This case study first appeared in the report Committed to Safety: Ten Case Studies on Reducing Harm to Patients by Douglas McCarthy and David Blumenthal, M.D.

Organization. The Johns Hopkins Hospital is a 900-bed academic medical center affiliated with the Johns Hopkins University School of Medicine and is one of three acute care hospitals in the Johns Hopkins Health System. Two of the hospital's intensive care units (ICUs) are discussed in this case study: a 14-bed, oncology surgical ICU (known as the Weinberg ICU or WICU), and a 15-bed surgical ICU (SICU) for general vascular surgery, trauma, and transplant patients. In both, patients are co-managed by intensivist-led multidisciplinary teams.

Objective and intervention. Intensive care physicians at Johns Hopkins developed the Comprehensive Unit-Based Safety Program (CUSP), a model for improving quality, safety, and communication. CUSP engages and empowers staff to identify and eliminate patient safety hazards by using the following eight steps:

• assess the unit's culture of safety;
• educate staff on the sciences of safety (e.g., anatomy of errors, systems thinking, interpersonal skills, blame vs. responsibility);
• identify safety concerns;
• meet regularly with a senior hospital executive who "adopts" the unit to provide support for removing system barriers and accountability for making safety improvements;
• prioritize improvements;
• implement improvements (teams adopt two or three simple, low-cost changes that can be made immediately and propose an additional two or three higher-cost changes that require hospital approval);
• share success stories and disseminate results; and
• reassess the unit's safety culture.

CUSP is part of a broader institutional commitment to improve patient safety at Johns Hopkins Hospital that has been informed, in part, by a partnership with the family of a pediatric patient, Josie King, who was a victim of medical error at the hospital. The King family donated funds and worked with Hopkins physicians to create a patient safety program in the hospital's Children's Center that has served as a model for improvement at the hospital and elsewhere. The CUSP model can be tested, adapted, and rolled-out sequentially among hospital units.

Date of implementation. CUSP was pilot tested in the Johns Hopkins Hospital WICU starting in July of 2001 and six months later (January 2002) in the SICU. Its design was influenced by participation in the Institute for Healthcare Improvement's Quantum Leaps in Patient Safety collaborative. Several other related safety improvement interventions were undertaken in these ICUs before and during CUSP, as described below.

Process of change. Unit improvement teams (physician, nurse, and administrator, plus other staff who wished to join) were given dedicated time each week to identify and champion safety improvement efforts. Interventions suggested by the safety assessment included creating a short-term patient goals form (Pronovost et al. 2003a), implementing a standardized process (known as medication reconciliation, see Case Study 10) for ensuring the accuracy of medication orders at ICU discharge (Pronovost et al. 2003b), and relabeling epidural catheters to prevent misidentification. The daily goals form was instituted after a survey found that nursing staff and residents frequently did not know the goals of therapy. The form is used as a checklist during physician intensivist–led rounds to identify tasks to be completed by the care team and to discover and mitigate safety risks.

A multidisciplinary team developed a related project to reduce catheter-related bloodstream infections in the ICU (Berenholtz et al. 2004a). The project included the following elements:

• Instituting a vascular access device (VAD) policy, which requires all providers to receive education on evidence-based infection control practices and successfully complete a post-test as a precondition to inserting catheters.
• Creating a catheter insertion cart—known as a "line cart"—with standardized supplies needed to meet infection control guidelines for sterile insertion of central lines. (Physicians previously had to find supplies located in eight different places, an unnecessary barrier to compliance.)
• Using a checklist to ensure adherence to evidence-based guidelines for safe catheter insertion: inserting a line only when needed, washing hands, using full barrier precautions, cleaning the insertion area with chlorohexadine, and avoiding the femoral site if possible.
• Empowering nurses to intervene if guidelines are violated, involving a negotiated change in teamwork behaviors on behalf of patient safety.
• Adding an item to the daily goals sheet that prompts the ICU team to ask the physician during daily patient rounds whether catheters can be removed (since central lines are sometimes left in the patient longer than clinically needed).

In a project designed to prevent hospital-acquired pneumonia by promoting evidence-based care of patients on mechanical ventilation, staff were surveyed to identify and overcome barriers to compliance. For example, after it was determined that four of five ICU nurses were not familiar with evidence for therapies to prevent complications, the staff members were educated about the therapies and the four care processes were added to the daily goals sheet as a checklist (Berenholtz et al. 2004b). (See Case Study 7, the VHA Transformation of the Intensive Care Unit National Collaborative, for additional discussion on preventing ventilator-associated pneumonia.)

Results. The following results were reported across several studies of complementary interventions that took place both before and during the time of CUSP implementation (Berenholtz et al. 2004a, 2004b; Pronovost et al. 2003a, 2003b, 2005). Staff perceptions of safety culture were measured using the Safety Climate Scale, a validated instrument adapted from the aviation industry (Sexton and Thomas 2003).

ICU staff ratings of a positive safety culture increased from 35 percent to 52 percent of nursing and physician staff following a six-month implementation of CUSP in the WICU. Safety climate scores did not change significantly among staff in the SICU, which served as a control group during this period. CUSP was then implemented in the SICU. Six months later, ratings of positive safety climate had increased to 68 percent of SICU nursing staff as compared to the baseline rate of 35 percent one year earlier (Figure 7).

• By "adopting" the ICU, senior executives' involvement led to approval of structural changes, including creating specialized patient transport teams and the presence of pharmacists in ICUs.
• Self-reported understanding of goals of care increased from 10 percent of residents and nurses at baseline to 95 percent after implementing the daily goals form.
• One year after implementing CUSP, average ICU length-of-stay decreased from two days to one day in the WICU and from three days to two days in the SICU. Medication errors in transfer orders were eliminated (from 94 percent before the intervention).
• The proportion of days on which patients received all four evidence-based therapies to prevent complications of ventilator care increased from 30 percent to 96 percent during a six-week intervention period, resulting in an estimated 27 fewer deaths, 754 fewer ICU bed-days, and $825,000 in savings annually.
• Observed catheter-related bloodstream infections were eliminated (from 11.3/1,000 catheter-days before the intervention), representing the prevention of an estimated 43 infections and eight deaths and yielding an estimated $2 million in savings annually. The multifaceted intervention involved the implementation of several initiatives over a five year period: the VAD policy, line cart, daily goals sheet, guidelines checklist, and nursing empowerment (Figure 8).
• Nursing turnover showed a nonsignificant trend toward reduction.

Lessons learned. Although it may not be possible to stop and learn from every defect in care, staff can be given dedicated time each week to identify and develop remedies to fix defects, said Dr. Pronovost. A simple tool—the daily goal sheet—transformed daily rounds to a more patient-centered approach that improved communication, the sense of partnership between nurses and physicians, and the quality of patient care. ICU nurses reported that the daily goal checklist was especially helpful since immediate patient care needs often make it difficult to listen during daily rounds. The tool was repeatedly revised to improve its usefulness during the intervention. However, another implement adopted from the Veteran's Administration—a prioritization tool—was rarely used and dropped in favor of informal assessments.

The Johns Hopkins team discovered that promoting teamwork and basic human factors approaches, such as simplification of processes, is key to increasing the use of evidence-based practices associated with improved patient outcomes. For example, complex guidelines can be converted into simple checklists that double as data collection tools, with each item capturing a critical process measure of quality. Educating nurses on evidence for recommended therapies helps to enlist their support for providing evidence-based care. "When you create a system that reliably delivers the processes or interventions that work, spectacular performance improvement follows," Dr. Pronovost said in an interview for the Joint Commission Journal on Quality and Safety (Berman 2004).

In teaching different institutions how to improve, Dr. Pronovost and his team have found it helpful to adapt the well-known Plan-Do-Study-Act (PDSA) model of process improvement to make it easier to communicate and promote change. These steps include: engaging through stories of harm, educating about evidence-based interventions, executing improvements using simple tools, and evaluating and sharing results to promote culture change.

Pronovost and Berenholtz (2004) conclude: "It seems that knowledge of performance does not actually translate into better care unless all of the stakeholder are committed, work together to redesign the processes of care and implement those new processes consistently."

Replication and related results. Within Johns Hopkins Hospital, CUSP is being used as a framework for patient safety improvement within 26 hospital units, involving over 140 personnel (Paine et al. 2004). Each of the hospital's executives has adopted a care unit and works collaboratively with the unit's interdisciplinary teams to overcome barriers to improvement.

CUSP and the interventions it has generated are being transferred to hundreds of ICUs nationwide through collaborative projects with state hospital associations in Michigan, New Jersey, Maryland, and other states. In Michigan, for example, 70 participating hospitals have succeeded in reducing the rate of catheter-related bloodstream infections by 50 percent, on average, across 127 ICUs. Of these, 68 ICUs have eliminated bloodstream infections or ventilator-associated pneumonias for six months or longer. The Johns Hopkins researchers estimate that these improvements have collectively saved 1,578 lives, 81,020 hospital days, and $165,534,736 in costs over a 15-month period, according to the Michigan Hospital Association (MHA 2005). Based on his experience, Dr. Pronovost said that state hospital associations provide a promising means of disseminating improvement methods throughout the country. An online version of CUSP, known as eCUSP, has been made available by the Patient Safety Group, a not-for-profit organization created by the Josie King Foundation and Johns Hopkins Medicine (Pronovost et al. 2006).

A "bundle" of evidence-based practices to prevent catheter-related bloodstream infections, such as those described above, is one of six high-impact interventions being promoted by the Institute for Healthcare Improvement as part of its 100,000 Lives Campaign (Berwick et al. 2006).

Implications. CUSP is successful because it provides a bridge between scientific validity and feasibility of adoption, said Dr. Pronovost. Many quality-improvement approaches emphasize feasibility but measurement is weak so validity cannot be assessed. Conversely, academic approaches that stress validity are often not easily replicated. CUSP is designed to provide sound principles and a flexible structure that rely on local wisdom to determine priorities for improvement, he said. Staff involvement in rapid process change and a sense of making a positive difference creates satisfaction and promotes culture transformation, creating demand for further adoption. Initiating change efforts within a single work unit and then replicating successful approaches in other units is a promising approach to building an organizational culture of safety.

back to top>

---

This study was based on publicly available information and self-reported data provided by the case study institution(s). The aim of Fund-sponsored case studies of this type is to identify institutions that have achieved results indicating high performance in a particular area, have undertaken innovations designed to reach higher performance, or exemplify attributes that can foster high performance. The studies are intended to enable other institutions to draw lessons from the studied organizations' experiences in ways that may aid their own efforts to become high performers. The Commonwealth Fund is not an accreditor of health care organizations or systems, and the inclusion of an institution in the Fund's case studies series is not an endorsement by the Fund for receipt of health care from the institution.