Despite these struggles, in 2004 Denver Health’s leadership was inspired to begin a quality improvement journey in part because of these substantial challenges that it faces as a safety-net institution. A number of foundational elements were already in place, including a vertically integrated health care system. Denver Health believes that this integrated system is key for achieving high levels of quality and safety because it provides people with geographically convenient access to care, seamless continuity of care across a person’s life and health care needs, and the right care, at the right time, with the right provider.

Another foundational element is that the system is staffed by almost 300 employed and salaried physicians, all of whom have academic appointments at the University of Colorado School of Medicine. This employed-physicians model promotes the alignment of goals across the enterprise and helps implement quality and safety interventions, perhaps more so than a hospital whose physicians have medical staff privileges but who are not employed by the hospital system. Nationally, there is a definite trend towards more employed physician models of hospital staffing.

In addition, the delivery of safe, high-quality, and efficient health care depends on the provider’s having comprehensive patient care information at the point of care. Denver Health is an advanced user of health information technology, which is becoming even more widespread at Denver Health, in response to incentives in the Health Information Technology for Economic and Clinical Health (HITECH) provisions in the American Recovery and Reinvestment Act of 2009. These foundations of an integrated system, employed academic physicians, and health information technology, provided a springboard for Denver Health’s structured approach to health care quality and patient safety.

An important step in Denver Health’s approach to creating high-quality care and patient safety was to identify a responsible person and department to lead this effort. Although decentralizing and integrating these safety and quality strategies into every clinical department is important, Denver Health saw a need for a centralized and distinct department of patient safety and care quality to facilitate the application of a broad array of changes in process, organization, and teamwork. An associate medical director position was created, with the responsibility of developing safety goals and an agenda for leading the department. This arrangement drew on the quality improvement literature, which demonstrates the association between developing broad and shared improvement goals and achieving substantial quality improvement, through the provision of administrative support to mine data fields for quality improvement purposes, having strong physician leadership and using credible and timely data feedback [11].

The inclusion of the director of medical education within Denver Health’s Department of Patient Safety and Quality reflects the criticality of oversight of medical education in effectuating improvement in health care quality. Physician-trainees are at the hub of many care delivery systems, especially in safety net hospitals and academic medical centers. Housestaff education is therefore an important ongoing topic which is inextricably related to the provision of high quality and safe care. Thus, medical trainees must work in concert with evidence-based quality initiatives. This has been facilitated at Denver Health by the components of team rounding, increased emphasis on attending oversight, checklists and computerized physician order entry (CPOE) with standard order sets. Moreover, the Accreditation Council for Graduate Medical Education (ACGME) now mandates housestaff work hour restrictions and therefore an increased level of attending supervision.

The inclusion of infection control in Denver Health’s Department of Patient Safety and Quality reflects a growing recognition of the severity of hospital acquired infections. An infectious disease physician with epidemiology training was appointed to head infection control and was supported by qualified infection control nurses. This structure facilitated the implementation of interventions in high risk areas discussed below. Previously, infection control was in the Department of Medicine. Of note, in this regard, sterile processing was just recently moved from the nursing department into the Department of Patient Safety and Quality, to further maximize this national synergy.

The third element in Denver Health’s approach to creating quality and safety was programs to manage high risk and/or high opportunity areas. This reflects the notion that safety is not only freedom from injury or damage but also the freedom from the risk of injury or damage. Some of the high risk/ high opportunity areas chosen were those identified from the literature which were of clinical relevance at Denver Health (Table 18.1). Each is discussed below.

“Failure to rescue” refers to failure to identify patients who are deteriorating and to intervene in a timely manner to prevent their deterioration. In a recent study of postoperative morality, “failure to rescue,” rather than the number of complications, was the key variable in explaining differences in mortality rates across hospital [12]. Thus, we opted to institute a rapid response system (RRS) to identify and intervene for such patients and intervene in their care. Given that the literature only showed modest evidence of success for the commonly accepted rapid response team approach, Denver Health opted for a variation therein. We defined our own “clinical triggers,” such as a systolic blood pressure less than 90 mmHg, which would trigger activation of RRS. However, our system did not involve a separate team of responders. Instead, it utilized the patient’s intern and resident teams who were called by the patient’s nurse, in response to the presence of a clinical trigger. Then in a structured and ordered sequence team members were expected to evaluate the patient at his or her bedside within 10 min of the nurse’s call. Using this new rapid response system, Denver Health reduced its cardiopulmonary arrest rate from a median of 5.9 per 1,000 discharges to 2.2 per 1,000 discharges (p < 0.001). The number of patient who required transfer back to the intensive care unit within 48 h after being moved to hospital floor units also decreased significantly, from 4.62 to 3.27 per 100 intensive care unit transfers (p = 0.03) [13].

Furthermore, we initiated hospitalist co-management or consultation for all patients on the orthopedic service, patients on low-volume inpatient surgical specialty services, such as oral maxillofacial, bariatric surgery and urology, as well as patients on the psychiatric ward with medical comorbidities. This facilitated the care of their medical problems, such as diabetes or cardiac disease, by providers whose expertise was in these areas.

Another Denver Health quality and safety initiative was related to infectious disease care. Antibiotic use is considered one of the most important aspects of infection control and their overuse and underuse have both been deemed by the Joint Commission to be a significant barrier to quality improvement. Almost 60 % of Denver Health’s inpatients were being treated with an antibiotic during their hospital stay. Therefore, a formal and robust antibiotic stewardship program was established to provide careful oversight and guidance to our clinical services who were using antibiotics. The approach spawned new programs, including mandatory infectious disease consultation for certain common and serious infections; concurrent and timely feedback to a prescribing team when multiple antibiotics were used for the same patient; new rules-driven guidelines embedded within our computerized physician order entry system for common inpatient infections such as pneumonia and cellulitis; and formal weekly infectious disease consultant rounds with intensive care unit teams.

As a result, Denver Health’s antibacterial drug use, in days of therapy per 1,000 patient days, was the lowest of 35 US academic health centers reporting through the University HealthSystem Consortium [14]. Moreover, proper treatment has increased and adverse consequences from illness have decreased for the highly prevalent Staphylococcus aureus bacteremia [15].

Another high risk hospital acquired condition is venous thromboembolism (VTE), blood clots occurring after surgery. These blood clots are the most common preventable cause of hospital deaths, and their prevention results in a cost avoidance of $25,000–$40,000 for each VTE prevented. A week long Lean ‘rapid improvement event’ (RIE) focused on the proper and cost-efficient utilization of prophylactic anticoagulation in high risk inpatients. Low molecular weight heparin (LMWH), a blood thinning medication used to prevent this complication, had become the most costly line item in the hospital pharmacy’s budget. Yet, Denver Health incidence of postoperative VTE was significantly worse than national benchmarks. The RIE produced an evidence-based risk assessment tool and clinical practice guideline which were embedded into CPOE admission order sets. Thereafter, compliance with the guideline began approaching 100 %, overall utilization of LMWH decreased more than 60 % and Denver Health’s occurrence of VTE achieved a UHC ranking for VTE in the top 10 % of outcomes [16]. Indeed many of the aforementioned patent safety strategies, which have been instituted at Denver Health, are amongst the recently released group of strategies which have been endorsed for immediate application by the AHRQ and an international panel of stakeholders [1], intended to promote optimal patient safety in patient care settings.

The aforementioned interventions have all focused on hospitalized patients. Improving ambulatory care poses very unique challenges [17]. Despite the fact that there are 900 million outpatient visits versus 35 million hospital discharges [18], there has been less effort directed towards outpatient quality improvement. However, with the growing focus on medical homes and health reform’s emphasis on accountable care organization, it is crucial that high quality safe care is also delivered to outpatient populations which heretofore had been relatively neglected [19]. Indeed, the very nature of ambulatory care safety lapses is different than those in the hospital practice both in the nature of the errors, i.e, diagnostic versus treatment errors and in the nature of the patient provider relationship. Yet, ambulatory patient safety was also a priority at Denver Health because the initiatives in this area have been encouraged by Denver Health’s integrated delivery system and HIT system, along with a robust data warehouse and dynamic patient registries. For example, Denver Health has a very mature immunization registry enabling an 88 % immunization rate in the 1 year olds served by our system. Denver Health was awarded the prestigious Codman Award by the Joint Commission for this effort. There are similar registries for asthma, trauma, cancer screening, hypertension, diabetes, anticoagulation and an obstetric care, with the newest one being in the area of chronic narcotic usage. These registries effectuate improved quality by providing aggregated point of care performance data by clinic site and by clinician to avail data to audit and provide timely, concurrent and specific feedback. The cancer registries’ patient-specific data serve as a visual prompt to the physician, during a patient encounter, to bring current breast, cervical and rectal cancer screening at the appropriate time and generate automatic notifications of patients if they miss scheduled testing. Thus, these registries are also tools for the proactive management and outreach to patients between visits to improve clinical indicators which may be suboptimal such as reminders about cancer screening. As an example, almost 75 % of Denver Health patients with hypertension have their blood pressure controlled compared to 50 % of Americans, and Denver Health record of hypertension control has exceeded the national rate for many years running.