In industry and manufacturing, there has been a long-standing focus on improving quality as it relates to production of products or services for sale in the marketplace. A number of well-recognized quality programs and approaches have been developed in manufacturing to meet this need, including Total Quality Management (TQM) , ISO 9000, Six Sigma, and the Toyota Production System (TPS) upon which Kaizen and Lean management approaches have been based. Although these quality improvement programs were developed in industry, much of the theory, knowledge, and practices developed under these programs have been applied to health systems generally and to diagnostic facilities in particular [9].

Comprehensive quality programs have traditionally had components grouped under the terms quality control and quality assurance. Quality control (QC) is product or production focused and refers to the activities that are performed to fill the requirements for quality. QC entails failure prevention systems that ensure that defined standards are followed through the process. Quality assurance (QA) is process focused and refers to defined, planned activities to set quality requirements and ensure that they are met. QA defines the standards for each aspect of production and the policies and procedures to meet customer requirements. QA and QC are complementary, overlapping, and interdependent. In many settings QA and QC are combined into a single structure and do not function as separate processes. Regardless, both of these functions are static, in the sense that they are designed to maintain standards of production. Today, quality programs must be characterized by a more comprehensive and dynamic approach referred to as continuous quality improvement (CQI) . CQI encompasses much of traditional QA/QC and is directed to constantly review, refine, and improve processes and product, rather than to simply maintain a given level of quality (see below). In this context, Sollecito and Johnson define CQI in healthcare as: “a structured organizational process for involving personnel in planning and executing a continuous flow of improvements to provide quality health care that meets or exceeds expectations” [10]. These authors point out that TQM/CQI activities form a continuum “manufacturing at one end of the continuum and professional services at the other.” Some aspects of performance in the vascular lab resemble manufacturing with strict standards of performance and repetitive steps leading to a reliable diagnostic assessment; others have a more professional character. A comprehensive quality program must incorporate features that address each of these complementary spheres of quality management (QC, QA, and CQI).

An effective CQI program requires structure which should address the relevant metrics, the processes for evaluation, the tools for data collection and analysis, the performance thresholds, the selection of specific quality initiatives, the methods of communication, and the parties responsible for managing and directing these activities. This structure should be developed collaboratively among all staff and should be documented formally in a CQI policy.

W. Edwards Deming, a seminal figure in quality improvement systems in industry and a major contributor to the development of TQM said: “if you can’t describe what you’re doing as a process, you don’t know what you are doing.” Thus, one of the key steps in achieving quality in the diagnostic medical facility is to look closely at the actual activities of the laboratory and break down lab processes into sequential steps each of which can be defined, clarified, measured, and potentially improved. Some of these steps in lab processes must be monitored regularly to meet basic QC/QA; others may become the focus of quality improvement initiatives. The process of quality improvement begins with a review of the current state through data and setting a goal to be accomplished. The lab can then identify changes that can lead to improvement and the metrics that can identify a change in performance.

TQM/CQI programs then involve a cycle of steps known as the Shewhart cycle often abbreviated as: Plan, Do, Study, and Act, or the PDSA cycle [11]. The steps are (1) Plan, identify the objective and develop a plan to reach the objective; (2) Do, execute the process change as a test; (3) Study, (or check) evaluate the data assessing the impact of the process change; and (4) Act, implement the process change into routine practice or further modify the plan and repeat the cycle. There are several variants of the PDSA (Fig. 3.2). One commonly used in healthcare is FOCUS-PDCA and was developed by the Hospital Corporation of America (HCA) [12]. The FOCUS piece precedes the PDCA cycle with a logical five-part lead-in: (1) Find a process to improve, (2) Organize an improvement team, (3) Clarify current knowledge on the process, (4) Understand the source of the process variation, and (5) Select the process improvement. Many view the FOCUS component of the FOCUS-PDCA program as clear delineation of the requisite elements in the Plan phase of the standard Shewhart cycle. These kinds of tools provide formal mechanisms to address performance shortcomings in an organized way that can support continued process improvement. Once the goals of a specific initiative have been reached, the thresholds can be set more stringently, or other areas for improvement can be identified.

In considering lab functioning as a process and identifying metrics for evaluation and potential improvement, it is useful to employ Donabedian’s framework to categorize elements of lab performance: structure, process, and outcome [7]. Structure metrics include the resources available to the lab including the physical plant, the diagnostic and support instrumentation, and the capacity and competencies of the staff (education, training, certification, licensure). Lab structure also includes the comprehensive lab policies and procedures related to safety (general, equipment maintenance, infection control), human resources (personnel management), as well as the detailed technical protocols for each type of exam (including equipment, preparation, positioning, techniques, minimal data requirements, proper documentation, and annotation) and appropriate validated diagnostic criteria to be employed by the interpreting physicians. In the vascular lab, process metrics include compliance with the policies, diagnostic protocols, and exam procedures formulated through the lab’s structure but extend beyond to interpersonal interactions, availability, and timeliness of services (access and scheduling). The accuracy and reproducibility of lab output are ultimate results of effective lab processes and are essential overall metrics of lab performance. The outcomes of a diagnostic unit occur primarily through contributions to caregiver decision-making and the relative efficiency (quality/cost) of this process. To contribute to patient outcomes, the lab output must be appropriate to the clinical concern (appropriateness of use), accurate, and communicated to the clinician meaningfully in a relevant time frame. Patient-centered outcomes include courtesy, respect, and timeliness, all of which contribute to patient satisfaction. The nature of the particular vascular lab and its role in the health system determine the relative weights and emphasis placed on the three domains of unit performance.

In a diagnostic facility, there is an overriding and appropriate emphasis on process metrics evaluating the accuracy (efficiency) and clinical utility (effectiveness) of the lab’s output, and a more in-depth consideration of metrics to measure these aspects of lab performance is useful here. Efficiency is measured by assessments of precision, bias, sensitivity, and reproducibility of the evaluations performed by the facility. These terms have different meanings depending on the types of testing that are performed in a laboratory setting. Certain tests will produce numerical results and others more qualitative results that are classified into categories or bins. Accuracy describes the ability of the measure to approximate the true value. Bias describes the deviation of the result from the true value. Precision refers to the variability of repeated measures.

In the vascular diagnostic laboratory, accuracy and bias are the most commonly assessed by comparison of the findings of the vascular noninvasive test to an alternative and perhaps more established method of assessing the condition under study (a “gold standard”). Naturally, this approach requires the existence of a true gold standard testing modality that produces data comparable to the vascular exam and a clear understanding of the accuracy of the “gold standard” in detecting the true value for the condition under study. This comparative approach becomes more difficult when the tests employed in these comparisons produce somewhat different information (e.g., anatomic vs. physiological). Some free standing or independent vascular labs do not have ready access to comparator examinations that may be performed in other facilities, and evaluation of lab performance by this method may not be possible. Alternative comparators, such as blinded repeat examination by other lab personnel (sonographer and interpreting physician), surgical specimens, or long-term clinical outcomes may be employed, but they have several clear limitations. Finally, there are vexing issues in the situation where the test to be evaluated is, in fact, the gold standard and the available comparators are known to be less valid. This is the case with venous duplex for the diagnosis of deep venous thrombosis and carotid duplex for the diagnosis of carotid atherosclerosis.

When appropriate external comparators are available, the validity, accuracy, and ultimate clinical utility of specific diagnostic tests can be evaluated by calculating relatively simple parameters describing the agreement between the two methods of testing. These parameters are:

These parameters are usually expressed as percentages. Calculation of these parameters is straightforward for dichotomous outcomes (positive or negative). The calculation of the parameters for exams with graded outcomes is somewhat more complicated and involves construction of a comparison matrix, as illustrated in Fig. 3.3. A complete review of the statistical calculations can be found in several standard reference works [14, 15].

The types of evaluation noted above represent variants of external validation of the accuracy of the lab work product and may allow the detection of bias related to equipment, policies, personnel, or methodology. Precision is usually assessed through repeated testing within the lab of the same sample (patient) over time or in different settings. The precision of a test is influenced by both underlying temporal biological variation (a well-recognized component of physiological testing and less of a component of anatomic exams) and the natural history of disease as well as variation based upon processes in the laboratory. Quality programs in the diagnostic lab are designed to minimize variability due to intrinsic lab performance. Although noninvasive testing has low risk and excellent patient acceptance, multiple repeated evaluations for internal quality control purposes are possible only to a limited extent. Issues of precision, reproducibility, and comparability are important in assuring the quality of the output of a given vascular lab, but these issues are magnified when results are compared across different labs, different sonographers, different interpreting physicians, and different time frames. These concerns have led to a major drive to make lab processes more standardized and consistent across the health system.

The effectiveness of a diagnostic test is assessed by the contribution of the findings to appropriate clinical decision-making, the ultimate clinical outcome in the treatment of the patient, and the cost entailed in achieving the desired clinical result. The role of any diagnostic test in clinical decision support is quite complicated and there are wide disparities in the application of diagnostic testing to patients presenting with similar conditions. There is some guidance regarding the utility of any particular diagnostic test in reaching effective clinical decisions arising from recommendations and guidelines based upon evidence-based research, consensus documents by subject matter experts, or recommendations of specialty societies with deep knowledge and understanding of the clinical entities of concern. These guidelines take the form of appropriate use criteria (AUC) documents. In some cases, there are several sets of AUC from different organizations, but most often these criteria are quite similar or nearly identical. There are several useful AUC documents that apply to noninvasive vascular imaging specifically and may be important in increasing effectiveness of the noninvasive vascular lab [16–18]. The impact of a single noninvasive vascular test on the overall clinical outcome of a patient may be difficult to determine accurately. However, the satisfaction of the patient with the experience of the vascular test and the patient’s perception of its role in his or her overall care are very important features in patient-centered care and can be assessed through surveys and other instruments. There is a great deal of interest among payers and patients in the efficient and effective use of medical resources and in the elimination of duplication and potentially wasteful processes. There is a strong argument that increased diagnostic efficiency as defined above and achieved by effective comprehensive quality programs will markedly improve effectiveness through a decrease in false-positive and false-negative findings, a reduction of needless duplicative testing, and more timely and appropriate treatment.