A Policy Statement on Cardiovascular Test Substitution and Authorization

For the evaluation of patients with known or suspected heart disease, the right test for the right patient at the right time is a key clinical tenet. Test substitution is a troublesome trend in recent years. After a physician has used scientific data and his or her best judgment to recommend a course of action, has discussed with the patient the rationale for a specific approach, and has considered as well the patient’s preferences, a third-party payer will often direct a different approach. This redirection, often driven by incomplete or inaccurate information, leads to increased administrative burden, delays appropriate care, adds to diagnostic uncertainty, erodes patient confidence in the chosen course of action, and can result in harm to patients from both care delays and the patient receiving the wrong test. For any given patient, the choice of test is highly nuanced and depends on patient and institutional factors, including local expertise and availability. The informed physician and his/her patient are best positioned to make these decisions. Representatives from cardiovascular imaging societies (The American College of Cardiology Imaging Council, American Society of Echocardiography [ASE], American Society of Nuclear Cardiology [ASNC], Society of Cardiovascular Computed Tomography [SCCT], Society of Cardiovascular Magnetic Resonance [SCMR], Society of Nuclear Medicine and Molecular Imaging [SNMMI]) met to discuss the challenges with test substitution; the proceedings of this summit are summarized in these consensus recommendations, which were endorsed by the leadership of the respective societies.

EVIDENCE REGARDING ANATOMIC VERSUS FUNCTIONAL STRATEGIES IN THE EVALUATION OF CORONARY ARTERY DISEASE

Even in the very common condition of coronary artery disease (CAD), there are few randomized trials comparing the utility of the various imaging modalities. PROMISE (PROspective Multicenter Imaging Study for Evaluation of chest pain) was a multicenter study at 193 North American sites in which 10,003 symptomatic patients (mean age: 60.8 years) requiring stress testing were randomized to an anatomic test (coronary computed tomography angiography [coronary CTA]) or a functional test (exercise electrocardiogram, stress echocardiogram, or stress nuclear based on site preference) (1). In this North American study, there was no significant difference in clinical outcomes with functional testing compared to anatomic testing (median follow-up: 2 years), and the coronary CTA strategy was associated with higher downstream utilization of invasive angiography and coronary revascularization. In contrast, the United Kingdom multicenter SCOT HEART (Scottish COmputed Tomography of the HEART) trial (2) showed that the addition of coronary CTA to usual care was associated with a significant reduction in congenital heart disease death or myocardial infarction over a mean follow-up of 5 years, and the use of coronary CTA was not associated with higher use of downstream invasive testing. However, the results of the SCOT HEART trial may be less applicable to North American cohorts, because, for example, the usual care arm was primarily exercise testing alone (3). In a prespecified subgroup analysis from the PROMISE study, an abnormal functional test in subjects >65 years and CTA in subjects <65 years predicted risk of cardiovascular death/myocardial infarction. The recent European Society of Cardiology (ESC) chronic coronary syndromes guidelines suggest a stratified approach based on the pretest likelihood of obstructive CAD as well as patient characteristics and preference, availability, and local expertise. Comparison trials of the diagnostic accuracy of cardiovascular imaging studies and meta-analyses have indeed shown differences in sensitivity and specificity but generally describe benefits and limitations of each modality. Concordant with these findings, current major guidelines and appropriate use criteria include individualized risk assessment as a necessary prerequisite to deciding whether or not testing is needed, as well as test selection. This is a critical feature of testing decisions and a core component of the patient-centeredness (4,5).

THE NEGATIVE IMPACT OF PRIOR AUTHORIZATION AND TEST SUBSTITUTION

Clinicians can often describe anecdotal stories of undue delays in diagnosis, excess expense, and even harm caused to patients by the prior authorization/test substitution process. Radiologic benefits managers are able to quantify the number of tests avoided (and thus the “savings” from the point of view of the insurer). However, they seem much less able to estimate the secondary costs, such as those from inaccurate or uncertain diagnoses, avoidable trips to the emergency department, or hospital admissions related to delays of the authorization process. There is also the very real issue of the negative impact on patient satisfaction and provider well-being. Such issues merit detailed study and will require a high level of transparency and cooperation between insurance providers and physicians.

There are emerging published reports supporting the adverse impact of preauthorization procedures on patient outcomes, although most of them pertain to noncardiovascular applications. The results of one small study suggest that telephone preauthorization for emergency care resulted in worse outcomes of death or disability or the near miss of such events, delays in the use of biologics for severe allergic disease/asthma (6,7), and more. However, the frequency of adverse outcomes is not known. This remains an underexplored area, particularly in cardiovascular care, and more research is needed to study the impact of test substitution and associated delays on cardiovascular outcomes.

IN THE ABSENCE OF FIRM EVIDENCE ABOUT THE BENEFITS OF TEST SUBSTITUTION, WHAT IS THE BEST APPROACH?

For any given clinical presentation, when multiple alternative tests are appropriate based on practice patterns and appropriate use criteria, and with each cardiovascular testing modality undergoing steady technological advancement, the physician’s role is to use his or her best judgment to decide on the optimal approach, factoring in the pretest likelihood of CAD, patient preference, patient characteristics, equipment availability, and technical and reader expertise (Figure 1).

FIGURE 1. A Patient-Centered and Personalized Approach to Chest Pain Imaging.

The choice of cardiovascular diagnostic evaluation in any given patient is highly nuanced. It is a shared decision between physician and patient, based on numerous individual patient factors, local expertise and access, and technology-related specifics. The key issue is the overall impact on health, taking into account patient preferences and the long-term cost-effectiveness of the strategy. ACS = acute coronary syndrome; CABG = coronary artery bypass graft; CAC = coronary artery calcium; CAD = coronary artery disease; CMR = cardiac magnetic resonance; CTA = computed tomography angiography; ETT = exercise treadmill test; ICD = implantable cardioverter-defibrillator; MINOCA = myocardial infarction with nonobstructive coronary arteries; PCI = percutaneous coronary intervention; PET = positron emission tomography; SPECT = single-photon emission computed tomography.

PRETEST PROBABILITY OF CAD.

The pretest probability of CAD not only determines the relative performance for each test but also the health care impact. For example, for suspected CAD, anatomic testing in individuals with chest pain syndromes and a low to intermediate probability is likely to be impactful because of the strong negative predictive value and positive long-term benefit of initiating primary preventive therapies in the setting of premature coronary atherosclerosis. Conversely, for those with established coronary disease or in older patients, functional testing is more likely to inform treatment decisions (8).

PATIENT PREFERENCE.

Patient preference is emerging as an important consideration for management decisions in medicine. Patient preferences for any given test may be driven by a number of factors, including cultural beliefs often related to race and ethnicity, insight into disease process, out-of-pocket costs, radiation aversion, aversion to invasive procedures, and claustrophobia. Men and women sometimes view these decisions differently, although the reasons for this are yet to be determined (9). Moreover, the current system does not explicitly factor in out-of-pocket costs for patients when 2 tests are equivalent (10). We recommend further study of these factors, especially race and ethnicity and economic barriers to care to further patient-centered imaging.

WHAT ARE THE IMPLICATIONS OF VARIABILITY IN “LOCAL EXPERTISE” AND AVAILABILITY OF EQUIPMENT ON TEST SUBSTITUTION?

Technology in all modalities is rapidly evolving. The availability of state-of-the-art equipment and levels of local expertise usually, but not always, imply high-quality imaging. Few institutions are able to accommodate the continual capital investment required for high-quality equipment; excessive prior authorization hinders such investment and innovation. The problem with test substitution is that these factors are hard to quantify because they may be quickly outdated with the advent of new equipment. Also, given the complexity of information provided by advanced imaging, at many medical centers, even a well-informed cardiologist benefits from a consultation with an imaging expert to select the best choice of test. The insurance companies, while substituting tests, may not be aware of these nuances in imaging. A poor-quality test often adds unnecessary expense and can even cause harm. When 2 tests are appropriate, a good-quality test is always better than a limited-quality test. The importance of local physician expertise in test selection is endorsed by 2 major guidelines. The 2019 ESC chest pain guidelines list as a Class I indication “selection of an initial non-invasive test based on the clinical likelihood of CAD and other patient characteristics that influence test performance, local expertise, and the availability of tests” (8). The 2014 American College of Cardiology/multisocietal guideline states, “If more than 1 modality falls into the same appropriate use category, it is assumed that physician judgment and available local expertise will be used to determine the correct test for an individual patient” (4).

AN APPROACH TO CARDIOVASCULAR TESTING TAKING INTO ACCOUNT FACTORS BASED ON PRETEST LIKELIHOOD OF CAD, PATIENT CHARACTERISTICS AND PREFERENCE, AVAILABLE EQUIPMENT, AND EXPERTISE

In the absence of definitive evidence, the choice of the test should be a shared decision for the patient and the physician rather than directed by the insurance provider. The use of prior authorization and radiologic benefits managers adds to the cost structure and administrative burden for physicians without evidence for improved patient care. Physicians are willing to accept guideline-directed management, relieving insurance providers of liability and reducing administrative burdens and cost. Thus, there is an upside, not only for the patient but also for the health insurance organizations, to pursuing algorithms where patients’ physicians follow guideline-directed management strategies without such intrusion. Physicians would welcome shared decision making directly with insurance companies to cut administrative costs and to improve patient outcomes.

CONSENSUS PRINCIPLES.

• Optimal patient care should be the priority for physicians and insurance providers.

• Given equivalent diagnostic accuracy and appropriate use criteria, there is no clear justification for routinely and indiscriminately recommending one diagnostic strategy over another.

• Test substitution driven by short-term cost savings should not be endorsed.

• In alignment with the principles of “patient-centered medicine,” a personalized approach to testing based on clinical information is better than one test for all patients.

• Physicians are willing to accept guideline-directed management, relieving insurance providers of liability and reducing administrative burdens and cost.

ABBREVIATIONS AND ACRONYMS

CAD

coronary artery disease

CTA

computed tomography angiography