SummaryKaiser Permanente–Northern California has a comprehensive program used by the region's 21 medical centers to reduce mortality from sepsis by focusing on early diagnosis and timely initiation of appropriate therapy. Program features include proactive screening in the emergency department, algorithms to guide evidence-based treatment, and an array of paper and electronic tools, including standard order sets, an electronic medical record alert, documentation tools, and training resources, that support the provision of appropriate care. Physicians and nurses within each hospital incorporate the program into site-specific workflows and receive support from onsite advisers and regular educational programs. The program has improved the ability to diagnose and provide appropriate care to patients with sepsis, leading to more patients meeting clinical targets, fewer deaths, and less need for inpatient care.Moderate: The evidence consists of pre- and post-implementation comparisons of key process and outcomes measures, including the proportion of patients diagnosed with sepsis, the percentage of those diagnosed that receive appropriate therapeutic interventions and achieve established clinical targets, mortality rates, and length of stay.
Developing OrganizationsKaiser Permanente-Northern California
Date First Implemented2009
Problem AddressedSepsis, a range of conditions resulting from the body's systemic response to infection,1 is common and often leads to lengthy hospitalizations and death. The impact can be reduced when patients are diagnosed and treated quickly and appropriately based on well-established recommendations, but many hospitals fail to do so, particularly with emergency department (ED) patients.
- A common, fatal condition: Severe sepsis occurs in approximately 750,000 intensive care unit (ICU) patients each year in the United States, with almost 40 percent dying as a result.2 Within Kaiser Permanente–Northern California's 21 hospitals, sepsis caused 24 percent of all inpatient deaths, though only 2.7 percent had sepsis as a primary or secondary diagnosis at admission.3
- Failure to diagnose in timely manner: Sepsis patients often experience poor outcomes because the condition goes undetected in its early stages, particularly for patients entering the hospital through the ED. Prior to 2008, few Kaiser Permanente–Northern California hospitals measured the success of sepsis resuscitation or provided hemodynamic monitoring of patients outside the ICUs. ED physicians and nurses did not consistently evaluate patients for the early signs of sepsis, which, combined with long ED stays, could delay both the identification and definitive treatment of patients with sepsis.3
- Inadequate treatment: Once diagnosed, sepsis patients often do not receive appropriate, evidence-based treatment. For example, studies of trauma patients with sepsis show that outcomes are poorer if they are not adequately resuscitated within 60 minutes of the event.4 In addition, timely antibiotic administration has been shown to improve outcomes for patients with sepsis.5 However, suboptimal treatment remains common. For example, though expert guidelines have been disseminated by the Surviving Sepsis Campaign (a collaborative initiative of the European Society of Intensive Care Medicine, the International Sepsis Forum, and the Society of Critical Care Medicine) that call for administration of antibiotics within an hour of suspected bacterial infection, the average time to first infusion of antibiotics exceeds 3 hours. Other treatment problems that contribute to poor sepsis patient outcomes include use of narrow-spectrum (affecting a limited number of organisms) rather than broad-spectrum (affecting a wide variety of organisms) antibiotics; inconsistent monitoring of physiologic parameters such as blood pressure, central venous pressure, and mixed venous oxygen saturation; and lack of treatment protocols and processes to support early, goal-directed therapy.6
Description of the Innovative ActivityKaiser Permanente–Northern California has a comprehensive program used across the region's 21 medical centers to reduce mortality from sepsis by focusing on early diagnosis and timely initiation of appropriate treatment. Program features include proactive screening in the ED, evidence-based treatment according to algorithms, and a set of written and electronic tools that support the provision of appropriate care. Physicians and nurses within each hospital incorporate the program into site-specific workflows and receive support from onsite advisers and regular educational programs. While program implementation varies by hospital, typical elements include the following:
- Screening and evaluation in the ED: ED nurses screen patients according to an algorithm, as follows:
- Screening for symptoms: Using a tool incorporated into the triage intake process, ED nurses screen all patients for symptoms of infection. The tool lists complaints by system that could suggest infection, such as fever, cough, and pelvic or abdominal pain.
- Screening for sepsis: If a patient has a suspected infection, nurses screen for systemic inflammatory response syndrome (SIRS). The four SIRS screening criteria include elevated heart rate, elevated respiratory rate, elevated or low temperature, and high or low white blood cell count. Nurses consider a diagnosis of sepsis if at least two SIRS criteria are met or if the patient has an altered level of consciousness.
- Confirmation of diagnosis: For any patient with suspected sepsis, the nurse rapidly notifies the ED physician, who orders a complete blood count, blood cultures, and serum lactate concentration (since an elevated level of lactic acid is a key sign of sepsis).
- Algorithm and order sets to guide evidence-based treatment: Sepsis patients receive treatment according to an algorithm and related order sets based on guidelines promulgated by the Surviving Sepsis Campaign and the Institute for Healthcare Improvement (see the Tools section for more information). The algorithm calls for the following:
- Rapid fluid administration: Patients with low blood pressure or high lactate receive rapid administration of intravenous (IV) fluid.
- Early administration of antibiotics: Physicians order administration of antibiotics within an hour of sepsis diagnosis, using electronic-based order sets that designate optimal antibiotic choice and dose for the infected organ system. Kaiser ED physicians and infectious disease specialists developed the order sets, which have been approved for region-wide use.
- Early goal-directed therapy: This part of the algorithm depicts the recommended order of treatment steps, including central line administration within 2 hours of diagnosis and use of vasopressor agents and other therapies. It also specifies associated therapeutic goals, such as reduction in lactate levels and target levels for mean arterial pressure, central venous pressure, and central venous oxygen saturation. All goals are timed to maintain focus on early hours of resuscitation whether in the ED or other parts of the hospital.
- Serial serum lactate testing: All patients with elevated serum lactate concentration are retested every 3 to 6 hours, with tests being prompted by an automatic order generated by the electronic medical record (EMR).
- Written and electronic tools to support appropriate care: A 100-page written “playbook” supports the delivery of appropriate care in each hospital. The playbook includes a list of expectations for hospitals related to implementation of the program (such as the composition of an onsite implementation team), staff meeting agendas and presentations, and tools that hospitals can use to improve sepsis care. Tools include care algorithms and standardized order sets for all diagnostic tests and treatments, along with tools to support nurse documentation, such as vital sign flowcharts, an ED admission handoff report, and a charge nurse checklist. Most tools are also available electronically through Kaiser’s EMR, which was rolled out within all Northern California region hospitals by 2010.
- Onsite advisers: Kaiser staff trained in performance improvement (known as "improvement advisers") oversee implementation at each hospital. In addition, health system leaders assign physician champions to serve as peer mentors.
- Ongoing review to facilitate improvement: A Web-based semiautomated chart abstraction tool identifies all ED patients with low blood pressure or elevated lactate levels and displays critical treatment parameters from the EMR to support case review. A nurse reviews cases from the last few days to determine if they met the criteria for early goal-directed therapy and whether clinical targets were met. This information helps to identify barriers to care, inform performance improvement, and promote standardized care.
- Communication and education: Improvement advisers and physician champions participate in monthly conference calls to solve problems and share best practices. Kaiser Permanente–Northern California also sponsors annual sepsis summits; twice-a-month sepsis conference calls for physician champions; a monthly sepsis newsletter; and a designated intranet site with access to tools, journal articles, and performance data related to sepsis care.
References/Related ArticlesWhippy A, Skeath M, Crawford B, et al. Kaiser Permanente's performance improvement system, part 3: multisite improvements in care for patients with sepsis. Jt Comm J Qual Patient Saf. 2011 Nov;37(11):483-93. [PubMed]
Contact the InnovatorAlan Whippy, MD
Medical Director of Quality and Safety
The Permanente Medical Group
1 Kaiser Plaza
Melinda Skeath, RN, CNS
Executive Director Quality & Regulatory Services
1950 Franklin Street, 14th Floor
Innovator DisclosuresDr. Whippy and Ms. Skeath reported having no financial interests or business/professional affiliations relevant to the work described in this profile other than the funders listed in the Funding Sources section.
ResultsThe program improved the ability to diagnose and provide appropriate care to patients with sepsis, leading to more patients meeting clinical targets, fewer deaths, and less need for inpatient care.
Moderate: The evidence consists of pre- and post-implementation comparisons of key process and outcomes measures, including the proportion of patients diagnosed with sepsis, the percentage of those diagnosed that receive appropriate therapeutic interventions and achieve established clinical targets, mortality rates, and length of stay.
- Improved ability to diagnose sepsis: Between July 2009 and May 2011, the ability to diagnosis sepsis improved significantly, with the number of cases diagnosed increasing from 35.7 to 119.4 per 1,000 admissions.
- More appropriate care: Over the same time period, the proportion of eligible patients with sepsis receiving appropriate treatments increased significantly, with improvements in serum lactate testing (from 27 to 97 percent), antibiotic administration within an hour of diagnosis (69.5 to 88.6 percent), and central venous pressure or central venous oxygen saturation recorded within 2 hours of diagnosis (41.5 to 85.1 percent). There were also significant improvements in attaining hemodynamic targets to improve perfusion: arterial pressure (52 to 93.9 percent), central venous pressure (41.5 to 86 percent), central venous oxygen saturation (30.8 to 75.4 percent), and lower serum lactate levels within 12 hours (52 to 92.1 percent). The percentage of eligible patients receiving all 6 components of early goal-directed therapy increased from 7.3 to 60.5 percent. This performance has been sustained over the subsequent year, with 70 percent of patients successful in all elements of care in May 2012.
- Fewer deaths: For the early goal-directed therapy population, composed of 100 to 140 patients each month with septic shock who are very high risk, mortality dropped from an initial 27 percent in the summer of 2009 to 17 percent in 2011. For the broader sepsis population, mortality dropped from an initial 24 percent in 2007 and 2008 to 10 percent in 2011. The ratio of observed to expected overall sepsis mortality dropped from 1.07 in 2008 to 0.64 in second quarter 2011. The sepsis mortality rate is 28 percent lower than the U.S. average.
- Lower length of stay (LOS): Over the same time period, the ratio of observed to expected LOS decreased from 1.08 to 0.8. Overall (principal and secondary) LOS is 4.8 days or 39 percent lower than the U.S. average.
Context of the InnovationKaiser Permanente–Northern California operates 21 medical centers and 226 clinics that collectively serve 3.25 million members. The 21 hospitals handle about 800,000 ED visits annually. The impetus for this program came in early 2008, when senior executives began noticing variations in mortality rates across the region's medical centers and subsequently initiated a review to identify opportunities for improvement. This analysis identified sepsis as a key factor responsible for inpatient mortality; as noted earlier, only 2.7 percent of the more than 205,000 patients admitted to one of the region's hospitals in 2007 had sepsis as a primary or secondary diagnosis, but sepsis was responsible for 24 percent of all inpatient deaths.3 This finding led senior executives to make reducing sepsis-related mortality a top priority for performance improvement.
Planning and Development ProcessSelected steps included the following:
- Holding summit to identify priority areas: The region held a summit in May 2008 to identify priority areas for reducing mortality. Approximately 350 individuals from the regional office and 21 medical centers attended. The summit included a breakout session on sepsis during which participants reviewed sepsis mortality data and discussed ways to improve performance in this area.
- Hiring performance improvement experts: Regional leaders hired three performance improvement experts to serve as mentors in developing a sepsis care initiative.
- Selecting pilot sites: At the time of the summit, one of the region’s medical centers—Kaiser Foundation Hospital in Vallejo—had been planning to launch an initiative to promote early goal-directed therapy for ED patients with sepsis, making this hospital a natural pilot site for the regional initiative. Regional leaders selected Kaiser Permanente San Jose Medical Center to serve as a second site.
- Forming steering committee: In the summer of 2008, representatives from both pilot sites formed a sepsis steering committee, with committee members including ED and ICU physicians, hospitalists, nurses, a nurse educator, a project manager, and data analysts. The steering committee ensured ED physician engagement by identifying a strong ED clinical champion and gaining support in advance by all ED physician directors.
- Designing and testing algorithms and tools: Each site created a multidisciplinary team made up of physicians and frontline staff and co-chaired by a physician and nurse. The teams developed and tested the sepsis care algorithm and related tools. With the help of the mentor and regional clinical leaders, the teams conducted several plan-do-study-act (PDSA) cycles with the following goals:
- Develop and refine a sepsis screening tool
- Refine the sepsis care algorithm
- Develop additional tools for clinician use
- Design, test, and collect staff education strategies
- Design and test measurement strategies, including a tool to abstract data from the EMR to facilitate performance improvement
- Creating playbook of tools: All tools developed and finalized by the pilot sites became part of a written playbook, including treatment algorithms, standardized order sets, vital sign flowcharts, alerts prompting a serum lactate order, nursing documentation tools, and training resources. Since the program expanded regionwide (see bullet below), additional tools continue to be added to the playbook as they are developed by other sites. As noted earlier, many of these tools have also been integrated into the EMR system since its introduction in 2010.
- Holding sepsis summit: The steering committee held a sepsis summit in November 2008 to introduce the program and distribute the playbook; representatives from each of the two pilot hospitals attended.
- Implementing regionwide: Between November 2008 and January 2009, representatives of the other 19 hospitals in the region created performance improvement teams responsible for implementing the program, including training staff on the algorithm and tools and setting up systems to monitor care processes. A simulation mannequin that supports ultrasound-guided central line placement was purchased for each hospital, and regional "train the trainer" programs supported central line training for all ED physicians. Team members typically included ED and ICU chiefs, nurses, a hospitalist, a nurse educator, and an improvement adviser. A high-level physician leader whose role bridged disciplines (i.e., ED, ICU, operating room) helped ensure that coordination was successful. The improvement advisers also met regularly with each other to share best practices, and regional sepsis expert mentors conducted monthly site visits at each medical center to help with implementation. Implementation milestones and early quality outcomes were tracked on the organization's high-level dashboard.
- Collecting data for evaluation: In June 2009, the hospitals began collecting data to evaluate the program's impact on care processes and outcomes.
Resources Used and Skills Needed
- Staffing: The program required no new staff at the hospital level. At the regional level, Kaiser hired three performance improvement experts to serve as mentors to the hospitals and an analyst to develop the automated chart abstraction tool and manage data review. The regional office already had improvement advisers on staff who devoted some time to the initiative as part of their regular responsibilities.
- Costs: Data on program costs are unavailable. Major expenses included the purchase of at least one hemodynamic monitor for every ED in the region to allow clinicians to monitor central venous pressure, oxygenation, and mean arterial pressure (ICUs already had such monitors). In addition, the region purchased a central venous catheterization ultrasound-guided insertion simulator for each hospital to enable training on insertion of central lines. Other costs include the salary and benefits of personnel added at the regional level, as described above.
Funding SourcesKaiser Permanente-Northern California; Gordon and Betty Moore Foundation
A grant from the Gordon and Betty Moore Foundation funded the salaries of the performance improvement mentors for 2 years.
Tools and Other ResourcesThe Surviving Sepsis Campaign guidelines for the management of severe sepsis and septic shock are available at: http://www.guideline.gov/summary/summary.aspx?ss=15&doc_id=12231&string.
The IHI sepsis resuscitation bundle is available at: http://www.ihi.org/knowledge/Pages/Changes/ImplementtheSepsisResuscitationBundle.aspx.
Various tools and resources developed by Kaiser Permanente–Northern California related to this program are publicly available, including the algorithm for early identification and treatment of sepsis, a tool to ensure the timely delivery of goal-directed therapy, the ED handoff report to the ICU, and a charge nurse checklist. These are available at: Whippy A, Skeath M, Crawford B, et al. Kaiser Permanente's performance improvement system, part 3: multisite improvements in care for patients with sepsis. Jt Comm J Qual Patient Saf. 2011 Nov;37(11):483-93. [PubMed]
Getting Started with This Innovation
- Document problem and develop program based on evidence: By analyzing and highlighting the impact of sepsis on hospital-specific mortality and then developing program elements according to evidence-based guidelines, program leaders helped to ensure that representatives from individual hospitals within the region, and from relevant departments within those hospitals, were willing to adopt standard components of sepsis care.
- Ensure appropriate infrastructure: A common infrastructure helps to create alignment across the entire organization. It should include an executive group or steering committee (comprising representatives from the different hospitals) to support and oversee the work, and hospital-based teams to make and test changes, ideally through PDSA cycles.
- Clearly define measurement strategy: Program leaders should clearly define the process and outcome measures to be tracked by hospital participants at the beginning of the initiative, thus allowing for cross-institution comparisons of performance.
- Allow flexibility: Once program elements and measures have been established, individual sites should be given the flexibility needed to implement those elements based on their own cultures and other organization-specific considerations.
- Hold regular meetings: During the adoption phase, regular meetings help to ensure that participants provide feedback, share best practices, analyze new processes, and make adjustments as necessary.
- Consider implications of organization size: In many ways, implementing this program at one hospital would be much easier than at a large, multihospital system, which requires very significant engagement at every level—from peers to hospital leadership.
- Ensure adequate support: At any size, strong support from the highest level of the hospital or system is critical. Small investments such as creating analytic support to track and display performance, having nursing staff review charts, collecting and telling stories, and supporting physician champions to engage in administrative activities are good investments. Because physicians and department managers are extremely busy and have daily urgent concerns, investment in project management with experience in supporting teams and doing "small tests of change" is very valuable.
Sustaining This Innovation
- Communicate constantly: Constant communication about the program and the need for better sepsis care helps to maintain the importance of the initiative and sends the message that organizational leaders, managers, and frontline staff support it.
- Evaluate and report on program impact: As part of the ongoing communication effort, data demonstrating the positive impact of the program should be widely disseminated and promoted. These data should be displayed on units to engage frontline staff in continuing to incorporate the new processes into their workflow.
- Train new staff: Integrate training on the program into new employee and medical resident orientation sessions and other relevant training programs.
- Standardize and automate care wherever possible: Standardizing and automating care through algorithms and order sets helps to ensure that care remains consistent despite staff turnover.
- Regularly review and refine program based on evidence: Maintaining strong performance requires regular monitoring of adherence to program elements and performance in sepsis care, particularly after new processes or elements are introduced. In addition, program staff should regularly review the published literature for new evidence-based findings or expert consensus statements on sepsis diagnosis and treatment, and then revise the program as appropriate.
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. [PubMed]
Whippy A, Skeath M, Crawford B, et al. Kaiser Permanente's performance improvement system, part 3: multisite improvements in care for patients with sepsis. Jt Comm J Qual Patient Saf. 2011 Nov;37(11):483-93. [PubMed]
Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008;36:296-327. [PubMed]
Nobre V, Sarasin FP, Pugin J. Prompt antibiotic administration and goal-directed hemodynamic support in patients with severe sepsis and septic shock. Curr Opin Crit Care. 2007;13(5):586-91. [PubMed]
Claessens Y, Dhainaut J. Diagnosis and treatment of severe sepsis. Crit Care. 2007;11(Suppl 5):S2. [PubMed]
|Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.|
Service Delivery Innovation Profile
Original publication: January 23, 2013.
Original publication indicates the date the profile was first posted to the Innovations Exchange.
Last updated: January 23, 2013.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.