The Institute of Medicine report on cancer care: Is the system "in crisis"?

Americans love to fight, traditionally. All real Americans love the sting and clash of battle...When you, here, everyone of you, were kids, you all admired the champion marble player, the fastest runner, the toughest boxer, the big league ball players, and the All-American football players. Americans love a winner. Americans will not tolerate a loser.

General George S. Patton, Jr. was famous for his flamboyance and aggressiveness going on the attack, among other things. He was also known for a number of pithy quotes made throughout his lifetime, particularly during World War II, such as the one above in which he declared how much Americans love to fight. I sometimes wonder whether he was more correct in that assessment than he knew in that we seem to view almost everything through the lens of war and a crisis that calls for a war. In medicine, for instance, we have the "war on cancer" and the "war on obesity." We have a propensity for likening problems and their solutions to war, where the goal is to destroy the enemy.

Similarly, but less dramatic (although only slightly so), we have various crises. Indeed, Americans seem to love to compare problems to crises as much as they love to compare them to war. Of course, the two are closely related, as a crisis of some sort is a necessary prelude to a war. We can argue about the definition of a "crisis," but one thing is certain. No matter how one defines it the word "crisis" implies an urgent problem and further implies that something must be done now—or at least very soon—to prevent the catastrophic consequences of that problem, which usually involve a breakdown of a current system. It was with these thoughts in mind that I approached the Institute of Medicine (IOM) consensus report released last week, Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis. It's a behemoth of a report, weighing in at 360 pages, and it does indeed paint a picture of some very serious problems in cancer care that should be addressed. It is subdued in that it doesn't engage in that oh-so-American tendency to declare "war" on every problem, but is cancer care "in crisis"? Of that, I'm not so sure, but it's definitely got problems. But, as I've said, the system's had problems for a long time, as the IOM itself documented in its 1999 report Ensuring Quality of Cancer Care, which concluded that "for many Americans with cancer, there is a wide gulf between what could be construed as the ideal and the reality of their experience with cancer care." The report also recommended steps to improve cancer care and the evidence base for cancer care, and to overcome barriers of access to high-quality cancer care. As the introduction to the current report states:

These recommendations led to a number of efforts targeted at improving the delivery of cancer care. The Secretary of the U.S. Department of Health and Human Services (HHS) established the Quality of Cancer Care Committee to work on issues identified in the report. A number of organizations used the report to develop core indicators of quality of cancer care and recommendations for improving the quality of cancer care, including the Agency for Healthcare Research and Quality (AHRQ), the National Quality Forum (NQF), and the National Dialogue on Cancer (a collaboration organized by former President George H.W. Bush and Senator Dianne Feinstein, now known as C-Change). In response to the report, the American Society of Clinical Oncology (ASCO) undertook a national study of the quality of care delivered by oncologists, called the National Initiative on Quality Cancer Care (ASCO, 2013). In addition, the Cancer Quality Alliance, a diverse group of stakeholders committed to advocating for improvements in the quality of cancer care, used this report and several other IOM reports to develop five cancer case studies depicting a vision for high-quality cancer care and a blueprint for action to realize that vision (Rose et al., 2008). The report provided major input into quality of cancer care legislation drafted by the Senate Health, Education, Labor, and Pension Committee

Not to blow my own horn (too much, anyway), I happen to be involved with two major quality improvement initiatives, one national through the American College of Surgeons Committee on Cancer and another statewide initiative seeking to improve the quality of breast cancer care throughout the state, not just at the two NCI-designated comprehensive cancer centers in the state and the several large health systems but at community hospitals as well. After all, most cancer care in this country does not occur at the 41 NCI-designated comprehensive cancer centers. It is clear to me that, since 1999, there has been a sea change in the attitudes of physicians regarding outcomes tracking to improve quality and adherence to evidence-based guidelines for the better. Physicians in the past were all too frequently wedded to a view in which a physician's autonomy and every decision was unquestioned (and the choice of the word "his" was intentional, given how male-dominated the profession was) and they viewed attempts to persuade them to adopt evidence-based guidelines as intolerable infringements on their god-like autonomy. Today, they are much more receptive to following the guidelines, particularly in oncology. (Unfortunately, when it comes to the CDC's recommended vaccine schedule we've seen an example of a physician who still follows the old-school, haughtily dismissing the recommended schedule because he thinks his clinical experience trumps expert, evidence-based guidelines. But I digress.)

Here's a typical example of how the report is being discussed in the press:

Cancer treatment has grown so complex, many U.S. doctors can't keep up with new information and are offering incorrect treatment, failing to explain options and leaving patients to coordinate their own care, according to a report released on Tuesday by the Institute of Medicine, part of the National Academy of Sciences.

The 315-page report, "Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis," identifies a long list of reasons for the crisis, including a growing demand for cancer care as more people receive a diagnosis and a shrinking oncology workforce.

"Probably on average the quality of care is not bad, but we know there are people who are not getting the (highest) standard of care," said Dr Patricia Ganz, who chaired the institute's panel that wrote the report and is a professor at the Schools of Medicine and Public Health, University of California, Los Angeles. "Patients need to be asking, Is my doctor giving me appropriate treatment?"

On its web page for the report, the IOM even includes a video designed to zero in on the issues discussed in the report:

The first thing I noticed about this video is that it does indeed touch on a number of prominent problems with the cancer care system, but that most of the problems discussed are in reality problems with the entire health care system that tend to be magnified in cancer care. The IOM identifies six items that it used as a conceptual framework on which to drape its solutions:

1. Engaged patients
2. Adequately staffed, trained, and coordinated workforce
3. Evidence-based cancer care
4. A learning health care IT system for cancer
5. Translation of evidence into clinical practice, quality measurement, and performance improvement
6. Accessible, affordable health care

The IOM also articulates ten goals for its recommendations:

1. Provide patients and their families with understandable information about cancer prognosis, treatment benefits and harms, palliative care, psychosocial support, and costs.
2. Provide patients with end-of-life care that meets their needs, values, and preferences.
3. Ensure coordinated and comprehensive patient-centered care.
4. Ensure that all individuals caring for cancer patients have appropriate core competencies.
5. Expand the breadth of data collected in cancer research for older adults and patients with multiple comorbid conditions.
6. Expand the depth of data collected in cancer research through a common set of data elements that capture patient-reported outcomes, relevant patient characteristics, and health behaviors.
7. Develop a learning health care information technology system for cancer that enables real-time analysis of data from cancer patients in a variety of care settings.
8. Develop a national quality reporting program for cancer care as part of a learning health care system.
9. Implement a national strategy to reduce disparities in access to cancer care for underserved populations by leveraging community interventions.
10. Improve the affordability of cancer care by leveraging existing efforts to reform payment and eliminate waste.

I don't think I'll be able to address all the issues, because each of the six items in the conceptual framework and goals have several recommendations to achieve them and the discussion would quickly balloon beyond the already prodigious verbiage of a typical one of these posts. (Also, I'm trying to make like Harriet and keep my posts a bit brief, at least under 4,000 words for a change.) So I'll basically pick and choose, since a lot of these principles and recommendations are related. Let's start with the issues that are basically the same issues that plague all of medicine, for instance an adequately staffed and trained workforce, a better IT system, and accessible, affordable care. All three of these are clearly interrelated, and tend to boil down to money, perverse incentives in the reimbursement system, and systemic inefficiencies. For example, if we want an adequately staffed, trained, and coordinated workforce, it's going to require cash. From somewhere.

Fortunately, there are some actions that can be taken legislatively, and recommendations to change the system so that reimbursement for team-based care is more favorable are appreciated. Of course, recommendations for a shift to team-based care will conflict with recommendations to improve the number of trained providers, because team-based care can be quite inefficient. Multidisciplinary tumor care groups often see far fewer patients in a day because each patient is seen by all the specialties on the same day. For example, such systems can make it hard for surgeons to see enough patients to generate enough cases to generate enough revenue be financially viable. As long as surgeons are paid by the procedure, this will be a problem. Also, multidisciplinary teams require a significant infrastructure to maintain, and that doesn't come cheap. Come to think of it, two of these goals (better IT, an adequately staffed and trained workforce) will likely conflict with making cancer care more affordable. True, it's argued that this won't be the case because better coordination of care, better cancer IT, and better staffing will reduce costs, but the evidence is rather thin to support these assertions.

In the case of IT and a better electronic medical record (EMR), certainly there would be benefits, as described in the report and greatly touted in the video, where one of the patients complained that lack of coordination between her doctors led to her being overdosed with an antihypertensive medication. The problem with coordinating care in the US is that there is no single unified health system, nor is there a single, unified medical record. Large health systems have common medical records, but that only helps if a patient is receiving all of his or her care in a single health care system. EMRs from different health care systems often don't talk to each other. Transferring data from EMRs can be a nightmare. Worse, many of the existing EMRs are not exactly what one would call "user friendly." Indeed, the EMR I'm stuck using can best be described as "user hostile." After five years, I can honestly say that I still don't fully understand much of it, and I'm not exactly computer illiterate. In fairness, I will point out that I haven't put the time in to learn all the ins and outs of the system, but that brings up another point, namely that the up-front costs, in terms of money and time to implement an EMR, are enormous. Between lost productivity and the costs of implementation and maintenance, many private practices can't afford it. Worse:

To make matters tougher on physicians, today's EHRs are not really designed to "optimize speed of throughput for the clinician," Haughton says. In many instances, a quick handwritten note in a patient's chart is all that's needed to record the encounter. Today's EHRs, however, with their quality reporting and meaningful use integrations, make it challenging for doctors to complete their notes swiftly. They must enter, and often re-enter again and again, patient data that are not immediately germane to the medical issue at hand. Thus, it can take physicians a significantly longer time to input their notes into an electronic record than it took with their old paper processes—thereby increasing practice costs after EHR implementation.

Moreover, the advantages of adopting a health IT, for cancer or any other disease, are blunted if the systems can't communicate with each other. Politically, it might never be possible to have a single EMR as there is an inherent distrust of the government having too much of our health information, something that would be facilitated by a single, unified EMR.

Of course, the EMR is only part of the better health IT so I don't want to dwell on it too long. The overall recommendation was to have a "learning IT," but I think that my brief discussion of EMRs is quite relevant, because this is what the IOM's vision for a "learning health care system" includes:

Science and Informatics

• Real-time access to knowledge – A learning health care system continuously and reliably captures, curates, and delivers the best available evidence to guide, support, tailor, and improve clinical decision making and care safety and quality.
• Digital capture of the care experience – A learning health care system captures the care experience on digital platforms for real-time generation and application of knowledge for care improvement.

"Digital capture of the care experience" is, basically, the EMR. However, the IOM goes beyond that in wanting this digital capture to be able to be fed into a system that can track outcomes and be used for "real time" care improvement and improved clinical decision-making. This is a very tall order that would require considerable allotment of resources. The IOM states that many of the elements necessary are already in place, but if they are I can't see them, and I'm involved in quality initiatives in my own institution, with the American College of Surgeons, and with our statewide collaborative. Data are not entered into our databases in anything resembling "real time," despite our best efforts. I'd love to see such a system, although I'm perhaps a bit more skeptical of the claimed benefits for it; indeed, it would be highly useful to have such a system for all of medicine, not just cancer. Also in fairness, the IOM does recognize the difficulties and challenges involved in

Of most interest to me are the IOM recommendations regarding evidence-based oncology. Ironically, it's hard not to point out that the evidence base for a lot of the health IT recommendations being able to do what they are claimed to be able to do is currently thin to nonexistent, but in terms of actual treatment there has been a major initiative over the last 14 years to improve the adherence to evidence-based guidelines in treating cancer patients. Although this has largely been fairly successful, the IOM points out that there are areas where current oncology practices fall short, particularly in the elderly, for whom the evidence base supporting some common treatments is disappointingly thin.

However, the IOM does make a good point that cancer care has become more complex in the last 14 years. That is arguably true, with the proliferation of genomic tests for various cancers, the attempts to administer "personalized" cancer treatment (or, at it's now more often called "precision oncology"). Few diseases, or classes of diseases (which is what cancer really is, a large class of diseases) require as many specialties to treat it. Nearly every patient with a solid tumor (as opposed to hematologic malignancies) can expect to be treated—or at least seen—by a minimum of two specialties (medical oncology and surgical oncology) and most see a third specialist as well (radiation oncology).

Once the province of surgeons and local-regional therapies, cancer treatment has evolved rapidly in recent decades. Systemic treatments emerged in the 1950s and 1960s, initially as relatively nonspecific chemotherapies with limited efficacy in some human cancers. Empiricism dominated oncology drug development in this era, rather than an understanding of tumor biology. In recent years, researchers have developed treatments targeting specific molecular aberrations (e.g., imatinib for chronic myelogenous leukemia, trastuzumab for breast cancer). Molecularly targeted treatments have pervaded Food and Drug Administration (FDA) approvals in oncology in the past decade and have improved patient outcomes for many cancers. These agents commonly require a test to assess the drug target in the patient’s tumor. As such, companion diagnostic testing (e.g., estrogen receptor [ER] and human epidermal growth factor receptor-2 [HER2] in breast cancer, anaplastic lymphoma kinase [ALK] and epidermal growth factor receptor [EGFR] in non small-cell lung cancer) has increased in importance. The sheer number of targeted agents has increased the educational burden for cancer care clinicians and the financial burden for the healthcare system. In the near future, the implementation of genome-based diagnostics will likely alter both our ability to deliver precision medicine and the complexity of cancer treatment (IOM, 2010, 2012b; NRC, 2011).

Indeed, even the surgical treatment of breast cancer has become more complex. When we do a sentinel lymph node biopsy, for instance, what findings lead us to go back and take the rest of the lymph nodes? Should we do breast-conserving surgery or a mastectomy? When should neoadjuvant chemotherapy be given to shrink the tumor, and how do we predict the likelihood of success?

In any case, the recommendations of the IOM are on more solid ground here, and include requiring researchers evaluating the role of standard and novel interventions and technologies used in cancer care to include a plan to study a population that mirrors the age distribution and health risk profile of patients with the disease, as well as recommending that the National Cancer Institute should build on ongoing efforts and work with other federal agencies, the Patient Centered Outcomes Research Institute, clinical and health services researchers, clinicians, and patients to develop a common set of data elements that captures patient-reported outcomes, relevant patient characteristics, and health behaviors that researchers should collect from randomized clinical trials and observational studies. And, of course, the IOM is big on comparative effectiveness research (CER), which basically involves doing research that compares the effectiveness of two existing treatments. One problem with CER, however, as I've pointed out before, is that it sometimes takes so long to do clinical trials that by the time the results of CER studies are published their results might very well no longer be relevant to cancer care because the standard of care has moved to something different than the treatments being compared. So, yes, I like CER, too. However, I don't know that its power is as great as its adherents claim.

I'm also very impressed with the IOM's recommendations for improved palliative care and increased emphasis on palliative care. Five years ago, my mother-in-law was diagnosed with a recurrence of a particularly nasty form of breast cancer. This recurrence was in the form of widespread metastatic disease, clearly incurable. If there's one thing that resonates with me it's the deficiencies in palliative care in this country, particularly with respect to home care and home hospice. Once my mother-in-law was in inpatient hospice, the care she received was excellent, but while she was still at home the burden fell nearly entirely on my wife and her sister, with very little guidance. If there's a gap in US cancer care, it's there. I learned more than I ever had before about the "real world" during those months.

Finally, one very important part of the IOM report is its emphasis on patient-centered care. I've written about patient-centered care on multiple occasions before (for example, here), and I definitely believe that we do not always study the patient outcomes that matter to patients. The video above featured a woman who saw a "famous" oncologist about her cancer who recommended a very aggressive course of therapy even though she had stage IV (and therefore incurable) disease, and a man with stage I prostate cancer whose first doctor recommended very aggressive surgery. Neither, we're told, were asked what they wanted or what they valued. Although I like to think that this sort of paternalism and "go for cure no matter what" attitude is becoming less prevalent, it still exists in a lot of oncology. That's why it's hard not to agree that patient-centered care is important. However, I always raise one caveat. The decisions and the trade-offs patients decide to accept should be informed by the science. The options presented to the patient, and their cost in terms of potential complications and impact on the patient’s ability to go about his daily activities and in essence live his life, must be based on science. The key is that we, as clinicians, must make sure that the patient has accurate, up-to-date, science-based information upon which to base his or her choice.

Let me just finish by pointing out that, my skepticism towards parts of this report notwithstanding, the IOM has demonstrated once again the difference between science- and evidence-based medicine and CAM. SBM rigorously questions everything about how it does everything and is continually trying to improve. Nothing is off the table, and we discuss our deficiencies and failures openly. I can't help but contrast that to certain other areas of "medicine" that I discuss frequently on this blog. Even if I don't agree with everything in the IOM report and suspect that the use of the word "crisis" was a means of getting attention more than an accurate description of the situation, the IOM itself deserves plaudits for its work in this area.