Week 4

1114237 - Jones & Bartlett Learning ©

C

© ALMAGAMI/Shutterstock

CHAPTER 4

Understanding Variation, Tools, and Data Sources for CQI in Health Care William A. Sollecito and David Hardison

Information is random and miscellaneous, but knowledge is orderly and cumulative.

—Daniel Boorstin

Continuous quality improvement (CQI) requires knowledge about the behavior of systems and

1114237 - Jones & Bartlett Learning ©

processes, which is most often obtained from data and other sources of information collected for that specific purpose and through the application of tools and techniques that have proven to be effective throughout the evolution of CQI as described earlier in this text. The use of data and tools is part of the concept of statistical thinking, which is a key component of both the philosophy and processes of CQI. Central to decisions about innovations and especially CQI in health care is the need to improve our knowledge about health care process performance through an understanding of the context in which processes behave, asking the right questions, and then the collecting and analyzing appropriate data. The goal is not simply to analyze data, but to gain knowledge by utilizing data to understand processes, sources of variation, and the impact of improvements. Most important in CQI applications is the requirement to understand variability and the predictability of health care processes. As described in Chapters 1 and 2, CQI requires a systems approach that analyzes assumptions and focuses on customers/ patients, processes, and interdependencies based on knowledge. This approach is most efficiently accomplished using stepwise, yet cyclical, procedures that span planning, implementation, and an assessment of the impact of the decisions made in identifying and applying changes—including how to determine when a change is needed and whether it leads to an improvement (Langley et al., 2009). Although many different methods can be applied to carry out these steps, CQI efforts in the past have spawned methods that have proven to be robust and efficient when focused

1114237 - Jones & Bartlett Learning ©

on listening to the “voice of the process” underpinned with an understanding of variation (Wheeler & Chambers, 2010). These methods include a continuous assessment of process performance, including an understanding of sources of variation, followed by decision making based on the application of knowledge about systems and measurements. Over the years, a set of tools have been developed to carry out these steps in the most efficient manner; as with CQI itself, these tools have undergone a continuous improvement process that enhances our learning about how to apply them, and when to introduce new tools. Especially in health care, which itself has undergone an evolutionary process rooted in new knowledge and the greater use of technology, new tools and data sources have been developed to ensure that the “Quadruple Aim” of health care—improving the experience of care, improving the health of populations, and reducing the per capita costs of health care while improving the work life of clinicians and health care staff (discussed in Chapter 2)—has been met (Bodenheimer & Sinsky, 2014; Sikka, Morath, & Leape, 2015). At the heart of these procedures is the use of evidence-based approaches that rely on health information technology that provides high-quality data to ensure that correct decisions are made in the efficient application of the “data-to-decision cycle” (McLaughlin & Kibbe, 2013). The greater use of technology to collect and analyze health care data presents greater opportunities for both converting data into information or knowledge and applying powerful methodologies for implementing improvement, but it also presents

1114237 - Jones & Bartlett Learning ©

challenges to ensure the most appropriate application of tools and techniques to carry out CQI initiatives.

Many valuable texts are available that explain the mechanics of CQI analysis methods and tools (e.g., Balestracci, 2009; Bialek, Duffy, & Moran, 2009; Carey & Lloyd, 2001; Langley et al., 2009; Lighter & Fair, 2004; Lloyd, 2019; Streibel, Sholtes, & Joiner, 2003). This chapter will not duplicate the information provided in those texts. Rather, the purpose of this chapter is to assist in understanding the role of variation in quality improvement, why measurement and statistical thinking are vital to quality improvement efforts, and to illustrate a few fundamental CQI tools that are particularly useful in health care to enhance our learning about process changes and facilitate improvement in health care processes. A minimum, but sufficient amount of detail, with relevant examples, will be provided to enhance understanding and enable the reader to develop and apply these tools. Finally, the chapter presents a brief overview of the strengths and weaknesses of various forms of health information data sources that are currently being used to carry out CQI initiatives.

1114237 - Jones & Bartlett Learning ©

▶ Health Care Systems and Processes

The early work of Donabedian is foundational in understanding health care systems and processes (Donabedian, 1980). He described the process of care as a system involving three components: structure, process and outcomes that comprise a set of activities requiring interaction between patients and providers. More recently, there has been greater attention paid to patient centered care and the Donabedian model has been expanded to include greater emphasis on the patient experience (Spath & Kelly, 2017). This is reflected to some degree in the greater emphasis on patient satisfaction data and its role in measuring quality of care; this source of data and information has strengths and weaknesses in CQI, which will be discussed in greater detail later in this chapter. The main emphasis of this section is the central role of concepts such as the Donabedian Model in understanding health care processes and outcomes.

1114237 - Jones & Bartlett Learning ©

Process Capability To understand the expected output of a process relative to a particular outcome or quality characteristic, or process capability, the process must be stable, and, hence, predictable within limits. Benefits of a stable process (Deming, 1986) include:

■   The process has an identity (capability); it is predictable. Therefore, there is a rational basis for planning.

■   Costs and quality are predictable. ■   Productivity is at a maximum and costs at a

minimum under the present system. ■   The effect of changes in the process can be

assessed with greater speed and reliability. In an unstable process, it is difficult to separate changes to the process from special causes of variation. Therefore, it is more difficult to know when a change results in improvement.

1114237 - Jones & Bartlett Learning ©

Interpreting Process Requirements and Performance Variation exists in every process and always will. It is the manager’s job to understand variation and continuously improve the processes so that they manage to meet the needs of the customers who depend upon their processes. The customer of each process has requirements which specify the level of performance that is needed. A provider working in the emergency department (ED) would require more rapid turnaround time (TAT) of laboratory results than a provider working in a primary care office. A 30-minute TAT in the ED may be unacceptable, whereas a 24-hour TAT may be acceptable for a primary care office. There may also be specific requirements for analyzing the laboratory specimen. Technology used for one type of test may require 30 minutes to be processed, while a different type of test can be completed within seconds. Customer and technical requirements must both be taken into account in order to interpret whether the performance of the process is acceptable or whether the process needs to be improved. That said, every process can be improved in the spirit of CQI.

Process requirements may be thought of as the criteria from which the effectiveness of a process may be evaluated from the process customer’s perspective. They function as both inputs to designing a process and outputs from executing a process. It is essential to first identify the customers of a process. Chapter 7 provides an in-depth discussion of customers of health care organizations (i.e., the role of patients in CQI). Brief

1114237 - Jones & Bartlett Learning ©

definitions, using a systems perspective, are provided here. A customer is defined as anyone who has expectations regarding a process operation or outputs. In health care delivery organizations, the primary customer is the patient, while the community may be the primary customer for a public health agency. Internal customers are those within the organizations and are sometimes thought of as those departments or coworkers “downstream” from the process. In other words, who the customers of a process are depends on the outputs of a process. For example, the recovery room or postanesthesia care unit may be thought of as the customer of the operating room. Patient care units may be thought of as customers of diagnostic departments (e.g., laboratory, radiology). Payers may be considered as external customers—those outside the provider organization. A stakeholder is anyone with an interest in or affected by the work you do. Regulatory bodies such as The Joint Commission (TJC), formerly known as the Joint Commission on Accreditation of Healthcare Organizations, or the National Commission on Quality Assurance (NCQA) would be considered stakeholders for hospitals and health insurance companies, respectively. Professional societies that define practice standards may also be thought of as stakeholders. The market refers to the environment in which you operate and do business and may include socioeconomic, demographic, geographic, and competitive considerations.

Once customers, stakeholders, and markets have been identified, it is essential to identify and understand what

1114237 - Jones & Bartlett Learning ©

they require of your services. For example, patients may require access and competent, courteous providers; payers may require a certain level of clinical results delivered in a cost-effective manner; regulatory bodies require compliance; and markets may require a culturally diverse approach to delivering services.

These requirements are vital to determining what services are needed and how the processes comprising the services are designed and improved. These requirements also provide the basis for selecting variables or attributes that will measure the process performance from the perspective of the customer.

TABLE 4.1 illustrates how one health-services organization operationalizes the links between customer requirements, process design, measurement, and goals.

TABLE 4.1 Links Between Customer Requirements, Process Design, and Measurement

TABLE 4-2 illustrates the core processes for each phase of the continuum of inpatient care. The patients’

1114237 - Jones & Bartlett Learning ©

interface with this organization follows the following path:

Admission → Assessment → Care Delivery/Treatment → Discharge

1114237 - Jones & Bartlett Learning ©

TABLE 4.2 Links Between Process Stages, Requirements, and Measures

Process Key Requirements Key Measures

Admit Admitting– registration

Timeliness ■   Time to admit

patients to the setting of care

■   Timeliness in admitting– registration rate on patient satisfaction survey questions

Assess Patient assessment

Timeliness ■   Percentage of

histories and physicals charted within 24 hours and/or prior to surgery

■   Pain assessed at appropriate intervals per hospital policy

Clinical laboratory and radiology services

Accuracy and timeliness

■   Quality control results–repeat rates

■   Turnaround time

■   Response rate on medical staff satisfaction survey

Care delivery–

Nurse responsiveness,

■   Response rate on

1114237 - Jones & Bartlett Learning ©

treatment Provision of clinical care

pain management, successful clinical outcomes

patient satisfaction and medical staff survey questions

■   Wait time for pain medications

■   Percentage of CHF patients received medication instruction–weighing

■   Percentage of ischemic heart patients discharged on proven therapies

■   Unplanned readmissions–return to ER or operating room mortality

Pharmacy– medication use

Accuracy ■   Use of dangerous

abbreviations in medication orders

■   Medication error rate of adverse drug events resulting from medication errors

Surgical services– anesthesia

Professional skill, competences, communication

■   Clear documentation of informed surgical and anesthesia consent

1114237 - Jones & Bartlett Learning ©

■   Peri-operative mortality

■   Surgical site infection rates

Discharge Case management

Appropriate utilization

■   Average length of stay (ALOS)

■   Payment denials

■   Unplanned re-admits

Discharge from setting of care

Assistance and clear directions

■   Discharge instructions documented and provided to patient

■   Response rate on patient satisfaction survey

The core process(es) for each phase of care are shown in the first column. The second column lists the key requirements for the process, derived from a variety of methods targeted toward understanding requirements of patients, internal customers, stakeholders, and the market in which the organization operates. The third column lists the attributes or variables that the organization measures to understand the degree to which its processes are meeting stakeholder requirements.

It is management’s job to ensure that the process requirements (also referred to as the Voice of the Customer) and the process performance (also referred

1114237 - Jones & Bartlett Learning ©

to as the Voice of the Process) are aligned (Carey & Lloyd, 2001; Wheeler, 2000; Wheeler & Chambers, 2010). If the two are not in alignment, then the process must be improved. The analysis of process performance must be based upon knowledge of the sources of variation in the process (i.e., common and special causes). This process analysis which will be described in detail later in this chapter, is accomplished by collecting data on process outcomes, plotting that data over time, constructing process behavior (also known as control) charts, and interpreting them based upon an understanding of variation. This is essentially converting those data into information and knowledge (Ross, 2014). Other considerations, to be discussed later in this chapter, include choosing the optimal sources of data with consideration of data quality and choosing appropriate tools and techniques that are useful for carrying out these analyses.

Each of these methods has a common approach to be successful in analyzing process performance; that is the need for careful measurement and an understanding of sources of variation, which will be covered in the next sections.