Communication and Care Coordination

December 20, 2013

Optimizing Patient Care Between Primary Care and Subspecialist Clinicians


Molly Cooke, MD, FACP

President, American College of Physicians; Professor of Medicine, Director of Education and Global Health Sciences, University of California, San Francisco


In this guest column, at the invitation of AGA’s president, Anil K. Rustgi, MD, AGAF, I would like to explore the issue of care coordination between generalist physicians and subspecialists, using internists and gastroenterologists as my case in point. I am an internist at University of California, San Francisco (UCSF). While I focus on HIV, my patient panel runs the full gamut on internal medicine and I have ample opportunity to collaborate with my GI colleagues. Undoubtedly, my most common referral to GI is for screening colonoscopy but even in my “smallish” practice, I have many patients with serious chronic disorders of the liver and gut, including ulcerative colitis, hepatitis C, HIV–HBV co-infection with cirrhosis and recurrent iron deficient anemia with intermittent guaiac-positive stools in elderly patients.

As a resident in internal medicine, I was taught that there were four reasons to consult a subspecialist:

  1. The patient presents a true diagnostic dilemma.
  2. A procedure is required that internists don’t do.
  3. There is a diagnostic or therapeutic decision to be made that requires more experience-derived judgment than a non-subspecialist can possess.
  4. The patient has an interesting finding, an unusual presentation or a rare disease that the fellows should have a chance to see.

These still seem like pretty good rules to me. But many teachers have commented that residents are partitioning their patients, delegating their problems to the relevant subspecialist. Anemia? Heme consult. Ulcerative colitis? GI. Of course, this apportioning of the patient creates all kinds of problems: consultants from different subspecialties may make conflicting recommendations, access to subspecialists for patients who truly need subspecialty input is impaired, patients can’t figure out who their doctor is, to name just a few. And then there are problems in the other direction: the subspecialist schedules a follow-up appointment when I was just asking a question, the subspecialist refers our patient to another subspecialist for a problem I was managing, the subspecialist makes a host of medication changes without letting me know the rationale.

How can we work together to make the ever-increasing coordination and collaboration our patients require work better while we are getting busier and busier? The solution is not likely to be found in increasing individual effort. Rather we need systematic and structural approaches. At UCSF we are experimenting with structured consultation/referral forms and what Justin Sewell, a UCSF gastroenterologist, has called “preconsultation exchange.” Both of these approaches have similar goals:

  1. To “right-size” the intensity of the subspecialist’s involvement.
  2. To make sure that the patient arrives at the subspecialty appointment with the studies needed to address the clinical question.
  3. To ensure that both the referring physician and the consultant have a shared understanding of what the consultant is being asked to do and who is responsible for what going forward.

Structured referrals are electronic templates within our EHR. When I indicate that I would like to refer a patient to GI, I am offered a drop-down menu of scenarios, such as “abdominal pain” and “hepatic mass.” Once I have selected the circumstance, the template “coaches” me to provide results of relevant tests. If I refer a patient for advice about the further evaluation of patient with elevated transaminases, the template will auto populate with the results of pertinent tests I have already obtained and remind me that I should order an antinuclear antibodies test and an international normalized ratio test before referring. (The negotiations between the general internists and the subspecialists as the templates were being developed about what constitutes high-value, routine evaluation prior to consultation were sometimes quite animated.) In addition, the template requires me to state a question or make an explicit request of my consultant and to indicate my intentions with respect to further care: one-time consultation, co-management or transfer of the subspecialty problem. The referring doctor is also asked to say whether it would be acceptable for the consultant to comment on the problem without seeing the patient.

Our experience with these approaches has been very positive, from the perspectives of both the referring doctor and the consultant. Perhaps even more important, patients appreciate that they don’t make unnecessary visits to subspecialists when an e-consult will do, and that they don’t waste their time at a subspecialty visit without the labs or imaging that the consultant needs to address the clinical question. The involved physicians have a clear and shared understanding of who is responsible for what in the ongoing care of the problem.

In addition to institutional experimentation with increasing the efficiency and value of consultations and referrals, national initiatives are underway. Professional societies are collaborating on the issue — on Sept. 30, 2013, the ACP’s Council of Subspecialty Societies convened the second annual Subspecialty Summit. The day-long meeting, attended by 23 societies including the AGA, focused on exactly this issue. Meanwhile, the increasing focus on value-based payment structures, including bundled payments and shared savings, will reward both the referring physician and the subspecialist for effective, high-value referrals and consultations.

One last thing — and it is a sore point among many internists, as our gastroenterology colleagues surely know — when we refer a complex patient for a nuanced expert opinion and the patient is seen by a nurse practitioner (NP) or physician assistant (PA), rather than a physician. Before I take this further, let me say that in the years that I worked in the AIDS clinic at San Francisco General Hospital (SFGH) my favorite consultant for difficult points of HIV management was a PA. What I know about learning in the workplace has convinced me that the right individual, regardless of their formal training, can develop true expertise and nuanced judgment. But not everyone does. If having thought about a patient long and hard and deciding that I want to the opinion of Dr. Doe, I want Dr. Doe’s opinion, just as I would look for my PA colleague when I was working at SFGH. While several factors provoke the referring physician’s irritation when she refers a patient to a gastroenterologist and the patient is seen by an NP, the solution is simple: forging shared expectations prior to the visit, expectations that are shared by the referring doctor, the consultant and the patient.

I greatly appreciate the opportunity to write a guest column for AGA Perspectives and look forward to your comments. Let’s continue the conversation.

Dr. Cooke is the current president of the American College of Physicians.

View from Gastroenterology


John I Allen, MD, MBA, AGAF

President Elect, American Gastroenterological Association; Professor of Medicine, Section of Digestive Diseases, Yale University School of Medicine, New Haven, CT


I have been asked by Anil Rustgi, MD, AGAF, to reflect on the excellent guest column written by Dr. Molly Cooke, president of the American College of Physicians (ACP), about care coordination between generalist physicians and gastroenterologists. I am currently in an academic medical center which uses an enterprise-wide electronic medical record (EMR) and have reduced my personal practice to hospital attending with minimal ambulatory clinics. Recently, however, I moved from a busy community practice in the Minneapolis/St. Paul area, also known as the Twin Cities, where communication with primary care was critical to focus resources and effective medical services.

The pressures of diminishing fee-for-service reimbursement, increasing patient numbers and the movement towards value-based payments (think capitation and total cost-of-care risk contracting) are combining to force us to think differently about care coordination. It is relatively simple to manage a request for a screening colonoscopy for a healthy 50-year-old, but coordinating care of a complex patient with multiple other providers and ongoing comorbid conditions often stresses both our consultative skills and our practice infrastructure. The goal of “right-sizing” the intensity of our involvement is attractive, especially as new shared financial risk models develop.

I would be a strong advocate for structured referrals and alternative modes of patient visits coordinated through our EMR. Should a primary care physician in Stamford, CT really force a patient to see me in New Haven (after fighting the traffic along Interstate 95) only to find out they came with incomplete records or for a condition that I could manage remotely? I would much prefer to help a colleague manage our mutual patient with asynchronous or even synchronous electronic or video conferencing. The promise of EMRs really would be fulfilled when this comes to fruition. Unfortunately, since the inception of the current Medicare physician payment methodology in the 1980s, we have been bound by the tyranny of the face-to-face visit (we do not get reimbursed for remote consultation with rare exception). AGA leadership agrees with the ACP that we need to advocate for removal of this barrier to remote consultation in the spirit of better coordination of care.

One of our goals is to develop better lines of communication and solve issues of shared expectations, coordination of care and boundary definition. We have initiated work on shared clinical decision support tools, clinical care algorithms and hopefully will begin developing consultation templates that will advance our care coordination, especially of the complex patients that need our mutual skills. We hope that policy makers, Medicare, EMR vendors and others who create regulations that so heavily influence our day-to-day practices will notice that positive, patient-centered results can be achieved.

Dr. Allen is president elect of the American Gastroenterological Association. He has no other conflicts to disclose.

4 out of 4 users found this page helpful.

Was this page helpful?