with severe sepsis. Given the frequency with which hyperlactatemia is present in patients with suspected infection [43], operationalizing serum lactate measures with the initial phlebotomy draw is an important step to risk-stratify patients. This step can be coupled later with intravenous fluid resuscitation for those with marked elevations (≥ 4 mmol/L), in accord with guideline recommendations [4]. Screening of initial lactate values can be further expedited by utilizing fingerstick point-of-care lactate devices [44]. Last, while serial lactate measures can be incorporated into triage decisions, there is no clear threshold that warrants ICU admission. Rather, persistent elevations in serum lactate can be used to identify patients who require close observation regardless of their admission location.
Several scoring systems have been developed to augment sepsis risk stratification within the ED. The most prominent of these are the Predisposition Insult Response and Organ failure (PIRO), Sequential Organ Failure Assessment (SOFA), and Mortality in the Emergency Department Sepsis (MEDS) scores, and the National early warning score (NEWS) [45-48]. The MEDS score incorporates host factors including age and co-morbid illness, as well as physiologic and laboratory tests which can be obtained rapidly in an ED setting. Multiple prospective and retrospective examinations of the MEDS scoring systems have demonstrated that it performs optimally in ED patients with sepsis but not those with severe sepsis, in terms of predicting 30-day mortality [46,47]. The PIRO score more extensively incorporates predisposing co-morbidities, physiologic and laboratory parameters, and has been modified to consider presumed source of infection, leading to a stronger predictive ability for mortality in more severely ill patients. In patients presenting to the ED with severe sepsis and septic shock, a prospective observational study found the PIRO to be the best predictor of mortality, compared to SOFA and MEDS scores [45]. In a recent study by Corfield et al, sepsis patients with a higher NEWS, according to initial ED vital signs (temperature, pulse, respiratory rate, systolic blood pressure, oxyhemoglobin saturation) and consciousness level, were significantly more likely to be admitted to an ICU within 48 hours or to experience in-hospital mortality [48].
Timely and Appropriate Antibiotics
In a landmark study published by Kumar and colleagues in 2006, the relationship between timing of antibiotics and mortality was established [49]. In 2731 adult septic shock patients, mortality increased 7.6% for every hour delay in effective antimicrobial administration. A striking finding, given that the study population was limited to patients cared for in the ICU, was the fact that only 50% of patients received appropriate antibiotics within 6 hours of onset of shock and nearly one-quarter of patients did not receive antibiotics until the 15th hour. As a direct result, in-hospital mortality was observed to be 58% in this study.
Over the ensuing decade, a series of studies have demonstrated a narrowing of the quality gap in this regard, and the result has coincided with a significant improvement in survival. In 2010, Gaieski and colleagues demonstrated a significant improvement in the prompt administration of antibiotic delivery in patients presenting to an ED with severe sepsis, with the median time from shock onset (sustained hypotension or lactate ≥ 4 mmol/L) to antibiotics down to 42 minutes [50]. Importantly, consistent with the Kumar study, time to appropriate antibiotics, rather than simply initial antibiotics, remained associated with in-hospital mortality independent of initiating early goal-directed therapy. In 2011, Puskarich and colleagues revealed that time to antibiotics continued to improve and, as a result, the investigators did not identify a relationship between time from triage to antibiotics and in-hospital mortality [51]. However, when antibiotics were delayed until after shock recognition, consistent with the study by Kumar and colleagues, survival decreased. Until recently, this important observation was challenging to operationalize clinically as little was known about how to facilitate risk-stratification of those at risk to develop shock. However, Capp and colleagues recently found that deterioration to septic shock 48 hours after ED presentation occurs in approximately one out of eight patients and identified gender (female), transient hypotension, and/or hyperlactatemia upon presentation as risk factors associated with such a deterioration [52].
As an essential element of sepsis care bundles, a focus on timely use of antibiotics in patients with suspected infection, has the potential to increase the use of antibiotics in the ED in patients determined subsequently to not be infected. To combat this acknowledged downstream effect, reconsideration of the utility of empiric antibiotics 48 to 72 hours after admission is required. This step can be accomplished through the use of a sepsis care pathway and/or a formal antibiotic stewardship program.
Quantitative Resuscitation
Rivers and colleagues, in a landmark 2001 trial, examined the effectiveness of a protocolized resuscitation strategy in the most proximal phase of severe sepsis and septic shock [53]. A distinguishing characteristic between the usual care arm and the intervention in this ED-based study, in addition to whether mixed central venous oxygen saturation was measured as a resuscitation end-point, was the inclusion of an ED provider at the bedside to attend to clinical management. The intervention, aimed at achieving physiologic targets, resulted in significantly more fluid resuscitation (3.5 L vs. 5.0 L within the first 6 hours) and a significant decrease in in-hospital mortality compared to the usual care arm (46.5 vs. 30.5%). The study revolutionized the culture and practice of sepsis care, in part by shining a light on the importance of timely resuscitation at the most proximal point of contact between the patient and the healthcare system. It also highlighted the importance of integrating serum lactate measurement into the early screening and risk stratification processes for sepsis care delivery.
The 2014 randomized trial of Protocol-Based Care for Early Septic Shock (ProCESS) revisited this concept, comparing the Rivers 2001 protocol to both a current guideline-based non-invasive algorithmic protocol and what had become usual ED care in the interim [54]. The ProCESS trial, which operationalized a team of bedside providers to direct care for each of the 3 distinct arms, found no significant difference between the arms in terms of 90-day and 1-year mortality, but mortality was approximately 10% less in all arms compared with the intervention arm of the Rivers trial. Further, subjects in each of the 3 arms received in excess of 2 L intravenous fluid resuscitation pre-randomization and 4.4–5.5 L when resuscitation spanned from pre-randomization to 6 hours post-randomization. The conclusion drawn is that the commonalities between the arms—early fluid resuscitation, early antibiotics, and the option to use physiologic measures as markers of the adequacy of treatment, all guided by bedside ED providers—are the most important factors for surviving sepsis. And the result is that practitioners have refined these tools over a decade, leading to steady improvements in survival.
Consistent with the ProCESS trial, a recent Australia and New Zealand trial confirmed no significant difference in 90-day mortality between protocolized EGDT and current usual care for septic shock within an ED [55]. Consistent with ProCESS and ProMISe [56], subjects enrolled in ARISE received in excess of 2.5 L in resuscitation pre-randomization, which when paired with fluid resuscitation in the 0-6 hour post-randomization period (1.96 L in the EGDT arm and 1.71 in the usual-care arm) resulted in resuscitation in the 4.5 to 5L range during the initial resuscitation. The ARISE trial was unique in that appropriate antibiotic administration was a requirement prior to randomization, ensuring that this important driver of mortality reduction was standardized between the two arms of the trial. In summary, while the ideal fluid resuscitation amount is unknown, requires a personalized approach, and further investigation is required to effectively incorporate non-invasive measures to guide fluid responsiveness, early and aggressive resuscitation paired with early antibiotic administration are essential aspects of effective sepsis management.
The Efferent Arm: Structure And Processes To Improve Outcomes
Personnel and Staffing
Quality care for the septic patient requires immediate availability of a multidisciplinary care team, including physicians and nurses with critical care experience who can be rapidly deployed to the bedside. The location of care provision may include on-going care in the initial ED room assignment or transfer to a dedicated area for the care of the critically ill patient within the ED.
To provide optimal care in the era of overcrowding and delayed transfer to an ICU, a movement towards ED intensive care units (ED-ICUs) has emerged [57]. The models of practice range from a model based upon ED intensivists, with expertise in critical care medicine, providing care within the traditional structure of an ED, to a model wherein a portion of the ED is assigned for the care of the critically ill for extended periods of time beyond the initial resuscitation. As these models mature from resuscitation bays capable of scaling up based on need to dedicated ED-ICUs, investments in shared Unit leadership (physician and nursing), staffing (physician, critical care nursing, respiratory therapy, critical care pharmacist) and processes of care (eg, multidisciplinary rounds) in line with established ICUs will be necessary.
While attractive conceptually, large-scale implementation of this movement is unlikely to occur outside of tertiary care academic medical centers. In the many EDs across the US without ED intensivists, and confronted with limited clinician resources, flexible physician and nursing staffing models will be necessary to ensure that care provisions are in accord with established guidelines. Potential solutions to provide the resources to meet the needs of these high-intensity patients include critical care consultation and a strategy traditionally applied to the ICU, telemedicine [58]. Last, given the relationship between hospital volume and mortality in severe sepsis [59,60], timely transfer to a high-volume center for specific cases may be appropriate, although the optimal timing, case selection, and impact of transfer on outcomes warrant further examination.
Clinical Decision Support Strategies
To complement the identification and risk-stratification available by screening and scoring systems, clinical decision support systems are novel tools to improve outcomes in the era of electronic medical records (EMR). Specific to sepsis care delivery, performance improvement initiatives including audit-and-feedback practice can increase severe sepsis guideline adherence, and even modest improvements in adherence appear to lead to sustained improvements that contributed to a 25% relative risk reduction in the observed mortality rate [61,62]. Clinical decision support tools can be used to link early recognition to optimal care processes, such as the Surviving Sepsis Campaign resuscitation and management bundles. The use of prompts as strategies to ensure that bundles of care are ordered and carried out is an important aspect to operationalize during the design phase [63].
Significant preparation is required to effectively carry out the clinical decision support design strategy. For example, to ensure timely antibiotic dispensing, a number of process steps will be required, including prompt notification to a central pharmacist or preferably, an ED pharmacist with access to a local pharmacy pre-stocked with commonly used antibiotics [64]. In addition, the use of an institution-specific antibiogram within the physician computer-order entry sepsis order set, that includes site-specific recommendations (eg, pulmonary, gastrointestinal source) and susceptibility patterns, is an essential aspect of optimal sepsis processes of care. Last, the antibiogram will need to be frequently updated to include season-specific (eg, oseltamivir administration for high-risk cases during influenza season) recommendations to ensure that providers are prompted with the most up-to-date clinical information.
Audit and Feedback and Continuous Performance Improvement
The multimodal approach required to translate knowledge (eg, guidelines) into sepsis care implemented at the bedside is an iterative process. An ED armed with a robust track-and-trigger system and an effective efferent arm, including sophisticated clinical decision support strategies, will require frequent auditing in the plan-do-study-act model of quality improvement to yield clinical effectiveness [61,62,65]. Auditing, paired with feedback to frontline providers, is essential to refine and improve the complex process required to provide expert care to the septic patient [29,65]. Sustained success in optimizing sepsis care delivery is the goal, yet significant work is required to determine the best strategies to achieve this endpoint.
Conclusion
Severe sepsis affects millions of individuals each year in the United States. Delays in recognition result in increased morbidity and mortality, at a tremendous cost to the patient and society. By designing strategies to identify sepsis in a timely, efficient, and effective manner, and by implementing ED structures and processes to increase adherence with sepsis-based guidelines, improved patient-centered outcomes can be realized.
Corresponding author: Mark E. Mikkelsen, MD, MSCE, Gates 05.042, 3400 Spruce St., Philadelphia, PA 19104, mark.mikkelsen@uphs.upenn.edu.
Financial disclosures: None.
Author contributions: conception and design, JHM, MEM; analysis and interpretation of data, DFG; drafting of article, JHM, DFG, MEM; critical revision of the article, JHM, MEM.
1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41:1167–74.
2. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–10.
3. Lagu T, Rothberg MB, Shieh MS, et al. Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007. Crit Care Med 2012;40:754–61.
4. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med 2013;39:165–228.
5. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013;369:840–51.
6. Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis in United States emergency departments. Crit Care Med 2007;35:1928–36.
7. Dombrovskiy VY, Martin AA, Sunderram J, et al. Rapid increase in hospitalization and mortality rates for severe sepsis in the United States: a trend analysis from 1993 to 2003. Crit Care Med 2007;35:1244–50.
8. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA 2014;312:90–2.
9. Kaukonen KM, Bailey M, Suzuki S, et al. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA 2014;311:1308–16.
10. Yende S, Angus DC. Long-term outcomes from sepsis. Curr Infect Dis Rep 2007;9:382–6.
11. Iwashyna TJ, Ely EW, Smith DM, et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA 2010; 304:1787–94.
12. Ortego A, Gaieski DF, Fuchs BD, et al. Hospital-based acute care use in survivors of septic shock. Crit Care Med 2015;43:729–37.
13. Prescott HC, Langa KM, Liu V, et al. Increased 1-year healthcare use in survivors of severe sepsis. Am J Respir Crit Care Med 2014;190:62–9.
14. Liu V, Lei X, Prescott HC, et al. Hospital readmission and healthcare utilization following sepsis in community settings. J Hosp Med 2014;9:502–7.
15. Jones AE, Brown MD, Trzeciak S, et al. The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta-analysis. Crit Care Med 2008;36:2734–9.
16. Seymour CW, Rea TD, Kahn JM, et al. Severe sepsis in pre-hospital emergency care: analysis of incidence, care, and outcome. Am J Respir Crit Care Med 2012;186:1264–71.
17. Seymour CW, Cooke CR, Mikkelsen ME, et al. Out-of-hospital fluid in severe sepsis: effect on early resuscitation in the emergency department. Prehosp Emerg Care 2010;14:145–52.
18. Seymour CW, Cooke CR, Heckbert SR, et al. Prehospital intravenous access and fluid resuscitation in severe sepsis: an observational cohort study. Crit Care 2014;18:533
19. Studnek JR, Artho MR, Garner CL, Jones AE. The impact of emergency medical services on the ED care of severe sepsis. Am J Emerg Med 2012;30:51–6.
20. Guss DA, Meyer FT, Neuman TS, et al. The impact of a regionalized trauma system on trauma care in San Diego County. Ann Emerg Med 1989;18:1141–5.
21. Liberman M, Mulder DS, Jurkovich GJ, Sampalis JS. The association between trauma system and trauma center components and outcome in a mature regionalized trauma system. Surgery 2005;137:647–58.
22. Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working toward a prioritized world agenda. Stroke 2010;41:1084–99.
23. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31:1250–6.
24. Sibbald W, Doig G, Inman K. Sepsis, SIRS, and infection. Intensive Care Med 1995;21:299–301.
25. Kaukonen KM, Bailey M, Pilcher D, et al. Systemic inflammatory response syndrome criteria in defining severe sepsis. N Engl J Med 2015; online March 17, 2015.
26. Shapiro NI, Howell MD, Bates D, et al. The association of sepsis syndrome and organ dysfunction with mortality in emergency department patients with suspected infection. Ann Emerg Med 2006;48:583–90.
27. Berger T, Green J, Horeczko T, et al. Shock index and early recognition of sepsis in the emergency department: pilot study. West J Emerg Med 2013;14:168–74.
28. Farley H, Zubrow MT, Gies J, et al. Emergency department tachypnea predicts transfer to a higher level of care in the first 24 hours after ED admission. Acad Emerg Med 2010;17:718–22.
29. Sinuff T, Muscadere J, Adhikari NK, et al. Knowledge translation interventions for critically ill patients: a systematic review. Crit Care Med 2013;41:2627–40.
30. Hoot NR, Aronsky D. Systematic review of emergency department crowding: causes, effects, and solutions. Ann Emerg Med 2008;52:126–36.
31. Hsia RY, Kellermann AL, Shen YC. Factors associated with closures of emergency departments in the United States. JAMA 2011;305:1978–85.
32. Chalfin DB, Trzeciak S, Likourezos A, et al. Impact of delayed transfer of critically ill patients from the emergency department to the intensive care unit. Crit Care Med 2007;35:1477–83.
33. Subbe CP, Kruger M, Rutherford P, et al. Validation of a modified early warning score in medical admissions. Q J Med 2001;94:521–6.
34. Umscheid CA, Betesh J, VanZandbergen C, et al. Development, implementation, and impact of an automated early warning and response system for sepsis. J Hosp Med 2015;10:26–31.
35. Churpek MM, Yuen TC, Winslow C, et al. Multicenter development and validation of a risk stratification tool for ward patients. Am J Respir Crit Care Med 2014;190:649–55.
36. Whittaker SA, Fuchs BD, Gaieski DF, et al. Epidemiology and outcomes in patients with severe sepsis admitted to the hospital wards. J Crit Care 2015;30:78–84.
37. Delgado MK, Liu V, Pines JM, et al. Risk factors for unplanned transfer to intensive care within 24 hours of admission from the emergency department in an integrated healthcare system. J Hosp Med 2013;8:13–9.
38. Valentini I, Pacilli AM, Carbonara P, et al. Influence of the admission pattern on the outcome of patients admitted to a respiratory intensive care unit: does a step-down admission differ from a step-up one? Respir Care 2013;58:2053–60.
39. Kompanje EJO, Jansen TC, van der Hoven B, Bakker J. The first demonstration of lactic acid in human blood in shock by Johann Joseph Scherer (1814-1869) in January 1843. Intensive Care Med 2007;33:1967–71.
40. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014;371:2309–19.
41. Jansen TC, van Bommel J, Mulder PG, et al. The prognostic value of blood lactate levels relative to that of vital signs in the pre-hospital setting: a pilot study. Crit Care 2008;12:R160.
42. del Portal DA, Shofer F, Mikkelsen ME, et al. Emergency department lactate is associated with mortality in older adults admitted with and without infections. Acad Emerg Med 2010;17:260–8.
43. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med 2009;37:1670–7.
44. Gaieski DF, Drumheller BC, Goyal M, et al. Accuracy of handheld point-of-care fingertip lactate measurement in the emergency department. West J Emerg Med 2013;14:58–62.
45. Macdonald SP, Arendts G, Fatovich DM, Brown SG. Comparison of PIRO, SOFA, and MEDS scores for predicting mortality in emergency department patients with severe sepsis and septic shock. Acad Emerg Med 2014;21:1257–63.
46. Carpenter CR, Keim SM, Upadhye S, Nguyen HB, Group BEiEMI. Risk stratification of the potentially septic patient in the emergency department: the Mortality in the Emergency Department Sepsis (MEDS) score. J Emerg Med 2009;37:319–27.
47. Sankoff JD, Goyal M, Gaieski DF, et al. Validation of the Mortality in Emergency Department Sepsis (MEDS) score in patients with the systemic inflammatory response syndrome (SIRS). Crit Care Med 2008;36:421–6.
48. Corfield AR, Lees F, Zealley I, et al. Utility of a single early warning score in patients with sepsis in the emergency department. Emerg Med J 2014;31:482–7.
49. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34:1589–96.
50. Gaieski DF, Mikkelsen ME, Band RA, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergency department. Crit Care Med 2010;38:1045–53.
51. Puskarich MA, Trzeciak S, Shapiro NI, et al. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011;39:2066–71.
52. Capp R, Horton CL, Takhar SS, et al. Predictors of patients who present to the emergency department with sepsis and progress to septic shock between 4 and 48 hours of emergency department arrival. Crit Care Med 2015 Jan 30.
53. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–77.
54. The ProCESS Investigators. A ranodmized trial of protocol-based care for early septic shock. N Engl J Med 2014;370:1683–93.
55. The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directed resuscitation for patients with early septic shock. N Engl J Med 2014;371:1496–506.
56. Mouncey PR, Osborn TM, Power GS, et al. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med 2015; online March 17, 2015.
57. Weingart SD, Sherwin RL, Emlet LL, et al. ED intensivists and ED intensive care units. Amer J Emerg Med 2013;31:617–20.
58. Lilly CM, Cody S, Zhao H, et al. Hospital mortality, length of stay, and preventable complications among critically ill patients before and after tele-ICU reengineering of critical care processes. JAMA 2011;305:2175–85.
59. Walkey AJ, Wiener RS. Hospital case volume and outcomes among patients hospitalized with severe sepsis. Am J Respir Crit Care Med 2014;189:548–55.
60. Gaieski DF, Edwards JM, Kallan MJ, et al. The relationship between hospital volume and mortality in severe sepsis. Am J Respir Crit Care Med 2014;190:665–74.
61. Levy MM, Dellinger RP, Townsend SR, et al. The surviving sepsis campaign: results of an international guideline-based performance improvement program targeting severe sepsis. Intensive Care Med 2010;36:222–31.
62. Levy MM, Rhodes A, Phillips GS, et al. Surviving sepsis campaign: association between performance metrics and outcomes in a 7.5-year study. Crit Care Med 2015;43:3–12.
63. Weiss CH, Moazed F, McEvoy CA, et al. Prompting physicians to address a daily checklist and process of care and clinical outcomes: a single-site study. Am J Respir Crit Care Med 2011;184:680–6.
64. Weant KA, Baker SN. Emergency medicine pharmacists and sepsis management. J Pharm Pract 2013;26:401–5.
65. Marwick CA, Guthrie B, Pringle JE, et al. A multifaceted intervention to improve sepsis management in general hospital wards with evaluation using segmented regression of interrupted time series. BMJ Qual Saf 2014;23:e2.
From the Department of Medicine, University of Pennsylvania, and the Department of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, PA.
Abstract
- Objective: To detail strategies to improve sepsis recognition and the quality of care provided to the septic patient.
- Methods: Review of the literature.
- Results: Severe sepsis affects nearly 3 million individuals each year in the United States, and cost estimates for these hospitalizations exceed $24 billion. Effective management is predicated on timely recognition. In this review, we detail strategies to improve early identification of potentially septic patients as well as the quality of care provided to the septic patient in the emergency department (ED). The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients at risk for clinical deterioration.
- Conclusion: ED structures and processes can be used to increase adherence with sepsis management guidelines to improve patient outcomes.
Severe sepsis affects nearly 3 million individuals each year in the United States and cost estimates for these hospitalizations exceed $24 billion [1–3]. Sepsis is a life-threatening condition characterized by a suspected or identified infection accompanied by a vigorous host inflammatory response. In severe sepsis, end-organ dysfunction manifests in myriad forms, including altered mental status, acute kidney injury, liver dysfunction, pulmonary dysfunction, and hemodynamic compromise [4,5]. This protean presentation of a deadly condition makes identification and risk stratification both challenging and essential to improving patient outcomes. The majority of patients with severe sepsis will receive their initial care within an emergency department (ED) [6,7]. It is essential that emergency medicine providers have the means to appropriately identify patients presenting with severe sepsis in a timely manner—thus facilitating life-saving measures such as early intravenous fluid resuscitation and administration of timely and appropriate antimicrobials.
In this review, we detail strategies to improve sepsis recognition and the quality of care provided to the septic patient in the ED. The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients for clinical deterioration. Then, we review suggested ED structures and processes to increase adherence with sepsis-based guidelines to improve patient outcomes. Successful implementation is predicated on hospital administrative support towards the efforts given the time and resources required and strong and committed leadership across the health care system.
Epidemiology of Severe Sepsis
Estimates of annual cases of severe sepsis vary, ranging from 1 million to 3 million cases in the United States [1–3]. In-hospital mortality for this condition ranges from 14% to 30% [5]. The incidence of severe sepsis in the United States has been increasing at a rate of 13% annually, with an estimated cost of greater than $24 billion per year [1,2]. In 2 large cohorts of hospitalized patients, it was found that sepsis contributed to 1 in every 2 to 3 deaths following inpatient admission [8]. Coincident with these increased estimates, advances in the early identification and treatment of sepsis have led to decreasing mortality rates over the past decade [1,9].
Of importance to the ED clinician, an episode of sepsis has long-term effects on cognitive and physical function, quality-of-life, and survival [10,11]. Post-discharge, approximately one-quarter of sepsis survivors will be readmitted within 30 days [12–14]. In as many as half of these instances, another life-threatening infection is the cause for readmission, making the past medical history, including a detailed accounting of recent episodes of sepsis, an important part of the initial ED evaluation [12]. Furthermore, severe sepsis survivors spend a large proportion of their time following discharge within a health care facility, and will frequently present to the ED with an acute condition from such an environment. Important factors for predicting readmission after a sepsis hospitalization include patient age, severity of illness, hospital length of stay, and the need for intensive care during the initial hospitalization [12–14].
Principles of Effective Sepsis Management
The principles of effective sepsis management begin with early identification in the pre-hospital setting, at triage, or when a patient begins to decompensate in the hospital. After the point of initial recognition, core principles include risk stratification, timely and appropriate antimicrobial administration, initial intravenous fluid boluses and ongoing resuscitation guided by physical examination and objective resuscitation end-points [4,5]. These practices have been operationalized in the care bundles of the Surviving Sepsis Campaign Guidelines [4]. Within 3 hours, the resuscitation bundle includes measuring serum lactate to risk stratify patients, obtaining blood cultures, administering broad-spectrum antibiotics, and administering 30 mL/kg crystalloid in patients with hypotension or hyperlactatemia [4]. The 6-hour bundle expands upon these initial measures and includes additional management recommendations based on resuscitation end-points.
As effective management is predicated on timely recognition, an understanding of the impact of delayed recognition is essential to provide optimal care for the severe sepsis patient in the ED. Decades of research has revealed that certain markers predict adverse outcomes, including transition to septic shock and death, as do delayed processes of care. Importantly, while early quantitative resuscitation was demonstrated to improve outcomes in a meta-analysis, there was no demonstrable benefit when resuscitation was initiated late (> 24 hours) in the course in the ICU (odds ratio of death, 1.16 [95% confidence interval, 0.60–2.22]) [15].
Strategies To Improve Recognition
Pre-Hospital Environment
From EMS to ED Triage
Borrowing the principle “time equals tissue” from a variety of time sensitive conditions (eg, myocardial infarction management [“time equals muscle”] and stroke care [“time equals brain”]), clinicians and researchers have realized that expedited recognition of severe sepsis patients begins at the time of initial contact with the health care system. For severe sepsis patients, clinicians need to think “time equals organ function.” Given the frequency with which sepsis patients arrive to the ED via EMS, effective communication between EMS and ED providers could be leveraged to prepare the ED team to provide timely care for the sepsis patient via a “sepsis alert.” While confirmation of its applicability to sepsis care is required in the absence of a regionalized network of sepsis centers, the rationale is based on the experience of the effectiveness of trauma and stroke alert systems [20–22]. For patients not recognized as potentially being infected by EMS providers during transport, repeat vital signs during ED triage can be screened to identify patients exhibiting signs of the systemic inflammatory response syndrome (SIRS) [4,23]. The same principles of effective communication apply for patients being sent from medical clinics to the ED for evaluation and treatment of potential severe sepsis. For patients arriving independent of EMS, focused triage and initial vital signs are the starting point for identifying severe sepsis at the most proximal phase of entry into the health care system.
Vital Signs and SIRS Criteria in the ED
The Afferent Arm: Multimodal Screening Strategies
While institutional practice improvement initiatives to facilitate sepsis recognition and care should incorporate educational strategies, led by champions with expertise in sepsis, the complex presentation of sepsis requires multimodal approaches [29]. These multimodal approaches, beginning at the time of ED triage, should be designed to harness information technology to screen patients to improve severe sepsis recognition (the afferent arm) and to utilize structures and processes of care efficiently and effectively (the efferent arm) to guide severe sepsis management according to sepsis-care bundles espoused by guidelines (Figure) [4].
Operational processes to screen for sepsis in the ED will need to account for ED organizational flow (eg, average time from registration to triage, average time from triage to being seen by a physician, average length of stay in the ED, number of hospital beds) and hand-off practices (eg, care transition from ED team to floor or ICU team, or within ED at shift change). For ED organizations with shorter ED lengths of stay (eg, < 2 hours), screening practices at ED triage will serve as the focal point to identify cases of sepsis. Boarding, defined as caring for a patient in the ED pending transfer, is common, increasing as a result of ED closures [30,31], and associated with prolonged hospital length of stay and increased in-hospital mortality when ICU transfer is delayed [32]. Sepsis patients in particular appear to be a vulnerable group of patients. While many explanations exist to account for the relationship between delayed transfer and adverse outcomes, timely recognition and management of the septic patient could be compromised with prolonged boarding. To combat this potential effect, continual assessment during the entire ED stay may unmask an initially unclear presentation of sepsis.
One strategy to identify sepsis in ED organizations with prolonged ED lengths of stay is through the use of a track-and-trigger system, or early warning system. Traditionally, track-and-trigger systems were implemented on the hospital wards, as means to identify physiological deterioration in a timely manner to prevent clinical deterioration [33]. More recently, early warning systems have been used to identify patients with sepsis on the hospital wards and within EDs, as these systems rely on physiological parameters such as SIRS that are cardinal features of sepsis [34]. However, given the potential for alert fatigue, designing a system that operates with high accuracy is imperative.
Efforts are underway to redefine sepsis, using a simplified approach and readily available physiological variables, with the main goal of targeting those most at-risk of an adverse outcome during the hospitalization. Simultaneously, an understanding of the overt and more occult manifestations are essential to incorporate into the clinical decision-making and pattern recognition required to identify sepsis in a timely and accurate manner. In Table 2, the signs and symptoms that may serve as flags for severe sepsis are presented.
Mature early warning systems, designed to leverage the electronic medical record (EMR) by capturing vital signs, laboratory measures, (eg, elevated serum creatinine compared to a recent hospitalization) and symptoms (eg, altered mental status), are well-positioned to herald clinical deterioration (eg, cardiac arrest) with improved accuracy [35] and to be applied to sepsis specifically [34]. While sophisticated analytical strategies, such as machine learning, are being used to improve the test characteristics of these early warning systems, iterative, prospective chart review is an essential and complementary performance improvement step to refine the process. Further, chart review affords the opportunity to ensure compliance with sepsis care bundles.
Knowledge of the risk factors associated with development of sepsis is critical for the front-line emergency physician and nurse. Additionally, as many of these risk factors are associated with adverse outcomes, including unplanned ICU transfer and in-hospital mortality, which occur in as many as one out of 8 patients admitted directly to the ward, they have utility for early risk-stratification and triaging purposes in the ED. Advanced age and pre-existing comorbid conditions, particularly an oncologic diagnosis and/or chronic organ dysfunction, are major risk factors for sepsis and worse outcomes result in those who develop sepsis [2]. Further, illness severity, including an elevated serum lactate level, is associated with adverse outcomes. These factors can be incorporated into triage decisions and/or close monitoring for patients admitted to the general ward [36]. Conversely, because patients admitted to the ICU setting and subsequently stepped down through their hospitalization may experience better outcomes compared to patients admitted to the general ward who then require step-up to an ICU setting (37,38), attention to triage practices is critical.
These complementary strategies, which serve as the afferent arm of the system, summon health care providers to the bedside of a vulnerable patient. However, clinical effectiveness in the management of severe sepsis requires a robust, sophisticated, and mature efferent arm capable of delivering expert care to the now recognized septic patient.
Principles of Effective Management Post-Recognition
Risk Stratification
An elevated serum lactate level was initially described in pathological states in the mid 19th century by Johann Joseph Scherer [39] and has long been associated with increased mortality in hospitalized patients [40]. Lactate is a useful biomarker for risk stratification in a variety of patients arriving to the ED, particularly those who have been identified at high risk for sepsis. Jansen and colleagues examined the measurement of pre-hospital serum lactate at the time of paramedic on-scene assessment in a group of acutely ill patients [41]. Patients with point-of-care lactate levels of 3.5 mmol/L or greater were found to have an in-hospital mortality of 41% versus 12% for those with lactate levels less than 3.5 mmol/L. Within the population with an elevated lactate, patients with a systolic blood pressure greater than 100 mgHg experienced a mortality of nearly 30%, while it was greater than 50% in hypotensive patients with an elevated lactate, highlighting the value of both hemodynamic and serum lactate measures. Upon arrival to the ED, lactate measurements have a strong correlation with mortality. In one retrospective cohort, lactate level was linearly associated with mortality in a broad array of patients older than age 65 years [42]. An initial serum lactate level in the ED in the intermediate (2.0 – 3.9 mmol/L) or high range (≥ 4 mmol/L) has been associated with increased odds of death 2 to 5 times higher independent of organ dysfunction in severe sepsis specifically [43].
As the association between serum lactate levels and death is independent of organ dysfunction, serum lactate is a simple and reliable tool to both enhance detection and risk-stratify patients presenting to the ED with severe sepsis. Given the frequency with which hyperlactatemia is present in patients with suspected infection [43], operationalizing serum lactate measures with the initial phlebotomy draw is an important step to risk-stratify patients. This step can be coupled later with intravenous fluid resuscitation for those with marked elevations (≥ 4 mmol/L), in accord with guideline recommendations [4]. Screening of initial lactate values can be further expedited by utilizing fingerstick point-of-care lactate devices [44]. Last, while serial lactate measures can be incorporated into triage decisions, there is no clear threshold that warrants ICU admission. Rather, persistent elevations in serum lactate can be used to identify patients who require close observation regardless of their admission location.
Several scoring systems have been developed to augment sepsis risk stratification within the ED. The most prominent of these are the Predisposition Insult Response and Organ failure (PIRO), Sequential Organ Failure Assessment (SOFA), and Mortality in the Emergency Department Sepsis (MEDS) scores, and the National early warning score (NEWS) [45-48]. The MEDS score incorporates host factors including age and co-morbid illness, as well as physiologic and laboratory tests which can be obtained rapidly in an ED setting. Multiple prospective and retrospective examinations of the MEDS scoring systems have demonstrated that it performs optimally in ED patients with sepsis but not those with severe sepsis, in terms of predicting 30-day mortality [46,47]. The PIRO score more extensively incorporates predisposing co-morbidities, physiologic and laboratory parameters, and has been modified to consider presumed source of infection, leading to a stronger predictive ability for mortality in more severely ill patients. In patients presenting to the ED with severe sepsis and septic shock, a prospective observational study found the PIRO to be the best predictor of mortality, compared to SOFA and MEDS scores [45]. In a recent study by Corfield et al, sepsis patients with a higher NEWS, according to initial ED vital signs (temperature, pulse, respiratory rate, systolic blood pressure, oxyhemoglobin saturation) and consciousness level, were significantly more likely to be admitted to an ICU within 48 hours or to experience in-hospital mortality [48].
Timely and Appropriate Antibiotics
In a landmark study published by Kumar and colleagues in 2006, the relationship between timing of antibiotics and mortality was established [49]. In 2731 adult septic shock patients, mortality increased 7.6% for every hour delay in effective antimicrobial administration. A striking finding, given that the study population was limited to patients cared for in the ICU, was the fact that only 50% of patients received appropriate antibiotics within 6 hours of onset of shock and nearly one-quarter of patients did not receive antibiotics until the 15th hour. As a direct result, in-hospital mortality was observed to be 58% in this study.
Over the ensuing decade, a series of studies have demonstrated a narrowing of the quality gap in this regard, and the result has coincided with a significant improvement in survival. In 2010, Gaieski and colleagues demonstrated a significant improvement in the prompt administration of antibiotic delivery in patients presenting to an ED with severe sepsis, with the median time from shock onset (sustained hypotension or lactate ≥ 4 mmol/L) to antibiotics down to 42 minutes [50]. Importantly, consistent with the Kumar study, time to appropriate antibiotics, rather than simply initial antibiotics, remained associated with in-hospital mortality independent of initiating early goal-directed therapy. In 2011, Puskarich and colleagues revealed that time to antibiotics continued to improve and, as a result, the investigators did not identify a relationship between time from triage to antibiotics and in-hospital mortality [51]. However, when antibiotics were delayed until after shock recognition, consistent with the study by Kumar and colleagues, survival decreased. Until recently, this important observation was challenging to operationalize clinically as little was known about how to facilitate risk-stratification of those at risk to develop shock. However, Capp and colleagues recently found that deterioration to septic shock 48 hours after ED presentation occurs in approximately one out of eight patients and identified gender (female), transient hypotension, and/or hyperlactatemia upon presentation as risk factors associated with such a deterioration [52].
As an essential element of sepsis care bundles, a focus on timely use of antibiotics in patients with suspected infection, has the potential to increase the use of antibiotics in the ED in patients determined subsequently to not be infected. To combat this acknowledged downstream effect, reconsideration of the utility of empiric antibiotics 48 to 72 hours after admission is required. This step can be accomplished through the use of a sepsis care pathway and/or a formal antibiotic stewardship program.
Quantitative Resuscitation
Rivers and colleagues, in a landmark 2001 trial, examined the effectiveness of a protocolized resuscitation strategy in the most proximal phase of severe sepsis and septic shock [53]. A distinguishing characteristic between the usual care arm and the intervention in this ED-based study, in addition to whether mixed central venous oxygen saturation was measured as a resuscitation end-point, was the inclusion of an ED provider at the bedside to attend to clinical management. The intervention, aimed at achieving physiologic targets, resulted in significantly more fluid resuscitation (3.5 L vs. 5.0 L within the first 6 hours) and a significant decrease in in-hospital mortality compared to the usual care arm (46.5 vs. 30.5%). The study revolutionized the culture and practice of sepsis care, in part by shining a light on the importance of timely resuscitation at the most proximal point of contact between the patient and the healthcare system. It also highlighted the importance of integrating serum lactate measurement into the early screening and risk stratification processes for sepsis care delivery.
The 2014 randomized trial of Protocol-Based Care for Early Septic Shock (ProCESS) revisited this concept, comparing the Rivers 2001 protocol to both a current guideline-based non-invasive algorithmic protocol and what had become usual ED care in the interim [54]. The ProCESS trial, which operationalized a team of bedside providers to direct care for each of the 3 distinct arms, found no significant difference between the arms in terms of 90-day and 1-year mortality, but mortality was approximately 10% less in all arms compared with the intervention arm of the Rivers trial. Further, subjects in each of the 3 arms received in excess of 2 L intravenous fluid resuscitation pre-randomization and 4.4–5.5 L when resuscitation spanned from pre-randomization to 6 hours post-randomization. The conclusion drawn is that the commonalities between the arms—early fluid resuscitation, early antibiotics, and the option to use physiologic measures as markers of the adequacy of treatment, all guided by bedside ED providers—are the most important factors for surviving sepsis. And the result is that practitioners have refined these tools over a decade, leading to steady improvements in survival.
Consistent with the ProCESS trial, a recent Australia and New Zealand trial confirmed no significant difference in 90-day mortality between protocolized EGDT and current usual care for septic shock within an ED [55]. Consistent with ProCESS and ProMISe [56], subjects enrolled in ARISE received in excess of 2.5 L in resuscitation pre-randomization, which when paired with fluid resuscitation in the 0-6 hour post-randomization period (1.96 L in the EGDT arm and 1.71 in the usual-care arm) resulted in resuscitation in the 4.5 to 5L range during the initial resuscitation. The ARISE trial was unique in that appropriate antibiotic administration was a requirement prior to randomization, ensuring that this important driver of mortality reduction was standardized between the two arms of the trial. In summary, while the ideal fluid resuscitation amount is unknown, requires a personalized approach, and further investigation is required to effectively incorporate non-invasive measures to guide fluid responsiveness, early and aggressive resuscitation paired with early antibiotic administration are essential aspects of effective sepsis management.
The Efferent Arm: Structure And Processes To Improve Outcomes
Personnel and Staffing
Quality care for the septic patient requires immediate availability of a multidisciplinary care team, including physicians and nurses with critical care experience who can be rapidly deployed to the bedside. The location of care provision may include on-going care in the initial ED room assignment or transfer to a dedicated area for the care of the critically ill patient within the ED.
To provide optimal care in the era of overcrowding and delayed transfer to an ICU, a movement towards ED intensive care units (ED-ICUs) has emerged [57]. The models of practice range from a model based upon ED intensivists, with expertise in critical care medicine, providing care within the traditional structure of an ED, to a model wherein a portion of the ED is assigned for the care of the critically ill for extended periods of time beyond the initial resuscitation. As these models mature from resuscitation bays capable of scaling up based on need to dedicated ED-ICUs, investments in shared Unit leadership (physician and nursing), staffing (physician, critical care nursing, respiratory therapy, critical care pharmacist) and processes of care (eg, multidisciplinary rounds) in line with established ICUs will be necessary.
While attractive conceptually, large-scale implementation of this movement is unlikely to occur outside of tertiary care academic medical centers. In the many EDs across the US without ED intensivists, and confronted with limited clinician resources, flexible physician and nursing staffing models will be necessary to ensure that care provisions are in accord with established guidelines. Potential solutions to provide the resources to meet the needs of these high-intensity patients include critical care consultation and a strategy traditionally applied to the ICU, telemedicine [58]. Last, given the relationship between hospital volume and mortality in severe sepsis [59,60], timely transfer to a high-volume center for specific cases may be appropriate, although the optimal timing, case selection, and impact of transfer on outcomes warrant further examination.
Clinical Decision Support Strategies
To complement the identification and risk-stratification available by screening and scoring systems, clinical decision support systems are novel tools to improve outcomes in the era of electronic medical records (EMR). Specific to sepsis care delivery, performance improvement initiatives including audit-and-feedback practice can increase severe sepsis guideline adherence, and even modest improvements in adherence appear to lead to sustained improvements that contributed to a 25% relative risk reduction in the observed mortality rate [61,62]. Clinical decision support tools can be used to link early recognition to optimal care processes, such as the Surviving Sepsis Campaign resuscitation and management bundles. The use of prompts as strategies to ensure that bundles of care are ordered and carried out is an important aspect to operationalize during the design phase [63].
Significant preparation is required to effectively carry out the clinical decision support design strategy. For example, to ensure timely antibiotic dispensing, a number of process steps will be required, including prompt notification to a central pharmacist or preferably, an ED pharmacist with access to a local pharmacy pre-stocked with commonly used antibiotics [64]. In addition, the use of an institution-specific antibiogram within the physician computer-order entry sepsis order set, that includes site-specific recommendations (eg, pulmonary, gastrointestinal source) and susceptibility patterns, is an essential aspect of optimal sepsis processes of care. Last, the antibiogram will need to be frequently updated to include season-specific (eg, oseltamivir administration for high-risk cases during influenza season) recommendations to ensure that providers are prompted with the most up-to-date clinical information.
Audit and Feedback and Continuous Performance Improvement
The multimodal approach required to translate knowledge (eg, guidelines) into sepsis care implemented at the bedside is an iterative process. An ED armed with a robust track-and-trigger system and an effective efferent arm, including sophisticated clinical decision support strategies, will require frequent auditing in the plan-do-study-act model of quality improvement to yield clinical effectiveness [61,62,65]. Auditing, paired with feedback to frontline providers, is essential to refine and improve the complex process required to provide expert care to the septic patient [29,65]. Sustained success in optimizing sepsis care delivery is the goal, yet significant work is required to determine the best strategies to achieve this endpoint.
Conclusion
Severe sepsis affects millions of individuals each year in the United States. Delays in recognition result in increased morbidity and mortality, at a tremendous cost to the patient and society. By designing strategies to identify sepsis in a timely, efficient, and effective manner, and by implementing ED structures and processes to increase adherence with sepsis-based guidelines, improved patient-centered outcomes can be realized.
Corresponding author: Mark E. Mikkelsen, MD, MSCE, Gates 05.042, 3400 Spruce St., Philadelphia, PA 19104, mark.mikkelsen@uphs.upenn.edu.
Financial disclosures: None.
Author contributions: conception and design, JHM, MEM; analysis and interpretation of data, DFG; drafting of article, JHM, DFG, MEM; critical revision of the article, JHM, MEM.
From the Department of Medicine, University of Pennsylvania, and the Department of Emergency Medicine, Thomas Jefferson University Hospital, Philadelphia, PA.
Abstract
- Objective: To detail strategies to improve sepsis recognition and the quality of care provided to the septic patient.
- Methods: Review of the literature.
- Results: Severe sepsis affects nearly 3 million individuals each year in the United States, and cost estimates for these hospitalizations exceed $24 billion. Effective management is predicated on timely recognition. In this review, we detail strategies to improve early identification of potentially septic patients as well as the quality of care provided to the septic patient in the emergency department (ED). The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients at risk for clinical deterioration.
- Conclusion: ED structures and processes can be used to increase adherence with sepsis management guidelines to improve patient outcomes.
Severe sepsis affects nearly 3 million individuals each year in the United States and cost estimates for these hospitalizations exceed $24 billion [1–3]. Sepsis is a life-threatening condition characterized by a suspected or identified infection accompanied by a vigorous host inflammatory response. In severe sepsis, end-organ dysfunction manifests in myriad forms, including altered mental status, acute kidney injury, liver dysfunction, pulmonary dysfunction, and hemodynamic compromise [4,5]. This protean presentation of a deadly condition makes identification and risk stratification both challenging and essential to improving patient outcomes. The majority of patients with severe sepsis will receive their initial care within an emergency department (ED) [6,7]. It is essential that emergency medicine providers have the means to appropriately identify patients presenting with severe sepsis in a timely manner—thus facilitating life-saving measures such as early intravenous fluid resuscitation and administration of timely and appropriate antimicrobials.
In this review, we detail strategies to improve sepsis recognition and the quality of care provided to the septic patient in the ED. The strategies discussed are based upon an understanding of the signs and symptoms of sepsis and the clinical risk factors associated with sepsis, which can be used to design novel strategies to screen patients for sepsis and risk stratify patients for clinical deterioration. Then, we review suggested ED structures and processes to increase adherence with sepsis-based guidelines to improve patient outcomes. Successful implementation is predicated on hospital administrative support towards the efforts given the time and resources required and strong and committed leadership across the health care system.
Epidemiology of Severe Sepsis
Estimates of annual cases of severe sepsis vary, ranging from 1 million to 3 million cases in the United States [1–3]. In-hospital mortality for this condition ranges from 14% to 30% [5]. The incidence of severe sepsis in the United States has been increasing at a rate of 13% annually, with an estimated cost of greater than $24 billion per year [1,2]. In 2 large cohorts of hospitalized patients, it was found that sepsis contributed to 1 in every 2 to 3 deaths following inpatient admission [8]. Coincident with these increased estimates, advances in the early identification and treatment of sepsis have led to decreasing mortality rates over the past decade [1,9].
Of importance to the ED clinician, an episode of sepsis has long-term effects on cognitive and physical function, quality-of-life, and survival [10,11]. Post-discharge, approximately one-quarter of sepsis survivors will be readmitted within 30 days [12–14]. In as many as half of these instances, another life-threatening infection is the cause for readmission, making the past medical history, including a detailed accounting of recent episodes of sepsis, an important part of the initial ED evaluation [12]. Furthermore, severe sepsis survivors spend a large proportion of their time following discharge within a health care facility, and will frequently present to the ED with an acute condition from such an environment. Important factors for predicting readmission after a sepsis hospitalization include patient age, severity of illness, hospital length of stay, and the need for intensive care during the initial hospitalization [12–14].
Principles of Effective Sepsis Management
The principles of effective sepsis management begin with early identification in the pre-hospital setting, at triage, or when a patient begins to decompensate in the hospital. After the point of initial recognition, core principles include risk stratification, timely and appropriate antimicrobial administration, initial intravenous fluid boluses and ongoing resuscitation guided by physical examination and objective resuscitation end-points [4,5]. These practices have been operationalized in the care bundles of the Surviving Sepsis Campaign Guidelines [4]. Within 3 hours, the resuscitation bundle includes measuring serum lactate to risk stratify patients, obtaining blood cultures, administering broad-spectrum antibiotics, and administering 30 mL/kg crystalloid in patients with hypotension or hyperlactatemia [4]. The 6-hour bundle expands upon these initial measures and includes additional management recommendations based on resuscitation end-points.
As effective management is predicated on timely recognition, an understanding of the impact of delayed recognition is essential to provide optimal care for the severe sepsis patient in the ED. Decades of research has revealed that certain markers predict adverse outcomes, including transition to septic shock and death, as do delayed processes of care. Importantly, while early quantitative resuscitation was demonstrated to improve outcomes in a meta-analysis, there was no demonstrable benefit when resuscitation was initiated late (> 24 hours) in the course in the ICU (odds ratio of death, 1.16 [95% confidence interval, 0.60–2.22]) [15].
Strategies To Improve Recognition
Pre-Hospital Environment
From EMS to ED Triage
Borrowing the principle “time equals tissue” from a variety of time sensitive conditions (eg, myocardial infarction management [“time equals muscle”] and stroke care [“time equals brain”]), clinicians and researchers have realized that expedited recognition of severe sepsis patients begins at the time of initial contact with the health care system. For severe sepsis patients, clinicians need to think “time equals organ function.” Given the frequency with which sepsis patients arrive to the ED via EMS, effective communication between EMS and ED providers could be leveraged to prepare the ED team to provide timely care for the sepsis patient via a “sepsis alert.” While confirmation of its applicability to sepsis care is required in the absence of a regionalized network of sepsis centers, the rationale is based on the experience of the effectiveness of trauma and stroke alert systems [20–22]. For patients not recognized as potentially being infected by EMS providers during transport, repeat vital signs during ED triage can be screened to identify patients exhibiting signs of the systemic inflammatory response syndrome (SIRS) [4,23]. The same principles of effective communication apply for patients being sent from medical clinics to the ED for evaluation and treatment of potential severe sepsis. For patients arriving independent of EMS, focused triage and initial vital signs are the starting point for identifying severe sepsis at the most proximal phase of entry into the health care system.
Vital Signs and SIRS Criteria in the ED
The Afferent Arm: Multimodal Screening Strategies
While institutional practice improvement initiatives to facilitate sepsis recognition and care should incorporate educational strategies, led by champions with expertise in sepsis, the complex presentation of sepsis requires multimodal approaches [29]. These multimodal approaches, beginning at the time of ED triage, should be designed to harness information technology to screen patients to improve severe sepsis recognition (the afferent arm) and to utilize structures and processes of care efficiently and effectively (the efferent arm) to guide severe sepsis management according to sepsis-care bundles espoused by guidelines (Figure) [4].
Operational processes to screen for sepsis in the ED will need to account for ED organizational flow (eg, average time from registration to triage, average time from triage to being seen by a physician, average length of stay in the ED, number of hospital beds) and hand-off practices (eg, care transition from ED team to floor or ICU team, or within ED at shift change). For ED organizations with shorter ED lengths of stay (eg, < 2 hours), screening practices at ED triage will serve as the focal point to identify cases of sepsis. Boarding, defined as caring for a patient in the ED pending transfer, is common, increasing as a result of ED closures [30,31], and associated with prolonged hospital length of stay and increased in-hospital mortality when ICU transfer is delayed [32]. Sepsis patients in particular appear to be a vulnerable group of patients. While many explanations exist to account for the relationship between delayed transfer and adverse outcomes, timely recognition and management of the septic patient could be compromised with prolonged boarding. To combat this potential effect, continual assessment during the entire ED stay may unmask an initially unclear presentation of sepsis.
One strategy to identify sepsis in ED organizations with prolonged ED lengths of stay is through the use of a track-and-trigger system, or early warning system. Traditionally, track-and-trigger systems were implemented on the hospital wards, as means to identify physiological deterioration in a timely manner to prevent clinical deterioration [33]. More recently, early warning systems have been used to identify patients with sepsis on the hospital wards and within EDs, as these systems rely on physiological parameters such as SIRS that are cardinal features of sepsis [34]. However, given the potential for alert fatigue, designing a system that operates with high accuracy is imperative.
Efforts are underway to redefine sepsis, using a simplified approach and readily available physiological variables, with the main goal of targeting those most at-risk of an adverse outcome during the hospitalization. Simultaneously, an understanding of the overt and more occult manifestations are essential to incorporate into the clinical decision-making and pattern recognition required to identify sepsis in a timely and accurate manner. In Table 2, the signs and symptoms that may serve as flags for severe sepsis are presented.
Mature early warning systems, designed to leverage the electronic medical record (EMR) by capturing vital signs, laboratory measures, (eg, elevated serum creatinine compared to a recent hospitalization) and symptoms (eg, altered mental status), are well-positioned to herald clinical deterioration (eg, cardiac arrest) with improved accuracy [35] and to be applied to sepsis specifically [34]. While sophisticated analytical strategies, such as machine learning, are being used to improve the test characteristics of these early warning systems, iterative, prospective chart review is an essential and complementary performance improvement step to refine the process. Further, chart review affords the opportunity to ensure compliance with sepsis care bundles.
Knowledge of the risk factors associated with development of sepsis is critical for the front-line emergency physician and nurse. Additionally, as many of these risk factors are associated with adverse outcomes, including unplanned ICU transfer and in-hospital mortality, which occur in as many as one out of 8 patients admitted directly to the ward, they have utility for early risk-stratification and triaging purposes in the ED. Advanced age and pre-existing comorbid conditions, particularly an oncologic diagnosis and/or chronic organ dysfunction, are major risk factors for sepsis and worse outcomes result in those who develop sepsis [2]. Further, illness severity, including an elevated serum lactate level, is associated with adverse outcomes. These factors can be incorporated into triage decisions and/or close monitoring for patients admitted to the general ward [36]. Conversely, because patients admitted to the ICU setting and subsequently stepped down through their hospitalization may experience better outcomes compared to patients admitted to the general ward who then require step-up to an ICU setting (37,38), attention to triage practices is critical.
These complementary strategies, which serve as the afferent arm of the system, summon health care providers to the bedside of a vulnerable patient. However, clinical effectiveness in the management of severe sepsis requires a robust, sophisticated, and mature efferent arm capable of delivering expert care to the now recognized septic patient.
Principles of Effective Management Post-Recognition
Risk Stratification
An elevated serum lactate level was initially described in pathological states in the mid 19th century by Johann Joseph Scherer [39] and has long been associated with increased mortality in hospitalized patients [40]. Lactate is a useful biomarker for risk stratification in a variety of patients arriving to the ED, particularly those who have been identified at high risk for sepsis. Jansen and colleagues examined the measurement of pre-hospital serum lactate at the time of paramedic on-scene assessment in a group of acutely ill patients [41]. Patients with point-of-care lactate levels of 3.5 mmol/L or greater were found to have an in-hospital mortality of 41% versus 12% for those with lactate levels less than 3.5 mmol/L. Within the population with an elevated lactate, patients with a systolic blood pressure greater than 100 mgHg experienced a mortality of nearly 30%, while it was greater than 50% in hypotensive patients with an elevated lactate, highlighting the value of both hemodynamic and serum lactate measures. Upon arrival to the ED, lactate measurements have a strong correlation with mortality. In one retrospective cohort, lactate level was linearly associated with mortality in a broad array of patients older than age 65 years [42]. An initial serum lactate level in the ED in the intermediate (2.0 – 3.9 mmol/L) or high range (≥ 4 mmol/L) has been associated with increased odds of death 2 to 5 times higher independent of organ dysfunction in severe sepsis specifically [43].
As the association between serum lactate levels and death is independent of organ dysfunction, serum lactate is a simple and reliable tool to both enhance detection and risk-stratify patients presenting to the ED with severe sepsis. Given the frequency with which hyperlactatemia is present in patients with suspected infection [43], operationalizing serum lactate measures with the initial phlebotomy draw is an important step to risk-stratify patients. This step can be coupled later with intravenous fluid resuscitation for those with marked elevations (≥ 4 mmol/L), in accord with guideline recommendations [4]. Screening of initial lactate values can be further expedited by utilizing fingerstick point-of-care lactate devices [44]. Last, while serial lactate measures can be incorporated into triage decisions, there is no clear threshold that warrants ICU admission. Rather, persistent elevations in serum lactate can be used to identify patients who require close observation regardless of their admission location.
Several scoring systems have been developed to augment sepsis risk stratification within the ED. The most prominent of these are the Predisposition Insult Response and Organ failure (PIRO), Sequential Organ Failure Assessment (SOFA), and Mortality in the Emergency Department Sepsis (MEDS) scores, and the National early warning score (NEWS) [45-48]. The MEDS score incorporates host factors including age and co-morbid illness, as well as physiologic and laboratory tests which can be obtained rapidly in an ED setting. Multiple prospective and retrospective examinations of the MEDS scoring systems have demonstrated that it performs optimally in ED patients with sepsis but not those with severe sepsis, in terms of predicting 30-day mortality [46,47]. The PIRO score more extensively incorporates predisposing co-morbidities, physiologic and laboratory parameters, and has been modified to consider presumed source of infection, leading to a stronger predictive ability for mortality in more severely ill patients. In patients presenting to the ED with severe sepsis and septic shock, a prospective observational study found the PIRO to be the best predictor of mortality, compared to SOFA and MEDS scores [45]. In a recent study by Corfield et al, sepsis patients with a higher NEWS, according to initial ED vital signs (temperature, pulse, respiratory rate, systolic blood pressure, oxyhemoglobin saturation) and consciousness level, were significantly more likely to be admitted to an ICU within 48 hours or to experience in-hospital mortality [48].
Timely and Appropriate Antibiotics
In a landmark study published by Kumar and colleagues in 2006, the relationship between timing of antibiotics and mortality was established [49]. In 2731 adult septic shock patients, mortality increased 7.6% for every hour delay in effective antimicrobial administration. A striking finding, given that the study population was limited to patients cared for in the ICU, was the fact that only 50% of patients received appropriate antibiotics within 6 hours of onset of shock and nearly one-quarter of patients did not receive antibiotics until the 15th hour. As a direct result, in-hospital mortality was observed to be 58% in this study.
Over the ensuing decade, a series of studies have demonstrated a narrowing of the quality gap in this regard, and the result has coincided with a significant improvement in survival. In 2010, Gaieski and colleagues demonstrated a significant improvement in the prompt administration of antibiotic delivery in patients presenting to an ED with severe sepsis, with the median time from shock onset (sustained hypotension or lactate ≥ 4 mmol/L) to antibiotics down to 42 minutes [50]. Importantly, consistent with the Kumar study, time to appropriate antibiotics, rather than simply initial antibiotics, remained associated with in-hospital mortality independent of initiating early goal-directed therapy. In 2011, Puskarich and colleagues revealed that time to antibiotics continued to improve and, as a result, the investigators did not identify a relationship between time from triage to antibiotics and in-hospital mortality [51]. However, when antibiotics were delayed until after shock recognition, consistent with the study by Kumar and colleagues, survival decreased. Until recently, this important observation was challenging to operationalize clinically as little was known about how to facilitate risk-stratification of those at risk to develop shock. However, Capp and colleagues recently found that deterioration to septic shock 48 hours after ED presentation occurs in approximately one out of eight patients and identified gender (female), transient hypotension, and/or hyperlactatemia upon presentation as risk factors associated with such a deterioration [52].
As an essential element of sepsis care bundles, a focus on timely use of antibiotics in patients with suspected infection, has the potential to increase the use of antibiotics in the ED in patients determined subsequently to not be infected. To combat this acknowledged downstream effect, reconsideration of the utility of empiric antibiotics 48 to 72 hours after admission is required. This step can be accomplished through the use of a sepsis care pathway and/or a formal antibiotic stewardship program.
Quantitative Resuscitation
Rivers and colleagues, in a landmark 2001 trial, examined the effectiveness of a protocolized resuscitation strategy in the most proximal phase of severe sepsis and septic shock [53]. A distinguishing characteristic between the usual care arm and the intervention in this ED-based study, in addition to whether mixed central venous oxygen saturation was measured as a resuscitation end-point, was the inclusion of an ED provider at the bedside to attend to clinical management. The intervention, aimed at achieving physiologic targets, resulted in significantly more fluid resuscitation (3.5 L vs. 5.0 L within the first 6 hours) and a significant decrease in in-hospital mortality compared to the usual care arm (46.5 vs. 30.5%). The study revolutionized the culture and practice of sepsis care, in part by shining a light on the importance of timely resuscitation at the most proximal point of contact between the patient and the healthcare system. It also highlighted the importance of integrating serum lactate measurement into the early screening and risk stratification processes for sepsis care delivery.
The 2014 randomized trial of Protocol-Based Care for Early Septic Shock (ProCESS) revisited this concept, comparing the Rivers 2001 protocol to both a current guideline-based non-invasive algorithmic protocol and what had become usual ED care in the interim [54]. The ProCESS trial, which operationalized a team of bedside providers to direct care for each of the 3 distinct arms, found no significant difference between the arms in terms of 90-day and 1-year mortality, but mortality was approximately 10% less in all arms compared with the intervention arm of the Rivers trial. Further, subjects in each of the 3 arms received in excess of 2 L intravenous fluid resuscitation pre-randomization and 4.4–5.5 L when resuscitation spanned from pre-randomization to 6 hours post-randomization. The conclusion drawn is that the commonalities between the arms—early fluid resuscitation, early antibiotics, and the option to use physiologic measures as markers of the adequacy of treatment, all guided by bedside ED providers—are the most important factors for surviving sepsis. And the result is that practitioners have refined these tools over a decade, leading to steady improvements in survival.
Consistent with the ProCESS trial, a recent Australia and New Zealand trial confirmed no significant difference in 90-day mortality between protocolized EGDT and current usual care for septic shock within an ED [55]. Consistent with ProCESS and ProMISe [56], subjects enrolled in ARISE received in excess of 2.5 L in resuscitation pre-randomization, which when paired with fluid resuscitation in the 0-6 hour post-randomization period (1.96 L in the EGDT arm and 1.71 in the usual-care arm) resulted in resuscitation in the 4.5 to 5L range during the initial resuscitation. The ARISE trial was unique in that appropriate antibiotic administration was a requirement prior to randomization, ensuring that this important driver of mortality reduction was standardized between the two arms of the trial. In summary, while the ideal fluid resuscitation amount is unknown, requires a personalized approach, and further investigation is required to effectively incorporate non-invasive measures to guide fluid responsiveness, early and aggressive resuscitation paired with early antibiotic administration are essential aspects of effective sepsis management.
The Efferent Arm: Structure And Processes To Improve Outcomes
Personnel and Staffing
Quality care for the septic patient requires immediate availability of a multidisciplinary care team, including physicians and nurses with critical care experience who can be rapidly deployed to the bedside. The location of care provision may include on-going care in the initial ED room assignment or transfer to a dedicated area for the care of the critically ill patient within the ED.
To provide optimal care in the era of overcrowding and delayed transfer to an ICU, a movement towards ED intensive care units (ED-ICUs) has emerged [57]. The models of practice range from a model based upon ED intensivists, with expertise in critical care medicine, providing care within the traditional structure of an ED, to a model wherein a portion of the ED is assigned for the care of the critically ill for extended periods of time beyond the initial resuscitation. As these models mature from resuscitation bays capable of scaling up based on need to dedicated ED-ICUs, investments in shared Unit leadership (physician and nursing), staffing (physician, critical care nursing, respiratory therapy, critical care pharmacist) and processes of care (eg, multidisciplinary rounds) in line with established ICUs will be necessary.
While attractive conceptually, large-scale implementation of this movement is unlikely to occur outside of tertiary care academic medical centers. In the many EDs across the US without ED intensivists, and confronted with limited clinician resources, flexible physician and nursing staffing models will be necessary to ensure that care provisions are in accord with established guidelines. Potential solutions to provide the resources to meet the needs of these high-intensity patients include critical care consultation and a strategy traditionally applied to the ICU, telemedicine [58]. Last, given the relationship between hospital volume and mortality in severe sepsis [59,60], timely transfer to a high-volume center for specific cases may be appropriate, although the optimal timing, case selection, and impact of transfer on outcomes warrant further examination.
Clinical Decision Support Strategies
To complement the identification and risk-stratification available by screening and scoring systems, clinical decision support systems are novel tools to improve outcomes in the era of electronic medical records (EMR). Specific to sepsis care delivery, performance improvement initiatives including audit-and-feedback practice can increase severe sepsis guideline adherence, and even modest improvements in adherence appear to lead to sustained improvements that contributed to a 25% relative risk reduction in the observed mortality rate [61,62]. Clinical decision support tools can be used to link early recognition to optimal care processes, such as the Surviving Sepsis Campaign resuscitation and management bundles. The use of prompts as strategies to ensure that bundles of care are ordered and carried out is an important aspect to operationalize during the design phase [63].
Significant preparation is required to effectively carry out the clinical decision support design strategy. For example, to ensure timely antibiotic dispensing, a number of process steps will be required, including prompt notification to a central pharmacist or preferably, an ED pharmacist with access to a local pharmacy pre-stocked with commonly used antibiotics [64]. In addition, the use of an institution-specific antibiogram within the physician computer-order entry sepsis order set, that includes site-specific recommendations (eg, pulmonary, gastrointestinal source) and susceptibility patterns, is an essential aspect of optimal sepsis processes of care. Last, the antibiogram will need to be frequently updated to include season-specific (eg, oseltamivir administration for high-risk cases during influenza season) recommendations to ensure that providers are prompted with the most up-to-date clinical information.
Audit and Feedback and Continuous Performance Improvement
The multimodal approach required to translate knowledge (eg, guidelines) into sepsis care implemented at the bedside is an iterative process. An ED armed with a robust track-and-trigger system and an effective efferent arm, including sophisticated clinical decision support strategies, will require frequent auditing in the plan-do-study-act model of quality improvement to yield clinical effectiveness [61,62,65]. Auditing, paired with feedback to frontline providers, is essential to refine and improve the complex process required to provide expert care to the septic patient [29,65]. Sustained success in optimizing sepsis care delivery is the goal, yet significant work is required to determine the best strategies to achieve this endpoint.
Conclusion
Severe sepsis affects millions of individuals each year in the United States. Delays in recognition result in increased morbidity and mortality, at a tremendous cost to the patient and society. By designing strategies to identify sepsis in a timely, efficient, and effective manner, and by implementing ED structures and processes to increase adherence with sepsis-based guidelines, improved patient-centered outcomes can be realized.
Corresponding author: Mark E. Mikkelsen, MD, MSCE, Gates 05.042, 3400 Spruce St., Philadelphia, PA 19104, mark.mikkelsen@uphs.upenn.edu.
Financial disclosures: None.
Author contributions: conception and design, JHM, MEM; analysis and interpretation of data, DFG; drafting of article, JHM, DFG, MEM; critical revision of the article, JHM, MEM.
1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41:1167–74.
2. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–10.
3. Lagu T, Rothberg MB, Shieh MS, et al. Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007. Crit Care Med 2012;40:754–61.
4. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med 2013;39:165–228.
5. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013;369:840–51.
6. Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis in United States emergency departments. Crit Care Med 2007;35:1928–36.
7. Dombrovskiy VY, Martin AA, Sunderram J, et al. Rapid increase in hospitalization and mortality rates for severe sepsis in the United States: a trend analysis from 1993 to 2003. Crit Care Med 2007;35:1244–50.
8. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA 2014;312:90–2.
9. Kaukonen KM, Bailey M, Suzuki S, et al. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA 2014;311:1308–16.
10. Yende S, Angus DC. Long-term outcomes from sepsis. Curr Infect Dis Rep 2007;9:382–6.
11. Iwashyna TJ, Ely EW, Smith DM, et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA 2010; 304:1787–94.
12. Ortego A, Gaieski DF, Fuchs BD, et al. Hospital-based acute care use in survivors of septic shock. Crit Care Med 2015;43:729–37.
13. Prescott HC, Langa KM, Liu V, et al. Increased 1-year healthcare use in survivors of severe sepsis. Am J Respir Crit Care Med 2014;190:62–9.
14. Liu V, Lei X, Prescott HC, et al. Hospital readmission and healthcare utilization following sepsis in community settings. J Hosp Med 2014;9:502–7.
15. Jones AE, Brown MD, Trzeciak S, et al. The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta-analysis. Crit Care Med 2008;36:2734–9.
16. Seymour CW, Rea TD, Kahn JM, et al. Severe sepsis in pre-hospital emergency care: analysis of incidence, care, and outcome. Am J Respir Crit Care Med 2012;186:1264–71.
17. Seymour CW, Cooke CR, Mikkelsen ME, et al. Out-of-hospital fluid in severe sepsis: effect on early resuscitation in the emergency department. Prehosp Emerg Care 2010;14:145–52.
18. Seymour CW, Cooke CR, Heckbert SR, et al. Prehospital intravenous access and fluid resuscitation in severe sepsis: an observational cohort study. Crit Care 2014;18:533
19. Studnek JR, Artho MR, Garner CL, Jones AE. The impact of emergency medical services on the ED care of severe sepsis. Am J Emerg Med 2012;30:51–6.
20. Guss DA, Meyer FT, Neuman TS, et al. The impact of a regionalized trauma system on trauma care in San Diego County. Ann Emerg Med 1989;18:1141–5.
21. Liberman M, Mulder DS, Jurkovich GJ, Sampalis JS. The association between trauma system and trauma center components and outcome in a mature regionalized trauma system. Surgery 2005;137:647–58.
22. Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working toward a prioritized world agenda. Stroke 2010;41:1084–99.
23. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31:1250–6.
24. Sibbald W, Doig G, Inman K. Sepsis, SIRS, and infection. Intensive Care Med 1995;21:299–301.
25. Kaukonen KM, Bailey M, Pilcher D, et al. Systemic inflammatory response syndrome criteria in defining severe sepsis. N Engl J Med 2015; online March 17, 2015.
26. Shapiro NI, Howell MD, Bates D, et al. The association of sepsis syndrome and organ dysfunction with mortality in emergency department patients with suspected infection. Ann Emerg Med 2006;48:583–90.
27. Berger T, Green J, Horeczko T, et al. Shock index and early recognition of sepsis in the emergency department: pilot study. West J Emerg Med 2013;14:168–74.
28. Farley H, Zubrow MT, Gies J, et al. Emergency department tachypnea predicts transfer to a higher level of care in the first 24 hours after ED admission. Acad Emerg Med 2010;17:718–22.
29. Sinuff T, Muscadere J, Adhikari NK, et al. Knowledge translation interventions for critically ill patients: a systematic review. Crit Care Med 2013;41:2627–40.
30. Hoot NR, Aronsky D. Systematic review of emergency department crowding: causes, effects, and solutions. Ann Emerg Med 2008;52:126–36.
31. Hsia RY, Kellermann AL, Shen YC. Factors associated with closures of emergency departments in the United States. JAMA 2011;305:1978–85.
32. Chalfin DB, Trzeciak S, Likourezos A, et al. Impact of delayed transfer of critically ill patients from the emergency department to the intensive care unit. Crit Care Med 2007;35:1477–83.
33. Subbe CP, Kruger M, Rutherford P, et al. Validation of a modified early warning score in medical admissions. Q J Med 2001;94:521–6.
34. Umscheid CA, Betesh J, VanZandbergen C, et al. Development, implementation, and impact of an automated early warning and response system for sepsis. J Hosp Med 2015;10:26–31.
35. Churpek MM, Yuen TC, Winslow C, et al. Multicenter development and validation of a risk stratification tool for ward patients. Am J Respir Crit Care Med 2014;190:649–55.
36. Whittaker SA, Fuchs BD, Gaieski DF, et al. Epidemiology and outcomes in patients with severe sepsis admitted to the hospital wards. J Crit Care 2015;30:78–84.
37. Delgado MK, Liu V, Pines JM, et al. Risk factors for unplanned transfer to intensive care within 24 hours of admission from the emergency department in an integrated healthcare system. J Hosp Med 2013;8:13–9.
38. Valentini I, Pacilli AM, Carbonara P, et al. Influence of the admission pattern on the outcome of patients admitted to a respiratory intensive care unit: does a step-down admission differ from a step-up one? Respir Care 2013;58:2053–60.
39. Kompanje EJO, Jansen TC, van der Hoven B, Bakker J. The first demonstration of lactic acid in human blood in shock by Johann Joseph Scherer (1814-1869) in January 1843. Intensive Care Med 2007;33:1967–71.
40. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014;371:2309–19.
41. Jansen TC, van Bommel J, Mulder PG, et al. The prognostic value of blood lactate levels relative to that of vital signs in the pre-hospital setting: a pilot study. Crit Care 2008;12:R160.
42. del Portal DA, Shofer F, Mikkelsen ME, et al. Emergency department lactate is associated with mortality in older adults admitted with and without infections. Acad Emerg Med 2010;17:260–8.
43. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med 2009;37:1670–7.
44. Gaieski DF, Drumheller BC, Goyal M, et al. Accuracy of handheld point-of-care fingertip lactate measurement in the emergency department. West J Emerg Med 2013;14:58–62.
45. Macdonald SP, Arendts G, Fatovich DM, Brown SG. Comparison of PIRO, SOFA, and MEDS scores for predicting mortality in emergency department patients with severe sepsis and septic shock. Acad Emerg Med 2014;21:1257–63.
46. Carpenter CR, Keim SM, Upadhye S, Nguyen HB, Group BEiEMI. Risk stratification of the potentially septic patient in the emergency department: the Mortality in the Emergency Department Sepsis (MEDS) score. J Emerg Med 2009;37:319–27.
47. Sankoff JD, Goyal M, Gaieski DF, et al. Validation of the Mortality in Emergency Department Sepsis (MEDS) score in patients with the systemic inflammatory response syndrome (SIRS). Crit Care Med 2008;36:421–6.
48. Corfield AR, Lees F, Zealley I, et al. Utility of a single early warning score in patients with sepsis in the emergency department. Emerg Med J 2014;31:482–7.
49. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34:1589–96.
50. Gaieski DF, Mikkelsen ME, Band RA, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergency department. Crit Care Med 2010;38:1045–53.
51. Puskarich MA, Trzeciak S, Shapiro NI, et al. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011;39:2066–71.
52. Capp R, Horton CL, Takhar SS, et al. Predictors of patients who present to the emergency department with sepsis and progress to septic shock between 4 and 48 hours of emergency department arrival. Crit Care Med 2015 Jan 30.
53. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–77.
54. The ProCESS Investigators. A ranodmized trial of protocol-based care for early septic shock. N Engl J Med 2014;370:1683–93.
55. The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directed resuscitation for patients with early septic shock. N Engl J Med 2014;371:1496–506.
56. Mouncey PR, Osborn TM, Power GS, et al. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med 2015; online March 17, 2015.
57. Weingart SD, Sherwin RL, Emlet LL, et al. ED intensivists and ED intensive care units. Amer J Emerg Med 2013;31:617–20.
58. Lilly CM, Cody S, Zhao H, et al. Hospital mortality, length of stay, and preventable complications among critically ill patients before and after tele-ICU reengineering of critical care processes. JAMA 2011;305:2175–85.
59. Walkey AJ, Wiener RS. Hospital case volume and outcomes among patients hospitalized with severe sepsis. Am J Respir Crit Care Med 2014;189:548–55.
60. Gaieski DF, Edwards JM, Kallan MJ, et al. The relationship between hospital volume and mortality in severe sepsis. Am J Respir Crit Care Med 2014;190:665–74.
61. Levy MM, Dellinger RP, Townsend SR, et al. The surviving sepsis campaign: results of an international guideline-based performance improvement program targeting severe sepsis. Intensive Care Med 2010;36:222–31.
62. Levy MM, Rhodes A, Phillips GS, et al. Surviving sepsis campaign: association between performance metrics and outcomes in a 7.5-year study. Crit Care Med 2015;43:3–12.
63. Weiss CH, Moazed F, McEvoy CA, et al. Prompting physicians to address a daily checklist and process of care and clinical outcomes: a single-site study. Am J Respir Crit Care Med 2011;184:680–6.
64. Weant KA, Baker SN. Emergency medicine pharmacists and sepsis management. J Pharm Pract 2013;26:401–5.
65. Marwick CA, Guthrie B, Pringle JE, et al. A multifaceted intervention to improve sepsis management in general hospital wards with evaluation using segmented regression of interrupted time series. BMJ Qual Saf 2014;23:e2.
1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41:1167–74.
2. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–10.
3. Lagu T, Rothberg MB, Shieh MS, et al. Hospitalizations, costs, and outcomes of severe sepsis in the United States 2003 to 2007. Crit Care Med 2012;40:754–61.
4. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med 2013;39:165–228.
5. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013;369:840–51.
6. Wang HE, Shapiro NI, Angus DC, Yealy DM. National estimates of severe sepsis in United States emergency departments. Crit Care Med 2007;35:1928–36.
7. Dombrovskiy VY, Martin AA, Sunderram J, et al. Rapid increase in hospitalization and mortality rates for severe sepsis in the United States: a trend analysis from 1993 to 2003. Crit Care Med 2007;35:1244–50.
8. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA 2014;312:90–2.
9. Kaukonen KM, Bailey M, Suzuki S, et al. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA 2014;311:1308–16.
10. Yende S, Angus DC. Long-term outcomes from sepsis. Curr Infect Dis Rep 2007;9:382–6.
11. Iwashyna TJ, Ely EW, Smith DM, et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA 2010; 304:1787–94.
12. Ortego A, Gaieski DF, Fuchs BD, et al. Hospital-based acute care use in survivors of septic shock. Crit Care Med 2015;43:729–37.
13. Prescott HC, Langa KM, Liu V, et al. Increased 1-year healthcare use in survivors of severe sepsis. Am J Respir Crit Care Med 2014;190:62–9.
14. Liu V, Lei X, Prescott HC, et al. Hospital readmission and healthcare utilization following sepsis in community settings. J Hosp Med 2014;9:502–7.
15. Jones AE, Brown MD, Trzeciak S, et al. The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta-analysis. Crit Care Med 2008;36:2734–9.
16. Seymour CW, Rea TD, Kahn JM, et al. Severe sepsis in pre-hospital emergency care: analysis of incidence, care, and outcome. Am J Respir Crit Care Med 2012;186:1264–71.
17. Seymour CW, Cooke CR, Mikkelsen ME, et al. Out-of-hospital fluid in severe sepsis: effect on early resuscitation in the emergency department. Prehosp Emerg Care 2010;14:145–52.
18. Seymour CW, Cooke CR, Heckbert SR, et al. Prehospital intravenous access and fluid resuscitation in severe sepsis: an observational cohort study. Crit Care 2014;18:533
19. Studnek JR, Artho MR, Garner CL, Jones AE. The impact of emergency medical services on the ED care of severe sepsis. Am J Emerg Med 2012;30:51–6.
20. Guss DA, Meyer FT, Neuman TS, et al. The impact of a regionalized trauma system on trauma care in San Diego County. Ann Emerg Med 1989;18:1141–5.
21. Liberman M, Mulder DS, Jurkovich GJ, Sampalis JS. The association between trauma system and trauma center components and outcome in a mature regionalized trauma system. Surgery 2005;137:647–58.
22. Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working toward a prioritized world agenda. Stroke 2010;41:1084–99.
23. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31:1250–6.
24. Sibbald W, Doig G, Inman K. Sepsis, SIRS, and infection. Intensive Care Med 1995;21:299–301.
25. Kaukonen KM, Bailey M, Pilcher D, et al. Systemic inflammatory response syndrome criteria in defining severe sepsis. N Engl J Med 2015; online March 17, 2015.
26. Shapiro NI, Howell MD, Bates D, et al. The association of sepsis syndrome and organ dysfunction with mortality in emergency department patients with suspected infection. Ann Emerg Med 2006;48:583–90.
27. Berger T, Green J, Horeczko T, et al. Shock index and early recognition of sepsis in the emergency department: pilot study. West J Emerg Med 2013;14:168–74.
28. Farley H, Zubrow MT, Gies J, et al. Emergency department tachypnea predicts transfer to a higher level of care in the first 24 hours after ED admission. Acad Emerg Med 2010;17:718–22.
29. Sinuff T, Muscadere J, Adhikari NK, et al. Knowledge translation interventions for critically ill patients: a systematic review. Crit Care Med 2013;41:2627–40.
30. Hoot NR, Aronsky D. Systematic review of emergency department crowding: causes, effects, and solutions. Ann Emerg Med 2008;52:126–36.
31. Hsia RY, Kellermann AL, Shen YC. Factors associated with closures of emergency departments in the United States. JAMA 2011;305:1978–85.
32. Chalfin DB, Trzeciak S, Likourezos A, et al. Impact of delayed transfer of critically ill patients from the emergency department to the intensive care unit. Crit Care Med 2007;35:1477–83.
33. Subbe CP, Kruger M, Rutherford P, et al. Validation of a modified early warning score in medical admissions. Q J Med 2001;94:521–6.
34. Umscheid CA, Betesh J, VanZandbergen C, et al. Development, implementation, and impact of an automated early warning and response system for sepsis. J Hosp Med 2015;10:26–31.
35. Churpek MM, Yuen TC, Winslow C, et al. Multicenter development and validation of a risk stratification tool for ward patients. Am J Respir Crit Care Med 2014;190:649–55.
36. Whittaker SA, Fuchs BD, Gaieski DF, et al. Epidemiology and outcomes in patients with severe sepsis admitted to the hospital wards. J Crit Care 2015;30:78–84.
37. Delgado MK, Liu V, Pines JM, et al. Risk factors for unplanned transfer to intensive care within 24 hours of admission from the emergency department in an integrated healthcare system. J Hosp Med 2013;8:13–9.
38. Valentini I, Pacilli AM, Carbonara P, et al. Influence of the admission pattern on the outcome of patients admitted to a respiratory intensive care unit: does a step-down admission differ from a step-up one? Respir Care 2013;58:2053–60.
39. Kompanje EJO, Jansen TC, van der Hoven B, Bakker J. The first demonstration of lactic acid in human blood in shock by Johann Joseph Scherer (1814-1869) in January 1843. Intensive Care Med 2007;33:1967–71.
40. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014;371:2309–19.
41. Jansen TC, van Bommel J, Mulder PG, et al. The prognostic value of blood lactate levels relative to that of vital signs in the pre-hospital setting: a pilot study. Crit Care 2008;12:R160.
42. del Portal DA, Shofer F, Mikkelsen ME, et al. Emergency department lactate is associated with mortality in older adults admitted with and without infections. Acad Emerg Med 2010;17:260–8.
43. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med 2009;37:1670–7.
44. Gaieski DF, Drumheller BC, Goyal M, et al. Accuracy of handheld point-of-care fingertip lactate measurement in the emergency department. West J Emerg Med 2013;14:58–62.
45. Macdonald SP, Arendts G, Fatovich DM, Brown SG. Comparison of PIRO, SOFA, and MEDS scores for predicting mortality in emergency department patients with severe sepsis and septic shock. Acad Emerg Med 2014;21:1257–63.
46. Carpenter CR, Keim SM, Upadhye S, Nguyen HB, Group BEiEMI. Risk stratification of the potentially septic patient in the emergency department: the Mortality in the Emergency Department Sepsis (MEDS) score. J Emerg Med 2009;37:319–27.
47. Sankoff JD, Goyal M, Gaieski DF, et al. Validation of the Mortality in Emergency Department Sepsis (MEDS) score in patients with the systemic inflammatory response syndrome (SIRS). Crit Care Med 2008;36:421–6.
48. Corfield AR, Lees F, Zealley I, et al. Utility of a single early warning score in patients with sepsis in the emergency department. Emerg Med J 2014;31:482–7.
49. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34:1589–96.
50. Gaieski DF, Mikkelsen ME, Band RA, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergency department. Crit Care Med 2010;38:1045–53.
51. Puskarich MA, Trzeciak S, Shapiro NI, et al. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011;39:2066–71.
52. Capp R, Horton CL, Takhar SS, et al. Predictors of patients who present to the emergency department with sepsis and progress to septic shock between 4 and 48 hours of emergency department arrival. Crit Care Med 2015 Jan 30.
53. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–77.
54. The ProCESS Investigators. A ranodmized trial of protocol-based care for early septic shock. N Engl J Med 2014;370:1683–93.
55. The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directed resuscitation for patients with early septic shock. N Engl J Med 2014;371:1496–506.
56. Mouncey PR, Osborn TM, Power GS, et al. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med 2015; online March 17, 2015.
57. Weingart SD, Sherwin RL, Emlet LL, et al. ED intensivists and ED intensive care units. Amer J Emerg Med 2013;31:617–20.
58. Lilly CM, Cody S, Zhao H, et al. Hospital mortality, length of stay, and preventable complications among critically ill patients before and after tele-ICU reengineering of critical care processes. JAMA 2011;305:2175–85.
59. Walkey AJ, Wiener RS. Hospital case volume and outcomes among patients hospitalized with severe sepsis. Am J Respir Crit Care Med 2014;189:548–55.
60. Gaieski DF, Edwards JM, Kallan MJ, et al. The relationship between hospital volume and mortality in severe sepsis. Am J Respir Crit Care Med 2014;190:665–74.
61. Levy MM, Dellinger RP, Townsend SR, et al. The surviving sepsis campaign: results of an international guideline-based performance improvement program targeting severe sepsis. Intensive Care Med 2010;36:222–31.
62. Levy MM, Rhodes A, Phillips GS, et al. Surviving sepsis campaign: association between performance metrics and outcomes in a 7.5-year study. Crit Care Med 2015;43:3–12.
63. Weiss CH, Moazed F, McEvoy CA, et al. Prompting physicians to address a daily checklist and process of care and clinical outcomes: a single-site study. Am J Respir Crit Care Med 2011;184:680–6.
64. Weant KA, Baker SN. Emergency medicine pharmacists and sepsis management. J Pharm Pract 2013;26:401–5.
65. Marwick CA, Guthrie B, Pringle JE, et al. A multifaceted intervention to improve sepsis management in general hospital wards with evaluation using segmented regression of interrupted time series. BMJ Qual Saf 2014;23:e2.
EWRS for Sepsis
There are as many as 3 million cases of severe sepsis and 750,000 resulting deaths in the United States annually.[1] Interventions such as goal‐directed resuscitation and antibiotics can reduce sepsis mortality, but their effectiveness depends on early administration. Thus, timely recognition is critical.[2, 3, 4, 5]
Despite this, early recognition in hospitalized patients can be challenging. Using chart documentation as a surrogate for provider recognition, we recently found only 20% of patients with severe sepsis admitted to our hospital from the emergency department were recognized.[6] Given these challenges, there has been increasing interest in developing automated systems to improve the timeliness of sepsis detection.[7, 8, 9, 10] Systems described in the literature have varied considerably in triggering criteria, effector responses, and study settings. Of those examining the impact of automated surveillance and response in the nonintensive care unit (ICU) acute inpatient setting, results suggest an increase in the timeliness of diagnostic and therapeutic interventions,[10] but less impact on patient outcomes.[7] Whether these results reflect inadequacies in the criteria used to identify patients (parameters or their thresholds) or an ineffective response to the alert (magnitude or timeliness) is unclear.
Given the consequences of severe sepsis in hospitalized patients, as well as the introduction of vital sign (VS) and provider data in our electronic health record (EHR), we sought to develop and implement an electronic sepsis detection and response system to improve patient outcomes. This study describes the development, validation, and impact of that system.
METHODS
Setting and Data Sources
The University of Pennsylvania Health System (UPHS) includes 3 hospitals with a capacity of over 1500 beds and 70,000 annual admissions. All hospitals use the EHR Sunrise Clinical Manager version 5.5 (Allscripts, Chicago, IL). The study period began in October 2011, when VS and provider contact information became available electronically. Data were retrieved from the Penn Data Store, which includes professionally coded data as well as clinical data from our EHRs. The study received expedited approval and a Health Insurance Portability and Accountability Act waiver from our institutional review board.
Development of the Intervention
The early warning and response system (EWRS) for sepsis was designed to monitor laboratory values and VSs in real time in our inpatient EHR to detect patients at risk for clinical deterioration and development of severe sepsis. The development team was multidisciplinary, including informaticians, physicians, nurses, and data analysts from all 3 hospitals.
To identify at‐risk patients, we used established criteria for severe sepsis, including the systemic inflammatory response syndrome criteria (temperature <36C or >38C, heart rate >90 bpm, respiratory rate >20 breaths/min or PaCO2 <32 mm Hg, and total white blood cell count <4000 or >12,000 or >10% bands) coupled with criteria suggesting organ dysfunction (cardiovascular dysfunction based on a systolic blood pressure <100 mm Hg, and hypoperfusion based on a serum lactate measure >2.2 mmol/L [the threshold for an abnormal result in our lab]).[11, 12]
To establish a threshold for triggering the system, a derivation cohort was used and defined as patients admitted between October 1, 2011 to October 31, 2011 1 to any inpatient acute care service. Those <18 years old or admitted to hospice, research, and obstetrics services were excluded. We calculated a risk score for each patient, defined as the sum of criteria met at any single time during their visit. At any given point in time, we used the most recent value for each criteria, with a look‐back period of 24 hours for VSs and 48 hours for labs. The minimum and maximum number of criteria that a patient could achieve at any single time was 0 and 6, respectively. We then categorized patients by the maximum number of criteria achieved and estimated the proportion of patients in each category who: (1) were transferred to an ICU during their hospital visit; (2) had a rapid response team (RRT) called during their visit; (3) died during their visit; (4) had a composite of 1, 2, or 3; or (5) were coded as sepsis at discharge (see Supporting Information in the online version of this article for further information). Once a threshold was chosen, we examined the time from first trigger to: (1) any ICU transfer; (2) any RRT; (3) death; or (4) a composite of 1, 2, or 3. We then estimated the screen positive rate, test characteristics, predictive values, and likelihood ratios of the specified threshold.
The efferent response arm of the EWRS included the covering provider (usually an intern), the bedside nurse, and rapid response coordinators, who were engaged from the outset in developing the operational response to the alert. This team was required to perform a bedside evaluation within 30 minutes of the alert, and enact changes in management if warranted. The rapid response coordinator was required to complete a 3‐question follow‐up assessment in the EHR asking whether all 3 team members gathered at the bedside, the most likely condition triggering the EWRS, and whether management changed (see Supporting Figure 1 in the online version of this article). To minimize the number of triggers, once a patient triggered an alert, any additional alert triggers during the same hospital stay were censored.
Implementation of the EWRS
All inpatients on noncritical care services were screened continuously. Hospice, research, and obstetrics services were excluded. If a patient met the EWRS criteria threshold, an alert was sent to the covering provider and rapid response coordinator by text page. The bedside nurses, who do not carry text‐enabled devices, were alerted by pop‐up notification in the EHR (see Supporting Figure 2 in the online version of this article). The notification was linked to a task that required nurses to verify in the EHR the VSs triggering the EWRS, and adverse trends in VSs or labs (see Supporting Figure 3 in the online version of this article).
The Preimplementation (Silent) Period and EWRS Validation
The EWRS was initially activated for a preimplementation silent period (June 6, 2012September 4, 2012) to both validate the tool and provide the baseline data to which the postimplementation period was compared. During this time, new admissions could trigger the alert, but notifications were not sent. We used admissions from the first 30 days of the preimplementation period to estimate the tool's screen positive rate, test characteristics, predictive values, and likelihood ratios.
The Postimplementation (Live) Period and Impact Analysis
The EWRS went live September 12, 2012, upon which new admissions triggering the alert would result in a notification and response. Unadjusted analyses using the [2] test for dichotomous variables and the Wilcoxon rank sum test for continuous variables compared demographics and the proportion of clinical process and outcome measures for those admitted during the silent period (June 6, 2012September 4, 2012) and a similar timeframe 1 year later when the intervention was live (June 6, 2013September 4, 2013). To be included in either of the time periods, patients had to trigger the alert during the period and be discharged within 45 days of the end of the period. The pre‐ and post‐sepsis mortality index was also examined (see the Supporting Information in the online version of this article for a detailed description of study measures). Multivariable regression models estimated the impact of the EWRS on the process and outcome measures, adjusted for differences between the patients in the preimplementation and postimplementation periods with respect to age, gender, Charlson index on admission, admitting service, hospital, and admission month. Logistic regression models examined dichotomous variables. Continuous variables were log transformed and examined using linear regression models. Cox regression models explored time to ICU transfer from trigger. Among patients with sepsis, a logistic regression model was used to compare the odds of mortality between the silent and live periods, adjusted for expected mortality, both within each hospital and across all hospitals.
Because there is a risk of providers becoming overly reliant on automated systems and overlooking those not triggering the system, we also examined the discharge disposition and mortality outcomes of those in both study periods not identified by the EWRS.
The primary analysis examined the impact of the EWRS across UPHS; we also examined the EWRS impact at each of our hospitals. Last, we performed subgroup analyses examining the EWRS impact in those assigned an International Classification of Diseases, 9th Revision code for sepsis at discharge or death. All analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC).
RESULTS
In the derivation cohort, 4575 patients met the inclusion criteria. The proportion of those in each category (06) achieving our outcomes of interest are described in Supporting Table 1 in the online version of this article. We defined a positive trigger as a score 4, as this threshold identified a limited number of patients (3.9% [180/4575]) with a high proportion experiencing our composite outcome (25.6% [46/180]). The proportion of patients with an EWRS score 4 and their time to event by hospital and health system is described in Supporting Table 2 in the online version of this article. Those with a score 4 were almost 4 times as likely to be transferred to the ICU, almost 7 times as likely to experience an RRT, and almost 10 times as likely to die. The screen positive, sensitivity, specificity, and positive and negative predictive values and likelihood ratios using this threshold and our composite outcome in the derivation cohort were 6%, 16%, 97%, 26%, 94%, 5.3, and 0.9, respectively, and in our validation cohort were 6%, 17%, 97%, 28%, 95%, 5.7, and 0.9, respectively.
In the preimplementation period, 3.8% of admissions (595/15,567) triggered the alert, as compared to 3.5% (545/15,526) in the postimplementation period. Demographics were similar across periods, except that in the postimplementation period patients were slightly younger and had a lower Charlson Comorbidity Index at admission (Table 1). The distribution of alerts across medicine and surgery services were similar (Table 1).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of encounters | 15,567 | 15,526 | |
| No. of alerts | 595 (4%) | 545 (4%) | 0.14 |
| Age, y, median (IQR) | 62.0 (48.570.5) | 59.7 (46.169.6) | 0.04 |
| Female | 298 (50%) | 274 (50%) | 0.95 |
| Race | |||
| White | 343 (58%) | 312 (57%) | 0.14 |
| Black | 207 (35%) | 171 (31%) | |
| Other | 23 (4%) | 31 (6%) | |
| Unknown | 22 (4%) | 31 (6%) | |
| Admission type | |||
| Elective | 201 (34%) | 167 (31%) | 0.40 |
| ED | 300 (50%) | 278 (51%) | |
| Transfer | 94 (16%) | 99 (18%) | |
| BMI, kg/m2, median (IQR) | 27.0 (23.032.0) | 26.0 (22.031.0) | 0.24 |
| Previous ICU admission | 137 (23%) | 127 (23%) | 0.91 |
| RRT before alert | 27 (5%) | 20 (4%) | 0.46 |
| Admission Charlson index, median (IQR) | 2.0 (1.04.0) | 2.0 (1.04.0) | 0.04 |
| Admitting service | |||
| Medicine | 398 (67%) | 364 (67%) | 0.18 |
| Surgery | 173 (29%) | 169 (31%) | |
| Other | 24 (4%) | 12 (2%) | |
| Service where alert fired | |||
| Medicine | 391 (66%) | 365 (67%) | 0.18 |
| Surgery | 175 (29%) | 164 (30%) | |
| Other | 29 (5%) | 15 (3%) | |
In our postimplementation period, 99% of coordinator pages and over three‐fourths of provider notifications were sent successfully. Almost three‐fourths of nurses reviewed the initial alert notification, and over 99% completed the electronic data verification and adverse trend review, with over half documenting adverse trends. Ninety‐five percent of the time the coordinators completed the follow‐up assessment. Over 90% of the time, the entire team evaluated the patient at bedside within 30 minutes. Almost half of the time, the team thought the patient had no critical illness. Over a third of the time, they thought the patient had sepsis, but reported over 90% of the time that they were aware of the diagnosis prior to the alert. (Supporting Table 3 in the online version of this article includes more details about the responses to the electronic notifications and follow‐up assessments.)
In unadjusted and adjusted analyses, ordering of antibiotics, intravenous fluid boluses, and lactate and blood cultures within 3 hours of the trigger increased significantly, as did ordering of blood products, chest radiographs, and cardiac monitoring within 6 hours of the trigger (Tables 2 and 3).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of alerts | 595 | 545 | |
| 500 mL IV bolus order <3 h after alert | 92 (15%) | 142 (26%) | <0.01 |
| IV/PO antibiotic order <3 h after alert | 75 (13%) | 123 (23%) | <0.01 |
| IV/PO sepsis antibiotic order <3 h after alert | 61 (10%) | 85 (16%) | <0.01 |
| Lactic acid order <3 h after alert | 57 (10%) | 128 (23%) | <0.01 |
| Blood culture order <3 h after alert | 68 (11%) | 99 (18%) | <0.01 |
| Blood gas order <6 h after alert | 53 (9%) | 59 (11%) | 0.28 |
| CBC or BMP <6 h after alert | 247 (42%) | 219 (40%) | 0.65 |
| Vasopressor <6 h after alert | 17 (3%) | 21 (4%) | 0.35 |
| Bronchodilator administration <6 h after alert | 71 (12%) | 64 (12%) | 0.92 |
| RBC, plasma, or platelet transfusion order <6 h after alert | 31 (5%) | 52 (10%) | <0.01 |
| Naloxone order <6 h after alert | 0 (0%) | 1 (0%) | 0.30 |
| AV node blocker order <6 h after alert | 35 (6%) | 35 (6%) | 0.70 |
| Loop diuretic order <6 h after alert | 35 (6%) | 28 (5%) | 0.58 |
| CXR <6 h after alert | 92 (15%) | 113 (21%) | 0.02 |
| CT head, chest, or ABD <6 h after alert | 29 (5%) | 34 (6%) | 0.31 |
| Cardiac monitoring (ECG or telemetry) <6 h after alert | 70 (12%) | 90 (17%) | 0.02 |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Odds Ratio | Adjusted Odds Ratio | Unadjusted Odds Ratio | Adjusted Odds Ratio | |
| ||||
| 500 mL IV bolus order <3 h after alert | 1.93 (1.442.58) | 1.93 (1.432.61) | 1.64 (1.112.43) | 1.65 (1.102.47) |
| IV/PO antibiotic order <3 h after alert | 2.02 (1.482.77) | 2.02 (1.462.78) | 1.99 (1.323.00) | 2.02 (1.323.09) |
| IV/PO sepsis antibiotic order <3 h after alert | 1.62 (1.142.30) | 1.57 (1.102.25) | 1.63 (1.052.53) | 1.65 (1.052.58) |
| Lactic acid order <3 h after alert | 2.90 (2.074.06) | 3.11 (2.194.41) | 2.41 (1.583.67) | 2.79 (1.794.34) |
| Blood culture <3 h after alert | 1.72 (1.232.40) | 1.76 (1.252.47) | 1.36 (0.872.10) | 1.40 (0.902.20) |
| Blood gas order <6 h after alert | 1.24 (0.841.83) | 1.32 (0.891.97) | 1.06 (0.631.77) | 1.13 (0.671.92) |
| BMP or CBC order <6 h after alert | 0.95 (0.751.20) | 0.96 (0.751.21) | 1.00 (0.701.44) | 1.04 (0.721.50) |
| Vasopressor order <6 h after alert | 1.36 (0.712.61) | 1.47 (0.762.83) | 1.32 (0.583.04) | 1.38 (0.593.25) |
| Bronchodilator administration <6 h after alert | 0.98 (0.691.41) | 1.02 (0.701.47) | 1.13 (0.641.99) | 1.17 (0.652.10) |
| Transfusion order <6 h after alert | 1.92 (1.213.04) | 1.95 (1.233.11) | 1.65 (0.913.01) | 1.68 (0.913.10) |
| AV node blocker order <6 h after alert | 1.10 (0.681.78) | 1.20 (0.722.00) | 0.38 (0.131.08) | 0.39 (0.121.20) |
| Loop diuretic order <6 h after alert | 0.87 (0.521.44) | 0.93 (0.561.57) | 1.63 (0.634.21) | 1.87 (0.705.00) |
| CXR <6 h after alert | 1.43 (1.061.94) | 1.47 (1.081.99) | 1.45 (0.942.24) | 1.56 (1.002.43) |
| CT <6 h after alert | 1.30 (0.782.16) | 1.30 (0.782.19) | 0.97 (0.521.82) | 0.94 (0.491.79) |
| Cardiac monitoring <6 h after alert | 1.48 (1.062.08) | 1.54 (1.092.16) | 1.32 (0.792.18) | 1.44 (0.862.41) |
Hospital and ICU length of stay were similar in the preimplementation and postimplementation periods. There was no difference in the proportion of patients transferred to the ICU following the alert; however, the proportion transferred within 6 hours of the alert increased, and the time to ICU transfer was halved (see Supporting Figure 4 in the online version of this article), but neither change was statistically significant in unadjusted analyses. Transfer to the ICU within 6 hours became statistically significant after adjustment. All mortality measures were lower in the postimplementation period, but none reached statistical significance. Discharge to home and sepsis documentation were both statistically higher in the postimplementation period, but discharge to home lost statistical significance after adjustment (Tables 4 and 5) (see Supporting Table 4 in the online version of this article).
| Hospitals AC | ||||
|---|---|---|---|---|
| Preimplementation | Postimplementation | P Value | ||
| ||||
| No. of alerts | 595 | 545 | ||
| Hospital LOS, d, median (IQR) | 10.1 (5.119.1) | 9.4 (5.218.9) | 0.92 | |
| ICU LOS after alert, d, median (IQR) | 3.4 (1.77.4) | 3.6 (1.96.8) | 0.72 | |
| ICU transfer <6 h after alert | 40 (7%) | 53 (10%) | 0.06 | |
| ICU transfer <24 h after alert | 71 (12%) | 79 (14%) | 0.20 | |
| ICU transfer any time after alert | 134 (23%) | 124 (23%) | 0.93 | |
| Time to first ICU after alert, h, median (IQR) | 21.3 (4.463.9) | 11.0 (2.358.7) | 0.22 | |
| RRT 6 h after alert | 13 (2%) | 9 (2%) | 0.51 | |
| Mortality of all patients | 52 (9%) | 41 (8%) | 0.45 | |
| Mortality 30 days after alert | 48 (8%) | 33 (6%) | 0.19 | |
| Mortality of those transferred to ICU | 40 (30%) | 32 (26%) | 0.47 | |
| Deceased or IP hospice | 94 (16%) | 72 (13%) | 0.22 | |
| Discharge to home | 347 (58%) | 351 (64%) | 0.04 | |
| Disposition location | ||||
| Home | 347 (58%) | 351 (64%) | 0.25 | |
| SNF | 89 (15%) | 65 (12%) | ||
| Rehab | 24 (4%) | 20 (4%) | ||
| LTC | 8 (1%) | 9 (2%) | ||
| Other hospital | 16 (3%) | 6 (1%) | ||
| Expired | 52 (9%) | 41 (8%) | ||
| Hospice IP | 42 (7%) | 31 (6%) | ||
| Hospice other | 11 (2%) | 14 (3%) | ||
| Other location | 6 (1%) | 8 (1%) | ||
| Sepsis discharge diagnosis | 230 (39%) | 247 (45%) | 0.02 | |
| Sepsis O/E | 1.37 | 1.06 | 0.18 | |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Estimate | Adjusted Estimate | Unadjusted Estimate | Adjusted Estimate | |
| ||||
| Hospital LOS, d | 1.01 (0.921.11) | 1.02 (0.931.12) | 0.99 (0.851.15) | 1.00 (0.871.16) |
| ICU transfer | 1.49 (0.972.29) | 1.65 (1.072.55) | 1.61 (0.922.84) | 1.82 (1.023.25) |
| Time to first ICU transfer after alert, h‖ | 1.17 (0.871.57) | 1.23 (0.921.66) | 1.21 (0.831.75) | 1.31 (0.901.90) |
| ICU LOS, d | 1.01 (0.771.31) | 0.99 (0.761.28) | 0.87 (0.621.21) | 0.88 (0.641.21) |
| RRT | 0.75 (0.321.77) | 0.84 (0.352.02) | 0.81 (0.292.27) | 0.82 (0.272.43) |
| Mortality | 0.85 (0.551.30) | 0.98 (0.631.53) | 0.85 (0.551.30) | 0.98 (0.631.53) |
| Mortality within 30 days of alert | 0.73 (0.461.16) | 0.87 (0.541.40) | 0.59 (0.341.04) | 0.69 (0.381.26) |
| Mortality or inpatient hospice transfer | 0.82 (0.471.41) | 0.78 (0.441.41) | 0.67 (0.361.25) | 0.65 (0.331.29) |
| Discharge to home | 1.29 (1.021.64) | 1.18 (0.911.52) | 1.36 (0.951.95) | 1.22 (0.811.84) |
| Sepsis discharge diagnosis | 1.32 (1.041.67) | 1.43 (1.101.85) | NA | NA |
In a subanalysis of EWRS impact on patients documented with sepsis at discharge, unadjusted and adjusted changes in clinical process and outcome measures across the time periods were similar to that of the total population (see Supporting Tables 5 and 6 and Supporting Figure 5 in the online version of this article). The unadjusted composite outcome of mortality or inpatient hospice was statistically lower in the postimplementation period, but lost statistical significance after adjustment.
The disposition and mortality outcomes of those not triggering the alert were unchanged across the 2 periods (see Supporting Tables 7, 8, and 9 in the online version of this article).
DISCUSSION
This study demonstrated that a predictive tool can accurately identify non‐ICU inpatients at increased risk for deterioration and death. In addition, we demonstrated the feasibility of deploying our EHR to screen patients in real time for deterioration and to trigger electronically a timely, robust, multidisciplinary bedside clinical evaluation. Compared to the control (silent) period, the EWRS resulted in a marked increase in early sepsis care, transfer to the ICU, and sepsis documentation, and an indication of a decreased sepsis mortality index and mortality, and increased discharge to home, although none of these latter 3 findings reached statistical significance.
Our study is unique in that it was implemented across a multihospital health system, which has identical EHRs, but diverse cultures, populations, staffing, and practice models. In addition, our study includes a preimplementation population similar to the postimplementation population (in terms of setting, month of admission, and adjustment for potential confounders).
Interestingly, patients identified by the EWRS who were subsequently transferred to an ICU had higher mortality rates (30% and 26% in the preimplementation and postimplementation periods, respectively, across UPHS) than those transferred to an ICU who were not identified by the EWRS (7% and 6% in the preimplementation and postimplementation periods, respectively, across UPHS) (Table 4) (see Supporting Table 7 in the online version of this article). This finding was robust to the study period, so is likely not related to the bedside evaluation prompted by the EWRS. It suggests the EWRS could help triage patients for appropriateness of ICU transfer, a particularly valuable role that should be explored further given the typical strains on ICU capacity,[13] and the mortality resulting from delays in patient transfers into ICUs.[14, 15]
Although we did not find a statistically significant mortality reduction, our study may have been underpowered to detect this outcome. Our study has other limitations. First, our preimplementation/postimplementation design may not fully account for secular changes in sepsis mortality. However, our comparison of similar time periods and our adjustment for observed demographic differences allow us to estimate with more certainty the change in sepsis care and mortality attributable to the intervention. Second, our study did not examine the effect of the EWRS on mortality after hospital discharge, where many such events occur. However, our capture of at least 45 hospital days on all study patients, as well as our inclusion of only those who died or were discharged during our study period, and our assessment of discharge disposition such as hospice, increase the chance that mortality reductions directly attributable to the EWRS were captured. Third, although the EWRS changed patient management, we did not assess the appropriateness of management changes. However, the impact of care changes was captured crudely by examining mortality rates and discharge disposition. Fourth, our study was limited to a single academic healthcare system, and our experience may not be generalizable to other healthcare systems with different EHRs and staff. However, the integration of our automated alert into a commercial EHR serving a diverse array of patient populations, clinical services, and service models throughout our healthcare system may improve the generalizability of our experience to other settings.
CONCLUSION
By leveraging readily available electronic data, an automated prediction tool identified at‐risk patients and mobilized care teams, resulting in more timely sepsis care, improved sepsis documentation, and a suggestion of reduced mortality. This alert may be scalable to other healthcare systems.
Acknowledgements
The authors thank Jennifer Barger, MS, BSN, RN; Patty Baroni, MSN, RN; Patrick J. Donnelly, MS, RN, CCRN; Mika Epps, MSN, RN; Allen L. Fasnacht, MSN, RN; Neil O. Fishman, MD; Kevin M. Fosnocht, MD; David F. Gaieski, MD; Tonya Johnson, MSN, RN, CCRN; Craig R. Kean, MS; Arash Kia, MD, MS; Matthew D. Mitchell, PhD; Stacie Neefe, BSN, RN; Nina J. Renzi, BSN, RN, CCRN; Alexander Roederer, Jean C. Romano, MSN, RN, NE‐BC; Heather Ross, BSN, RN, CCRN; William D. Schweickert, MD; Esme Singer, MD; and Kendal Williams, MD, MPH for their help in developing, testing and operationalizing the EWRS examined in this study; their assistance in data acquisition; and for advice regarding data analysis. This study was previously presented as an oral abstract at the 2013 American Medical Informatics Association Meeting, November 1620, 2013, Washington, DC.
Disclosures: Dr. Umscheid's contribution to this project was supported in part by the National Center for Research Resources, grant UL1RR024134, which is now at the National Center for Advancing Translational Sciences, grant UL1TR000003. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors report no potential financial conflicts of interest relevant to this article.
- , , , . Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41(5):1167–1174.
- , , , et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637.
- , , , et al. The Surviving Sepsis Campaign: results of an international guideline‐based performance improvement program targeting severe sepsis. Crit Care Med. 2010;38(2):367–374.
- , , , et al. Early goal‐directed therapy in severe sepsis and septic shock revisited: concepts, controversies, and contemporary findings. Chest. 2006;130(5):1579–1595.
- , , , et al. Early goal‐directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.
- , , , , , . Severe sepsis cohorts derived from claims‐based strategies appear to be biased toward a more severely ill patient population. Crit Care Med. 2013;41(4):945–953.
- , , , et al. A trial of a real‐time alert for clinical deterioration in patients hospitalized on general medical wards. J Hosp Med. 2013;8(5):236–242.
- , , , , , . Bedside electronic capture of clinical observations and automated clinical alerts to improve compliance with an Early Warning Score protocol. Crit Care Resusc. 2011;13(2):83–88.
- , , , . Prospective trial of real‐time electronic surveillance to expedite early care of severe sepsis. Ann Emerg Med. 2011;57(5):500–504.
- , , , et al. Implementation of a real‐time computerized sepsis alert in nonintensive care unit patients. Crit Care Med. 2011;39(3):469–473.
- , , , et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101(6):1644–1655.
- , , , et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31(4):1250–1256.
- , , , , . Rationing critical care beds: a systematic review. Crit Care Med. 2004;32(7):1588–1597.
- . Delayed admission to intensive care unit for critically surgical patients is associated with increased mortality. Am J Surg. 2014;208:268–274.
- , , , et al. Impact of delayed admission to intensive care units on mortality of critically ill patients: a cohort study. Crit Care. 2011;15(1):R28.
There are as many as 3 million cases of severe sepsis and 750,000 resulting deaths in the United States annually.[1] Interventions such as goal‐directed resuscitation and antibiotics can reduce sepsis mortality, but their effectiveness depends on early administration. Thus, timely recognition is critical.[2, 3, 4, 5]
Despite this, early recognition in hospitalized patients can be challenging. Using chart documentation as a surrogate for provider recognition, we recently found only 20% of patients with severe sepsis admitted to our hospital from the emergency department were recognized.[6] Given these challenges, there has been increasing interest in developing automated systems to improve the timeliness of sepsis detection.[7, 8, 9, 10] Systems described in the literature have varied considerably in triggering criteria, effector responses, and study settings. Of those examining the impact of automated surveillance and response in the nonintensive care unit (ICU) acute inpatient setting, results suggest an increase in the timeliness of diagnostic and therapeutic interventions,[10] but less impact on patient outcomes.[7] Whether these results reflect inadequacies in the criteria used to identify patients (parameters or their thresholds) or an ineffective response to the alert (magnitude or timeliness) is unclear.
Given the consequences of severe sepsis in hospitalized patients, as well as the introduction of vital sign (VS) and provider data in our electronic health record (EHR), we sought to develop and implement an electronic sepsis detection and response system to improve patient outcomes. This study describes the development, validation, and impact of that system.
METHODS
Setting and Data Sources
The University of Pennsylvania Health System (UPHS) includes 3 hospitals with a capacity of over 1500 beds and 70,000 annual admissions. All hospitals use the EHR Sunrise Clinical Manager version 5.5 (Allscripts, Chicago, IL). The study period began in October 2011, when VS and provider contact information became available electronically. Data were retrieved from the Penn Data Store, which includes professionally coded data as well as clinical data from our EHRs. The study received expedited approval and a Health Insurance Portability and Accountability Act waiver from our institutional review board.
Development of the Intervention
The early warning and response system (EWRS) for sepsis was designed to monitor laboratory values and VSs in real time in our inpatient EHR to detect patients at risk for clinical deterioration and development of severe sepsis. The development team was multidisciplinary, including informaticians, physicians, nurses, and data analysts from all 3 hospitals.
To identify at‐risk patients, we used established criteria for severe sepsis, including the systemic inflammatory response syndrome criteria (temperature <36C or >38C, heart rate >90 bpm, respiratory rate >20 breaths/min or PaCO2 <32 mm Hg, and total white blood cell count <4000 or >12,000 or >10% bands) coupled with criteria suggesting organ dysfunction (cardiovascular dysfunction based on a systolic blood pressure <100 mm Hg, and hypoperfusion based on a serum lactate measure >2.2 mmol/L [the threshold for an abnormal result in our lab]).[11, 12]
To establish a threshold for triggering the system, a derivation cohort was used and defined as patients admitted between October 1, 2011 to October 31, 2011 1 to any inpatient acute care service. Those <18 years old or admitted to hospice, research, and obstetrics services were excluded. We calculated a risk score for each patient, defined as the sum of criteria met at any single time during their visit. At any given point in time, we used the most recent value for each criteria, with a look‐back period of 24 hours for VSs and 48 hours for labs. The minimum and maximum number of criteria that a patient could achieve at any single time was 0 and 6, respectively. We then categorized patients by the maximum number of criteria achieved and estimated the proportion of patients in each category who: (1) were transferred to an ICU during their hospital visit; (2) had a rapid response team (RRT) called during their visit; (3) died during their visit; (4) had a composite of 1, 2, or 3; or (5) were coded as sepsis at discharge (see Supporting Information in the online version of this article for further information). Once a threshold was chosen, we examined the time from first trigger to: (1) any ICU transfer; (2) any RRT; (3) death; or (4) a composite of 1, 2, or 3. We then estimated the screen positive rate, test characteristics, predictive values, and likelihood ratios of the specified threshold.
The efferent response arm of the EWRS included the covering provider (usually an intern), the bedside nurse, and rapid response coordinators, who were engaged from the outset in developing the operational response to the alert. This team was required to perform a bedside evaluation within 30 minutes of the alert, and enact changes in management if warranted. The rapid response coordinator was required to complete a 3‐question follow‐up assessment in the EHR asking whether all 3 team members gathered at the bedside, the most likely condition triggering the EWRS, and whether management changed (see Supporting Figure 1 in the online version of this article). To minimize the number of triggers, once a patient triggered an alert, any additional alert triggers during the same hospital stay were censored.
Implementation of the EWRS
All inpatients on noncritical care services were screened continuously. Hospice, research, and obstetrics services were excluded. If a patient met the EWRS criteria threshold, an alert was sent to the covering provider and rapid response coordinator by text page. The bedside nurses, who do not carry text‐enabled devices, were alerted by pop‐up notification in the EHR (see Supporting Figure 2 in the online version of this article). The notification was linked to a task that required nurses to verify in the EHR the VSs triggering the EWRS, and adverse trends in VSs or labs (see Supporting Figure 3 in the online version of this article).
The Preimplementation (Silent) Period and EWRS Validation
The EWRS was initially activated for a preimplementation silent period (June 6, 2012September 4, 2012) to both validate the tool and provide the baseline data to which the postimplementation period was compared. During this time, new admissions could trigger the alert, but notifications were not sent. We used admissions from the first 30 days of the preimplementation period to estimate the tool's screen positive rate, test characteristics, predictive values, and likelihood ratios.
The Postimplementation (Live) Period and Impact Analysis
The EWRS went live September 12, 2012, upon which new admissions triggering the alert would result in a notification and response. Unadjusted analyses using the [2] test for dichotomous variables and the Wilcoxon rank sum test for continuous variables compared demographics and the proportion of clinical process and outcome measures for those admitted during the silent period (June 6, 2012September 4, 2012) and a similar timeframe 1 year later when the intervention was live (June 6, 2013September 4, 2013). To be included in either of the time periods, patients had to trigger the alert during the period and be discharged within 45 days of the end of the period. The pre‐ and post‐sepsis mortality index was also examined (see the Supporting Information in the online version of this article for a detailed description of study measures). Multivariable regression models estimated the impact of the EWRS on the process and outcome measures, adjusted for differences between the patients in the preimplementation and postimplementation periods with respect to age, gender, Charlson index on admission, admitting service, hospital, and admission month. Logistic regression models examined dichotomous variables. Continuous variables were log transformed and examined using linear regression models. Cox regression models explored time to ICU transfer from trigger. Among patients with sepsis, a logistic regression model was used to compare the odds of mortality between the silent and live periods, adjusted for expected mortality, both within each hospital and across all hospitals.
Because there is a risk of providers becoming overly reliant on automated systems and overlooking those not triggering the system, we also examined the discharge disposition and mortality outcomes of those in both study periods not identified by the EWRS.
The primary analysis examined the impact of the EWRS across UPHS; we also examined the EWRS impact at each of our hospitals. Last, we performed subgroup analyses examining the EWRS impact in those assigned an International Classification of Diseases, 9th Revision code for sepsis at discharge or death. All analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC).
RESULTS
In the derivation cohort, 4575 patients met the inclusion criteria. The proportion of those in each category (06) achieving our outcomes of interest are described in Supporting Table 1 in the online version of this article. We defined a positive trigger as a score 4, as this threshold identified a limited number of patients (3.9% [180/4575]) with a high proportion experiencing our composite outcome (25.6% [46/180]). The proportion of patients with an EWRS score 4 and their time to event by hospital and health system is described in Supporting Table 2 in the online version of this article. Those with a score 4 were almost 4 times as likely to be transferred to the ICU, almost 7 times as likely to experience an RRT, and almost 10 times as likely to die. The screen positive, sensitivity, specificity, and positive and negative predictive values and likelihood ratios using this threshold and our composite outcome in the derivation cohort were 6%, 16%, 97%, 26%, 94%, 5.3, and 0.9, respectively, and in our validation cohort were 6%, 17%, 97%, 28%, 95%, 5.7, and 0.9, respectively.
In the preimplementation period, 3.8% of admissions (595/15,567) triggered the alert, as compared to 3.5% (545/15,526) in the postimplementation period. Demographics were similar across periods, except that in the postimplementation period patients were slightly younger and had a lower Charlson Comorbidity Index at admission (Table 1). The distribution of alerts across medicine and surgery services were similar (Table 1).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of encounters | 15,567 | 15,526 | |
| No. of alerts | 595 (4%) | 545 (4%) | 0.14 |
| Age, y, median (IQR) | 62.0 (48.570.5) | 59.7 (46.169.6) | 0.04 |
| Female | 298 (50%) | 274 (50%) | 0.95 |
| Race | |||
| White | 343 (58%) | 312 (57%) | 0.14 |
| Black | 207 (35%) | 171 (31%) | |
| Other | 23 (4%) | 31 (6%) | |
| Unknown | 22 (4%) | 31 (6%) | |
| Admission type | |||
| Elective | 201 (34%) | 167 (31%) | 0.40 |
| ED | 300 (50%) | 278 (51%) | |
| Transfer | 94 (16%) | 99 (18%) | |
| BMI, kg/m2, median (IQR) | 27.0 (23.032.0) | 26.0 (22.031.0) | 0.24 |
| Previous ICU admission | 137 (23%) | 127 (23%) | 0.91 |
| RRT before alert | 27 (5%) | 20 (4%) | 0.46 |
| Admission Charlson index, median (IQR) | 2.0 (1.04.0) | 2.0 (1.04.0) | 0.04 |
| Admitting service | |||
| Medicine | 398 (67%) | 364 (67%) | 0.18 |
| Surgery | 173 (29%) | 169 (31%) | |
| Other | 24 (4%) | 12 (2%) | |
| Service where alert fired | |||
| Medicine | 391 (66%) | 365 (67%) | 0.18 |
| Surgery | 175 (29%) | 164 (30%) | |
| Other | 29 (5%) | 15 (3%) | |
In our postimplementation period, 99% of coordinator pages and over three‐fourths of provider notifications were sent successfully. Almost three‐fourths of nurses reviewed the initial alert notification, and over 99% completed the electronic data verification and adverse trend review, with over half documenting adverse trends. Ninety‐five percent of the time the coordinators completed the follow‐up assessment. Over 90% of the time, the entire team evaluated the patient at bedside within 30 minutes. Almost half of the time, the team thought the patient had no critical illness. Over a third of the time, they thought the patient had sepsis, but reported over 90% of the time that they were aware of the diagnosis prior to the alert. (Supporting Table 3 in the online version of this article includes more details about the responses to the electronic notifications and follow‐up assessments.)
In unadjusted and adjusted analyses, ordering of antibiotics, intravenous fluid boluses, and lactate and blood cultures within 3 hours of the trigger increased significantly, as did ordering of blood products, chest radiographs, and cardiac monitoring within 6 hours of the trigger (Tables 2 and 3).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of alerts | 595 | 545 | |
| 500 mL IV bolus order <3 h after alert | 92 (15%) | 142 (26%) | <0.01 |
| IV/PO antibiotic order <3 h after alert | 75 (13%) | 123 (23%) | <0.01 |
| IV/PO sepsis antibiotic order <3 h after alert | 61 (10%) | 85 (16%) | <0.01 |
| Lactic acid order <3 h after alert | 57 (10%) | 128 (23%) | <0.01 |
| Blood culture order <3 h after alert | 68 (11%) | 99 (18%) | <0.01 |
| Blood gas order <6 h after alert | 53 (9%) | 59 (11%) | 0.28 |
| CBC or BMP <6 h after alert | 247 (42%) | 219 (40%) | 0.65 |
| Vasopressor <6 h after alert | 17 (3%) | 21 (4%) | 0.35 |
| Bronchodilator administration <6 h after alert | 71 (12%) | 64 (12%) | 0.92 |
| RBC, plasma, or platelet transfusion order <6 h after alert | 31 (5%) | 52 (10%) | <0.01 |
| Naloxone order <6 h after alert | 0 (0%) | 1 (0%) | 0.30 |
| AV node blocker order <6 h after alert | 35 (6%) | 35 (6%) | 0.70 |
| Loop diuretic order <6 h after alert | 35 (6%) | 28 (5%) | 0.58 |
| CXR <6 h after alert | 92 (15%) | 113 (21%) | 0.02 |
| CT head, chest, or ABD <6 h after alert | 29 (5%) | 34 (6%) | 0.31 |
| Cardiac monitoring (ECG or telemetry) <6 h after alert | 70 (12%) | 90 (17%) | 0.02 |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Odds Ratio | Adjusted Odds Ratio | Unadjusted Odds Ratio | Adjusted Odds Ratio | |
| ||||
| 500 mL IV bolus order <3 h after alert | 1.93 (1.442.58) | 1.93 (1.432.61) | 1.64 (1.112.43) | 1.65 (1.102.47) |
| IV/PO antibiotic order <3 h after alert | 2.02 (1.482.77) | 2.02 (1.462.78) | 1.99 (1.323.00) | 2.02 (1.323.09) |
| IV/PO sepsis antibiotic order <3 h after alert | 1.62 (1.142.30) | 1.57 (1.102.25) | 1.63 (1.052.53) | 1.65 (1.052.58) |
| Lactic acid order <3 h after alert | 2.90 (2.074.06) | 3.11 (2.194.41) | 2.41 (1.583.67) | 2.79 (1.794.34) |
| Blood culture <3 h after alert | 1.72 (1.232.40) | 1.76 (1.252.47) | 1.36 (0.872.10) | 1.40 (0.902.20) |
| Blood gas order <6 h after alert | 1.24 (0.841.83) | 1.32 (0.891.97) | 1.06 (0.631.77) | 1.13 (0.671.92) |
| BMP or CBC order <6 h after alert | 0.95 (0.751.20) | 0.96 (0.751.21) | 1.00 (0.701.44) | 1.04 (0.721.50) |
| Vasopressor order <6 h after alert | 1.36 (0.712.61) | 1.47 (0.762.83) | 1.32 (0.583.04) | 1.38 (0.593.25) |
| Bronchodilator administration <6 h after alert | 0.98 (0.691.41) | 1.02 (0.701.47) | 1.13 (0.641.99) | 1.17 (0.652.10) |
| Transfusion order <6 h after alert | 1.92 (1.213.04) | 1.95 (1.233.11) | 1.65 (0.913.01) | 1.68 (0.913.10) |
| AV node blocker order <6 h after alert | 1.10 (0.681.78) | 1.20 (0.722.00) | 0.38 (0.131.08) | 0.39 (0.121.20) |
| Loop diuretic order <6 h after alert | 0.87 (0.521.44) | 0.93 (0.561.57) | 1.63 (0.634.21) | 1.87 (0.705.00) |
| CXR <6 h after alert | 1.43 (1.061.94) | 1.47 (1.081.99) | 1.45 (0.942.24) | 1.56 (1.002.43) |
| CT <6 h after alert | 1.30 (0.782.16) | 1.30 (0.782.19) | 0.97 (0.521.82) | 0.94 (0.491.79) |
| Cardiac monitoring <6 h after alert | 1.48 (1.062.08) | 1.54 (1.092.16) | 1.32 (0.792.18) | 1.44 (0.862.41) |
Hospital and ICU length of stay were similar in the preimplementation and postimplementation periods. There was no difference in the proportion of patients transferred to the ICU following the alert; however, the proportion transferred within 6 hours of the alert increased, and the time to ICU transfer was halved (see Supporting Figure 4 in the online version of this article), but neither change was statistically significant in unadjusted analyses. Transfer to the ICU within 6 hours became statistically significant after adjustment. All mortality measures were lower in the postimplementation period, but none reached statistical significance. Discharge to home and sepsis documentation were both statistically higher in the postimplementation period, but discharge to home lost statistical significance after adjustment (Tables 4 and 5) (see Supporting Table 4 in the online version of this article).
| Hospitals AC | ||||
|---|---|---|---|---|
| Preimplementation | Postimplementation | P Value | ||
| ||||
| No. of alerts | 595 | 545 | ||
| Hospital LOS, d, median (IQR) | 10.1 (5.119.1) | 9.4 (5.218.9) | 0.92 | |
| ICU LOS after alert, d, median (IQR) | 3.4 (1.77.4) | 3.6 (1.96.8) | 0.72 | |
| ICU transfer <6 h after alert | 40 (7%) | 53 (10%) | 0.06 | |
| ICU transfer <24 h after alert | 71 (12%) | 79 (14%) | 0.20 | |
| ICU transfer any time after alert | 134 (23%) | 124 (23%) | 0.93 | |
| Time to first ICU after alert, h, median (IQR) | 21.3 (4.463.9) | 11.0 (2.358.7) | 0.22 | |
| RRT 6 h after alert | 13 (2%) | 9 (2%) | 0.51 | |
| Mortality of all patients | 52 (9%) | 41 (8%) | 0.45 | |
| Mortality 30 days after alert | 48 (8%) | 33 (6%) | 0.19 | |
| Mortality of those transferred to ICU | 40 (30%) | 32 (26%) | 0.47 | |
| Deceased or IP hospice | 94 (16%) | 72 (13%) | 0.22 | |
| Discharge to home | 347 (58%) | 351 (64%) | 0.04 | |
| Disposition location | ||||
| Home | 347 (58%) | 351 (64%) | 0.25 | |
| SNF | 89 (15%) | 65 (12%) | ||
| Rehab | 24 (4%) | 20 (4%) | ||
| LTC | 8 (1%) | 9 (2%) | ||
| Other hospital | 16 (3%) | 6 (1%) | ||
| Expired | 52 (9%) | 41 (8%) | ||
| Hospice IP | 42 (7%) | 31 (6%) | ||
| Hospice other | 11 (2%) | 14 (3%) | ||
| Other location | 6 (1%) | 8 (1%) | ||
| Sepsis discharge diagnosis | 230 (39%) | 247 (45%) | 0.02 | |
| Sepsis O/E | 1.37 | 1.06 | 0.18 | |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Estimate | Adjusted Estimate | Unadjusted Estimate | Adjusted Estimate | |
| ||||
| Hospital LOS, d | 1.01 (0.921.11) | 1.02 (0.931.12) | 0.99 (0.851.15) | 1.00 (0.871.16) |
| ICU transfer | 1.49 (0.972.29) | 1.65 (1.072.55) | 1.61 (0.922.84) | 1.82 (1.023.25) |
| Time to first ICU transfer after alert, h‖ | 1.17 (0.871.57) | 1.23 (0.921.66) | 1.21 (0.831.75) | 1.31 (0.901.90) |
| ICU LOS, d | 1.01 (0.771.31) | 0.99 (0.761.28) | 0.87 (0.621.21) | 0.88 (0.641.21) |
| RRT | 0.75 (0.321.77) | 0.84 (0.352.02) | 0.81 (0.292.27) | 0.82 (0.272.43) |
| Mortality | 0.85 (0.551.30) | 0.98 (0.631.53) | 0.85 (0.551.30) | 0.98 (0.631.53) |
| Mortality within 30 days of alert | 0.73 (0.461.16) | 0.87 (0.541.40) | 0.59 (0.341.04) | 0.69 (0.381.26) |
| Mortality or inpatient hospice transfer | 0.82 (0.471.41) | 0.78 (0.441.41) | 0.67 (0.361.25) | 0.65 (0.331.29) |
| Discharge to home | 1.29 (1.021.64) | 1.18 (0.911.52) | 1.36 (0.951.95) | 1.22 (0.811.84) |
| Sepsis discharge diagnosis | 1.32 (1.041.67) | 1.43 (1.101.85) | NA | NA |
In a subanalysis of EWRS impact on patients documented with sepsis at discharge, unadjusted and adjusted changes in clinical process and outcome measures across the time periods were similar to that of the total population (see Supporting Tables 5 and 6 and Supporting Figure 5 in the online version of this article). The unadjusted composite outcome of mortality or inpatient hospice was statistically lower in the postimplementation period, but lost statistical significance after adjustment.
The disposition and mortality outcomes of those not triggering the alert were unchanged across the 2 periods (see Supporting Tables 7, 8, and 9 in the online version of this article).
DISCUSSION
This study demonstrated that a predictive tool can accurately identify non‐ICU inpatients at increased risk for deterioration and death. In addition, we demonstrated the feasibility of deploying our EHR to screen patients in real time for deterioration and to trigger electronically a timely, robust, multidisciplinary bedside clinical evaluation. Compared to the control (silent) period, the EWRS resulted in a marked increase in early sepsis care, transfer to the ICU, and sepsis documentation, and an indication of a decreased sepsis mortality index and mortality, and increased discharge to home, although none of these latter 3 findings reached statistical significance.
Our study is unique in that it was implemented across a multihospital health system, which has identical EHRs, but diverse cultures, populations, staffing, and practice models. In addition, our study includes a preimplementation population similar to the postimplementation population (in terms of setting, month of admission, and adjustment for potential confounders).
Interestingly, patients identified by the EWRS who were subsequently transferred to an ICU had higher mortality rates (30% and 26% in the preimplementation and postimplementation periods, respectively, across UPHS) than those transferred to an ICU who were not identified by the EWRS (7% and 6% in the preimplementation and postimplementation periods, respectively, across UPHS) (Table 4) (see Supporting Table 7 in the online version of this article). This finding was robust to the study period, so is likely not related to the bedside evaluation prompted by the EWRS. It suggests the EWRS could help triage patients for appropriateness of ICU transfer, a particularly valuable role that should be explored further given the typical strains on ICU capacity,[13] and the mortality resulting from delays in patient transfers into ICUs.[14, 15]
Although we did not find a statistically significant mortality reduction, our study may have been underpowered to detect this outcome. Our study has other limitations. First, our preimplementation/postimplementation design may not fully account for secular changes in sepsis mortality. However, our comparison of similar time periods and our adjustment for observed demographic differences allow us to estimate with more certainty the change in sepsis care and mortality attributable to the intervention. Second, our study did not examine the effect of the EWRS on mortality after hospital discharge, where many such events occur. However, our capture of at least 45 hospital days on all study patients, as well as our inclusion of only those who died or were discharged during our study period, and our assessment of discharge disposition such as hospice, increase the chance that mortality reductions directly attributable to the EWRS were captured. Third, although the EWRS changed patient management, we did not assess the appropriateness of management changes. However, the impact of care changes was captured crudely by examining mortality rates and discharge disposition. Fourth, our study was limited to a single academic healthcare system, and our experience may not be generalizable to other healthcare systems with different EHRs and staff. However, the integration of our automated alert into a commercial EHR serving a diverse array of patient populations, clinical services, and service models throughout our healthcare system may improve the generalizability of our experience to other settings.
CONCLUSION
By leveraging readily available electronic data, an automated prediction tool identified at‐risk patients and mobilized care teams, resulting in more timely sepsis care, improved sepsis documentation, and a suggestion of reduced mortality. This alert may be scalable to other healthcare systems.
Acknowledgements
The authors thank Jennifer Barger, MS, BSN, RN; Patty Baroni, MSN, RN; Patrick J. Donnelly, MS, RN, CCRN; Mika Epps, MSN, RN; Allen L. Fasnacht, MSN, RN; Neil O. Fishman, MD; Kevin M. Fosnocht, MD; David F. Gaieski, MD; Tonya Johnson, MSN, RN, CCRN; Craig R. Kean, MS; Arash Kia, MD, MS; Matthew D. Mitchell, PhD; Stacie Neefe, BSN, RN; Nina J. Renzi, BSN, RN, CCRN; Alexander Roederer, Jean C. Romano, MSN, RN, NE‐BC; Heather Ross, BSN, RN, CCRN; William D. Schweickert, MD; Esme Singer, MD; and Kendal Williams, MD, MPH for their help in developing, testing and operationalizing the EWRS examined in this study; their assistance in data acquisition; and for advice regarding data analysis. This study was previously presented as an oral abstract at the 2013 American Medical Informatics Association Meeting, November 1620, 2013, Washington, DC.
Disclosures: Dr. Umscheid's contribution to this project was supported in part by the National Center for Research Resources, grant UL1RR024134, which is now at the National Center for Advancing Translational Sciences, grant UL1TR000003. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors report no potential financial conflicts of interest relevant to this article.
There are as many as 3 million cases of severe sepsis and 750,000 resulting deaths in the United States annually.[1] Interventions such as goal‐directed resuscitation and antibiotics can reduce sepsis mortality, but their effectiveness depends on early administration. Thus, timely recognition is critical.[2, 3, 4, 5]
Despite this, early recognition in hospitalized patients can be challenging. Using chart documentation as a surrogate for provider recognition, we recently found only 20% of patients with severe sepsis admitted to our hospital from the emergency department were recognized.[6] Given these challenges, there has been increasing interest in developing automated systems to improve the timeliness of sepsis detection.[7, 8, 9, 10] Systems described in the literature have varied considerably in triggering criteria, effector responses, and study settings. Of those examining the impact of automated surveillance and response in the nonintensive care unit (ICU) acute inpatient setting, results suggest an increase in the timeliness of diagnostic and therapeutic interventions,[10] but less impact on patient outcomes.[7] Whether these results reflect inadequacies in the criteria used to identify patients (parameters or their thresholds) or an ineffective response to the alert (magnitude or timeliness) is unclear.
Given the consequences of severe sepsis in hospitalized patients, as well as the introduction of vital sign (VS) and provider data in our electronic health record (EHR), we sought to develop and implement an electronic sepsis detection and response system to improve patient outcomes. This study describes the development, validation, and impact of that system.
METHODS
Setting and Data Sources
The University of Pennsylvania Health System (UPHS) includes 3 hospitals with a capacity of over 1500 beds and 70,000 annual admissions. All hospitals use the EHR Sunrise Clinical Manager version 5.5 (Allscripts, Chicago, IL). The study period began in October 2011, when VS and provider contact information became available electronically. Data were retrieved from the Penn Data Store, which includes professionally coded data as well as clinical data from our EHRs. The study received expedited approval and a Health Insurance Portability and Accountability Act waiver from our institutional review board.
Development of the Intervention
The early warning and response system (EWRS) for sepsis was designed to monitor laboratory values and VSs in real time in our inpatient EHR to detect patients at risk for clinical deterioration and development of severe sepsis. The development team was multidisciplinary, including informaticians, physicians, nurses, and data analysts from all 3 hospitals.
To identify at‐risk patients, we used established criteria for severe sepsis, including the systemic inflammatory response syndrome criteria (temperature <36C or >38C, heart rate >90 bpm, respiratory rate >20 breaths/min or PaCO2 <32 mm Hg, and total white blood cell count <4000 or >12,000 or >10% bands) coupled with criteria suggesting organ dysfunction (cardiovascular dysfunction based on a systolic blood pressure <100 mm Hg, and hypoperfusion based on a serum lactate measure >2.2 mmol/L [the threshold for an abnormal result in our lab]).[11, 12]
To establish a threshold for triggering the system, a derivation cohort was used and defined as patients admitted between October 1, 2011 to October 31, 2011 1 to any inpatient acute care service. Those <18 years old or admitted to hospice, research, and obstetrics services were excluded. We calculated a risk score for each patient, defined as the sum of criteria met at any single time during their visit. At any given point in time, we used the most recent value for each criteria, with a look‐back period of 24 hours for VSs and 48 hours for labs. The minimum and maximum number of criteria that a patient could achieve at any single time was 0 and 6, respectively. We then categorized patients by the maximum number of criteria achieved and estimated the proportion of patients in each category who: (1) were transferred to an ICU during their hospital visit; (2) had a rapid response team (RRT) called during their visit; (3) died during their visit; (4) had a composite of 1, 2, or 3; or (5) were coded as sepsis at discharge (see Supporting Information in the online version of this article for further information). Once a threshold was chosen, we examined the time from first trigger to: (1) any ICU transfer; (2) any RRT; (3) death; or (4) a composite of 1, 2, or 3. We then estimated the screen positive rate, test characteristics, predictive values, and likelihood ratios of the specified threshold.
The efferent response arm of the EWRS included the covering provider (usually an intern), the bedside nurse, and rapid response coordinators, who were engaged from the outset in developing the operational response to the alert. This team was required to perform a bedside evaluation within 30 minutes of the alert, and enact changes in management if warranted. The rapid response coordinator was required to complete a 3‐question follow‐up assessment in the EHR asking whether all 3 team members gathered at the bedside, the most likely condition triggering the EWRS, and whether management changed (see Supporting Figure 1 in the online version of this article). To minimize the number of triggers, once a patient triggered an alert, any additional alert triggers during the same hospital stay were censored.
Implementation of the EWRS
All inpatients on noncritical care services were screened continuously. Hospice, research, and obstetrics services were excluded. If a patient met the EWRS criteria threshold, an alert was sent to the covering provider and rapid response coordinator by text page. The bedside nurses, who do not carry text‐enabled devices, were alerted by pop‐up notification in the EHR (see Supporting Figure 2 in the online version of this article). The notification was linked to a task that required nurses to verify in the EHR the VSs triggering the EWRS, and adverse trends in VSs or labs (see Supporting Figure 3 in the online version of this article).
The Preimplementation (Silent) Period and EWRS Validation
The EWRS was initially activated for a preimplementation silent period (June 6, 2012September 4, 2012) to both validate the tool and provide the baseline data to which the postimplementation period was compared. During this time, new admissions could trigger the alert, but notifications were not sent. We used admissions from the first 30 days of the preimplementation period to estimate the tool's screen positive rate, test characteristics, predictive values, and likelihood ratios.
The Postimplementation (Live) Period and Impact Analysis
The EWRS went live September 12, 2012, upon which new admissions triggering the alert would result in a notification and response. Unadjusted analyses using the [2] test for dichotomous variables and the Wilcoxon rank sum test for continuous variables compared demographics and the proportion of clinical process and outcome measures for those admitted during the silent period (June 6, 2012September 4, 2012) and a similar timeframe 1 year later when the intervention was live (June 6, 2013September 4, 2013). To be included in either of the time periods, patients had to trigger the alert during the period and be discharged within 45 days of the end of the period. The pre‐ and post‐sepsis mortality index was also examined (see the Supporting Information in the online version of this article for a detailed description of study measures). Multivariable regression models estimated the impact of the EWRS on the process and outcome measures, adjusted for differences between the patients in the preimplementation and postimplementation periods with respect to age, gender, Charlson index on admission, admitting service, hospital, and admission month. Logistic regression models examined dichotomous variables. Continuous variables were log transformed and examined using linear regression models. Cox regression models explored time to ICU transfer from trigger. Among patients with sepsis, a logistic regression model was used to compare the odds of mortality between the silent and live periods, adjusted for expected mortality, both within each hospital and across all hospitals.
Because there is a risk of providers becoming overly reliant on automated systems and overlooking those not triggering the system, we also examined the discharge disposition and mortality outcomes of those in both study periods not identified by the EWRS.
The primary analysis examined the impact of the EWRS across UPHS; we also examined the EWRS impact at each of our hospitals. Last, we performed subgroup analyses examining the EWRS impact in those assigned an International Classification of Diseases, 9th Revision code for sepsis at discharge or death. All analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC).
RESULTS
In the derivation cohort, 4575 patients met the inclusion criteria. The proportion of those in each category (06) achieving our outcomes of interest are described in Supporting Table 1 in the online version of this article. We defined a positive trigger as a score 4, as this threshold identified a limited number of patients (3.9% [180/4575]) with a high proportion experiencing our composite outcome (25.6% [46/180]). The proportion of patients with an EWRS score 4 and their time to event by hospital and health system is described in Supporting Table 2 in the online version of this article. Those with a score 4 were almost 4 times as likely to be transferred to the ICU, almost 7 times as likely to experience an RRT, and almost 10 times as likely to die. The screen positive, sensitivity, specificity, and positive and negative predictive values and likelihood ratios using this threshold and our composite outcome in the derivation cohort were 6%, 16%, 97%, 26%, 94%, 5.3, and 0.9, respectively, and in our validation cohort were 6%, 17%, 97%, 28%, 95%, 5.7, and 0.9, respectively.
In the preimplementation period, 3.8% of admissions (595/15,567) triggered the alert, as compared to 3.5% (545/15,526) in the postimplementation period. Demographics were similar across periods, except that in the postimplementation period patients were slightly younger and had a lower Charlson Comorbidity Index at admission (Table 1). The distribution of alerts across medicine and surgery services were similar (Table 1).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of encounters | 15,567 | 15,526 | |
| No. of alerts | 595 (4%) | 545 (4%) | 0.14 |
| Age, y, median (IQR) | 62.0 (48.570.5) | 59.7 (46.169.6) | 0.04 |
| Female | 298 (50%) | 274 (50%) | 0.95 |
| Race | |||
| White | 343 (58%) | 312 (57%) | 0.14 |
| Black | 207 (35%) | 171 (31%) | |
| Other | 23 (4%) | 31 (6%) | |
| Unknown | 22 (4%) | 31 (6%) | |
| Admission type | |||
| Elective | 201 (34%) | 167 (31%) | 0.40 |
| ED | 300 (50%) | 278 (51%) | |
| Transfer | 94 (16%) | 99 (18%) | |
| BMI, kg/m2, median (IQR) | 27.0 (23.032.0) | 26.0 (22.031.0) | 0.24 |
| Previous ICU admission | 137 (23%) | 127 (23%) | 0.91 |
| RRT before alert | 27 (5%) | 20 (4%) | 0.46 |
| Admission Charlson index, median (IQR) | 2.0 (1.04.0) | 2.0 (1.04.0) | 0.04 |
| Admitting service | |||
| Medicine | 398 (67%) | 364 (67%) | 0.18 |
| Surgery | 173 (29%) | 169 (31%) | |
| Other | 24 (4%) | 12 (2%) | |
| Service where alert fired | |||
| Medicine | 391 (66%) | 365 (67%) | 0.18 |
| Surgery | 175 (29%) | 164 (30%) | |
| Other | 29 (5%) | 15 (3%) | |
In our postimplementation period, 99% of coordinator pages and over three‐fourths of provider notifications were sent successfully. Almost three‐fourths of nurses reviewed the initial alert notification, and over 99% completed the electronic data verification and adverse trend review, with over half documenting adverse trends. Ninety‐five percent of the time the coordinators completed the follow‐up assessment. Over 90% of the time, the entire team evaluated the patient at bedside within 30 minutes. Almost half of the time, the team thought the patient had no critical illness. Over a third of the time, they thought the patient had sepsis, but reported over 90% of the time that they were aware of the diagnosis prior to the alert. (Supporting Table 3 in the online version of this article includes more details about the responses to the electronic notifications and follow‐up assessments.)
In unadjusted and adjusted analyses, ordering of antibiotics, intravenous fluid boluses, and lactate and blood cultures within 3 hours of the trigger increased significantly, as did ordering of blood products, chest radiographs, and cardiac monitoring within 6 hours of the trigger (Tables 2 and 3).
| Hospitals AC | |||
|---|---|---|---|
| Preimplementation | Postimplementation | P Value | |
| |||
| No. of alerts | 595 | 545 | |
| 500 mL IV bolus order <3 h after alert | 92 (15%) | 142 (26%) | <0.01 |
| IV/PO antibiotic order <3 h after alert | 75 (13%) | 123 (23%) | <0.01 |
| IV/PO sepsis antibiotic order <3 h after alert | 61 (10%) | 85 (16%) | <0.01 |
| Lactic acid order <3 h after alert | 57 (10%) | 128 (23%) | <0.01 |
| Blood culture order <3 h after alert | 68 (11%) | 99 (18%) | <0.01 |
| Blood gas order <6 h after alert | 53 (9%) | 59 (11%) | 0.28 |
| CBC or BMP <6 h after alert | 247 (42%) | 219 (40%) | 0.65 |
| Vasopressor <6 h after alert | 17 (3%) | 21 (4%) | 0.35 |
| Bronchodilator administration <6 h after alert | 71 (12%) | 64 (12%) | 0.92 |
| RBC, plasma, or platelet transfusion order <6 h after alert | 31 (5%) | 52 (10%) | <0.01 |
| Naloxone order <6 h after alert | 0 (0%) | 1 (0%) | 0.30 |
| AV node blocker order <6 h after alert | 35 (6%) | 35 (6%) | 0.70 |
| Loop diuretic order <6 h after alert | 35 (6%) | 28 (5%) | 0.58 |
| CXR <6 h after alert | 92 (15%) | 113 (21%) | 0.02 |
| CT head, chest, or ABD <6 h after alert | 29 (5%) | 34 (6%) | 0.31 |
| Cardiac monitoring (ECG or telemetry) <6 h after alert | 70 (12%) | 90 (17%) | 0.02 |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Odds Ratio | Adjusted Odds Ratio | Unadjusted Odds Ratio | Adjusted Odds Ratio | |
| ||||
| 500 mL IV bolus order <3 h after alert | 1.93 (1.442.58) | 1.93 (1.432.61) | 1.64 (1.112.43) | 1.65 (1.102.47) |
| IV/PO antibiotic order <3 h after alert | 2.02 (1.482.77) | 2.02 (1.462.78) | 1.99 (1.323.00) | 2.02 (1.323.09) |
| IV/PO sepsis antibiotic order <3 h after alert | 1.62 (1.142.30) | 1.57 (1.102.25) | 1.63 (1.052.53) | 1.65 (1.052.58) |
| Lactic acid order <3 h after alert | 2.90 (2.074.06) | 3.11 (2.194.41) | 2.41 (1.583.67) | 2.79 (1.794.34) |
| Blood culture <3 h after alert | 1.72 (1.232.40) | 1.76 (1.252.47) | 1.36 (0.872.10) | 1.40 (0.902.20) |
| Blood gas order <6 h after alert | 1.24 (0.841.83) | 1.32 (0.891.97) | 1.06 (0.631.77) | 1.13 (0.671.92) |
| BMP or CBC order <6 h after alert | 0.95 (0.751.20) | 0.96 (0.751.21) | 1.00 (0.701.44) | 1.04 (0.721.50) |
| Vasopressor order <6 h after alert | 1.36 (0.712.61) | 1.47 (0.762.83) | 1.32 (0.583.04) | 1.38 (0.593.25) |
| Bronchodilator administration <6 h after alert | 0.98 (0.691.41) | 1.02 (0.701.47) | 1.13 (0.641.99) | 1.17 (0.652.10) |
| Transfusion order <6 h after alert | 1.92 (1.213.04) | 1.95 (1.233.11) | 1.65 (0.913.01) | 1.68 (0.913.10) |
| AV node blocker order <6 h after alert | 1.10 (0.681.78) | 1.20 (0.722.00) | 0.38 (0.131.08) | 0.39 (0.121.20) |
| Loop diuretic order <6 h after alert | 0.87 (0.521.44) | 0.93 (0.561.57) | 1.63 (0.634.21) | 1.87 (0.705.00) |
| CXR <6 h after alert | 1.43 (1.061.94) | 1.47 (1.081.99) | 1.45 (0.942.24) | 1.56 (1.002.43) |
| CT <6 h after alert | 1.30 (0.782.16) | 1.30 (0.782.19) | 0.97 (0.521.82) | 0.94 (0.491.79) |
| Cardiac monitoring <6 h after alert | 1.48 (1.062.08) | 1.54 (1.092.16) | 1.32 (0.792.18) | 1.44 (0.862.41) |
Hospital and ICU length of stay were similar in the preimplementation and postimplementation periods. There was no difference in the proportion of patients transferred to the ICU following the alert; however, the proportion transferred within 6 hours of the alert increased, and the time to ICU transfer was halved (see Supporting Figure 4 in the online version of this article), but neither change was statistically significant in unadjusted analyses. Transfer to the ICU within 6 hours became statistically significant after adjustment. All mortality measures were lower in the postimplementation period, but none reached statistical significance. Discharge to home and sepsis documentation were both statistically higher in the postimplementation period, but discharge to home lost statistical significance after adjustment (Tables 4 and 5) (see Supporting Table 4 in the online version of this article).
| Hospitals AC | ||||
|---|---|---|---|---|
| Preimplementation | Postimplementation | P Value | ||
| ||||
| No. of alerts | 595 | 545 | ||
| Hospital LOS, d, median (IQR) | 10.1 (5.119.1) | 9.4 (5.218.9) | 0.92 | |
| ICU LOS after alert, d, median (IQR) | 3.4 (1.77.4) | 3.6 (1.96.8) | 0.72 | |
| ICU transfer <6 h after alert | 40 (7%) | 53 (10%) | 0.06 | |
| ICU transfer <24 h after alert | 71 (12%) | 79 (14%) | 0.20 | |
| ICU transfer any time after alert | 134 (23%) | 124 (23%) | 0.93 | |
| Time to first ICU after alert, h, median (IQR) | 21.3 (4.463.9) | 11.0 (2.358.7) | 0.22 | |
| RRT 6 h after alert | 13 (2%) | 9 (2%) | 0.51 | |
| Mortality of all patients | 52 (9%) | 41 (8%) | 0.45 | |
| Mortality 30 days after alert | 48 (8%) | 33 (6%) | 0.19 | |
| Mortality of those transferred to ICU | 40 (30%) | 32 (26%) | 0.47 | |
| Deceased or IP hospice | 94 (16%) | 72 (13%) | 0.22 | |
| Discharge to home | 347 (58%) | 351 (64%) | 0.04 | |
| Disposition location | ||||
| Home | 347 (58%) | 351 (64%) | 0.25 | |
| SNF | 89 (15%) | 65 (12%) | ||
| Rehab | 24 (4%) | 20 (4%) | ||
| LTC | 8 (1%) | 9 (2%) | ||
| Other hospital | 16 (3%) | 6 (1%) | ||
| Expired | 52 (9%) | 41 (8%) | ||
| Hospice IP | 42 (7%) | 31 (6%) | ||
| Hospice other | 11 (2%) | 14 (3%) | ||
| Other location | 6 (1%) | 8 (1%) | ||
| Sepsis discharge diagnosis | 230 (39%) | 247 (45%) | 0.02 | |
| Sepsis O/E | 1.37 | 1.06 | 0.18 | |
| All Alerted Patients | Discharged With Sepsis Code* | |||
|---|---|---|---|---|
| Unadjusted Estimate | Adjusted Estimate | Unadjusted Estimate | Adjusted Estimate | |
| ||||
| Hospital LOS, d | 1.01 (0.921.11) | 1.02 (0.931.12) | 0.99 (0.851.15) | 1.00 (0.871.16) |
| ICU transfer | 1.49 (0.972.29) | 1.65 (1.072.55) | 1.61 (0.922.84) | 1.82 (1.023.25) |
| Time to first ICU transfer after alert, h‖ | 1.17 (0.871.57) | 1.23 (0.921.66) | 1.21 (0.831.75) | 1.31 (0.901.90) |
| ICU LOS, d | 1.01 (0.771.31) | 0.99 (0.761.28) | 0.87 (0.621.21) | 0.88 (0.641.21) |
| RRT | 0.75 (0.321.77) | 0.84 (0.352.02) | 0.81 (0.292.27) | 0.82 (0.272.43) |
| Mortality | 0.85 (0.551.30) | 0.98 (0.631.53) | 0.85 (0.551.30) | 0.98 (0.631.53) |
| Mortality within 30 days of alert | 0.73 (0.461.16) | 0.87 (0.541.40) | 0.59 (0.341.04) | 0.69 (0.381.26) |
| Mortality or inpatient hospice transfer | 0.82 (0.471.41) | 0.78 (0.441.41) | 0.67 (0.361.25) | 0.65 (0.331.29) |
| Discharge to home | 1.29 (1.021.64) | 1.18 (0.911.52) | 1.36 (0.951.95) | 1.22 (0.811.84) |
| Sepsis discharge diagnosis | 1.32 (1.041.67) | 1.43 (1.101.85) | NA | NA |
In a subanalysis of EWRS impact on patients documented with sepsis at discharge, unadjusted and adjusted changes in clinical process and outcome measures across the time periods were similar to that of the total population (see Supporting Tables 5 and 6 and Supporting Figure 5 in the online version of this article). The unadjusted composite outcome of mortality or inpatient hospice was statistically lower in the postimplementation period, but lost statistical significance after adjustment.
The disposition and mortality outcomes of those not triggering the alert were unchanged across the 2 periods (see Supporting Tables 7, 8, and 9 in the online version of this article).
DISCUSSION
This study demonstrated that a predictive tool can accurately identify non‐ICU inpatients at increased risk for deterioration and death. In addition, we demonstrated the feasibility of deploying our EHR to screen patients in real time for deterioration and to trigger electronically a timely, robust, multidisciplinary bedside clinical evaluation. Compared to the control (silent) period, the EWRS resulted in a marked increase in early sepsis care, transfer to the ICU, and sepsis documentation, and an indication of a decreased sepsis mortality index and mortality, and increased discharge to home, although none of these latter 3 findings reached statistical significance.
Our study is unique in that it was implemented across a multihospital health system, which has identical EHRs, but diverse cultures, populations, staffing, and practice models. In addition, our study includes a preimplementation population similar to the postimplementation population (in terms of setting, month of admission, and adjustment for potential confounders).
Interestingly, patients identified by the EWRS who were subsequently transferred to an ICU had higher mortality rates (30% and 26% in the preimplementation and postimplementation periods, respectively, across UPHS) than those transferred to an ICU who were not identified by the EWRS (7% and 6% in the preimplementation and postimplementation periods, respectively, across UPHS) (Table 4) (see Supporting Table 7 in the online version of this article). This finding was robust to the study period, so is likely not related to the bedside evaluation prompted by the EWRS. It suggests the EWRS could help triage patients for appropriateness of ICU transfer, a particularly valuable role that should be explored further given the typical strains on ICU capacity,[13] and the mortality resulting from delays in patient transfers into ICUs.[14, 15]
Although we did not find a statistically significant mortality reduction, our study may have been underpowered to detect this outcome. Our study has other limitations. First, our preimplementation/postimplementation design may not fully account for secular changes in sepsis mortality. However, our comparison of similar time periods and our adjustment for observed demographic differences allow us to estimate with more certainty the change in sepsis care and mortality attributable to the intervention. Second, our study did not examine the effect of the EWRS on mortality after hospital discharge, where many such events occur. However, our capture of at least 45 hospital days on all study patients, as well as our inclusion of only those who died or were discharged during our study period, and our assessment of discharge disposition such as hospice, increase the chance that mortality reductions directly attributable to the EWRS were captured. Third, although the EWRS changed patient management, we did not assess the appropriateness of management changes. However, the impact of care changes was captured crudely by examining mortality rates and discharge disposition. Fourth, our study was limited to a single academic healthcare system, and our experience may not be generalizable to other healthcare systems with different EHRs and staff. However, the integration of our automated alert into a commercial EHR serving a diverse array of patient populations, clinical services, and service models throughout our healthcare system may improve the generalizability of our experience to other settings.
CONCLUSION
By leveraging readily available electronic data, an automated prediction tool identified at‐risk patients and mobilized care teams, resulting in more timely sepsis care, improved sepsis documentation, and a suggestion of reduced mortality. This alert may be scalable to other healthcare systems.
Acknowledgements
The authors thank Jennifer Barger, MS, BSN, RN; Patty Baroni, MSN, RN; Patrick J. Donnelly, MS, RN, CCRN; Mika Epps, MSN, RN; Allen L. Fasnacht, MSN, RN; Neil O. Fishman, MD; Kevin M. Fosnocht, MD; David F. Gaieski, MD; Tonya Johnson, MSN, RN, CCRN; Craig R. Kean, MS; Arash Kia, MD, MS; Matthew D. Mitchell, PhD; Stacie Neefe, BSN, RN; Nina J. Renzi, BSN, RN, CCRN; Alexander Roederer, Jean C. Romano, MSN, RN, NE‐BC; Heather Ross, BSN, RN, CCRN; William D. Schweickert, MD; Esme Singer, MD; and Kendal Williams, MD, MPH for their help in developing, testing and operationalizing the EWRS examined in this study; their assistance in data acquisition; and for advice regarding data analysis. This study was previously presented as an oral abstract at the 2013 American Medical Informatics Association Meeting, November 1620, 2013, Washington, DC.
Disclosures: Dr. Umscheid's contribution to this project was supported in part by the National Center for Research Resources, grant UL1RR024134, which is now at the National Center for Advancing Translational Sciences, grant UL1TR000003. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors report no potential financial conflicts of interest relevant to this article.
- , , , . Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41(5):1167–1174.
- , , , et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637.
- , , , et al. The Surviving Sepsis Campaign: results of an international guideline‐based performance improvement program targeting severe sepsis. Crit Care Med. 2010;38(2):367–374.
- , , , et al. Early goal‐directed therapy in severe sepsis and septic shock revisited: concepts, controversies, and contemporary findings. Chest. 2006;130(5):1579–1595.
- , , , et al. Early goal‐directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.
- , , , , , . Severe sepsis cohorts derived from claims‐based strategies appear to be biased toward a more severely ill patient population. Crit Care Med. 2013;41(4):945–953.
- , , , et al. A trial of a real‐time alert for clinical deterioration in patients hospitalized on general medical wards. J Hosp Med. 2013;8(5):236–242.
- , , , , , . Bedside electronic capture of clinical observations and automated clinical alerts to improve compliance with an Early Warning Score protocol. Crit Care Resusc. 2011;13(2):83–88.
- , , , . Prospective trial of real‐time electronic surveillance to expedite early care of severe sepsis. Ann Emerg Med. 2011;57(5):500–504.
- , , , et al. Implementation of a real‐time computerized sepsis alert in nonintensive care unit patients. Crit Care Med. 2011;39(3):469–473.
- , , , et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101(6):1644–1655.
- , , , et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31(4):1250–1256.
- , , , , . Rationing critical care beds: a systematic review. Crit Care Med. 2004;32(7):1588–1597.
- . Delayed admission to intensive care unit for critically surgical patients is associated with increased mortality. Am J Surg. 2014;208:268–274.
- , , , et al. Impact of delayed admission to intensive care units on mortality of critically ill patients: a cohort study. Crit Care. 2011;15(1):R28.
- , , , . Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med. 2013;41(5):1167–1174.
- , , , et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637.
- , , , et al. The Surviving Sepsis Campaign: results of an international guideline‐based performance improvement program targeting severe sepsis. Crit Care Med. 2010;38(2):367–374.
- , , , et al. Early goal‐directed therapy in severe sepsis and septic shock revisited: concepts, controversies, and contemporary findings. Chest. 2006;130(5):1579–1595.
- , , , et al. Early goal‐directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.
- , , , , , . Severe sepsis cohorts derived from claims‐based strategies appear to be biased toward a more severely ill patient population. Crit Care Med. 2013;41(4):945–953.
- , , , et al. A trial of a real‐time alert for clinical deterioration in patients hospitalized on general medical wards. J Hosp Med. 2013;8(5):236–242.
- , , , , , . Bedside electronic capture of clinical observations and automated clinical alerts to improve compliance with an Early Warning Score protocol. Crit Care Resusc. 2011;13(2):83–88.
- , , , . Prospective trial of real‐time electronic surveillance to expedite early care of severe sepsis. Ann Emerg Med. 2011;57(5):500–504.
- , , , et al. Implementation of a real‐time computerized sepsis alert in nonintensive care unit patients. Crit Care Med. 2011;39(3):469–473.
- , , , et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101(6):1644–1655.
- , , , et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31(4):1250–1256.
- , , , , . Rationing critical care beds: a systematic review. Crit Care Med. 2004;32(7):1588–1597.
- . Delayed admission to intensive care unit for critically surgical patients is associated with increased mortality. Am J Surg. 2014;208:268–274.
- , , , et al. Impact of delayed admission to intensive care units on mortality of critically ill patients: a cohort study. Crit Care. 2011;15(1):R28.
© 2014 Society of Hospital Medicine