A Targeted Approach to Reducing Overutilization: Use of Percutaneous Coronary Intervention in Stable Coronary Artery Disease

Introduction

I mprove quality, reduce costs, and manage the growing burden of chronic conditions; these are lofty goals for an already ailing US health care system. A fundamental challenge underlying many of the system's woes can be summarized succinctly as too much care and too little value. This is particularly true when our system is viewed from a global perspective. The United States spends more than any other developed country on health care (almost 50% more per capita on an adjusted basis), but these dollars spent do not translate into better health outcomes. ¹

One major factor that contributes to excessive health care spending in the United States is overutilization. Overutilization can manifest in 2 ways: (1) a higher volume of care than is necessary (eg, office visits, hospitalizations, tests, procedures, prescriptions), or (2) a more costly mix of care than is necessary (eg, use of specialists, tests, procedures, prescriptions). ² In essence, this judgment encompasses the use of unnecessary care when more appropriate alternatives would have produced similar or potentially better outcomes.

The Centers for Disease Control and Prevention (CDC) has identified a number of factors that could influence increased health care utilization. These factors include: (1) new procedures and technologies, (2) consensus guidelines that recommend increases in utilization, (3) new drugs, (4) changes in practice patterns, and (5) changes in consumer preferences and demand. ³ Emanuel writes that, “US culture emphasizes the new and the fancy; old and plain are equated with deprivation,” noting that patients also equate more testing and treatment with better care. ²

Increasing fragmentation of care, defensive medicine due to medicolegal concerns, and marketing directly to physicians and consumers encourages use of more tests and procedures. ^2,4 Patients and physicians are further hindered by lack of price transparency, unaware of the true cost of an episode of care. These factors, coupled with financial incentives inherent in a third-party, fee-for-service system, help illustrate how overutilization has become ingrained in the health care system.

Overutilization is evidenced by variation in how the same condition is treated in similar populations. Although care must be personalized for the patient, there are well-documented patterns of unexplained variation in health care practice. ⁵ This is a clear indicator of the gap between health care practice and the evolving science of evidence-based health care.

Emerging models of care (eg, accountable care organizations) aim to reduce waste, enhance coordination, and promote shared decision making among key stakeholders, including the patient. When given the opportunity to become actively involved in the decision-making process, patients often select more conservative treatments. Factoring an informed patient's preferences into a care plan could have a positive impact on both clinical outcomes and cost of care. ⁶

Understanding how to transform the traditional patient into an informed consumer of health care services is a key strategy to address overutilization.

A pervasive issue such as overutilization is best addressed by identifying and focusing on specific areas in which overutilization is common. Interventional cardiovascular treatments, including coronary artery bypass grafts (CABGs), angioplasties, and revascularization with stents, have been identified as target areas for improvement. ²

Identifying a “Target”: Percutaneous Coronary Intervention Use for Stable Coronary Artery Disease

The high incidence of cardiovascular disease in the United States presents an ideal opportunity to evaluate the appropriateness of treatment and to highlight potential overutilization. All patients with coronary artery disease (CAD) are managed with medications and lifestyle changes. When clinically indicated, multidisciplinary outpatient cardiac rehabilitation is effective in promoting lifestyle change and has been shown to reduce morbidity and mortality when compared with usual care. ^{8 –10} It is increasingly recommended in recent evidence-based treatment guidelines for patients with CAD. ¹¹ In appropriate clinical settings, interventional procedures such as percutaneous coronary intervention (PCI) and CABG surgery also can reduce morbidity and mortality by revascularizing the heart muscle by opening or bypassing a blocked vessel.

Guidelines for the management of patients with chronic stable angina are evidence-based recommendations developed by the American College of Cardiology and American Heart Association Task Force. The goals are (1) prevention of heart attacks and death, and (2) control of angina symptoms that may limit typical daily activities. ⁷

The initial treatment of low-risk patients is medication—optimal medical therapy (OMT)—and lifestyle modification, elements that have been demonstrated to reduce the risk for cardiac events. All patients should be counseled about the importance of behavioral risk reduction, including smoking cessation, weight management, attainable exercise goals, blood pressure control, lipid management, diabetes management, and management of any other coexisting conditions. Stable angina symptoms are controlled primarily with nitroglycerin, followed by the addition of beta-blockers and/or calcium channel blockers, and long-acting nitrates as indicated. ⁷

From a value-based (ie, quality/cost) perspective, the decision to move from OMT and lifestyle modifications to an interventional procedure should be based on the severity of the symptoms and the degree of risk, with patient participation in the decision-making process whenever possible. Higher-risk patients include those with more advanced vessel disease (eg, left main, 3-vessel, or 2-vessel CAD with significant stenosis, and possible involvement of the proximal left anterior descending coronary artery) and unstable angina. These situations are governed by different recommendations compared to the stable patient, with PCI and CABG as potentially life-saving procedures. PCI is less likely to be of benefit in low-risk situations, such as chronic stable angina, because this intervention has not been shown to prevent a potential heart attack, stroke, or death. ¹²

Because of the advances in safety and efficacy over the last 30 years, PCI has become a common first-line procedure for the management of CAD in the United States. ¹² Estimates indicate that the procedure is performed 600,000 times in the United States each year at a cost of over $12 billion. ¹³

Results from the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial, published in 2007, raised questions about the value of PCI in stable CAD, with the finding that adding PCI to OMT does have some short-term benefit in controlling symptoms, but does not reduce mortality and other major cardiac events when compared to OMT alone. Initially, there was a statistically significant difference in angina control between the 2 groups: 7% in the first year and 6% in the third year. By the fifth year of the study, the difference was no longer significant. ¹²

Despite these findings, both patients and physicians are inclined to overestimate the benefits of the procedure in low-risk situations. Studies supporting this theory show that, even among patients with little or no angina, 71%–88% believe that PCI reduces the risk of death. ¹

In 2009, the American College of Cardiology Foundation published Appropriate Use Criteria (AUC) for cardiac revascularization, including both PCI and CABG, based on current clinical evidence, including the COURAGE trial, clinical practice guidelines, and expert opinion. These criteria were designed to translate evidence into a national guide for evidence-based decision making in routine clinical practice and represent an evolutionary movement that brings together multiple guidelines to address most clinical scenarios, including those for which there is no clear evidence base. The AUC have been incorporated into the National Cardiovascular Data Registry (NCDR), which contains data contributed by about 80% of US hospitals.

The first report on performance against AUC was published by Chan et al in July 2011. ¹³ Their analysis of NCDR data for 500,000 PCIs revealed that the majority were done in acute situations for appropriate indications. However, the data told a very different story for the 30% of PCIs done for stable CAD in the non-acute setting: only 50% of these PCIs could be clearly classified as “appropriate.” Of the remaining, 38% were classified as “uncertain,” meaning that there is currently insufficient clinical evidence to determine whether PCI was appropriate in a given situation. The study also reported that 12% of the PCIs done in these non-acute situations were classified as “inappropriate.” ¹³ The overwhelming majority of these patients had few or no symptoms, were not receiving OMT, and had low-risk findings on noninvasive tests. ^13,14

Chan et al also noted significant variation between hospitals, with the percentage of inappropriate PCIs ranging from a low of nearly 0% to a high of 50%. This study supports the assertion that PCI is overutilized in chronic stable CAD, and the degree of variation between hospitals highlights an important opportunity to examine the overutilization of non-acute PCI. ¹³

This type of unexplained variation has been well documented by the Dartmouth Atlas of Health Care, which cites financial incentives and the correlation between high numbers of physicians and overutilization.

A meta-analysis published in the Archives of Internal Medicine in February 2012 reexamined comparative outcomes in stable CAD with consideration for the impact of improvements in both PCI and OMT that might call previous reports into question. ¹⁵ This analysis notes that subsequent guidelines expand the definition of OMT to include statins, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers, in addition to aspirin, nitroglycerin, beta-blockers, and calcium channel blockers, while PCI also has evolved from balloon angioplasty to stent placement. ¹⁶

When including only studies that reflect current medical practice, the authors found no significant difference in mortality, nonfatal myocardial infarction (MI), unplanned revascularization procedures, or angina over a mean follow-up period of 4.3 years. Improvements in OMT have increased the proportion of patients who are angina free, and this was felt to be the reason for the absence of the short-term benefit previously seen in the COURAGE trial. Of particular note is that 70% of patients initially treated with OMT remained on OMT alone and did not require a subsequent revascularization procedure. This marked difference in the total number of revascularization procedures raises questions about the absolute benefit of lower revascularization rates associated with drug-eluting stents. ¹⁵

In fact, from an overall resource utilization perspective, both the Boden ¹² and Stergiopoulos ¹⁵ studies clearly illustrate the excess cost of relying on PCI as an initial therapy in stable angina (Tables 1 and 2).

In some instances, PCI is used when CABG may be a superior therapeutic alternative. A 2010 study showed that in hospitals with PCI but no CABG capability, patients received more recommendations for PCI and fewer for CABG than were indicated according to clinical guidelines. ¹⁷

Yet another factor in overutilization of PCI is a logistical one. The trend toward “ad hoc” PCI, in which PCI is done at the same time as the diagnostic study, limits the opportunity for multidisciplinary input and meaningful discussion of treatment options with the patient. Although “ad hoc” PCI eliminates the need for a second procedure, a recent publication calls for a “pause” between a diagnosis and treatment whenever clinically feasible. This pause allows time for multidisciplinary input and shared decision making based on a clear understanding of the risks and potential benefits of treatment options for an individual patient once the coronary anatomy is known. ¹⁸ The pause serves to avoid misperceptions on the part of the patient. In the case of PCI, research indicates that most patients perceive it as a 1-time procedure that reduces the need for medication when, in reality, stringent adherence to a medical regimen that includes 2 types of antiplatelet therapy is critical to achieving the benefits of PCI. ¹

Patient-Related Factors

There is no question that the stakes are high with heart disease. Patients diagnosed with a heart condition may feel overwhelmed and may have difficulty understanding their treatment options or the difference between stable and unstable angina. In a large, single-center survey of 8000 patients admitted to an internal medicine service, nearly all patients wanted to engage in a dialogue with their physicians about treatment options. However, two thirds of patients preferred that their physician make the ultimate choice. ¹⁹ At this critical point, it is important for the physician to consider questions that patients (and their families) have while balancing the need for comprehensible information. Several studies have concluded that outcomes improve when patients play a significant role in the decision making. ¹

Provider-Related Factors

Factors beyond symptom relief and prevention of heart attack and death play a role in the overutilization of PCI. Lin conducted a focus group study to explore physician-related factors that influence overutilization of PCI. ⁴ Both primary care physicians (PCPs) and cardiologists cited medicolegal concerns as the main driver of a bias to “do something,” believing that the risk of procedure-related complications was more acceptable than the risk of failing to intervene. In addition, both salaried and fee-for-service cardiologists tend to believe that opening a blocked artery is still beneficial, even with the knowledge that PCI does not prevent MI or death. ⁴

Direct-to-consumer marketing of cardiac screening tests adds another layer of complexity to primary care decision making. PCPs cited fear of missing the “widow-maker” when consulted by an asymptomatic, low-risk patient who self-referred for noninvasive screening and is now overly anxious about an unexpected abnormal result. PCPs also noted patient preferences for “less invasive” procedures as a factor when choosing PCI over CABG, seemingly without considering that “noninvasive” OMT might be a more appropriate consideration. ⁴

Discussion

In non-acute situations, the choice to pursue PCI is elective. Current guidelines recommend that OMT and lifestyle modification form the cornerstone of first-line treatment for low-risk patients. When faced with treatment decisions, both physicians and patients should factor in the overall risks associated with PCI, which include MI, restenosis, and death. ²⁰ Failure to comply with strict post-procedure medical therapy greatly increases the risk of fatal MI.

Understanding the patient's level of health literacy is critical to an effective explanation of the diagnosis and treatment options. Health literacy is the degree to which individuals have the capacity to obtain, process, and understand basic health information and the services needed to make appropriate health decisions. ²¹ Patients may not understand that adherence to medical therapy will continue to be important or that PCI is not a “fix” that obviates the need for medication or lifestyle modification. In fact, PCI often increases the requirement for chronic medical therapy.

A comparison of outcomes associated with OMT vs. PCI as a first-line strategy in stable CAD demonstrates how inappropriate overutilization raises cost without a commensurate improvement in health outcomes. Table 1 uses data from the COURAGE trial to illustrate the cost associated with potentially unnecessary PCI in stable CAD.

In addition to the cost of adding a second antiplatelet agent (ie, clopidogrel) to an OMT regimen, the patients in the PCI group incurred the cost of 88 additional revascularization procedures (PCI or CABG) per 100 patients. Despite the dramatic difference in cost between the 2 groups, there is no significant difference in health outcomes (Tables 1 and 2).

This illustration highlights the potential value of examining the PCI selection process and developing additional decision-making tools for physicians who treat low-risk patients. ¹³

The best examples of quality improvement programs incorporate measurement and evidence-based guidelines into the day-to-day work flow of clinical practice. They employ a team approach to target unexplained variation, using feedback to continually improve the process of care. ²² Clinicians and patients can benefit from decision-making tools that aid in balancing guidelines and individual patient characteristics within the context of a population health perspective.

Conclusion

PCI is a potentially lifesaving procedure that is safe and effective in situations where it is indicated to improve survival, alleviate symptoms not managed by OMT, and reduce the risk of MI. Shared decision making can reinforce the benefits of PCI in the acute setting, which usually requires adjunctive pharmacological, physical, and sociological rehabilitation to gain maximum benefit from the invasive procedure. The challenge of true shared decision making arises when the indication for PCI is less compelling and less intense strategies that also require adjunctive pharmacological, physical, and sociological rehabilitation are recommended. A clear understanding of the risks, benefits, and relative value associated with various treatment options is difficult to achieve in today's fragmented health care system, and the complex web of subtle and not-so-subtle contributing factors complicate easy solutions to the problem of overutilization. Patients must be counseled in a balanced manner about the benefits, risks, and follow-up strategies with either an invasive or noninvasive treatment strategy, as well as about the mandatory nature of adherence to medication and the potential risk and benefits of the procedure itself. Effective, patient-centric counseling is necessary to overcome fear, anxiety, and a wealth of misperceptions and mistaken beliefs about the relative value of therapeutic interventions versus OMT and lifestyle modification.

Shared decision making between patients and physicians often results in more conservative care, and tools should be developed to encourage physicians to incorporate shared decision making prior to performing PCI in non-acute situations.

Ideally, physicians should discuss treatment options with the patient well before a procedure is scheduled. Consideration should be given to the patient's level of health literacy, as well as his or her personal values, when making treatment recommendations. In this manner, patients are more likely to understand the implications of a procedure, the follow-up care required, and expected long-term outcomes. When patients are able to weigh the risks and benefits of each option as it applies to them, they become active partners in their own care.