6 Steps for Implementing Successful Performance Improvement Initiatives in Healthcare
registries and a starter set of common metrics further reduced the time required to just two weeks since the patient population (cohorts) were already defined (ICD codes, APR DRGs, clinical data, etc.,), and the teams could easily compare data (admissions, readmissions, LOS, etc.,) across the different patient cohorts to help identify the greatest opportunities.
In addition to speeding the development of performance improvement programs, an analytics application can help an organization identify priorities for improvement efforts by uncovering variation. Variation points to a potential for standardizing processes, because the existence of variation inherently means that some care practices are more efficient and produce higher-quality outcomes than others, while there also is a greater likelihood that some practices are not achieving optimum outcomes. Hospitals and health systems will have a significant opportunity for care improvement if they can identify their highest-performing practices and begin to make those practices and evidence-based practices the standards for all caregivers.
The Anatomy of Healthcare Delivery framework, shown in Figure 2, and developed by David A. Burton, MD demonstrates the potential pathways patients can go through in their interactions with the delivery system. It is a conceptual framework that enables one to organize their thinking about the care delivery process and to focus their attention on key processes and decision-making points. The degree to which an organization standardizes their approach in each of the knowledge asset categories (indicated by the orange and blue boxes shown in the diagram) will impact the degree of variation in care delivery.
Once an organization examines how patients flow through the care delivery system and its critical decision points, they can use the information to create a logical framework to organize a Clinical Integration hierarchy, as illustrated in Figure 3. The Clinical Integration hierarchy organizes clinical programs based on physician specialists and other clinicians who share management of care processes and who are responsible for the ordering of care for patients —versus traditional service lines that are mostly used for marketing purposes. The teams either work on things together or one team’s output is another team’s input (e.g., OB-GYN sub-specialists and neonatologists).
With clinical programs and clinical support services broken into categories that align with the way care is delivered, an organization can use a Pareto approach (also known as the 80/20 rule), to identify their highest opportunities: the clinical programs with the highest count, highest cost or those that have the highest variation. One can review the ranking to see which key clinical care processes make up the majority of the care provided.
Variation in cost can be a good surrogate for quality of care, because higher cost may result from delivery of inefficient or unnecessary services. As the prescribers of care, clinicians are one of the greatest influencers in managing variable cost, which represents direct cost in departments. By focusing on variable cost — looking at the volume of procedures and cost per procedure, in particular — they can identify avoidable cost and begin working with clinicians, using evidence-based practices, to address them.
The Health Catalyst Key Process Analysis application is based on the Pareto principle, and is used to prioritize performance improvement efforts. Cost is displayed on the x-axis, as shown in Figure 4; the y-axis shows the variation in resources consumed. The clinical programs with the highest cost and highest variation are in box one. Septicemia is one care process that shows both high cost and high variation.
Data governance is also a key component of the analytic strategy. A data governance committee should be responsible for understanding and implementing local data standards (facility codes, department codes, etc.); as well as regional and industry standards (CPT, ICD, SNOMED, LOINC, etc.). In addition to coded data standards, the committee is also involved in the standard use of algorithms to bind data into analytic algorithms that should be consistently used throughout the organization, such as calculating length of stay, defining readmission criteria, defining patient cohorts, and attributing patients to providers in accountable care arrangements.
Step 3: Prioritize programs using a combination of analytics and an adoption system
Successfully improving clinical outcomes and streamlining operations requires a strong organizational commitment and changes in culture, organizational structure, staff education, and workflow processes, what Health Catalyst calls an adoption system. Consequently, any organization that embarks on this performance improvement journey should first assess its readiness for change. Examples of criteria that are evaluated in an organizational readiness assessment include clinical leadership readiness, data availability, shared vision, and administrative support (e.g., data manager, outcomes analyst availability).
A readiness assessment helps the organization determine how ready the teams are to accept change, to estimate what, if any, impact there is on staffing, and the potential impact on front-line caregivers. Understanding the strategic objectives and integrating results from a readiness assessment, along with the analytics, help the organization prioritize which care families (clinical services) to begin with.
Step 4: Define the Performance Improvement Program’s Permanent Teams
The organization will require permanent performance improvement teams to review and analyze data, define evidence-based and best practices, and monitor ongoing result. Improvement teams should be created to coincide with an organization’s internal structure. One way to organize teams is described below and shown in Figure 5.
Guidance team. A guidance team should be assigned accountability for clinical quality across the continuum of care in a specific domain (such as Women and Children). The primary role of such a team should be to select goals, prioritize work, allocate resources, and remove barriers. The team should then delegate accountability to clinical improvement teams to improve care.
Clinical improvement teams. These teams typically are led by a physician and nurse and consist of