Literature Review of Modified Early Warning Scoring Tools Including Sepsis Screening Criteria

Purpose: The purpose of this comprehensive literature review was to evaluate current published evidence reporting decreased failure to rescue effects from modified early warning scoring (MEWS) system tools in hospitalized adult medical-surgical/telemetry patients from all-cause diagnoses including sepsis. Failure to rescue outcomes defined by researchers were rapid response team (RRT) activation and cardiopulmonary arrest (CPA) outside to an intensive care unit (ICU).  Severe sepsis exhibits worldwide clinical significance to practitioners and patients due to high associated mortality rates and costs to treat. Merely recording assessment findings may not be enough to identify patients at-risk for deterioration prior to clinical decline, especially with rapid decline in condition demonstrated by septic patients. Nurses’ thorough and timely clinical assessments, together with a willingness to ask for help in clinical management of deteriorating patients, are essential to survival of hospitalized patients during crucial changes in condition potentially leading to clinical worsening and death. Incomplete recording of vital signs (especially respiratory rate), misinterpretation of data, and a lack of urgency in communicating findings to other healthcare team members amidst the complexity of hospital systems contribute to adverse patient events. Incomplete recording of assessment data, misinterpretation of clinical findings, and delayed communication of discoveries to fellow healthcare workers within complex health delivery systems contribute to unanticipated negative patient events. The research team conducted this comprehensive literature review prior to recommending adaptations to early warning scoring system (EWSS) physiologic screening criteria for institutional use.  Researchers examined published findings reporting effect of MEWS tool usage (Intervention) on patient mortality (Outcome) and failure to rescue events (Outcome) in hospitalized (Time) adult medical-surgical/telemetry patients (Population).

Methods: A comprehensive review and evaluation of published peer-reviewed literature was conducted by a team of registered nurses employed by a nonprofit urban 883 bed hospital and private Christian university to establish interrater reliability in ranking of evidence findings. Electronic databases and clinical practice sources searched were The National Library of Medicine database (PubMed), MEDLINE, Cumulative Index to Nursing and Allied Health (CINAHL), the Cochrane Library of systematic reviews, and the Agency for Healthcare Research and Quality (AHRQ).  Search strategies combined the keywords “EWSS,” “MEWS”, and each term spelled out.  Literature was searched through 2014.  Two reference librarians cross verified the search strategy and confirmed inclusion of all applicable studies. All published, peer-reviewed articles (n=544) were initially included and evaluated by the research team leader and one research team member.  Inclusion criteria included research using adults (> 18 years of age) admitted to medical-surgical wards, telemetry units, or emergently transferred into the ICU due EWSS/MEWS instrument trigger mechanisms. Additional inclusion criteria included studies validating or examining existing EWSS/MEWS tool physiologic variable impact on mortality, RRT usage, or CPA. Articles not using adult (> 18 years of age), medical-surgical or telemetry monitored patients were excluded. Additionally, articles using patients with a specific disease focus were excluded for a final sample of (n=17) articles and one article using a pediatric population meeting all other criteria other than population of interest were used. 

Results: Articles (N=18) were identified for literature review inclusion. Two systematic reviews identified were rated Level I and Level V.   Evidence ratings by research team members included 6% Level I, 44% Level IV, 6% Level V, 33% Level VI, and 11% Level VII.  One systematic literature review of 33 descriptive studies was rated level V. Four studies reported a predictive value for mortality and two described mortality reduction from use of MEWS tools. Three studies measured MEWS tool usage impact on emergency calls to RRTs and medical emergency teams (METs) (failure to rescue). Four articles reported impacts on both mortality and RRT utilization (failure to rescue measurements) through MEWS tool usage. MEWS instruments are expansively adopted and implementation suggested for detecting inpatients at-risk for clinical deterioration; however, limited high-level evidence and no clinical trials linked use of EWSS/MEWS tools to robust outcomes measures through this literature review.  This comprehensive literature review identified no assessment tool and algorithm combining nursing assessment findings adjusted for systemic inflammatory response syndrome (SIRS) criteria with lab results to aid in identification of both the at-risk and septic patient. A systematic review of peer-reviewed literature reported evaluations of 33 unique aggregate weighted track and trigger systems (AWTTS) physiological scoring tools. The 33 AWTTS tools evaluated by researchers lacked inclusion of all four SIRS criteria used to diagnose sepsis (Smith, Prytherch, Schmidt, & Featherstone, 2008). Notably, seven AWTTS screened for only two of the four SIRS criteria.  Only four evaluated AWTTS tools scored oxygen source required for respiratory support instead of oxygen saturation measurements despite oxygen source representing a more accurate picture of respiratory effort worsening (Smith et al., 2008). Additionally, variability in scored physiological MEWS components and the score assigned for each physiologic derangement differed between all 33 examined tools.

Conclusion: Examined literature suggested physiological assessment findings can predict patient outcome.  Standards for validating MEWS criteria, organizational-specific reliability testing, and multi-site trials are recommended by this review team to establish robust evidence supporting MEWS tool usage. Systematic development and testing of all-cause illness (including sepsis) screening tools using sound research methods is imperative. MEWS instruments quantify patient clinical pictures with score weighted clinical decision algorithms to assist nurses’ clinical decision making.  Sound research methodology for tool development and testing potentially leads to improved outcomes and decreased incidence of failure to rescue. Clinical outcomes from tool development and testing should be measured, evaluated, and reported to publish MEWS impact on clinical outcomes.