6.1 On the Subject of Appropriateness

In Chapter 5 we wrote primarily about the language of conflict in health care, and how a tone of war pervades health care delivery, including reimbursement negotiations. The language is part and parcel of a set of entrenched behaviors and assumed interests. In this chapter, we turn our attention to the language of alignment and how stakeholders occasionally attempt to find common ground despite their fundamentally different approaches to health care delivery. We lean heavily on the phrase “health care delivery” in this chapter because it is really the only phrase that captures the contribution of the different stakeholders and the confluence of different influences on the enterprise of providing care for patients.

Employers, insurers, hospitals, and physicians all influence health care delivery, along with many other salient stakeholder groups. Harvard Business School Professor Michael Porter and University of Virginia’s Elisabeth Teisberg wrote about the “problematic” governance of physicians, especially when it comes to executing a more inclusive health delivery strategy in a facility like a hospital. The problem, as described by Porter and Teisberg (2006), is twofold. Firstly, some types of physicians may have stretched themselves thinly, perhaps treating a range of medical conditions (including some that fall outside their area of experience and expertise) out of financial necessity. Secondly, physicians, hospitals, and other providers are driven by areas of medicine that have “generous reimbursement levels.”

In my own analysis, I ask students and professionals alike to reflect on not only the business proposition of providing care in the United States, but the business and (p. 89 ) professional risks of providing care to an increasingly litigious and difficult-to-please patient-consumer population. In other words, physicians, and even hospitals, can be profoundly influenced by the stakeholder group commonly referred to as “patients.” No doubt that for many different types of providers and provider organizations, the benefits of practicing defensive medicine seem to far outweigh the (sometimes) theoretical hazards. Although many physicians might not openly reveal, some will admit (behind closed doors or among friends) that ordering unnecessary tests are a way of defending themselves against litigious patients, a behavior that is commonly seen as “defensive medicine” (Glabman, 2005; Studdert et al., 2005). A number of narratives from physicians interviewed on National Public Radio (NPR) capture these familiar sentiments well. Defensive medicine is not only a means of avoiding litigation; it is also a means of satisfying hard-to-please patient-consumers, who may have certain expectations about what constitutes good or even appropriate care (Glabman, 2005). When interviewed by radio reporter and host Melissa Block for NPR’s Morning Edition, a family physician from New Mexico, Dr. Greg Darrow, had this to say:

There’s an expectation for doctors employed by a large group to hit a production target…. Marketing, patient satisfaction, quarterly numbers, and all the business ends can intrude into the practice of medicine. (Block, 2009)

Dr. Darrow later added that physicians are not only pressured by their employers, or the partners that they work with, or their own practice’s bottom line; patients also are increasingly pushy. He has found that patient-consumers often demand expensive imaging tests for minor injuries such as a twisted knee, which would typically resolve with adequate rest and ice (Block, 2009). With this and the multitude of other examples, it is possible to see that the focus in health care delivery today has evolved to proportion a higher priority to patient satisfaction over patient outcomes. Often, in various health care debates, patients are portrayed as a sort of innocent bystander. However, the market force that has almost euphemistically become know as “patient satisfaction” can be a greater influence than any other in certain situations. No doubt, it is health outcomes and survivability indicators that are a better measure for the performance of health care delivery, and not necessarily patient, or even doctor, satisfaction.

Chasing satisfaction on both the physician side and the patient side in the United States may have caused an interesting dichotomy in the delivery of health care. This point has not escaped the attention of insurers, who are particularly interested in curbing the use of tests and procedures that are inappropriate and possibly dangerous, and are invariably costly. Yet instituting strict guidelines or introducing cost sharing are often seen as having deleterious effects on care. Simply curbing costs by means of rationing health care and introducing greater cost-sharing as a disincentive to inappropriate use (i.e., having patients absorb a higher proportion of the cost through higher deductibles, higher co-pays, and less coverage for pharmaceuticals and certain tests and procedures) has been associated with hastening the poor management of chronic conditions (Solomon, Goldman, Joyce, & Escarce, 2009).

(p. 90 ) Some insurers, employers, and other purchasers are instead focusing efforts on promoting appropriate care (Glabman, 2005). However, these stakeholders alone are not well-equipped to institute change through unilateral measures that may otherwise be embraced by additional stakeholders. Instead, an approach that encompasses all stakeholders would be far more appropriate. We are back at our Gordian Knot analogy, where the dynamics of different forces not only act against one another but also are intricately and inextricably woven together. Stories and data about unnecessary intervention have stood at the forefront of the health care debate in recent years. Medical and public health policy journals are full of examples of how runaway interventional treatment has cost purchasers and payers not only great expense, but also their health (J. E. Wennberg, Bronner, Skinner, Fisher, & Goodman, 2009; Yasaitis, Fisher, Mackenzie, & Wasson, 2009). The larger theme of these critiques is that we spend a lot on care that does not help us, and could even harm us (Abelson, 2006; Brownlee, 2003; Gawande, Berwick, Fisher, & McClellan, 2009). But deciphering exactly what tests and procedures are necessary and beneficial in different instances is not an exact science, and oversight mechanisms are often complicated by purchasers and payers attempts to control costs.

Nevertheless, ways to ensure better care and better outcomes may exist. Much has been written about the benefits of active engagement between patients and physicians. Communication does lead to improved outcomes. Without a doubt, patient education about preventative care is a cornerstone of public health intervention strategy. Promotion of disease prevention through better awareness is a major tool used by health providers, consultancy groups, and government entities to cut expense while also promoting appropriate care. Even so, incomplete or improper implementation can render preventative care incapable of changing patient behavior most responsible for mortality and morbidity rates.

There have been some great historical successes with investing public money in preventative medicine. More recently, a number of health consulting firms have been offering an outcome-based approach to sizable purchasers of health care, promising not only a healthier, more productive workforce but also major cost-savings down the road. Health consultancy firm Pitney Bowes extols the merits of this approach, dubbing it “value-based benefits design” (Mahoney & Hom, 2006).

I have witnessed firsthand a huge growth in the use of consultancy firms by health insurers in recent years. One such consultancy firm, CareCore National, has garnered significant accolades for their ability to root out inappropriate care. As one of the firm’s executives, Dr. Russell Amico, was quoted as saying in the New York Times, “We’re seeing layering of tests on top of each other.” He was suggesting that tests are often ordered together, unnecessarily, with more definitive tests occur alongside less useful and somewhat redundant tests (Berenson & Abelson, 2008). Yet many physicians will likely reject such a claim when shown a specific example in their area of expertise.

One area of medicine that has received much scrutiny lately is cardiac care. Contemporary computed tomography (CT) angiograms, for example, give physicians unprecedented moving images of the beating heart and its valves and arteries. (p. 91 ) Even so, the proliferation and use of CT angiograms is not necessarily associated with better outcomes, or with high success rates for preventing cardiac events or improving cardiac care (Berenson & Abelson, 2008; Ladapo, Horwitz, Weinstein, Gazelle, & Cutler, 2009). A brief yet informative look at the different scenarios that influence the decision to use a CT scan can be found on the New York Times Web site (DeSantis, 2008), and we reprint it here in Figure 6.1. This diagram is not perfectly definitive; however, we believe it will help readers zero in on the issues discussed in the next case vignette and the literature review that follows it.

Figure 6.1 points to the crux of what many media outlets, including the New York Times, have recently questioned in the debate about “unnecessary” tests. The New York Times took a cynical view of the new CT technology, arguing that the CT angiogram may produce nice, compelling images of the heart but does not yield as much information as other, older tests. While this diagram highlights some important questions about CT angiograms as compared to other diagnostic tests, the newspaper oversimplifies the issue here. Firstly, it may be too early to know the full potential and insight that the latest CT scan technologies add to cardiac care. Secondly, we feel that decisions regarding appropriateness are far more complex than is portrayed in the examples here. But the fundamental question remains: “do new tests and treatments always, or nearly always, lead to better outcomes?

To better understand the complexity behind how diagnostic decisions are made, defended by specialists but questioned by purchasers, I thought it best to supply you with a case vignette. Certainly this next case vignette does not cover all aspects of cardiac care, guidelines, purchasers’ concerns, or reimbursement negotiations. It does offer insight into how some clinical decisions and costs are viewed by stakeholder groups seeking alignment.

The case vignette is about a meeting of an insurance firm, a group of physicians specializing in cardiac care, and a hospital network that provides a highly specialized laboratory and clinic where the cardiologists practice medicine and carry out research. A close read of the vignette reveals that there are many other stakeholders that shape relationships and practices, even though they are not present at this meeting. There is, for example, a business consultancy group that the insurer has contracted but which is not represented at the meeting. There are competing cardiologists at other health systems and independent practices who indirectly determine the range of choices. There are testing centers that influence the cardiologists’ work but which are mentioned only in passing. At this meeting, the most discussed stakeholder, who is noticeably excluded, is the patient.

This absence is certainly a logical one, given that the meeting is one in a series of physician payment negotiations. But it is also notable that the invisible hand of patient-consumer satisfaction may be driving much physician behavior. What patients expect greatly influences the measure of appropriate care. Yet, without having a seat at the table, the patient stakeholder group has some of its influences funneled through the actions and negotiating stances of purchasers. Their pull is exercised through another party, too, because generally speaking the insurers are representing employers, one of the major subsidizers of health care (along with the (p. 92 )

(p. 93 ) government/taxpayers), as well as themselves. Even though this meeting is ultimately about reimbursements that are linked to physician services, it touches on a number of issues regarding decision making and care delivery, specifically in the area of appropriate imaging for cardiac patients.

As you read this case vignette, it is worth noting how stakeholders as diverse as physicians and insurers all seek to provide better value for the patients, yet they have strikingly different approaches. These approaches encompass their own strategic business needs as well as the patients’ interests. Indeed, as one student of stakeholder theory wrote, it is the relationships governing stake- holders that are both “complex and dynamic” (Grossi, 2003). In his thesis, submitted for a Masters of Science and Engineering Management at the Massachusetts Institute of Technology, Ignacio Grossi (2003) integrates important concepts and tools to analyze stakeholder relationship and saliency. He proffers a systems approach to understanding how stakeholders are structured as a means of first understanding the dynamics of how these stakeholders can work together. Grossi developed a classification to help characterize the relationship between stakeholders. He also proposes better ways to create value between those very same stakeholders. We will illustrate some of Grossi’s ideas toward the end of this chapter, after we have had a chance to unpack the many components of the case vignette.

The development of metrics to analyze stakeholder relationships has not been as succinctly developed as we might hope. To use these metrics, Grossi (2003) makes a number of assumptions. In at least one instance he assumes, for the purpose of modeling, that “relationships between stakeholders are bidirectional and balanced.” In another variation of his DSM, he is able to account for asymmetries, or, as he puts it, “particular characteristic[s] of the relationships between stakeholders.” Understanding the clustering and nature of stakeholder relationships with fine accuracy is not easily achieved, even when using specialized tools. With these shortcomings in mind, we still encourage the use of tools that facilitate a deeper understanding and foster a great deal of group discussion. Importantly, we want to stress the fact that analytical tools are only successful if the data used are accurate. In fact, we affectionately refer to the phenomenon of making decisions on bad data as “GIGO,” or garbage-in-garbage-out. It is important to not blame the tools while understanding that in some cases bad decisions are predicated on bad data.

(p. 94 ) We should make one last caveat before delving into the details of the meeting featuring the insurer, the cardiac specialists, and representatives of the hospital network. In order to accurately capture the true nature of a stakeholder relationship, a great amount of human knowledge is needed. Relationships are moving targets, and so it is possible to see only a snapshot of a relationship when discussing stakeholder alignment. As we will see in later sections of this chapter and book, before the possibility of value creation arises, stakeholder alignment is an absolute necessity. This alignment is only possible after the different parties take the time to fully appreciate the shifting nature of their relationships. It is with these thoughts that we continue on to our next case vignette.

We revisit Thomas Jones, the chief negotiator for the large East Coast health system we mentioned in the Chapter 5, as he oversees contractual relationships of mostly third-party payers. In this case vignette he is hosting one of a series of talks regarding reimbursements for cardiac specialists who work in his network. Within a hospital system, the many community-based physicians who work on the premises establish their own contracts with payers independent of the hospital itself. Yet at times, Mr. Jones explained, this can lead to tension between the hospital and the visiting doctors: “If the doctor is working in one of our facilities and is self-employed, or employed by a group that we don’t own or control, I can not set the price.” The doctor, however, may expect that his or her reimbursement is contingent on the hospital’s prenegotiated price. This is a problem, says Mr. Jones, because the expectations of the patients are far from being aligned with the hospital, the physicians, or the payers’ expectations. What they want regarding who treats them, where they get treated, and how they get treated often confounds the delicate dividing lines among the provider groups. Often, it is this lack of transparency that may be at the heart of the distrust and occasional conflict between stakeholders.

Another source of conflict according to Mr. Jones lies in the process of creating contracts between insurers and hospitals. A hospital negotiates reimbursement rates without including physicians who are not directly employed by the hospital. This exclusion may exist despite the fact that the nonemployed physician works on hospital grounds. In contrast, employed physicians often work side by side with community, nonemployed physicians who have different relationships with the payers and at times are treating the same patients. The employed physicians are represented by the hospital and the contract or arrangement of the payers. Mr. Jones notes that this is a possible area of tension and misalignment: “Some of the physicians in the community may be getting paid more than some of the physicians that work at the hospital, while some get less. No one really knows,” Jones said in a follow-up phone conversation. He explained that because all payment schedules are proprietary and confidential, no one is really sure how anyone else is getting paid, or whether a comparison would show reimbursement rates to be higher, the same, or lower.

When it comes to negotiating physician reimbursement, insurers have a vested interest in evaluating physician performance. They may focus their negotiations on indicators that try to characterize or measure efficient care. In recent years, one common measure involves whether physicians and other clinical staff are sticking to “appropriateness criteria” when ordering tests. In earlier chapters, we looked at how rapid spending paralleled a rise in imaging across the United States, and how there are large regional variations in ordering tests (Smith-Bindman, Miglioretti, & Larson, 2008; U.S. Government Accountability Office [GAO], 2008; (p. 95 ) D. E. Wennberg & Wennberg, 2003). Physicians defend their decisions by relating them to the circumstances in which they order diagnostic tests. The poor quality of a test, or the report generated from the test, and the need for further clarification are often cited by physicians as reasons to repeat or reorder tests. In this light, physicians argue that measures curtailing their ability to order a diagnostic test comprise a blatant interference with physician autonomy.

The discussion outlined in the following paragraphs is between cardiac specialists and researchers who are employed within this particular health system and are therefore considered to be part of the system’s core staff. The physicians, the hospital administrators, and the insurer are the three major players in this negotiation. As in our last case vignette, all the names of people and organizations have been changed here.¹

6.2 The Meeting

On a rainy summer day in 2009, Mr. Jones hosted high-level talks between his health system’s cardiac specialists and a major insurer. The agenda outlined a number of presentations; among them was a PowerPoint presentation from the insurer introducing new performance criteria and programs, followed by a presentation by one of the health system’s cardiac specialists summarizing findings from the health system’s own patient population. Time was also set aside to discuss a series of studies comparing guidelines of appropriateness criteria for cardiac testing. First to present was Dr. Peter Morrow, a medical director at Bigtoe Insurance. He introduced results from a cardiac imaging program intended to curtail inappropriate testing. The goal of this program, explained Dr. Morrow, is to try to maintain or improve quality for patients. He maintained that Bigtoe Insurance is looking at clinical performance in terms of measurable cardiac outcomes. He also stated that a major goal is to try to maintain satisfaction among physicians and patients while eliminating inappropriate testing. Dr. Morrow reported that the insurer’s approach to quality is to have heart specialists undergo precertification initiatives. For the health system, these initiatives imply that their cardiac specialists may need to take exams in order to prove a certain level of competence.

This program was not entirely novel. Bigtoe Insurance had introduced precertification requirements for nuclear stress testing, CT cardiography, and magnetic resonance imaging (MRI) in the past. But now they were also proposing to expand the program to encompass precertification for physicians ordering echocardiography, stress echocardiography, and diagnostic left heart catheterization. Dr. Morrow explained that the rationale behind the move was to utilize new criteria developed by the American College of Cardiology (ACC) and the Cardiology Foundation. He then posited that recent literature indicates that anywhere between 11% and 22% of testing ordered by physicians may be inappropriate, a percentage considered both too high and wasteful. To address this, Dr. Morrow reported that the insurer had (p. 96 ) started to collaborate with the ACC to craft a proposal that would address inappropriate echocardiography use, particularly the repeat echocardiography ordered for patients within one year of initial testing. In summary, Bigtoe Insurance was delicately trying to introduce a new protocol whereby they could identify a group of physicians as “high-performing” cardiologists within Mr. Jones’ health system. Those physicians labeled as being “high-performing” would not have the additional burden of going through a precertification process.

Next, Dr. Morrow detailed some of the criteria of how and under what circumstances Bigtoe Insurance identifies doctors and laboratories that meet their basic quality criteria. First, the doctors would have to be participating with Bigtoe Insurance already. Additionally, the laboratory where the echocardiology occurred would have to be credentialed under a recognized national accreditation program. He also mentioned that the practicing doctors needed to agree with the appropriateness criteria that had already been developed with the ACC, and if they were not going to be subject to precertification they would have to use some other method to justify the appropriateness of the echocardiogram that they were ordering. Individuals selected at higher-performing centers would not have to register the echocardiograms, but they would be asked to voluntarily enhance data collection and participate in a long-term study examining decisions to repeat any echocardiograms. Dr. Morrow contended that such data would allow the insurer and the ACC to better analyze all the usage patterns related to echocardiography.

In response to this presentation, the hospital’s leading physicians articulated a number of questions and comments. First to speak was an “in-house” specialist, Dr. Angelo DiNardo, who wondered just how objectively the insurer, or a consultant to the insurer, could conduct the precertification process. Dr. DiNardo spearheads some major quality- and process-improvement programs and manages an institute that conducts cutting-edge research within the health system. His colleague, Dr. Conrad Shultz, who oversees cardiovascular services for facilities across the entire health system, acknowledged that Bigtoe Insurance’s program was going in the right direction. He even mentioned that the program should encompass the staff interpreting the echo, adding that those qualified to do so were already subject to a rigorous fellowship program and a series of exams. He offered that Bigtoe Insurance could utilize the results from such a fellowship program.

To that point, Dr. Morrow said that there had been significant discussion between the insurer and the ACC, since the insurer had originally proposed a panel of board-certified “interpreters” that would assist in deciding which doctors may bypass the program altogether. Expanding on this point, Dr. Morrow’s colleague Dr. Carl Patel explained that Bigtoe Insurance had partnered with health consultancy firm Balloon Consulting to generate reports on physician performance as far as appropriateness criteria is concerned. Mr. Jones interjected that Balloon Consulting had been the recipient of criticism for not giving adequate and transparent feedback to explain their methods, data evaluation, and communication with physicians. To this point, Dr. Patel emphasized that in setting up the program, Bigtoe hoped to instill a sense of trust among the doctors who enrolled. However, it became clear that the specialists at the table were not especially comfortable with the details of the new program.

(p. 97 ) Dr. Conrad Shultz took a moment to relay an anecdote about how he was denied a test on a patient, despite having a legitimate reason to order the test. He suggested that he and his cardiology colleagues in the health network already faced difficulties in reordering necessary tests, and he feared that retesting was often disallowed by an insurer for arbitrary reasons:

I had a patient that had a dissecting aneurism of the aorta fix, and an aortic valve replacement that I’ve been following for several years. The last echo I had on her was more than two years ago, and I decided to get an echo to see if the aorta was expanding. And, unfortunately, six months ago, her internist sent her somewhere to get an echo. And that echo, I had the report, it described a normal aortic valve—not even a prosthetic valve! So, how could I possibly rely on the information? Balloon Consulting denied the echo because she had one six months ago, so what recourse [do I have] to that?

In response to this story, Dr. Patel commented that the program Bigtoe was introducing was intended not to question the decisions made by every doctor, but to figure out which doctors should be designated “high-performing.” He explained further:

In the bypass program we proposed no denials of anything, they [physicians] would tell us why they are getting an echo, and they would tell us the other echo doesn’t appear to be accurate. That would appear to be a legitimate reason to say: okay that’s why you’re getting another echo.

Dr. Patel went on to say that Bigtoe Insurance hopefully satisfied the concerns of the ACC by making the transparency of their own appropriateness guidelines evident to all practicing cardiologists. He said that the insurer had worked out specific algorithms surrounding echocardiology, stress echocardiology, and the diagnostic catheterization certification program. At the time of the meeting, Bigtoe Insurance was engaged in ongoing discussions with the ACC to formalize data sharing, particularly on echocardiology. Mr. Jones commented that Bigtoe Insurance should be commended for taking the time and effort to explain their methods to specialist staff within his health system, adding that most insurers were not as forward-thinking. He then handed the floor over to one of his health system’s cardiologists, Dr. Janet Dodd, who was presenting some of her team’s own data.

Dr. Dodd, who is in charge of nuclear cardiology at one of the health system’s specialty hospitals, started off her presentation by handing out a document comparing appropriateness criteria proposed by Bigtoe Insurance with that put forward by the ACC. She stated that the most recent version of the ACC’s appropriateness criteria was published on May 18, 2009, and that it had been expanded recently to guide the use of nuclear imaging. She then highlighted a study that supported her concerns about the lack of overlap of the appropriateness criteria proposed by the ACC with the appropriateness criteria being adopted by Bigtoe Insurance.

I just told you the number of criteria the initial document analyzed here honed in on 52 [guidelines]…. Current criteria will profile 66 [guidelines]. The overlap between (p. 98 ) those, and the original 50, and Bigtoe Insurance’s [criteria] stands at 27 [guidelines]. And if you read those carefully, if you take, for example, the first one—you will see that not only the wording, the language, [and] the profile is different, but the whole approach and premise is different. Why is that the case?

Dr. Dodd also cited a memo written by Dr. William Van Decker of the American Society of Nuclear Cardiology (ASNC) to the U.S. Senate Finance Committee on the topic of health reform (Decker, 2009). In Box 6.1, you can see an excerpt from this memo, which comprises mainly feedback to early government proposals for health care reform involving imaging utilization. In the memo, Dr. Decker, the chair for the ASNC Government Relations Committee, makes it clear that the ASNC is opposed to mandatory lab accreditation and the declaration of appropriateness criteria regarding imaging. Dr. Dodd said that while these words seem “harsh,” she believes the memo really does reflect the sentiments of most practicing cardiovascular physicians. She said this is the case because, in many ways, cardiologists see guidelines, particularly those laid out by insurers, as arbitrary. She argued that the Bigtoe guidelines were not something that practicing physicians could abide because the specialists believed the criteria failed to specifically include the multitude of clinical situations that require cardiologists to reorder tests.

Dr. Dodd then presented research she and her colleagues had conducted on appropriateness criteria and patient outcomes. She highlighted different studies (p. 99 ) (some unpublished) and gathered data with her colleagues on whether patients classified as inappropriate were in fact not deserving of the medical evaluations garnered through various imaging tests. She said that many of her findings were in line with published results from the Mayo Clinic. “An appropriateness criterion is a tool,” she cautioned. “You want to make sure that a tool is clinically relevant.” Dr. Dodd based her analysis on data pulled from her patient population. It showed that patients who had undergone tests deemed inappropriate by existing guidelines did have the lowest cardiac event rate. However, a small percentage of this population did experience a cardiac event shortly after being ruled as an inappropriate candidate for testing.

Inappropriate patients had the lowest event rate, it was a 2.4%, and certain patients came in the middle at a 7% [event rate]. So, as a practicing cardiologist, anything that exceeds 2% for us is a significant event rate. [The] 2.4% [event rate] is essentially our ground floor, and anything that gets above that is a serious event rate.

Dr. Dodd was making this point to demonstrate that she and her colleagues were not comfortable with seeing a higher percentage of patients labeled as inappropriate patients for testing. The 2% that would experience a later cardiac event was too much to accept. To this point, Dr. Shultz added his interpretation of the findings. He said that the kind of patient that he and his colleagues attract to their laboratory (which is also integrated into the health system) is very different from the patients seen by someone in a private practice. He maintained that a lot of patients are put through prescreening before they are seen by his staff, so they come in with known heart problems. He added that applying “blanket criteria” across a broad spectrum of patients (e.g., those who have already been prescreened versus those being seen for the first time) is, in his eyes, dangerous. Like Dr. Dodd, Dr. Shultz fears that people labeled as inappropriate by guidelines, or those denied coverage, may very well be good candidates for screening. Dr. Dodd added that her team was still investigating the reasons why appropriateness criteria were not sensitive enough to pick up patients who would otherwise be deemed appropriate by a specialist, or a patient who may go on to have a cardiac event despite having been ruled as inappropriate for certain screening tests.

As the specialists made their case for not being strictly tied to criteria that might curtail their use of screening tests, Mr. Jones, the hospital’s negotiator, was intent on highlighting the research his hospital was able to produce. He summarized this to the insurer in a comment he made as Dr. Janet Dodd finished her presentation:

What I think what we are trying to do is, [we are] trying to propose a way of our doing business in this area with you, if you are open to it… there is an enormous amount of content and research and depth in our organization. It’s to show you that we can control inappropriate use of this kind of testing.

Mr. Jones concluded his argument by reporting that he was proposing his health system’s doctors and specialists should self-manage as far as appropriateness criteria (p. 100 ) were concerned, while keeping Bigtoe Insurance in the loop as far as data gathering on physician performance was concerned. He added that self-managing appropriateness criteria would allow the hospital system to attract more specialized doctors. Before wrapping up the meeting with some final comments, Mr. Jones did take a moment to digress on the topic of health consultants that regularly advise and inform insurers about projected health outcomes and performance as far as appropriateness criteria are concerned: “The consultants come in and they have very bad data… and they do not have the vested interests of the patients in mind.” Mr. Jones made these comments perhaps as a cautionary note to Bigtoe to not rely too heavily on data gathered by Balloon Consulting.

Mr. Jones proposed several new programs and price-related negotiation tactics aimed at pending renewals of various contract deals with the insurer. He spoke of how his health system had sought to bring more doctors on board, and how it was reducing prices and adding wellness programs for patient-customers (and the employers of patient-customers) hoping to engage in more preventative care. To these comments, a representative of the insurer asserted that Bigtoe Insurance and the health system did indeed share a number of “goals” and “common ground.” Shortly thereafter, the meeting ended with a promise to continue talks on different program proposals to be discussed among special task forces consisting of the insurer, the physicians, and hospital administrators.

6.3 A “Stress Test” for Recognizing Potential Alignment

This case vignette is a moment in the life of several real health care stakeholders. The situation is a familiar one and is easily generalized for a wider understanding of stakeholder discussion. One of my graduate students summarized the literature that relates heavily to the above case vignette quite succinctly. We can distill the situation down to the bare bones of the literature that best summarizes the conflicts of interest these types of stakeholders face regularly. The following synopsis captures the concerns of the different stakeholders well; it also makes some assumptions that not everyone may agree with. This is simply one way of summarizing the stakeholders’ perspectives; it is not the only way. We use this summary because it relates to current literature. My colleagues and I have since adapted it to use in a stakeholder role-play exercise that complements the case vignette.

Disease of the heart is said to be the leading cause of death in the United States (Heron et al., 2009). Recent advances in technology, both old and new, have contributed to the cost of care of cardiac patients. Tests that make use of echocardiology, cardiac CT, myocardial perfusion imaging, and cardiac MRI have shown varying degrees of accuracy in diagnosis and have therefore prompted questions about how and when these tests are used, what other tests they are used in conjunction with, and how they are ordered and interpreted (Hendel, 2008). Insurers and purchasers worry about the rising costs of newer, more expensive diagnostic tests (Glabman, 2005). They are also concerned that some tests may be duplicated or redundant when ordered indiscriminately, and that quality controls may be low (p. 101 ) among some imaging centers (Glabman, 2005; Seidel & Nash, 2004). More-recent literature has highlighted the theoretical effect of repeated and high exposure to radiation, and the possibility of follow-up procedures that may encourage unnecessary therapeutic interventions (Brenner & Hall, 2007). More generally, researchers have begun to focus attention on how intensive medicine may add little health benefit or even detract from the patients’ health while adding great expense to care (Fisher, 2003; Fisher, Goodman, Skinner, & Bronner, 2009; Fisher et al., 2003a, 2003b).

Separately, and most alarmingly, the news media has picked up on pockets of cardiologists that demonstrate unusual interventionist trends and decision patterns, which stray from evidence-based medicine. In some of the worst-case scenarios, a handful of cardiologists demonstrating unusually extreme trends in practicing a high number of interventionist procedures have been investigated, sued, or even arrested by federal authorities. In one well-publicized case, a Louisiana doctor was charged with defrauding Medicare after having been accused of performing unnecessary procedures (Abelson, 2006).

Even when obvious fraud is not suspected, the news media has not let recent scholarly findings go unnoticed. Negative press is damaging not only to the specialists in question, but also to the hospitals where they practice, as well as to the field of cardiology as a whole. Patient–provider trust may be damaged or, worse still, irrevocably eroded. For example, the New York Times ran an article describing a group of Ohio cardiologists that appeared to perform angioplasties at four times the rate of the national average (Abelson, 2006). The information came to light following reports published by the Dartmouth Institute for Health Policy and Clinical Practice (2009). In recent years, the Institute’s oft-quoted reports have been referred to simply as the “Dartmouth Data” and have highlighted the eye-opening discrepancies in trends of care across the United States. The Dartmouth Data tease apart the reasons behind obvious differences in the patterns of care and have recently zeroed in on what has been termed as differences in “local ecology of health care” and the “social norms” among physicians as two of the leading factors guiding decisions in care (Fisher et al., 2009).

It is practically indisputable that the practice of cardiology is one of the most lucrative areas for hospitals and networks of providers and physicians. Investing in and promoting new technology has allowed many doctors to administer fast, seemingly definitive tests that tell patients, even asymptomatic ones, whether they may be at risk for a coronary event. The tests have great appeal to patient-consumers interested in thwarting a potential health problem because they are not only noninvasive, but also fast. At least one study has shown the latest version of the CT scan (the 64-slice scanners, which capture an uninterrupted image of the beating heart) to be accurate in giving definitive results faster than conventional stress testing among low-risk patients complaining of chest pains (Raff & Goldstein, 2007). Because of the noninvasive nature of the test and the compelling imagery decipherable to even the layperson, some suggest that tests such as a CT scan of the heart may also give patients better incentives to comply better with physician advice (Berenson & Abelson, 2008). Some physicians may even conclude that the latest (p. 102 ) CT scan can save time and money, aid patient education, and even save costs in the long run (Raff & Goldstein, 2007).

Despite this promise, the data on whether these tests can be seen as a value-added step in preventative medicine are hotly contested, and so is their use, especially when ordered in conjunction with older, more traditional tests (Ladapo et al., 2009). Hospitals may worry that patient-consumers will go elsewhere if they cannot offer the latest technology. Hospitals that find themselves in a favorable reimbursement environment might be more likely to invest in the technology (Berwick, 2003; Ladapo et al., 2009; Teplensky, Pauly, Kimberly, Hillman, & Schwartz, 1995).

Concerned that providers are pushing expensive technology that does not truly promote evidence-based medical decisions, insurers and the government have started to adopt guidelines and pay-for-performance schemes based upon physician and technician adherence to a set of established clinical guidelines (Swayne, 2005). Conflict often arises around which guidelines are appropriate, as different constituency groups promote different sets. Historically speaking, associations and institutes representing physicians and specialties such as cardiology have long published guidelines for their members; yet the joint cardiovascular practice guidelines of the ACC and the American Heart Association have come under increasing scrutiny. A recent study revealed that a large number of these guidelines (a little less than one-half) are based on evidence that is considered to be relatively weak (Tricoci, Allen, Kramer, Califf, & Smith, 2009). More pointedly, physicians and academics have criticized guidelines promoted by insurers and other purchasers as being only remotely fashioned after the most recent evidence-based medicine (Steinberg & Luce, 2005).

Societies that represent cardiologists and the technologies in question have vigorously resisted government and private efforts to impose guidelines or limitations on spending for these tests. A group of cardiologists recently sued Secretary of Health and Human Services Kathleen Sebelius over Medicare fee cuts for procedures such as echocardiograms and nuclear stress tests (Sternberg, 2009). Additionally, in response to fee cuts, the American Society of Echocardiography (ASE) ramped up its advocacy efforts to fight private-payer precertification for echocardiography. It is also seeking to preserve Medicare reimbursements for the diagnostic technology. In a statement on the ASE Web site, the Society thanked members for making generous contributions toward this effort (ASE, 2010).

Linking reimbursements to performance as measured by clinical outcomes has had mixed reviews as far as the literature is concerned. An early study on the influence of one pay-for-performance pilot project run by the Centers for Medicare & Medicaid Services (CMS) showed little to no improvement in quality of care for patients with acute myocardial infarction (Glickman et al., 2007). Subsequent studies have shown that pay-for-performance programs—and not only CMS pay-for-performance programs—have influenced notable improvements in other areas of medicine (O’Kane, 2007). Many other studies and literature reviews reveal that evidence on the matter is still scarce or inconclusive (Mehrotra, Damberg, Sorbero, & Teleki, 2009; Petersen, Woodard, Urech, Daw, & Sookanan, 2006; Rosenthal, Frank, Li, & Epstein, 2005). It is not surprising, therefore, that a hospital system and other types of providers may try to avoid certain aspects of the pay-for-performance (p. 103 ) programs. Negotiating contracts that specifically tie reimbursements to a set of guidelines could be simply putting a provider at a financial disadvantage. Finally, not all physicians unequivocally accept the conclusions of the Dartmouth Data without skepticism. Some would argue that the intensive medicine practiced in high-spending enclaves actually does extend life in chronically ill heart patients with many comorbidities (Langberg & Black, 2009).

Perhaps it is too early to know precisely how different approaches to medicine create or remedy inefficiencies. Despite the burgeoning interest in appropriate-care medicine, the literature and research targeting inequities, inefficiencies, and inappropriate care are not necessarily the drivers behind the appropriateness criteria. Thomas Jones, the health system negotiator from our case vignette, believes that, more often than not, the Dartmouth Data and similar research are misused by insurers to arbitrarily negotiate for lower reimbursements, without evidence of savings or better health outcomes. In more than one conversation that we had with Mr. Jones, he labeled pay-for-performance as “just another scam.”

The insurers may refer to the data as part of a drive to reduce reimbursements. And the hospitals that know what they are doing will use it as a way to increase their price.… There is an underlying criticism that doctors do too much, or that patients want too much because of defensive medicine. And if you took away defensive medicine everyone would get appropriate care. I’m telling you that we don’t know what appropriate care is.… I am going to fight every effort that limits care to people who aren’t sick, but who maybe [sic] sick. Because there is no one out there who definitively knows who should have that cardiogram, or that MRI, that surgery or that chemotherapy.

At this point in the exercise I ask a few volunteers to do a quick role-play simulation. I assign different groups to represent different stakeholders in a fictitious case vignette. Some represent hospital negotiators, others play a group of “in-house” cardiologists associated with a large network of hospitals, and some are tasked with representing the insurers. If the size of the group warrants it, I might also assign some volunteers to represent physicians working independently, the news media, government bodies, and patient advocacy groups. With a smaller group, however, I ask them to consider the value proposition of the following five influences, but rather than simply assigning them a certain value I ask them to prioritize the influences in order of importance. Sometimes I present the exercise like this:

As far as value creation is concerned, all of these influences are important. But as a stakeholder you have limited resources and time and must decide how to prioritize these items given that you are in the middle of a time-sensitive negotiation. These are the issues at hand: (a) better health outcomes, (b) how news media portrays your actions, (c) increasing coverage for treatment (which is also thought of as access), (d) patient, or “customer,” satisfaction, (e) adequate revenue for services rendered. (p. 104 )

Then I ask the participants to rank these in order of importance, with “5” being the strongest and “1” being the least important. No item can be given the same importance as any other in this exercise.

Figure 6.2 shows the result when I ran this simulation with a few of my colleagues playing the roles of different stakeholders. I would point out that this is an exercise of role-play, not the mapping of a real-life scenario. The first observation from the result is that there is very little overlap. Also, the overlap may be skewed, reflecting the various priorities of the different stakeholders. Overlap is represented by the shape at the center of the diagram. By using spider plots we have come across a unique visual effect. It is compelling because this mapping exercise lets each stakeholder see just how far away he or she is from alignment with other stakeholders and their priorities. Often, I find that a stakeholder may hold on too tightly to a particular agenda that he or she may feel must be prioritized above all other issues; or perhaps a stakeholder may not be able to “see” clearly the priorities of another stakeholder, making negotiations difficult due to a certain level of misunderstanding.

Past literature on stakeholder theory helps managers think strategically about how powerful and influential other stakeholders are in different scenarios. Saliency modeling also helped them become aware of those who may be smaller and therefore less obvious, but nonetheless influential, entities (Savage, Nix, Whitehead, & Blair, 1991). One book published in 1990 guided readers to create lists of various stakeholders. In different instances, the authors showed how stakeholder groups were varied and diverse, and the book offered comparisons of one type of hospital ecology to another (Blair & Fottler, 1990).

Other work on stakeholder alignment, such as that mentioned in Chapter 5, has concentrated more on saliency. Stakeholder saliency in business parlance has come to mean evaluating three elements: power, legitimacy, and criticality (Mitchell, Agle, & Wood, 1997). In many ways, stakeholder theory has stressed the importance of identifying the key stakeholders and thinking strategically about handling these stakeholders and their possible influences. Here, I want to revive the basic spirit in which stakeholder theory was first conceived by Edward Freeman and colleagues, one that assigned a normative and prescriptive aspect to stakeholder behavior.

(p. 105 ) Freeman et al. (2007) spoke of a “philosophy of voluntarism” in which stakeholders self-manage, without the hand of government or other large entities overseeing them. They also called for an “intensive dialogue” between stakeholders, and for the generalizing of the “marketing approach.” Among other things, they said that “a constant monitoring of processes” was needed to better serve all the stakeholders (Freeman, Harrison, & Wicks, 2007). Value is inherently knotted into the stakeholders and the stakeholder relationships, but accessing that value is challenging because many stakeholders cannot see past their conflicting priorities. Furthermore, smaller stakeholders may be at the periphery of the Gordian knot, but that does not mean they would not be able to derail the value-creation process.

We like to refer to a definition of “a stakeholder” put forward by Grossi (2003), a student of lean enterprise thinking and stakeholder theory whom we mentioned earlier in this chapter. His work centered on analyzing stakeholders from a “systems dynamics approach.” This is an approach that incorporates a global analysis of a particular ecosystem. In the business and science worlds, it has often been applied to analyze how various elements and entities may influence one another. In his Master’s thesis, Grossi (2003) looked at how to define stakeholders and the concept of value within a business environment that takes into account all the potential contributors, large and small. His definition is below:

[A] stakeholder is any group or individual who directly or indirectly affects or is affected by the level of achievement of an enterprise’s value creation process.

In this book, we relate the notion of stakeholder value and the value-creation process back to the health care environment. Here is our definition of value from a stakeholder perspective:

A stakeholder is a group or an individual who directly or indirectly influences health outcomes in striving to create value for the patient.

we want to be clear that the patient is not excluded from this definition. Indeed, the patient can be a very active stakeholder who may choose to derail the whole value-creation process, or augment it, by pursuing or shunning a healthier lifestyle or adhering to sound advice from health care providers. In the next section, we focus on the nature of value creation between health care stakeholders and how it can be augmented by Lean principles. If appropriated deliberately, Lean principles can help health care stakeholders align, or simply see the possibility for aligning their priorities.

6.4 Stakeholder Theory in the Context of Lean Enterprise Thinking

Negotiations between stakeholders (and the decisions that eventuate from the discussions) have the potential to either augment or erode value. Having clearly defined guiding principles can be useful in deciding whether certain decisions, attitudes, (p. 106 ) priorities, and approaches can have a positive or negative impact on creating value. When considering Lean enterprise, one has to be cautious, because often the audience may have very strong, preconceived ideas about what constitutes “Lean.” Many people may associate “Lean processes” with uncomfortable change, cutbacks in staff, and expensive consultants. For example, numerous hospitals across the United States have hired consultants to expedite the processing of patient admission to busy emergency departments. Some of these efforts have incorporated Lean analysis and proven successful. Some, however, have not been successful, depending on how the process changes were instituted and whether the stakeholders involved were actively brought into the efforts. Here, we are not going to discuss specifics of process improvement influenced by Lean thinking, which involves careful calculations and the continual study of the situation. Instead, we are taking a hard look at the philosophy behind Lean thinking when it comes to seeking stakeholder alignment, especially in the health care environment.

Classic Lean thinking incorporates ideas that are perhaps best summarized by James P. Womack and Daniel T. Jones, who co-wrote a series of books explaining the seemingly miraculous success of the Toyota Production System pioneered by Taiichi Ohno (Ohno, 1988; Womack & Jones, 1996; Womack, Jones, & Roos, 1990). To summarize (and to simplify slightly), they proposed that an enterprise engaging in Lean thinking was to first establish and specify “value” from the customer’s perspective. Secondly, enterprises and stakeholders are to identify and map out the steps that exist in the creation of value. Thirdly, these steps need to come in fast sequential order, and wasteful steps that do not add value need to be eliminated. Fourthly, customers need to be able to “pull” value from the producer. Finally, the process is perfected and repeated, with careful attention paid to wasteful steps and the pulling of value.

A simple circular diagram, which we interpret and reproduce in Figure 6.3, summarizes this breakdown, but this depiction can also be accessed in its original form on a Web site created by author James P. Womack to promote his Lean Enterprise Institute (The Lean Enterprise Institute, 2009). We want to emphasize that what we might distill from this diagram is where and how stakeholders can seek alignment with these Lean principles. In Chapter 1, we quoted Professor Earll Murman and his co-authors, who mentioned how adding value is only possible when there is mutual agreement, which may be either tacit or implicit (Murman et al., 2002a).

Getting started on true value creation in the world of health care is difficult. This is where a few tried-and-true Lean thinking tools can help. Borrowing from the five Lean principles diagrammed in Figure 6.3, we are posing the following six questions to health care stakeholders in any given scenario or situation:

1. 1. Can the health care stakeholders establish and specify “value” in terms of outcomes or a better quality of living for patients?

2. 2. Are the stakeholders able to identify and map out an agreed-upon route with specific steps toward the previously agreed-upon value?

(p. 107 )


Figure 6.3 Classic Lean Enterprise thinking. Adapted from “What is Lean” by J.P. Womack, 2009, The Lean Enterprise Institute. Principles of Lean. Retrieved August 26, 2010 from http://www.lean.org/WhatsLean/Principles.cfm.

3. 3. Are there barriers to these steps for one or more particular stakeholders (e.g., high costs, market-related barriers, legal and privacy barriers, etc.) that might prohibit a clear direction (e.g., the flow)?

4. 4. Can the stakeholders agree upon what steps may be wasteful and therefore need to be eliminated?

5. 5. Can each stakeholder group help patients better manage their health in the short and long term through better education and the coordination of care delivery?

6. 6. Are there reliable benchmarks that can be used to gauge health improvement in terms of patient well-being, morbidity and mortality outcomes, cost savings, participation, and feedback from the patient’s side? (This last question encompasses the Lean principle of “pull,” which seeks to let the customer [in this case the patient] add value to care.)

If we take a moment to think back to our case vignette, we can see how the stakeholders obviously had different motivations; yet their priorities were not too far from being closely aligned. In our role-play exercise, alignment seemed a distant possibility. This is something that we established by mapping out the fictitious stakeholders’ priorities. However, if we look back at the real stakeholders in the case vignette, they were constantly pursuing continued communication and were dedicated to finding common ground despite the strong differences in their approaches to creating value. If we consider that value for the patient can be found in improved health outcomes, and perhaps also in a better quality of life through decreased morbidity or improved chronic disease management, then all the stakeholders in our exercise may align themselves easily.

When we engage in a scenario of stakeholder role-play, it becomes clear that there are many paths up the same mountain; I have run this role-play exercise where all the volunteers prioritized health outcomes, for example. An interesting discussion ensued, with observers in the classroom saying that while everyone desires (p. 108 ) improved health outcomes, they saw no immediately actionable items in such a time-sensitive scenario that would really impact outcomes. Another observer commented that outcomes are particular to the population of a hospital or the community it serves, and therefore data on outcomes may be inconsistent or easily influenced by outside factors that providers may not be able to control. With that particular group of role-playing volunteers, I would say that there seemed to be great potential for creating long-term alignment in creating better value for patients. But I have also encountered a group where all the volunteers wanted to preserve customer satisfaction as the utmost priority because they were mainly motivated by the fear of losing market share.

As we mentioned earlier, customer satisfaction does not necessarily move in tandem with better patient outcomes. In such a situation, more long-term thinking needs to be applied, and a deeper analysis of customer satisfaction needs to be undertaken. Does this mean higher satisfaction among an increasingly dwindling group of individuals with excellent insurance coverage? We know from our literature review in the earlier chapters that an intensive, invasive, and even defensive type of medicine may satiate the demanding customer-patient—but this does not bring better health outcomes to the wider population, and it increases costs for all patients and all other stakeholders. Generally speaking, purchasers and the health insurance companies have not been effective at reigning in rising health costs while also ensuring improving outcomes (although there are some exceptions to this generalization, which we will discuss in the next chapter).

For a moment, we want to zero in on one key element of Lean enterprise thinking, and that is what is meant by the “value proposition” between stakeholders, summarized by Professor Murman and colleagues (2002b). With this proposition comes a question that begs an honest and aligned view of “value” among the stakeholders: does every stakeholder have the “sufficient capability” to succeed at every agreed-upon task at hand? Understanding a stakeholder’s “capacity” is a major step in arriving at the value proposition (Murman et al., 2002b). Oftentimes, mistrust between stakeholders comes not only from negative experiences of past interactions, but also from assumptions about fellow stakeholders’ present and future ability and capacity to fulfill a particular promise. Because doubts enter the fray about long-term planning and ability to commit to long-term planning, stakeholders are likely to focus on the short-term ideas on cost-cutting measures and boosting satisfaction.

Let us relate this back, once again, to the “Tragedy of the Commons” put forward by Garrett Hardin (1968). In the game-theory scenario suggested by Hardin, herdsmen share a common pasture. We would find that the difficulty of managing this common resource arises not because the herdsmen do not recognize the value in sharing the pasture; quite the contrary—they recognize that sharing the pasture promotes sustained growth. The difficulty arises because there is a temptation to “cheat,” and with this there is also mistrust. This mistrust arises because there is a chance that cheaters will initially be rewarded by higher profits if they add more cattle and then exit the game by selling their entire herd, thus escaping the consequences as the pasture collapses from overuse.

(p. 109 ) We want to emphasize that trust is an element in the sustained management of a common resource, but arriving at a point where trust is possible is the main challenge faced by stakeholders in health care (Shore, 2005, 2007). We are not asking stakeholders to blindly trust in one another. A major tenet of Lean thinking is continued and sustained interaction, communication, and verification (Murman et al., 2002b). This goes beyond simply building trust with other stakeholders—we are advocating that stakeholders do their homework. They must stay aware of the trends and developing values of their fellow stakeholders (Murman et al., 2002b; Shore, 2007). This requires a learned competency of fostering deep and sustained information exchange between stakeholders—even among stakeholders who have seen themselves as competitors in the past.

In the next chapter, we present another case vignette, in which a large purchaser is working very closely with other stakeholders not only to establish a great deal of trust, but also to control health care costs through the fostering of preventative medicine and a healthier lifestyle among its beneficiaries. Toward the end of the next chapter, we evaluate the value proposition among the stakeholders involved.