The thrust of the comments of our key informants both in New Jersey and nationally, plus our own assessment of the current state of cardiovascular care, led us to focus on the shifting boundary between invasive cardiology and cardiac surgery as potentially the most pressing issue facing the State. This shift is important not only in itself, but also because it is the current manifestation of what we see as a continuing and increasingly contentious evolution in the care of patients with cardiovascular disease over the next five years, and the increasing isolation of invasive cardiology in its pivotal role as "treatment arbiter" for patients with CAD. In this environment a combination of new technologies and treatment modalities will change not only clinical pathways and decision trees but also which clinical specialty is responsible for various aspects of patient care. This will have two important effects:
There will be continuing divergence of views and frequent changes in the allocation of resources among different specialties with respect to optimal treatment. Advances in basic science and the development and diffusion of new technologies will advance faster than prospective, randomized controlled trials can be used to determine clinical "gold standards" for diagnosis and treatment. Unless there is a process for measuring the outcomes of patient care, regardless of clinical pathway, it will be very difficult to assess either the quality or the appropriateness of care being provided.
- Clinical practice will change direction frequently within the next five years as new technologies affect which types of physicians are responsible for diagnosis and treatment decisions, the decision tree followed by those physicians, the techniques and equipment available to them, the demand for new non-invasive techniques, etc. Market forces will not assure quality and appropriateness - for example, managed care plans will not adopt stringent credentialing standards or practice guidelines. The "managed care backlash" has taught managed care executives "not to get between the physician and the patient," and they will avoid establishing their own standards, instead preferring to apply whatever state regulatory requirements and specialty society guidelines exist. State agencies will need to use foresight tempered by appropriate clinical judgment to craft regulations that facilitate desired outcomes.
One issue before the department is a re-assessment of its regulatory structure for coronary angiography, angioplasty, and the need for surgical back-up for these procedures. Nineteen ninety nine ACC/AHA guidelines state that many coronary angiography labs exist without on-site cardiac surgical back up. The guidelines add that there is no evidence that suggests that outcomes are aversely affected because of the absence of back up surgery. Regarding coronary angioplasty, these same guidelines state "to do elective angioplasty without surgical back-up exposes both the patient and the physician to unnecessary risk and should not be done routinely." However, these are not the only issues the Department faces with respect to these two procedures. Through our interviews with key informants in New Jersey and nationally we found several additional points, not addressed in the clinical literature, that warrant consideration on the issue of regulation of catheter-based procedures.
The majority of our respondents felt that there is not only a danger of over-utilization of catheter-based procedures, but that there is indeed some degree of over-utilization currently based on the potential for supplier-induced demand. There was disagreement on the extent of this problem, but most respondents pointed to the dissolution of the traditional clinical decision tree and the infrequency with which internists, non-invasive cardiologists and cardiac surgeons are involved in treatment decisions as factors behind inappropriate utilization - that is, the frequency with which decisions to perform angioplasty, for example, are "self-referrals" for invasive cardiologists. "On the fly," "ad hoc" or "look-see" procedures (a diagnostic catheterization followed immediately by angioplasty) are not only commonplace in New Jersey, but constitute a high volume of angioplasty in other markets (although we could find no state data on the proportion of all angioplasties performed this way, so our information is anecdotal from respondents). The "oculo-dilatory reflex" ("see a lesion, dilate it"), which several respondents characterized as a common clinical philosophy among invasive cardiologists, plus the very broad clinical guidelines promulgated by the ACC and others were reasons mentioned for concern about the appropriate use of catheter-based procedures. In a clinical area like cardiovascular care where patient symptoms often drive treatment, subjective choices can always be defended if they are dependent on patient descriptions of pain.
Indications for angioplasty are divided into three categories by the American College of Cardiology and American Heart Association:
- Class I: Conditions for which there is general agreement that coronary angioplasty is justified. Class I does not mean that angioplasty is the only acceptable therapy.
- Class II: Conditions for which there is a divergence of opinion with respect to the justification for coronary angioplasty in terms of value and appropriateness.
- Class III: Conditions for which there is general agreement that coronary angioplasty is not ordinarily indicated.
Respondents characterized the current practice environment as one in which Class I is dominated by the use of angioplasty (as opposed to medical alternatives), and Class II contains an increasing volume of angioplasty procedures.
One New Jersey respondent illustrated the appropriateness problem by using the analogy of breast cancer treatment. Two decades ago "it was not uncommon to do a breast biopsy, get pathology results from a frozen section, and proceed immediately to mastectomy in one continuous series of procedures. Now, of course, this would never occur." Time would be taken to consider and discuss various treatment options, then the patient would be brought back to the facility to initiate treatment. But, suggested this respondent, is this example so different from the what happens in "look-see" angioplasty today? Issues of clinical decision making, self-referral, and informed consent are seldom raised during every-day practice - the only contra-indication mentioned in our discussions had to do with dye load for patients with impaired kidney function.
Because these issues deal with the appropriateness of clinical decision-making all along patients' clinical pathways from initial diagnosis through treatment options, the example of Kaiser Permanente (KP) Southern California continuing to use "cath conferences" is particularly relevant. KP is able to look at the patient from the vantage point of a full-service cardiovascular care program where diagnosis and treatment protocols are created cooperatively among specialists who share a common set of financial incentives, rather than specialists competing in a "zero-sum game" for patients. There was some feeling among our respondents that this model produced better decision-making for patients, and possibly a lower level of inappropriate use. Although the Permanente Medical Group is a unique organization, the lesson to be taken from this example is that considering patient diagnosis and treatment options in toto, and from a collaborative (objective) perspective rather than from the perspective of a single specialty, produces better care.
The Effect of New Technology
With coronary MRI on the horizon, two important issues will arise. First will be the de-coupling of angiography and angioplasty if MRI replaces coronary angiography as the diagnostic imaging method of choice. It might well be that the adoption of MRI, even if it is the equal of or superior to angiography (in addition to being non-invasive) could be slowed by the current close relationship of the catheter-based diagnostic and treatment procedures, as represented by the increase in volume of "look-see" procedures. A second issue, however, will be the potential for a "turf war" over control of the diagnostic procedure, since MRI is traditionally the responsibility of radiologists and it has the potential, if it replaces coronary angiography, to reshape the clinical decision tree for diagnosing coronary artery disease (replacing also current non-invasive tests that provide only indirect evidence of arterial lesions - stress tests, etc.). On the other hand, this reshaping of the decision tree could be the opportunity to encourage multi-stage review of patient options that includes perspectives from primary care, diagnostic cardiology, invasive cardiology and surgery. This could, in turn, lead to more emphasis on thrombolytic therapy, or at least an increase in the diffusion of "combination therapies" that combine the early administration of new thrombolytics with primary PTCA for AMI.
Reimbursement and cost issues will be significant determinants in the emergence of coronary MRI. The price of the equipment retrofit (or new machine) necessary for coronary MRI and the level of reimbursement for the procedure will influence how this new tool develops. One might assume, however, reimbursement that approximates current payment for MRI with contrast medium, which would price the procedure below coronary angiography. How it will effect the total cost of cardiovascular care, however, will depend on how it affects the use of other services. Also, a non-invasive diagnostic procedure will be a powerful draw for patients. Some observers can even envision coronary MRI being used on a regular basis for asymptomatic patients whose family history, for example, puts them at risk for coronary artery disease. Two additional points about the potential effects of coronary MRI: its emergence will do away with the debate about the need for cardiac surgery back-up for what are currently invasive diagnostic procedures, and it could vastly expand the use of telemedicine as studies could be transmitted and read anywhere, so central locations could be established "around the world" to interpret studies from any facility.
With respect to telemedicine, we do not see similar scenarios developing with respect to robotic cardiac surgery. The potential for complications and the uncertainty involved with distant surgical operators makes this an unlikely scenario over the next five years.
We do see, however, the development of more complex catheters that combine rotoblader extraction of plaques with deployment of catheter "shields" beyond the obstructing plaque to protect against migration of plaque and clot fragments.
New Jersey's academic medical centers (AMCs) play a relatively minor role (compared to neighboring states such as New York and Pennsylvania) in providing clinical leadership within the state. Large, aggressive community hospitals, not AMCs, set the standards for clinical practice. This means that new techniques, procedures and equipment do not have an informal but widely accepted portal into the state's practice community where they can first be properly assessed before spreading throughout the hospital industry - something that does not happen in community hospitals, even those with teaching programs.
Treatment for AMI
One area on which there was unanimity regarding appropriate use and the emerging importance of a treatment modality was primary angioplasty for AMI patients. Instead of appropriateness, the issues here are: first, whether this procedure should be performed only in facilities with cardiac surgery back-up (currently the standard of care in ACC guidelines, although the guidelines indicate that the primary angioplasty experience of the facility is possibly more important that surgical back-up); second, what minimum volumes or other proxy measures are appropriate safeguards of quality (respondents told us that primary angioplasty is qualitatively different from elective angioplasty and a high volume of the latter is a necessary but not sufficient condition of ability to perform the former); and third, given the time sensitivity of the procedure, where should New Jersey policy be on the continuum between regionalization and ubiquity - that is, between locating this procedure "everywhere" to allow speedy access for AMI patients versus placing it in high-volume facilities each located to serve a certain population center. A factor in this decision must also be the emergency medical services transport system in the state, which is extremely variable according to our sources, and should receive more attention as an integral part of the state's total infrastructure for cardiovascular care. The issue of primary angioplasty for AMI could be the opportunity to re-organize the transport system (establish a more consistent and efficient transport system state-wide) in such a way that it makes possible the establishment of designated primary angioplasty sites serving major population centers and others for less densely-populated areas of the state.
Quality and Appropriateness of Care
Regarding the relationship between patient volume and the outcomes of treatment, studies performed using data from New York State indicate that patients receiving a CABG in a facility doing fewer than 500 procedures per year, or an angioplasty in a facility doing fewer than 400 per year, have a 35 to 40 percent increased risk of mortality.25, 26 These and similar research findings are the basis for the increasing popularity of volume based quality standards in the health care purchasing community. Objections to the use of volume as a proxy measure for quality in cardiac care have been raised regarding measurement at the facility level versus surgeon-specific volumes for cardiac surgery or operator-specific volumes for angioplasty (or team-specific volumes). When data such as risk-adjusted outcomes for cardiac surgery are available and the components of the relationship between volume and outcomes are analyzed, direct measures of quality can be superior to proxies (not all low-volume hospitals have poor CABG outcomes, for example), in addition to being more acceptable to the provider community (for example, useful in quality assessment programs). In the absence of additional information about CABG or angioplasty, however, most respondents said that volume is still the best proxy for quality.
For cardiac surgery programs we wish to emphasize a distinction between the use of volume standards or outcome measures to address quality issues on the one hand, and certificate of need regulation, on the other, for the following reason. With population-based rates of CABG surgery across the country steady or declining, it might be unwise to allow the proliferation of new programs in markets where volumes and reimbursement will not be sufficient to support new programs. Standards for regionalizing cardiac surgery services should be based on the epidemiology of cardiovascular disease, and should be population-based to ensure that programs of questionable viability do not proliferate to the point where there are incentives to provide services to generate revenues just to keep the programs open. In addition, existing inner city cardiac surgery programs depend on the payer mix they enjoy by attracting Medicare and commercially-insured patients from the suburbs. If new suburban programs took those paying patients away, the financial viability of inner-city programs would be threatened, and with it the access to services for minority populations who we know, from a growing body of research, have problems gaining access to tertiary cardiac care.22, 23
Our discussions with respondents about interventional cardiology focused more on the issue of appropriateness than on quality and outcomes. The best care processes can be used on a patient who then has a good short-term outcome, but if the patient does not need the procedure (if the procedure does not improve functional status or longevity, for example, or is less cost-effective than medical treatment which could produce the same outcome), the procedure is inappropriate. The appropriateness issue in the minds of respondents was directly tied to disagreement over long-term benefits of angioplasty and stenting, under-appreciation of medical therapy, and a perception among cardiologists that patients have a strong preference for angioplasty over CABG, and for "look-see" procedures ("get it all taken care of in one session") over separation of angiography and angioplasty.
Education is paramount in dealing with uncertainties in the quality and appropriateness of current treatment, and the rate at which new therapies are developed and disseminated throughout the health care system. An educated patient (or, as some would put it, an informed consumer) is better equipped to question, to learn and to see the importance of both negotiating a path through the medical care system and being cognizant of the lifestyle plus preventive and therapeutic medical options available to patients with or at high risk for CAD. Physician education is needed to ensure that patients receive the full benefit of currently-available medical therapies, which research has shown is not currently the case.20 Technologically, both the patient and physician populations will depend more on use of the Internet to address these issues. Not only will patients use the Internet to become better informed about their own medical conditions and the care options available to them, but physicians will use Internet technology for communication of the vast amounts of information (individual patient records, epidemiologic data on patient populations, analysis of process-of-care and outcomes data for their own patients, current peer-reviewed literature, etc.) necessary to meet evolving standards of patient-centric care and the expectations of an increasingly empowered patient population.