'The Other': Confronting the Tension Between Academic and Community Oncologists

Ravi B. Parikh, MD, MPP


February 11, 2019

"Well...why did they do that?" asked my attending.

It was a good question. We were meeting a patient with metastatic prostate cancer on rounds at our academic hospital. There were some treatment choices that were difficult to explain in the sparse medical records available. Why didn't he receive chemotherapy for his rapidly progressing, symptomatic disease? Why hadn't he received germline testing, given his strong family history of early cancers?

Unfortunately, the patient himself didn't know, and we couldn't reach his local oncologist by phone. We gave our recommendations and moved on, resigning ourselves to the fact that oncologists practiced differently "in the community."

This wasn't the first time I'd felt a tension between community and academic practice. Throughout medical school, residency, and even in fellowship, I'd heard phrases like "out in the community" or "over in the Ivory Tower," used to refer pejoratively to the way community and academic providers practiced. Having trained mostly in academic centers, where patients were referred for second opinions and clinical trials, I often thought that academic medicine was the "purer" form of practice, where practitioners made evidence-based decisions free from the pressures of fee-for-service medicine.

For standard-of-care therapies, it is probably best to have a local oncologist.

But as a hematology-oncology fellow, I've seen patients in both academic and community settings, training under specialized and general oncologists. I've learned that the relationship between community and academic oncology is much more nuanced—and practice is perhaps not so different. I've seen several examples of how academic and community oncologists can not only collaborate but also capitalize on each other's unique resources. Here are a few of the most powerful lessons I've learned.

A local oncologist is usually the best oncologist. In our academic oncology clinics, we often consult on patients seeking a second opinion. Last year I met one woman with early-stage breast cancer whose oncologist recommended that she start hormonal therapy after her surgery. It was standard of care; nevertheless, the patient delayed starting her treatment until our breast cancer specialists could weigh in.

When we met her, she strongly favored transferring her care to our facility, despite living several hours away. She wanted treatment from an "expert," someone who was involved in cutting-edge treatments and research in breast cancer. She also wanted to be eligible for clinical trials, even though there were none that she was eligible for that day.

To be fair, oncologists and advocacy groups encourage second opinions. And patients often seek second opinions because they are worried that they are receiving too little or biased information. However, in many—perhaps even most—cases, we agree with a local oncologist's recommendations.

What has been harder is convincing some patients that their local cancer clinic is actually the best place for them to receive their care. For standard-of-care therapies, representing the vast majority of what oncologists prescribe patients, it is probably best to have a local oncologist who is able to respond quickly and in-person to urgent symptoms when the patient is unable to travel.

I've realized just how difficult it is to be a 'jack of all trades' in oncology.

Perhaps more worrisome, there is some evidence that second opinions result in more aggressive cancer care without significant benefit. This may be because patients seek second opinions when they prefer aggressive treatment, and so consulting oncologists may be biased toward recommending aggressive treatment even when a local oncologist doesn't think that's the best course of action.

Ultimately, I've found that community and academic oncologists usually agree on the most important aspects of a patient's treatment plan, and often community oncologists consider important treatment aspects (notably cost) that subspecialists do not.

Being a general oncologist is becoming impossible. While I've always envisioned a career in academic medicine, I've also always been in awe of the community oncologist who "sees everything." Prior to fellowship, I strongly considered being a general oncologist. I loved the idea of treating patients with cancer through their disease course, and I wasn't sure whether I'd ever like a disease area enough to focus on one set of tumors. And if new treatments like immunotherapies were going to replace specific chemotherapies, should I really be focusing on one disease group anyway?

However, since week 1 of fellowship, I've realized just how difficult it is to be a "jack of all trades" in oncology.

Every clinic required hours of reading, learning standard-of-care staging and treatment of illness. And it seemed like in every disease area (melanoma, breast, lung, etc.), there had been a recent paper published that had completely revolutionized care. Even some of my disease-specific attendings acknowledged how much work it was to keep up.

Oncologists are realizing the limits of general oncology.

As one of my co-fellows put it, being in multiple disease clinics was like drinking from a firehose. It seemed impossible for a general oncologist to keep up with a changing landscape.

This becomes a bigger problem when local oncologists see limited volumes of certain cancers. As Larry Shulman, MD, deputy director for clinical services of the Abramson Cancer Center at the University of Pennsylvania, and chair of the Commission on Cancer, told me, "Some small community hospital oncologists only see three to four patients a year with stage III breast cancer, and even fewer with rarer cancers." How can we expect local oncologists to provide similar quality for rare cancers compared with subspecialists? Indeed, for some cancers, there are early signals of a quality gap between specialized and community cancer sites.

Specialized oncology is one solution. But as Shulman noted, "The solution to improving the quality gap in oncology can't be that everyone gets seen at an academic center. We need to figure out how to support high-quality care at community hospitals, close to patients' homes and where most patients with cancer in the United States are cared for."

First of all, it's very possible that there is not a meaningful difference in outcomes for common cancers, when controlling for case mix and cancer-specific variables. But even if there was, only 8%-10% of patients with cancer are cared for at large academic centers. Significantly increasing the patient load at already-strained academic centers would probably wipe out any quality difference that existed.

The lines between general and academic oncologists are beginning to blur.

In nonacademic oncology clinics, whether it be at the VA or a private community site, oncologists are realizing the limits of general oncology. The need to keep up with emerging biomarker-based and precision oncology approaches only complicates the issue. As a result, many community oncologists now only see a few tumor types or develop a certain expertise.

Len Lichtenfeld, MD, interim chief medical officer of the American Cancer Society and a longtime medical oncologist, told me how the number of general oncologists in practice has been declining. "As community oncology practices are increasingly owned by hospitals or become part of large organized groups, oncologists within those groups have begun to specialize in different disease areas."

As such, the lines between general and academic oncologists are beginning to blur in an unprecedented way, particularly as phase 1, 2, and 3 trials move into the community and outside of the confines of academic medical centers. Large national networks like US Oncology provide a potential model for specialization in large community centers, in that they allow for sharing of best practices and resources in community settings. And as large academic centers affiliate with or acquire community practices, perhaps some academic oncologists will even move out into the community.

We need better models for communication between academic and community oncologists. Even if large networks like US Oncology can improve access to subspecialized oncology, many patients will receive care with general or community oncologists who are highly qualified to treat a majority of cancers. We need to do better in recognizing the unique strengths of each so that patients with cancer are receiving the highest-quality care as close to home as possible.

Shulman has one vision that follows the model of US trauma centers, which are certified as level 1, 2, or 3 depending on the amount of resources available. Clinicians at level 3 trauma centers frequently perform initial stabilization of a patient and then transfer patients to level 1 centers if more specialized care is needed. There is a well-known referral protocol and geographic distribution of such centers, and each level has different accreditation criteria and resourcing standards.

Shulman's group, the Commission on Cancer, has established similar accreditation programs for rectal cancers, for example, where surgery requires high levels of expertise to be done well with low perioperative mortality. What if we envisioned a similar framework for treating other cancers, such as rare, orphan cancers? It's a promising idea; however, only about 35% of oncologists agree with automatic referrals to high-volume centers for rare cancers.

Most oncologists favor better dissemination of knowledge and best practices into the community, rather than forcing patients to be treated at high-volume (generally academic) centers. Lichtenfeld agrees that, in the future, "academic centers may fill a more consultative role than they did in the past," but forcing patients to be treated in such centers is certainly not desired.

Of note, as Lichtenfeld pointed out, technology has democratized certain services that were previously only available at academic centers. A good example is next-generation sequencing (NGS) from companies like Foundation Medicine, Inc. For all of the concerns about overutilization of high-cost NGS, companies like Foundation Medicine made NGS available to a wide array of oncologists who use it regularly to influence practice.

Telemedicine and remote care-sharing platforms may be another option to offer specialized care in the community. Our hospital, the University of Pennsylvania, offers virtual consults in areas like genetics and bone marrow transplant pre-evaluation. Project ECHO is a virtual care-sharing platform that has been shown to disseminate evidence-based care for diseases like hepatitis C to rural communities without compromising quality of care.

Bridging the academic-community divide starts in training. No matter how many technology solutions, accreditation programs, or large physician networks we develop, general oncology will not, and should never, disappear. Most patients with cancer should receive their primary treatment close to home. And there will always be a need for communication between academic and community oncologists.

I think we have the greatest room for improvement in these domains. Most oncology trainees learn to practice primarily at academic centers or affiliated sites and have little experience interacting with community oncologists. For those of us who choose academic medicine after fellowship, our interactions with community providers are limited to coordinating care or giving second opinions. For those of us who choose community or private-practice oncology after fellowship, our days are primarily spent seeing patients, making the occasional referral to an academic center, and sometimes not even talking to the academic provider in person.

You can see how our practice and learning environments force us to think of academic or community oncologists as "the other." How can we improve this relationship?

Perhaps one answer is increasing exposure to community practice during fellowship so that we understand the time, financial, and other pressures that community oncologists face. This could continue into practice if academic oncologists increased the amount of time they spend—either virtually or in person—seeing patients in the community. And by all means, we need to open electronic and other communication channels between academic and community oncologists, so that getting a specialist's opinion or coordinating a patient's care does not involve jumping through a series of hoops.

An Epilogue

I eventually did manage to get in touch with the community oncologist for our patient with prostate cancer, informing her of what we recommended while the patient was in the hospital. She clarified many of our previous questions. She had strongly recommended chemotherapy initially, but the patient had declined, preferring pain control and novel antiandrogen therapy instead. And germline testing had been sent and was pending. We discussed other aspects of the patient's care.

She had already done pretty much exactly what we recommended.

But, perhaps most important, she had recognized that this patient needed to hear the opinion of another oncologist, and she was happy that we had consulted on the case and agreed with her plan. "You'd be surprised at how rarely I get a chance to talk to you guys up there!" she said.

Perhaps we could all do a better job at that.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.