Friday, November 20, 2009

What is the optimal "pre-informatics" education?

Readers of this blog and other writings of mine know that I have written a great deal about the optimal education and training needed to attain the knowledge and skills to work in the biomedical and health informatics profession. However, I am increasingly asked by people what is the best education and background to have before commencing informatics education. In other words, what is the most appropriate "pre-informatics" education?

Unfortunately, this is probably not a question that has a single answer, due to the heterogeneous nature of the jobs carried out by people educated in informatics and their equally heterogeneous backgrounds going in. While many in education leadership these days talk of "career pathways," the reality is that there are many pathways that feed into informatics as well as many pathways out into a variety of jobs (as demonstrated in Figure 3 of my "A stimulus to define informatics and health information technology" paper)

It is also important to remember that biomedical and health informatics is not just the intersection of healthcare and information technology (IT). Rather, it is the unique synergy and interaction that takes place when those and other disciplines intersect. I have made this point in other postings in this blog and others.

As such, you cannot be optimally trained in informatics just by having a background in its constituent disciplines. (This is one reason why I have trouble with educational institutions that are quickly creating informatics programs merely by combining, for example, healthcare and IT courses, as noted in the above Healthcare Informatics posting.)

But let's try to develop some notion of what advice we might give to someone considering education in biomedical and health informatics. My discussion is predicated on my view (bias?) that informatics is best taught at the graduate level, where one brings together a variety of competencies into a final common pathway. Therefore, one should likely have a baccalaureate degree in one of the areas I describe next, although we have seen plenty of examples in our program of those with prior degrees in completely different fields, such as law and economics.

Clearly first and foremost on the list of advice is having some knowledge or a prior degree in the underlying biomedical or health domain of one's interest. In the case of clinical informatics, this is an understanding of healthcare and its knowledge, way of thinking, and workflow. One does not necessarily need to have a formal healthcare degree (e.g., medicine, nursing, pharmacy, etc.), but there is a clear advantage to having one.

In the case of other areas of biomedical and health informatics, analogous reasoning applies. In bioinformatics, for example, one should have a strong background or prior degree in biology and/or other life sciences. In public health informatics, one needs a substantial background or prior degree in public health or a related area.

Since informatics is often (incorrectly, in my view) equated as IT or computer science (CS) in health care or biology, the next question is, how much of an IT or CS background is required? The answer to this question is that it depends on the career pathway desired. Clearly everyone in biomedical and health informatics needs to be facile and competent with IT. They must have an aptitude for quick learning of IT systems, i.e., be a power user of computers, especially in areas like productivity applications, searching, Web applications, and the like. They must also understand "information," and have skills in its application to further goals of healthcare, biomedical research, public health, and the like.

Beyond that, the amount of IT or CS knowledge depends on one's career goals. Certainly someone who wishes to engage in tasks such as data mining, text mining, and computational biology must have a deeper knowledge. These individuals must know how to program, understand information and system architecture, and be able to adapt to new technologies as they emerge to solve specific tasks. But if someone's focus is going to be leading an electronic health record (EHR) implementation in a healthcare organization or helping healthcare teams analyze data for quality, a deep understanding of IT and CS is less necessary.

Related to IT and CS is mathematics. Again, the amount necessary depends on one's career objectives. I personally believe that every "knowledge worker" in the world should have a basic understanding of statistics. This is not just the various statistical tests and when they are appropriately applied, but also the foundational knowledge of descriptive and inferential statistics. This is not just a requirement for being a good informatics professional, but also a good citizen, and appropriately understanding research results, risk analysis, and other important issues of the world. Certainly anyone who is going to do any kind of analysis of data in their informatics work needs to have a basic knowledge of statistics.

Another set of skills that are important for many informaticians to have are business skills, soft skills, and other abilities to work with people to achieve organizational and/or project goals. Perhaps the person coding bioinformatics algorithms or data mining routines might not need much of these (though good project management skills never hurt anyone!), it is more the individuals involved in management and leadership of IT in biomedical and health settings who need these skills.

Some informaticians need other specific backgrounds and skills. For example, anyone who is going to become a researcher needs education not only in the specific research methods they hope to apply, but also exposure to larger aspects of critical thinking, study design, and related topics.

So my advice to those seeking to develop or further their careers in biomedical and health informatics through education is to have a general sense of your career direction, bring as much as the above pre-requisites as you can into the educational program, and then be prepared to learn about the rest while bringing them all synergistically together to be the best informatician you can be. While a previous degree in one of the foundational areas of informatics is helpful, it is not an absolute requirement.

Monday, November 9, 2009

Academia = Education AND Research

The American Medical Informatics Association (AMIA) Annual Symposium is clearly the best informatics meeting of the year. It is rigorously academic, so the quality is top-notch, but not too much, so you can glean plenty of practical information as well.

The AMIA symposium is also an opportunity for us because we can showcase our department. As always, our faculty and students will pepper the program with great papers, panels, and posters. In addition, the annual OHSU banquet is a gratifying display of the energy and passion of our program, not to mention quite fun. I look forward to this year's meeting in San Francisco later this month.

This meeting always give me a chance to reflect on the importance of a comprehensive academic program that values both education and research. A vibrant graduate-level program cannot thrive without both. Being at the cutting edge of research allows faculty to be the knowledge and thought leaders in their respective areas.

This was borne out a couple years ago when we hosted a focus group that assembled a number of what we call "local distance" students, which are students who live in the Portland area but prefer to enroll in our on-line program. We wanted to know why they preferred that instead of coming "up the hill" to the OHSU campus. The answers were obvious in retrospect: they appreciate the convenience of being able to carry out their studies at their preferred hours (usually evenings and weekends) and they did not want to deal with the hassle of driving to and parking on our campus (which everyone knows can be a pain, at least during regular working hours).

There was, however, another interesting finding that came from the focus group. These students told us they were drawn to our program not only because of its local connection, but also because they valued the faculty and their leadership roles in the field, especially their research. Even though they were unlikely to become researchers themselves, or for some to even do research, they believed it was important to obtain their education in a department that was known for being a leader in research as well.

As always, I look forward to catching up with students, alumni, and old friends at the AMIA meeting.

Wednesday, November 4, 2009

The workforce for meaningful use

The discussion about health IT workforce continues to heat up as health care organizations realize that achieving meaningful use of electronic health records will require not only hardware and software, but people who have the expertise to make it happen. That expertise requires as much an understanding of information use and analysis, clinical organization and workflow, and business and management as it does IT, i.e., the substance of informatics.

One recent article describes a developing "war on talent" for health IT workers. The same publication features another article about how health care organizations are "racing" to fill CIO positions.

Finally, an IT publication describes why "your next job may be in health care."

By the way, many people ask me where they can read a succinct overview about "meaningful use," and I have found a nice 6-pager by David Classen of CSC. Of course, the "ground truth" comes from the matrix recommended by the Office of the National Coordinator to CMS, who will set the final rules in the near future.