Bariatric Surgery Initiative Also Module 04 Writt

LEADERSHIP DETERMINES WHAT TO MEASURE

Leaders lead according to a value system, defining the kind of organization the institution should be. It is up to the senior leadership of the hospital to define the level of quality that is acceptable and the level that is not. Leadership defines priorities by answering such questions as these:

• • What aspects of the organization are critical to its success?
• • What expenses are most and least profitable?
• • How can excellent patient outcomes be achieved efficiently and economically?
• • What variables influence patient satisfaction?

These and many other factors need to be understood and balanced—through measures.

With objective criteria in hand, administrators have access to quality variables and can use factual information to make decisions. Becoming familiar with and using quality measures to deliver quality care helps the health care leader to do the right thing for the patient and to increase financial efficiency for the organization. The better the care, the fewer the complaints, complications, and incidents. When administrators understand how measures of quality reflect operational processes, clinical care, and patient services, as well as underlie good financial management, they become more comfortable about monitoring the delivery of care they are responsible for. Leadership and a strong quality management department should collaborate on using measures to understand the processes, procedures, and operations that have positive and negative impacts on patient care and organizational processes.

MEASURES DEFINE QUALITY CARE

Prevention is good medicine and helps the organization maintain its financial stability. Measures should be used to establish benchmarks for preventive processes—processes such as monitoring sterilization to prevent infection, providing fall prevention, preventing skin injuries, or reducing length of stay (LOS) through appropriate and timely antibiotic administration. For example, to decrease expenses, increase efficiency, and produce good to excellent outcomes, leadership needs to control nosocomial (hospital-acquired) infection.

Specifying the numerator and denominator of the measure ensures that it accurately reflects the information you want to collect. For instance, the general infection rate can be computed as the relationship between the number of patients who contract any infection within a month (the numerator, or N) divided by the number of patients admitted to the hospital per month (the denominator, or D). However, if the information you want is more specific, you define the measurement accordingly. If you are concerned about the incidence of sternal wound infections postsurgery, N becomes the number of postsurgical patients with wound infections over a specific period of time divided by the total number of surgical patients over that same time period (D). Once the measure is defined and the rate can be calculated, the information can be tracked over time. Collecting such measures allows an administrator to monitor trends, such as whether infection is rising, decreasing, spiking, or comparable to the national benchmark. Figure 4.1 illustrates the rate of surgical site infection in one hospital over a twenty-two-month period and shows that its rate is, by and large, lower than the national benchmark.

By carefully defining a measure, with the specific numerator for the objective of the study and the denominator delimiting the population of which the numerator is a subset, leaders can objectively and productively study performance, success, and opportunities for improvements. The data in Figure 4.1, for example, show that spikes in infection occur in the same months each year (January–February and September). With that information leadership can drill down in their data and attempt to analyze what might be contributing to the rise of infection during those months.

MEASURES INFORM FINANCIAL DECISIONS

Data regarding the specifics of care help administrators make efficient financial decisions. For example, the nursing shortage in this country has resulted in staff vacancies that have had an impact on patient care. CEOs and senior administrative staff are expected to make hiring decisions, but how? Using what information? In other words, what are the criteria for evaluating long-term versus short-term investment decisions? Hiring decisions obviously have an impact on the budget, but unless administrative leadership uses objective measures to look at the specifics of operations, evaluates the effectiveness of services, and gauges the effect of staff-patient ratios, how can they understand staffing requirements and the relationship between staffing and patient outcomes?

Many health care institutions are in financial difficulty because important decisions are being made without adequate understanding and information. Think of open-heart surgery and its huge requirements in terms of the operating room (OR), intensive care unit (ICU), specialized staff, and ancillary services and then compare those requirements to, for example, the treatment of patients with pneumonia, a far less resource-intensive hospitalization, assuming, that is, that the patient does not develop complications. Variables for both these conditions can be measured. Information (that is, data) about these variables gives administrators insights into the relationships among services, outcomes, and resource needs.

Tracking several potentially related variables can offer leadership important information. Figure 4.2 combines two variables, LOS and readmission within thirty days, across eight hospitals. If a patient requires readmittance within thirty days of discharge, it is possible that that patient was discharged prematurely or that the care was in some way deficient or inadequate. If administrators examine only LOS, they might believe that the shorter the LOS, the more efficient the hospital. However, if the hospital with a short LOS has a high rate of readmittance, as Hospital B does, then leaders may want to investigate and target improvements. Hospital D has both a long LOS and a high rate of admittance, suggesting inefficiencies of care that have financial consequences. Hospital G is providing the most efficient and effective care.

Because the government reimburses institutions according to the complexity of each case (using the case mix index, or CMI) and the procedures required to treat specific diseases, financial resources are dependent on clinical considerations and operational processes. For open-heart surgery cases, a measurable variable, such as turnaround time in the operating room, can have a financial impact for the institution. If the first procedure of the day is postponed due to operational issues, then for the rest of the day procedures are late. Late procedures have implications. It may become necessary to hire extra staff to work into an evening or night shift. Any complication during a procedure tends to cause expensive delays. Therefore good clinical supervision with clinical support can reduce such expenses. Ideally, a finance officer and a senior administrator learn enough about the delivery of care to ask intelligent questions and establish appropriate measures for data collection and analysis.

Tools and technology and even staff cannot be evaluated as a unidimensional financial expense. An administrator or financial officer can collect data in order to understand the complexity of services. For example, in the ICU there is usually a one-to-one patient-staff ratio. However, administrators may want to know if that ratio is crucial to the welfare of the patient, if the expense results in improved outcomes, or if it is simply a high degree of (perhaps unnecessary) monitoring. Analyzing measures helps an administrator discover the clinical as well as the financial value of a service. When leadership understands clinical care, financial decisions are not made in a void.

MEASURES AND PURCHASING DECISIONS

The financial implications of purchasing decisions are entwined with various aspects of patient care, and intelligent decisions cannot be made without an understanding of other expenditures and the impact on patient outcomes.

Administrators should consider using their quality management departments to mediate information between finance and the medical requirements of care. Quality indicators can help administrators determine the value of specific services, such as whether an elaborate (and expensive) CAT scan will result in better patient outcomes. Without data there is no way to assess whether more sophisticated technology should be purchased. With data, leadership can expect answers to such reasonable questions as what are the financial and clinical implications of a 64-slice CAT scan, and how will it be better for patient care than a 34-slice scan? The medical staff may request new equipment, but it is up to leadership to understand that equipment’s relative value to the organization. Measures improve administrative understanding by providing detailed information.

Some decisions regarding expenses may have far-reaching implications, others may be of less consequence. Purchasing improved cardiac stents, for example, may reduce bleeding and complications from the stent procedure, so although this purchase is expensive it may result in fewer complications, a shorter LOS, and therefore a better financial situation than the hospital would have if the purchase were not made. Data collected over time would reveal the value, and leadership would be able to intelligently monitor costs and benefits. Likewise, robotics technology is very costly. Without objective data it would be difficult to determine if such an expense is of worth to the patients and to the hospital. Information can be collected about the volume of patients who might be attracted to the institution if robotic surgical procedures were in place and the outcomes were excellent. A financial assessment could be projected based on those numbers. Obviously, numbers provide a great deal of crucial information for decision making.

An example of a quality variable that reveals a great deal about operational and financial efficiency is mortality. Administrators should collect these measures monthly in order to monitor the delivery of care and the services being offered. If there are problems, for example, if there were three unexpected mortalities in the OR, there may be a problem that requires addressing. Mortalities cost money. Reports have to be filed with appropriate agencies; malpractice suits can occur; peer reviews have to be conducted. If the source of the mortality is infection, then corrective actions have to be put in place. If the source of the mortality is clinical incompetence, intervention or reeducation can be conducted.

But it is most important to know that the mistakes occurred and then to ascertain the causes in order to develop appropriate improvements. Administrators look at mortality reports and often go looking for someone to blame, rather than considering the situation as an opportunity to improve the delivery of care. If the hospital reports a high mortality rate for a specific procedure, such as cardiac bypass surgery, or for a particular patient population, such as heart failure patients, there might be a financial impact associated with that report because patients with these conditions or who need these procedures may be less attracted to the hospital. The public understands mortality data. (Physicians may say the data are flawed or not risk adjusted, but if the data are out there and the public is afraid, people won’t come to the hospital for treatment.) Operationally, it may be important to understand why the rate is high so that specific processes can be targeted for improvement.

Quality issues and operational issues are interdependent. If data reveal that patients with certain conditions, such as elderly patients with AMI, have a higher incidence of mortality than others, then the care of that patient population has to be carefully reviewed. If patients from certain nursing homes die at a higher rate than others because those patients have comorbidities that are having an impact on mortality, then improving risk assessment might increase safety for those patients. These questions can be empirically tested through developing measures, collecting data, and analyzing trends.

MEASURES AND PATIENT SAFETY

Quality management data are required by agencies for accreditation and for compliance with regulations, but data are also collected as part of various national programs to assess and improve the quality of care, such as the CMS core measures, the Institute for Healthcare Improvement (IHI) 100,000 Lives Campaign, and the National Patient Safety Goals initiative of the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) (see Figure 4.3). JCAHO mandates that each of its goals be implemented; the individual organization determines how to implement each goal. For example, to improve accuracy of patient identification, an organization is required to check two patient identifiers before administering medication, blood products, or performing clinical testing, treatments, or procedures. The hospital determines which two identifiers it will use. Improving communication involves ensuring that phone and verbal orders are properly understood; JCAHO recommends that hospitals require a read-back by the person receiving the order. Medication safety involves several improvements: limit drug concentrations, review look-alike and sound-alike drugs to prevent interchanges, and label all medications. For infections, comply with CDC guidelines for hand hygiene. These goals and their implementation recommendations can be found at the JCAHO Web site (jcaho.org).

The data about safety are collected, and administrators should use the information to understand their operations; furthermore, because quality management data are benchmarked against national standards, administrative and other leaders can evaluate how their operations compare to other institutions. Through measures and benchmarks the data provide relevant information about daily performance and about areas where improvements should be instituted.

The IHI 100,000 Lives Campaign is the first national initiative to prevent avoidable deaths in hospitals and to implement change to improve patient care. The goal is to save 100,000 lives as of June 14, 2006. Highlights of the prevention program include the creation of rapid responses teams, using evidence-based care for AMI, preventing ventilator-acquired pneumonia, preventing indwelling venous catheter infections, preventing surgical site infections, and preventing severe drug events.

QUALITY METHODOLOGY FOR PERFORMANCE IMPROVEMENT

Collecting data on an operational variable, such as blood administration, waiting time in the ED, turnaround time in the OR, or time to receive consultations or laboratory reports, reveals information about efficiency; efficiency has an impact on the financial success of the institution. In addition to using the quality management department to establish databases and benchmarks for best practices, the organization can use quality methodologies, such as PDCA and Six Sigma, that help analysts to inform administrators about services and to improve the delivery of care.

Using quality methodologies may enhance the assumption that excellent care is equal to a sound business plan and economic success. However, a simple economic model might even be in opposition to the mission of a hospital, which may be to serve the poor and the underserved. Such patients may not have the luxury of focusing on health prevention in the way that individuals with economic means and health insurance do. This lack of prevention might result in more sickness, which might in turn burden the hospital because it will be providing expensive care without reimbursement. Such expense can be anticipated, however. Therefore those expenses within the organization’s control should be maximally efficient.

As long as the CEO is using a methodology that is based on data and statistical analysis, measures help employees and managers and administrators and members of the governance committees to share clearly defined goals that stem from a specific philosophical position and to share a commitment to excellence and improvement. Using any deliberate methodology creates a focus for addressing the process of care or product or service. With numbers, administrators can suggest, for example, improving the volume (that is, raising the numbers), eliminating wasteful services (as measured through volume and finance), improving productive services, and targeting specific goals.

Six Sigma is a methodological tool designed to reduce the negative economic impact of inefficient services. Based on the concept of the normal curve, Six Sigma was initially used as a measurement standard in product variation. In the 1920s, Walter Shewhart showed that three sigma from the mean is the point where a process requires correction. As a quality management tool for health care, Six Sigma is useful for analyzing and improving operational processes through measuring how far specific data vary from the mean.

For example, to understand turnaround time in the OR, data can be gathered about timeliness of patient preparation, OR readiness, equipment reliability, surgeon start time, readiness of appropriate ancillary staff, availability of required documentation, causes of delays, if any, and analysis of morbidity that might require extra OR time or an unanticipated return for repair. All of these variables can and should be measured, and each has a financial analogue. Once the inefficient process is identified, improvements can be developed.

The Plan Do Check Act (PDCA) cycle is a robust performance improvement methodology, and one that works particularly well in a health care setting. This model was also developed for monitoring quality improvement in industrial settings and is designed to standardize processes and minimize variation, that is, eliminate mistakes and rework. The PDCA cycle, by breaking function and role into variables that can be measured, helps leadership understand the clinical and medical environment and the method of providing care.

Using the PDCA cycle to continuously improve quality allows current performance to be measured, processes to be analyzed, and improvement actions to be identified (Plan). Improvement actions are then implemented (Do), and the benefits of the actions are measured (Check). Once measured, improvements can be standardized and communicated and reassessed (Act). The PDCA cycle provides for the systematic acquisition of knowledge through focused data collection and, through measurements, validates that improvements are effective (see Figure 4.4).

There are many advantages to using an industrial performance improvement model, such as PDCA, to continuously evaluate improvement and determine variation from the standard. The PDCA cycle provides a continuous loop of quality monitoring, based on data from measures. By defining the numerator and denominator of a measure, leadership can objectively understand the product being delivered, and by holding staff accountable to these measures, leadership clearly anticipates a uniform standard of excellence.

As with most complex activities, doing something according to a plan is more productive than simply reacting to some stimulus on the spur of the moment. In health care, planning involves collecting information and analyzing current processes, identifying gaps in care, establishing improvements, and monitoring their effectiveness. Making improvements or changing processes is often met with resistance and confusion over accountability (who is in charge) and details of the process changes (who is doing what).

My experience shows that to improve a process, adopt new information, and actually change the delivery of care, the unit manager and the clinicians benefit by working within a methodology, such as the PDCA, that continuously and objectively reviews and evaluates their actions. The PDCA method allows the professionals to pause and consider the workload with a critical eye. Working with many patients, with multiple diagnoses and treatment plans, caregivers require a method that directs and prioritizes activity. Daily planning must be continuously communicated, from the beginning to end of shift, through the changes in shift, and to the end of the shift to maximize efficiency and reduce potential for errors.

DEVELOPING A PERFORMANCE IMPROVEMENT PLAN

Every aspect of the PDCA cycle depends on measurements, not of an individual’s experience but of a population of patients. The first stage, Plan, requires that stakeholders, who have similar goals, formulate an assumption about care, in other words, develop a hypothesis. The hypothesis may be derived from external or internal sources. For example, because the CMS requires smoking cessation counseling for pneumonia patients, administrators may assume that most of the patients are receiving the recommended counseling. Their assumption may be that clinicians are incorporating patient education about smoking into their practice.

Data can be collected to confirm that assumption. Quality management can develop a methodology for chart review and determine the percentages of patients who have had the counseling and of those who haven’t. With this information in hand, further analysis can drill down in the data and examine the records of those patients who did not receive counseling to see if they have any areas in common, such as physician, unit, secondary diagnoses, and so forth. However, without quantifying the process, it is hard to convince anyone that there is a problem, let alone that it should be fixed.

The assumption or hypothesis should reflect areas of concern to the investigating team. Another assumption might be that patients who are given antibiotics before surgery have fewer infections than patients who are not given this medication. This is a testable assumption. Other testable assumptions are that patients who develop pneumonia on ventilators were not properly weaned off the ventilators, and that patients who fall do so because of a desire to be mobile when there are insufficient staff to assist them. Administrators and staff should meet together to determine which assumption to measure and which care process to improve.

In the planning stage organizational culture should be evaluated to determine whether there are possibilities for change and whether a structure exists to implement changed practices. Leadership chooses which battles deserve to be fought; not every process needs to be changed, and different stakeholders may be interested in different issues. Physicians may be concerned with high mortality, surgeons with infections, nurses with falls, and respiratory therapists with ventilator-associated pneumonias. It is up to the administrative leadership to determine priorities, perhaps based on the goals, mission, and vision of the institution or derived from external pressures from the public and the media or revealed on some scalar dimension by such questions as which problem poses the highest risk, where can the impact of improvement efforts be greatest, or where can financial gains be seen? The senior staff of the organization decides priorities for improvement, what outcomes to look at, what processes to change, which measures to use, and what process to develop to monitor, assess, analyze, and communicate the results of the data collection activities.

Before you determine your measures it is essential to define your clinical or operational goals, establish priorities, and understand the patients (the organization’s customers) and their concerns and priorities. The quality management department at our health system developed a prioritization matrix to help decision makers evaluate competing issues for performance improvement (see Table 4.1). Competing issues for improvement are listed across the top of the matrix. Each issue is evaluated by the criteria listed along the side of the matrix—such as alignment with leadership goals and vision, impact on the delivery of care, or outcomes showing a negative trend. Different organizations will define their criteria differently, but it is useful to think about prioritization in terms of structure, process, and outcome. For each issue a value is entered in each cell of the matrix, from 0 to 3 (no application to maximum concern) and these values are totaled. A comparison of the totals defines the most pressing priorities. By objectifying and quantifying priority options, stakeholders have an opportunity to evaluate and consider how to allocate resources.

Table 4.1. Prioritization Matrix.

In the Plan stage the stakeholders should be able to realize that change is possible and that change would be good for the institution, the patients, and themselves. Even this initial point may be difficult because often caregivers see no need for improvement, an attitude that there is no reason to fix what isn’t broken. Tradition—doing things the way they have always been done—makes people comfortable. However, acquiring data usually reveals that improvements should be made.

When the senior staff agree on priorities, develop assumptions about performance improvement, and assign responsibilities for roles and functions within the organization, the Do phase of the cycle begins. The stakeholders of a process or procedure determine the improvement. For example, surgeons may want a better assessment for administering antibiotics in a timely way. When weaning protocols are being reviewed, the pulmonary physicians and the respiratory therapists are the stakeholders, as well as the nursing staff. If falls are being investigated, perhaps a multidisciplinary committee can develop an improved risk assessment screen for patients who are at high risk for falls. The Do phase is where a change is designed and relevant measures (numerators and denominators) are defined to monitor the process of change and the improvements. Also in this phase, procedural details are developed, such as which staff members will be collecting data for the measure, how the data will be collected (in what form) and reported (to whom), who will analyze the data, over what period of time, and how the results of the analysis will be reported out and to whom.

As always, the measure is defined by what the stakeholders want to know. If mortality rates are at issue, then the group may want to look at various procedures and have analysts analyze mortality according to various clinical services, patient populations, treatment, and diagnoses, whatever is of interest to leadership and staff. It is a good idea to review the literature for existing methods of data collection and analyses. Established studies can become benchmarks for the standard of care.

Once the design of the measure and the data collection efforts have been accomplished, improvements and changed practices are designed and implemented. The Check phase of the PDCA cycle is used to monitor the new procedures and to see if they are successful. New measures may need to be developed, such as the timing of antibiotic administration. During the Check phase it is important to keep monitoring the improvements to ensure that they are maintained. This phase is the evaluation phase, in which the program under study is assessed. It is useful to ask the stakeholders and the medical board for input, in order to increase confidence in the improvement efforts.

In the Act phase, changes are implemented, a procedure that requires administrative commitment. During this phase a table of measures can be developed that will provide a snapshot of improvements (or the lack thereof) over time. In this stage it is also important to effectively communicate information about changed processes throughout the organization, from the bedside workers to the members of the highest governance committees.

CASE EXAMPLE: PLAN DO CHECK ACT FOR BARIATRIC SURGERY

In the health care system where I work, because we have a strong quality management department with databases that promote measures and that are respected by physicians and administrative leadership, quality management was able to conduct an improvement initiative related to bariatric surgery, that is, surgery for the treatment of obesity. Improvement was driven in this case from within the organization.

The more an experience can be quantified, the better it can be understood. Quantifying experience also promotes accountability of staff because expectations are clear, standards are defined, variation is discouraged, and most important, it is obvious that someone cares and is monitoring what is happening. When a multidisciplinary system task force determined that various aspects of bariatric surgery needed to be carefully evaluated, the task force realized that there would be an advantage to using a deliberative process, such as the PDCA cycle for performance improvement.

In the Plan phase the objective was to protect the safety of this patient population, which is highly complex physically, socially, and psychologically, and to develop guidelines. This relatively new surgery has risks that can result in complications, a dysfunctional life, and even death. Specific standards of care had to be explicitly defined for this procedure. For example, although many general surgeons wanted to perform the procedure in their hospitals, as it is an innovative and high-demand surgery and has the potential to be lucrative for the physician and for the hospital, not all surgeons were qualified to perform bariatric surgery. Therefore guidelines for physician credentialing needed to be developed. Moreover, the results of surgery were not only related to the technical ability of the physician but to the patient’s ability to comply with dietary protocols for weight management. Therefore communication among the nutritionist, social worker, psychologist, and surgical team was as important for a successful outcome as the procedure itself.

In the Do phase, assessment and credentialing issues were addressed. A multidisciplinary task force was convened and charged with developing a method to implement a safe and low-risk environment for the patient, determining the specific requirements for credentialing physicians, and establishing a consistent methodology for appropriate patient identification, selection, and assessment. The task force was composed of specialists from quality management; physicians, including bariatric surgeons; community physicians; the chief medical officer; the chief of surgery; pulmonologists, for input into sleep apnea; anesthesiologists, for input into airway management; radiologists, regarding the limitations of and alternatives to diagnostic testing equipment for this patient population; intensivists; nurses; nutritionists; psychologists; psychiatrists; and social workers. The health care team ensuring patient safety wanted patients and families to recognize that this procedure alters a patient not only biologically but also through its powerful impact on psyche and lifestyle as well.

For over two years the team researched the available clinical literature, brainstormed many issues, and came to consensus on specifications of appropriate and safe care. This effort resulted in the development of guidelines for volume-based credentialing of physicians, for assessment for appropriate patient selection, and for patient counseling, institutional requirements, and staff education. Because the guidelines were based on evidence-based practice, care was standardized and measurable, from the presurgical physician’s office visit to one year postoperatively. A specific algorithm for clinicians was developed, the bariatric surgery clinical pathway, which could be used to standardize care (see Figure 4.5). This CareMap documents, on a single page for each day, whether or not specific consultations, tests, treatments, medications, and much more have been met or remain unmet.

Once guidelines were established, the Check phase was begun. The quality management department, in collaboration with the multidisciplinary task force, created a database to monitor relevant indicators from presurgery to one-year postsurgery for ongoing review. The outcome database, approved by the system hospitals’ medical boards, is used to evaluate each program through the use of common data definitions and uniform numerators and denominators; it tracks and trends patient demographics, outcomes, and complications (see Table 4.2 for one example of an outcome database).

Objective outcome measures analyze patient outcomes. The guidelines seek to prevent serious postoperative complications (such as deep vein thromboses). A standardized program was created through consensus; lessons learned and best practices were shared and then implemented across the health care system. By reviewing the data and employing objective standardized definitions that were compared to internal and external benchmarks, accountability was increased, as was communication among the board of trustees, the hospital medical boards, and the physicians. Improved communication also served as a tool for teaching about the risk and benefits of the procedure. The process spurred the medical boards to develop specific standards of care around the procedure, especially around different weight categories.

Table 4.2. Sample Bariatric Table of Measures.

The Act phase includes ongoing education for the relevant specialists across the system. The Center for Weight Management, under the psychiatry department, provides comprehensive weight management services for patients and their families and offers educational programs to staff. Psychological and sensitivity training and education are also made available to ensure competency among nonphysicians. The latter program is directed specifically toward this special patient population and its physiological as well as psychosocial risks. Teleconferences have been held that addressed sleep-disordered breathing and obesity and also morbidity and mortality.

The result of this initiative was improved patient care and more efficient organizational processes that resulted in decreased cost to the hospitals and the system. Because complications went down, the LOS for patients was shorter. The readmission and reoperation rates were low. Patients were appropriately screened, assessed, and educated and received follow-up counseling and support. Staff were objectively credentialed, and education was provided for staff in various specialties—nursing, surgery, anesthesia, nutrition, and psychology. Staff were encouraged to share their experiences and lessons learned in an open and blame-free environment, and this facilitated discussion about such controversial issues as the particular requirements for adolescent surgery and surgery on the super morbidly obese (patients weighing over 500 pounds).

The bariatric surgery initiative also informed capital investment decisions to upgrade facilities and to acquire appropriate equipment according to a principled and deliberate set of criteria. As each hospital created a bariatric center, new physical environments were established with special operating rooms, beds, and wheelchairs. Patient support groups were created and the transfer of information from the physician’s office to the hospital is now seamless. Goals were developed for each institution to obtain JCAHO Disease Specific Certification or designation for having standards of excellence as endorsed by the American Society of Bariatric Surgery Centers of Excellence Program, a designation that attracts patients. Some hospitals sought both.

MONITORING VARIATION FROM THE STANDARD

The PDCA cycle for performance improvement was originally designed to minimize defects in production. The theory is that if products are made according to a standard, every product will be perfect. Lack of perfection is equivalent to variation from the standard. In health care, also, variation from the standard serves as a red flag that there may be a defect in the process, that is, a defect in the delivery of an intervention or an unanticipated outcome. As in industry, the goal of performance improvement is to minimize defects in processes. Delivering safe and effective care is good business. Trying to react to problems after they occur or to correct bad practices that have become entrenched costs more money than adhering to standard (evidence-based) guidelines.

Through the use of measures processes can be explained. The population is clearly defined, as is the service under evaluation. The measure is a proxy for the specific service performed, with the numerator defining what is being done and the denominator defining the group that the service is being performed on.

Organizations can also ensure compliance with standards by spending money on consultants. But if regulations and evidence-based indicators are understood for what they are—standards of excellence—everyone involved can be encouraged to internalize each standard and to do it right, before expensive events or complications occur. Measures create efficiencies and better care outcomes. By measuring variation, leadership develops criteria with which to objectify the distance between the gold standard of care (defined by evidence-based medicine) and actual practice. The wider the gap between the two, the poorer the care, and the more expensive it is to provide this substandard care.

Let’s define the health care product as removing an infected appendix before it bursts. If you remove it in a timely way, that’s the standard of care; if you do not, there could be serious complications. The standard of care is also to avoid removing a healthy appendix. With measures, it can be determined how many false positives occurred in the hospital, how many erupted appendixes happened, and why and with what outcomes. Before determining how best to approach patients with appendicitis, it is important to evaluate the current practice. A measure can be developed and tracked over time. Once you have a sense of the scope of the problem, and the volume of patients involved, a multidisciplinary group might study the literature on appropriate standards of care for appendicitis. Research is valuable because relying on the experience of one or two physicians may be inadequate. Their experience may involve too few patients to make accurate generalizations. If data from evidence-based medicine are used, you have the advantage of learning from large numbers of patients, from many physicians, and from many reports of the best treatments and the adverse events that may occur.

If data reveal that indeed there have been instances of erupted appendixes in your organization, it may be useful to develop an algorithm of care with the goal of avoiding this terrible situation. The algorithm would detail criteria for identifying the problem, and outline the appropriate actions to take. Consensus can be established on whether to use physical impressions, such as abdominal pain, lab results, such as elevated white count, or radiological results, such as CAT scans, to determine the diagnosis. Health care experts may decide that the algorithm should include three indicators for a diagnosis of appendicitis and that if the patient has two then the surgeon might consider the evidence and act quickly. The algorithm would then be monitored to see if it is successful, and if so, it becomes the standard of care throughout the organization. Such work can be based on even one occurrence of a burst appendix.

Even one adverse event costs a hospital large amounts of money in follow-up care for complications, malpractice claims, and poor public relations. By internalizing the idea of providing patients with value, everyone benefits. When an adverse event occurs, it is important to do a root cause analysis to determine the gaps in care and the risk points. You can be sure that if a problem occurs once, unless it is fully understood, it will happen again. You don’t want to be in the position of having to face the patients, families, and media and explain why there are problems that endanger patients in your hospital. You want to always anticipate potential problems, using measures to monitor care. With measures, as soon as you see a blip in the data—a rise in infection, for example—you can send in a SWAT team of analysts to figure out what is going wrong and develop corrective actions. Senior leadership should not only commit to measuring variations in care but also build the PDCA, a planned and deliberate approach to continuous quality improvement, into the culture. When leaders expect forethought, the staff will deliver.

CASE EXAMPLE: MOVING BETWEEN LEVELS OF CARE

Maintaining standards and monitoring variation promotes improved organizational processes as well as better clinical care. The following example illustrates how recognizing defects in care and establishing improvements benefits the patient and the organization.

Management of LOS for elderly medical patients can be predicted to some extent based on past experience (that is, data) with this population. Their needs in terms of mobility, posthospitalization care, and physical and psychosocial issues can and should be planned for. However, care of the elderly is usually not planned for well. Often their needs, other than for medical intervention, are not identified upon admission. This is unfortunate because any issue or problem may increase over time, especially if it interacts with medical problems.

One of our community hospitals receives a large number of admissions from a particular nursing home. Several years ago the hospital received a letter of complaint from that nursing home saying that when its residents had to be hospitalized, they were returning to the home in worse shape than they were when they left. Leadership responded to this issue by replying that indeed, over time, the elderly patient does get worse, regardless of medical intervention. In other words, the care the hospital was delivering was appropriate and the patient population was at risk. However, when a second complaint was made, warning that the nursing home would refer patients to another institution, the CEO asked the quality management department to look into the matter more scientifically.

Data revealed that the nursing home complaint was valid, that patients were indeed returning to the nursing home with less mobility, with decubiti, with infections, and with depression—objectification of the notion of “worse.” Although the hospital physicians had adequately dealt with the specific medical problem that brought each elderly patient into the hospital—for example, the patient who arrived with a fever or a high white blood count did receive antibiotics on time—there was little, if any, attention to any other factor. The patient’s physical condition deteriorated because there was no communication between the nursing home and the hospital about the multiple needs of the patient.

Further quality management research found that the physical environment of the hospital was not particularly suitable for the elderly patient, and therefore patients were at increased risk for falls. The food was not appealing and was perhaps left out of reach, and therefore patients were not eating. With their nutrition suffering, patients were not properly absorbing their medication. Care providers were not monitoring that the patients were mobile enough to ward off skin injuries, and therefore patients suffered from decubiti and complications of decubiti.

Once these specific issues were identified, changes were made in the process of care. Clinical staff received education about caring for the elderly patient, specifically improving environmental factors, risk assessment for falls and decubiti, and nutritional counseling. In addition to these improvements communication between the nursing home and the hospital was improved. Information was transferred about the physical and psychosocial needs of the patient as well as the medical problem that required hospitalization.

The transition between the nursing home and the hospital became smooth, and the patients returned to the nursing home with appropriate and improved health status. It was the CEO who, responding to the nursing home complaint, led the charge to change the process, to identify the problem, and to correct it. The physicians and the nursing staff changed their clinical outlook to improve the delivery of care. The intervention of the CEO was focused on the process and the operation of providing care among institutions, that is, patient flow from one level of care to another.

UNDERSTANDING PATIENT FLOW

Patient flow has a financial impact on the hospital, and patient flow can be deconstructed into individual and measurable parts that can be monitored for improvement. It is obvious that the more efficiently patients are moved through their episode of hospitalization, the greater will be the advantage to the hospital and the higher will be patient satisfaction. Leadership needs to supervise patient flow, and the most productive way to do that is via measures. Figure 4.6 outlines the levels of care in a typical episode of hospitalization. When administrators understand patient flow and can identify bottlenecks in the process, they can then collect information about the impact on services, on the budget, and on clinical outcomes. Once problems are identified, relevant improvements can be implemented.

Generally, patients enter the hospital through the ED. Hospitals don’t get paid for extended ED stays, yet many EDs function almost like a hospital unit, because, for a host of reasons, it is difficult to move patients from the ED onto one of the hospital’s regular units. Measures can be collected that reveal waiting time in the ED, the number of potential patients who left without being evaluated, and the time from triage to diagnosis to admittance to a unit. If these data reveal that the time is prolonged, other data can be collected about the cause of the delay. Are delays caused by waiting for consults, for lab work, or for reports or by other technical issues, or are they due to housekeeping or transport bottlenecks? If the ED is overcrowded and patient care is delayed, what contributes to the congestion and impedes efficiency? Without measures, one might conclude that the ED is short-staffed, but there are many other possibilities. Table 4.3 shows a table of measures with examples of ED variables that could be collected and reported to senior leadership to assess the delivery of service. If the ED is crowded and patients have to wait for a long time or are otherwise dissatisfied with their care, they may not return. By tracking information over time, leadership can locate where the delivery of care has fallen short of the standard. It’s a buyer’s market, and so if leadership wants to attract patients, care has to be competent.

If housekeeping is not able to make up a clean room so that a patient can be moved from the ED onto the unit in a timely way, improvements can be made. Once the problem is identified, a new process can be developed to improve turnaround time, resulting in better care for the patients, greater efficiency for the ED, and financial improvement as patients move appropriately to different levels of care. Data can be collected on how long it takes for laboratory test results to be received and how long patients remain in the ED awaiting those results.

Table 4.3. Sample Emergency Department Table of Measures.

With everyone working independently to meet the goals of his or her own department or service, regardless of the other departments or services, interdepartmental communication may be weak. Each department’s objective might be met, but not the whole organization’s. To change this, staff in departments involved in any way with ED patients must work with consciousness of their impact on patient flow, rather than focusing solely on their own department’s goal.

There may be many reasons that patients remain in the ED longer than clinically appropriate. Measures can help administrators pinpoint where processes should be improved. For example, if the discharge planning process is not begun appropriately, it may be another source of extended stays. Data on processes will inform administrators about bottlenecks. All these measures of care are also measures of effectiveness, efficiency, and thus financial viability.

It is important for administrators to supervise the throughput process and not let segments of the process act independently of each other. This means that administrators are not focused just on the ED but on radiology, housekeeping, dietary, and so forth, as well. All services have to understand their role in patient flow. The movement of patients along the continuum of care within the hospital must be analyzed daily in order to locate points that force the patient to stay in one place longer than necessary.

Quality and operational measures lead to financial success. When the care is smooth and timely, then the laboratories; technical processes; ancillary services; and environmental, housekeeping, and nutritional services—as well as clinical services—are all working effectively and efficiently. In addition, the communication structure of the performance improvement committees, if used appropriately, reinforces this success.

SUMMARY

Measures and databases

• • Reflect leadership priorities.
• • Respond to external and internal requirements.
• • Reflect best practices.
• • Are founded on evidence-based research.
• • Should be realistically accessible for collection and analysis, with explicit numerators and denominators.

The quality management department should help administrators develop databases and consistent measures and help leaders determine best practices in care. When the leadership supports the data collection and analysis efforts, clinical staff will follow suit. Valid measures help to standardize assessment across various units of the hospital or across institutions in the health care system. Everyone agrees on the same numerator and denominator. Apples are always being measured against apples, and not against anything else.

It is important to involve the relevant stakeholders in the definition, collection, and analysis of the measures and to use a deliberate methodology, such as the PDCA, for performance improvement. Consistent measures can be replicated over and over again—in different environments and for varied periods of time. Measures have to be reasonable and collectible and about something someone cares about.

Things to Think About

There is a sudden spike in infection among surgical patients. The local newspaper is warning people to stay away from the hospital where you are a top administrator. How would you manage this issue?

• • What measures would you develop to identify the source of the infection?
• • How would you interpret the results of the data about the source of the infection, and according to what standard?
• • What process would you use to develop an improvement plan?
• • Which stakeholders would you involve in the process?
• • What data would you collect to monitor the improvements?
• • How would you communicate with the media about the improvements?