The Bon Secours Health System Description

The Bon Secours Health System set an enterprise wide, (13 hospitals), 3 year (2007-2009) Strategic Quality Plan goal of Zero Preventable Deaths that expanded the Institute of Healthcare Improvement’s mortality review processes by including the following:

• Incorporation of Center to Advance Palliative Care Screening Tool and Palliative Care Bundle
• Increased palliative care service in the ICU
• Increased focus on advanced directive determination and bundle compliance
• Increased focus on sepsis identification, intervention, and prevention through the ED and acquired after surgery

Goal

To promote system-wide changes to improve patient outcome and process measures in Mortality, Palliative Care and Sepsis by the end of FY 2009 through:

Measures

Mortality

• Reduction in the hospital-specific mortality rate to <75%, or
• 25% reduction in raw mortality for all facilities already below 75%

Palliative Care

• Increase in Palliative Care consults in the ICU from 0 to 5%
• Increase in obtaining DNR orders from within 1 to 2 days prior to death to 5 days prior to death
• Increased completion of Palliative Care Bundle from 0% to 50%

Sepsis

• Reduction in the Sepsis Mortality Rate of patients admitted through the Emergency Department from 13.35% (FY 2007) to 11.11%
• Reduction in Hospital-Acquired Sepsis in Post-Surgical Patients from 5.42% (FY 2007) to 5.02%

Data Collected: For all 13 hospitals

Mortality

• Hospital Mortality Rate and their Raw Mortality

Palliative Care

• Percent of Palliative Care Consults in the ICU
• Number of days from DNR to day of death
• Percent of Palliative Care Bundle compliance

Sepsis

• Sepsis Mortality Rate of patients admitted through ED
• Hospital Acquired Sepsis – Post surgical patients

Summary of Results / What Worked and Why

Mortality

• 2.37 lives saved per month since 2006

Palliative Care

• Increased Palliative Care consults in the ICU from 0 to 7.52%
• Increased acquisition of DNR orders from within 1 to 2 days prior to death to 10 days prior to death
• Increased completion of Palliative Care Bundle from 0 to 90%

Sepsis

• 13.38% reduction in sepsis mortality for sepsis patients admitted through the Emergency Department
• Incidence of hospital-acquired post-surgical sepsis decreased to 3.5%

What Worked and Why

The reduction of mortality and preventable harm has been achieved through the strong collaborative teamwork, commitment, and leadership at both the system and local level. At the same time, palliative care access and palliative care bundle compliance have increased, advance directive planning has occurred earlier, and sepsis has received substantial attention as a key area of preventing mortality. Significant changes and improvements resulted from re-defining, expanding, and improving on review processes, tools, data collection, reporting, analysis, and improvement strategies. Below are specific steps that improved outcomes.

• Established Bon Secours definition of “Zero Preventable Deaths”, ZPD, by using 3 categories addressing death: anticipated and unanticipated deaths with no care failures and unanticipated deaths (potentially preventable), these categories are defined below.
• Anticipated: Natural consequence of advanced chronic or end of life disease.

Goal

Early palliative care service intervention preventing unnecessary suffering through end of life.

• Unanticipated Deaths but no care failures: No variation in care with choice to allow natural death. Goal: deliver evidence-based medical care supporting life or a dignified death according to patient’s wishes.• Unanticipated Deaths with improvement opportunities: Result of a known / unknown variation in care, and evidence-based medicine. Goal: Increase identification and interventions that save lives.
• Ensured committed leadership – Executive team commitment was needed to facilitate change and remove barriers to process changes. To achieve buy-in and engagement, different change/improvement strategies needed to be developed and implemented at the corporate, local administration, and clinician levels.
• Made mortality intervention multi-dimensional – This multi-dimensional approach decreased mortality rates by focusing on strategies that prevent patient deaths that were potentially preventable or “unanticipated.” At the same time, patients who were at the natural end-of-life stage and not preventable deaths were deemed “anticipated deaths” and were supported through palliative care service.
• Broke down the ICU silo – The multi-dimensional mortality metrics identified non-preventable ICU deaths that established a corporate benchmark metric requiring a percentage of palliative care referrals in the ICU. Palliative care increased education and rounding as part of the interdisciplinary ICU team, the palliative care screening tool was incorporated in the admission process, and the palliative care bundle was initiated for patients with an ICU length of stay over five days.
• Involved multi-disciplinary teams – Involving physicians, quality assurance staff, palliative care staff, and the care delivery team at the bedside for monthly mortality reviews was important to understand process breakdowns and to develop appreciation for processes by discipline.
• Improved education and training processes – Developing written tools and definitions were critical at the beginning stages of the project but enhanced with the Corporate Mortality Leadership Team traveling to hospitals and reviewing cases with the teams to validate understanding of processes and increasing knowledge transfer opportunities. Quarterly system-wide “Lunch-and-Learn” teleconferences were another mechanism to share and increase knowledge and validate standardized processes.
• Improved education on sepsis – Education was provided for nurses and physicians on the signs and symptoms of sepsis focusing on early recognition, diagnosis, use and implementation of evidence-based sepsis order sets and rapid response teams. Routine monthly system wide “Lunch and Leans” improved education, allowed for a routine review of data and outcomes, and fostered the development of action items supporting quality improvement processes.

Lessons Learned

• Sustaining processes – It is essential to plan ahead to ensure processes are sustained during staff turnover, times of limited resources, and lack of physician availability.
• Standardization of hospice patient types – Although during the Mortality Review process all hospice patients were removed from the review based on chart details, the lack of a specific hospice patient type created problems with data integrity.
• Standardization of the palliative care order entry process – It is important to have a standard palliative care order entry process that has a palliative care patient type linked to hospital information systems to increase the accuracy and efficiency of data abstraction and outcome data reporting.