Monday, November 11, 2013

Continued Concerns for Building the Capacity of the Clinical Informatics Subspecialty

About a year ago, I wrote a posting expressing concerns about how we will build capacity for the clinical informatics subspecialty. A year later, the subspecialty has moved forward, with the first offering of the certification exam this past October as well as release of draft rules for fellowships accredited by the Accreditation Council for Graduate Medical Education (ACGME).

I still, however, have many of the concerns raised last year, especially now that some of the challenges for standing up ACGME-accredited fellowships have come into sharper focus with the release of the draft rules, and some of us actually trying to figure out how we will develop these fellowships within our own institutions. While the current "grandfathering" period is likely to result in a good starting number of certified individuals, the subspecialty will only succeed in the long run if there are enough training programs to meet the workforce needs of the subspecialty going forward. If only a small number of training programs develop, or they are very difficult for individuals to enter for reasons having nothing to do with their qualifications, then the subspecialty may not be sustianable in the long run.

I have had many of these concerns since the very idea of a subspecialty was first developed by the American Medical Informatics Association (AMIA) around 2006. While I still greatly support the notion of professional recognition of those who work in informatics, I nonetheless have concerns about trying to do so using the mechanism of the medical board subspecialty. In this posting, I will update my thinking on these concerns and also present an alternative for addressing them, even though I am well aware that it may be incompatible with the rules, or at least the traditions, of ACGME.


For those not familiar with the conventional approach to medical subspecialty training, it is important to remember that clinical fellowships differ somewhat from the graduate education model under which most informatics education programs have historically operated. In graduate education, progress is made in units of courses. While most courses are based on specific subjects and/or competencies, there can also be courses for internship/practicum work, theses/dissertations, and integrative learning experiences. With the world's growing number of adult learners in all fields, graduate education can be pursued on a part-time basis, at the appropriate pace for the learner.

Medical training (including clinical fellowships), however, has historically progressed in time-based units, typically in units of years. As such, internal medicine or family medicine residencies are three years long, with other specialties having similar or longer training periods. Those who pursue subspecialty training do so in fellowships lasting one to several years. Sometimes a training program will be extended for research or other work, but almost always in one-year increments.

It is therefore no surprise that the draft ACGME rules propose a time-based approach to training. According to the draft rules, the clinical informatics fellowship will be two years long. While ACGME is currently reviewing public comments, including ours, it is unlikely they will change from that two-year length.


Whereas my concerns about clinical informatics fellowships last year were conceptual, we now have real (draft) rules to figure out how to develop fellowships. My concerns are not lessened, and I see potential problems. These include the funding model of fellowships, the requirement to align in a given institution with a single medical specialty, and the arbitrary two-year time frame. I do believe, however, that there are means to accomplish the same ends by a different approach, which I will explain after elaborating more on my concerns.

The funding problem for clinical informatics fellowships will stem from the fact that these fellowships will be different from most other subspecialty fellowships. This derives from the regulations of the Centers for Medicare and Medical Services (CMS) that when one is a clinical fellow, he or she is viewed as a trainee and not allowed to perform any clinical practice without supervision, even if fully board-certificated in the area of practice.

In a typical clinical fellowship, this mostly makes sense, as the clinical work that a fellow is doing is related to the subspecialty in which he or she is training. A clinical informatics fellow will be functioning somewhat differently, as his or her fellowship work will be done in informatics, whereas he or she will be expected to maintain their clinical skills in the practice of the specialty in which he or she was originally board-certified. I am not aware of any fellowship that asks a clinical fellow to maintain practice in their original specialty. Cardiology or rheumatology fellows are not asked, for example, to maintain their practice of general internal medicine, which is the specialty in which they trained before entering their subspecialty fellowship.

Continuing this example, a person entering a cardiology or rheumatology fellowship after completing an internal medicine residency will perform his or her clinical work in their fellowship in cardiology or rheumatology. A clinical informatics fellow training after an internal medicine residency, on the other hand, will be doing his or her "clinical" work in informatics and practicing general internal medicine, mostly to maintain his or her clinical skills. A challenge is that this practice will need to be supervised as if the fellow were a trainee, even if he or she is already board-certified. This will incur a cost to institutions in that attending physicians are limited in the number of residents and fellows they can supervise at any one time. By requiring the clinical informatics fellow to have supervision, it is one less other resident or fellow that an attending physician can supervise under CMS rules.

This underscores the bigger challenge of figuring out how to fund the fellowship. Unlike a graduate education program, where someone (the student, a training grant, a scholarship, etc.) pays the tuition, which in turn supports the program, a clinical fellow earns a salary, which is funded by either CMS graduate medical education (GME) money or the institution itself. Most clinical departments in academic medical centers can fund such fellowships out of other revenues, including the practice revenues of the attending physicians. The challenge for clinical informatics fellows will be how to fund the fellows, not only their stipends, but also the other costs, such as faculty supervision time, tuition for the courses offered, scholarship expenses (e.g., travel to meetings), and administrative costs.

One option suggested by some is to allow the fellows to "moonlight," a tried and true way for medical trainees to augment their income. A problem with moonlighting for clinical experience and revenue is that it will disconnect the clinical informatics fellow clinically from the setting where he or she is doing their informatics training, thus making them less integrated with the environment whose informatics systems they are learning to improve.

A second major problem is the ACGME requirement for clinical informatics fellowships to be aligned with one of six (perhaps increasing to nine) residency programs. This is mainly being done for efficiency reasons, so that new residency review committees (RRCs) that review residencies and fellowships for accreditation will not need to be created. It is stressed that programs can accept anyone from any specialty, but in reality programs will likely resemble the specialties to whose programs they are administratively aligned, if for no other reason due to the normal give and take of academic medical center negotiations and politics.

A third major problem is the two-year time frame. While I do believe that most clinical informatics subspecialists are likely to require two years to complete their training, the arbitrary two-year time frame is at odds with the growing change in medical education from time-based tp competency-based education. Learning informatics is no more like "steeping tea" than medical education [1].

One of the outcomes from these problems is that it may be too difficult for there to be an adequate number of fellowship programs for qualified individuals from all medical specialties to make the subspecialty truly viable. If each fellowship has a one-off situation (i.e., only able or willing to take trainees from one or a small number of specialties) or must be funded, to use the words of my institution's chief medical officer, "creatively," this may undermine the long-term viability of the subspecialty.

Alternative Approach

I cannot criticize the proposed approach without offering an alternative, and I believe there are approaches that could be rigorous enough to ensure an equally if not more robust educational and training experience than the proposed fellowship model. It would no doubt test the boundaries of a tradition-bound organization like ACGME but could also show innovation reflective (and indeed required) of modern education.

We must remember that there will be three basic activities of clinical informatics subspecialty trainees:
  1. Clinical informatics education to master the core knowledge of the field
  2. Clinical informatics project work to gain skills and practical experience
  3. Clinical practice to maintain their skills in their primary medical specialty
How can we insure that these three general areas are met without requiring a trainee to be in a fixed location or to train over a fixed period of time? The first activity area is easy. There is no question any more that essentially any learning experience can be delivered using online learning. There are many educational programs in informatics, including ours, that have a strong track record in successfully imparting knowledge to students remotely [2].

Second, how would trainees get their practical hands-on project work? Again, many informatics programs, certainly ours, have developed mechanisms by which students can do internships or practicums in remote location through a combination of affiliation agreements, local mentoring, and remote supervision. While our program currently has students performing 3-6 months at a time of these, I see no reason why the practical experience could not be expanded to a year or longer. Strict guidelines for experience and both local and remote mentoring could be put in place to insure quality.

Finally, what about clinical practice? This may be easiest of all. Requiring a trainee to perform a certain volume of clinical practice, while adhering to all appropriate requirements for licensure and maintenance of certification, should be more than adequate to insure practice in their primary specialty. Many informatics distance learning students are already maintaining their clinical practices to maintain their livelihood. Making clinical practice explicit, instead of as something requiring supervision, will also allow training to be more financially viable for the fellow. Any costs of tuition and practical work could easily be offset by clinical practice revenue.

There would obviously need to be some sort of national infrastructure to set standards and monitor progress of clinical informatics trainees. There are any number of organizations that could perform this task, such as AMIA, and it could perhaps be a requirement of accreditation.

In fact, ACGME and the larger medical education community may learn from alternative approaches like this for training in other specialties. One major national concern these days is that number of residency positions for medical school graduates is not keeping up with the increases of medical school enrollment or, for that matter, the national need for physicians [3]. It is possible that alternative approaches like this could expand the capacity of all medical specialties and subspecialties, and not just clinical informatics.


1. Hodges BD, A tea-steeping or i-Doc model for medical education? Academic Medicine, 2010. 85: S34-S44.
2. Hersh WR, The full spectrum of biomedical informatics education at Oregon Health & Science University. Methods of Information in Medicine, 2007. 46: 80-83.
3. Iglehart JK, The residency mismatch. New England Journal of Medicine, 2013. 369: 297-299.


  1. How about dumping the whole idea of fellowship? Like you mentioned, a 20th century approach will not help for developing a 21st century specialty. Online study + project experience will be the ideal approach...

    1. I agree, and online study plus project experience is essentially what our graduate program currently offers. However, if we want to be an ABMS-approved subspecialty, we must have training via ACGME-accredited fellowships. Hopefully the model will change over time.