Section 3016 Has Arrived!

The long-anticipated Funding Opportunity Announcements (FOAs) for health information technology (HIT) workforce development, as specified in Section 3016 of the American Recovery and Reinvestment Act (ARRA), have been released by the Office of the National Coordinator for Health IT (ONC). These FOAs reveal the overall plans of ONC for quickly ramping up the workforce to meet the ARRA electronic health record (EHR) adoption goals. This will be done by two specific FOAs, one that funds development of five regional consortia of community colleges around the country, who in turn will implement curricula developed by the other FOA that funds five national curriculum development centers.

ONC has also provided its own analysis and plan for the workforce. Both FOAs state a need to train 51,000 workers to implement the ARRA HIT agenda. They list specifically six job roles for this workforce, noting that preparation for any them will typically require six months of intense training for individuals with appropriate backgrounds. These individuals could work for a variety of entities, including health care organizations, regional extension centers, government organizations, vendors, and others. (It would be great for those of us with an interest in workforce research and education to see their data and analysis of the workforce needs. I hope they release it at some point.)

Four of the job roles are "mobile adoption support positions" and involve personnel who carry out the initial EHR implementation at a site and then move on to the next one. The specific roles for these positions are:

• Practice workflow and information management redesign specialists
• Clinician/practitioner consultants
• Implementation support specialists
• Implementation managers

The other two job roles are "permanent staff of healthcare delivery and public health sites" and provide ongoing support after implementation. The specific roles for these positions include:

• Technical/software support staff
• Trainers

Although this process presents a coherent and specific plan to quickly ramp up the workforce, an approach with which I concur, I also hope it also raises the visibility of clinical informatics as an important component of EHR adoption and a career option for people who aspire to work professionally (and academically) in HIT. I am somewhat disappointed that the programs developed will not have pathways that articulate with more formal degree programs (e.g., baccalaureate and master's degrees, or even associate degrees for that matter). I do hope that educators who participate will be given the option to find ways to make further articulation and career enhancement happen. In addition, while I know that community colleges have historically risen to the challenge of quickly implementing skills-based training, I also wonder whether those with no experience or expertise teaching this content will be able to ramp up quickly enough. Where, for example, will they find faculty with sufficient expertise to do more than just deliver a packaged curriculum?

One also has to wonder whether those with short-term training will have the complex array of knowledge and skills in areas like healthcare workflow, change management, quality measurement, and oral and written communication (among even more) that are so essential for successful EHR implementation in complicated entities like healthcare organizations. We will no doubt be witnessing a great workforce and education "experiment" (just like, I suppose, all of the ARRA EHR agenda). I look forward to being a "subject" in that experiment.

On to the details. The first FOA (EP-HIT-10-001) is entitled, Community College Consortia to Educate Health Information Technology Professionals in Health Care Program. This FOA divides the country into five regions, each of which will have one consortium. Each consortium has a target amount of funding, number of students, and number of institutions. (Oregon is in Region A, the smallest of the five regions in terms of population served.) The overall allocation is $70 million.

Each consortium will have a lead organization and targeted number of member community colleges (varying from 5-8 in Region A to 17-23 in Region E in the Northeast). Each will have an expected minimum number of students each year, with the overall goal of training 10,500 students annually. Programs will not award degrees, but instead award a six-month certificate. It is assumed many students will have some background in healthcare and/or IT, and will need to flexibly learn more of what they do not know. Each member institution does not have to offer training for all six job roles, but each consortium as a whole must cover all the roles. The number of institutions must be within the specified range (unless an exemption is given), though each member does not need to train an equal numbers of students. Training must be commence by September 30, 2010.

Funding is for two years. The FOA does not address sustainability, either as requiring it in the plan or its being assessed in evaluation of proposals. It does say that the funding is a one-time opportunity. Up to $1M per consortium can be budgeted for consortium expenses. Up to 8% indirect (F&A) costs will be awarded to the lead institution and any others that have federal F&A agreements. My reading of the FOA is that any institution of higher learning, including a university, that offers HIT training (e.g., OHSU) can be a lead institution. The amount of funding works out to about $3,000-$4,000 per student trained (depending on the region and how much is allocated for consortium expenses).

The second FOA (EP-HIT-10-003) is entitled, Curriculum Development Centers Program. The goal of this FOA is to develop curricula that will be deployed by the community college consortia formed in the other FOA to train HIT professionals in 20 topics the cover the six specific job roles.

This FOA will fund five centers to develop materials in some number (applicants need to specify 7-10 where they have expertise) of areas from a list of 20. They are required to be academic training programs in informatics, HIT, or HIM, and engage both community college educators as well as instructional design experts. One center will also be designated the National Training and Dissemination Center, and additionally carry out training as well as establishing the repository for download of materials.

Each center will be funded up to $1.82M over two years ($910,000 per year), including indirect costs. The National Training and Dissemination Center will receive an additional $900,000 ($450,000 per year), including indirect costs.

Like the other FOA, there is a great amount of prescriptive detail provided. The 20 curricular areas from which to choose include:

• Introduction to Health Care and Public Health in the U.S
• The Culture of Health Care
• Terminology in Health Care and Public Health Settings
• Introduction to Information and Computer Science
• History of Health Information Technology in the U.S.
• Health Management Information Systems
• Working with Health IT Systems
• Installation and Maintenance of Health IT systems
• Networking and Health Information Exchange
• Fundamentals of Health Workflow Process Analysis & Redesign
• Configuring EHRs
• Quality Improvement
• Public Health IT
• Special Topics Course on Vendor-Specific Systems
• Usability and Human Factors
• Professionalism/Customer Service in the Health Environment
• Working in Teams
• Planning, Management and Leadership for Health IT
• Introduction to Project Management
• Training and Instructional Design

This whole approach is not without risk, but as I said above, it is a great plan for quickly ramping up the workforce. I do hope that it will allow for professional growth and development for those undertake the training and want to work as professionals in this extremely important component of improving healthcare. I also hope that the process will be evaluated well, so we can make corrections to the process when we find areas it does not work. In any case, I will stop writing now and get back to working on my proposals!

Two new international workforce studies

Things are really heating up in the health information technology (HIT) workforce arena! Not only have two new workforce analyses from abroad been published, but in the US, the long-anticipated Funding Opportunity Announcements (FOAs) for HIT workforce development, as specified in Section 3016 of the American Recovery and Reinvestment Act (ARRA), have been released by the Office of the National Coordinator for Health IT (ONC).

The first workforce study comes from Australia (Legg, M. and Lovelock, B., 2009. A Review of the Australian Health Informatics Workforce. Melbourne, Australia: Health Informatics Society of Australia). Commissioned by the Health Informatics Society of Australia (HISA), this analysis looked at the "health informatics" workforce in that country. The authors defined health informatics for the purposes of its analysis, although also noted that this definition was likely incomplete. The study was carried out in two steps consisting of two focused workshops with a small number of people and then a larger survey that was distributed to all members and anyone who ever participated in any HISA activity. A total of 1,279 (out of 6,434 possible) people completed the latter.

The study classified jobs into two broad categories:

1. Those who work "in the system," e.g., to use their words: records, analysis, direct, decision, communications, and training
2. Those who work "on the system," e.g., to use their words: systems, infostructure, improvement, education, resource, and administration

The study characterized the work of these individuals by categories (many performed more than one) and also captured data on perceived needs for education, training, and expansion of the workforce. The authors concluded with a number of recommendations for expansion and improvement of the workforce.

On the issue of workforce size, the authors made several estimates, all of which came out with a relatively consistent range. This included the use of government employment statistics for health information management and IT professionals in health care. The authors also made a quick calculation based on my own previous research (of other research studies) finding a ratio of 1 IT worker per 50 non-IT workers in healthcare settings. The bottom line is an estimate of around 12,000 health informatics professionals (range 9,000-15,000) in a country with a population of approximately 21 million. (This number is not that far off the proportionate number of people identified for the United States in my HIMSS Analytics Database study, i.e., 108,390 IT personnel, perhaps along with 10,000 informaticians and 50,000 need for the ARRA EHR agenda, in a country of over 300 million.)

The study also reported that participants consistently expressed concern about a pending shortage of personnel, for which the authors proposed a number of solutions. These included efforts to increase the supply of workers through training and re-training, improving their productivity, and reducing demand through better design of systems and processes.

The second workforce study comes from Canada (O'Grady, J., 2009. Health Informatics and Health Information Management: Human Resources Report. Toronto, Ontario, Prism Economics and Analysis). It focused on professionals defined as working in the fields of health informatics and health information management . This study attempted to characterize job roles and activities, quantify the workforce, and anticipate future needs based on low, medium, and high growth scenarios. (They also noted that Canada was similar to the US and different from most Western European countries in being a laggard in the adoption of HIT.) Similar to other studies, they used a variety of techniques, including government employment statistics. They calculated needs based not only on growth in use of HIT but also replacement of those retiring or otherwise leaving the workforce.

The study looked at seven categories of workers, assessing job roles and specific challenges for each. In aggregate, the authors estimated current HIT employment in Canada at 32,450, broken down among the job categories as follows (approximately):

1. Information Technology - public sector 11,000-13,000, private sector 5,000-6,000
2. Health Information Management - 4,300-5,800
3. Canadian Health System Management and Administration (counted in Analysis and Evaluation)
4. Project Management - public sector 760-900, private sector 1,100-1,300
5. Organizational and Behavioral Management - public sector 1,100-1,300, private sector 1,900-2,300
6. Analysis and Evaluation - 3,300-3,900
7. Clinical Informatics - public sector 600-710, private sector 380-450

The growth scenarios were based on the following (interesting) assumptions:

• Low growth - across the board expenditure cuts by the Canadian health system
• Medium growth - deferring of some capital investments
• High growth - HIT investment unaffected by changes in health care spending

The cumulative employment growth for all categories combined to 2014 would then be:

• Low growth - 7.6% growth to 35,020
• Medium growth - 14.3% growth to 37,200
• High growth - 26.1% growth to 41,030

For a country with a population of 33 million, these numbers are similarly proportionate to the Australian and US data!

The study also noted that Canada, like most countries, will face a substantial training need for many current and future personnel. Not only does 27% of the current 2009 workforce require additional formal training and experience, but 39%, 59%, or 78% of the 2014 workforce will require additional training under the low, medium, and high growth scenarios respectively.

So clearly Australia and Canada have major HIT training needs. Of course, so does the United States, as noted by my HIMSS Analytics Database study as well as recent estimates by ONC. To that end, two FOAs were recently released by the ONC to address workforce needs, as specified by Section 3016. These will be addressed in another posting soon.

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.

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.

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.

Informatics Is Not Just For Clinicians

I receive a steady stream of emails from people who are interested in careers and/or education in the biomedical and health informatics field. To the extent I can, I try to reply, giving advice and steering them to more information.

One inquiry I recently received was from someone who has an information technology (IT) background and noted that most of my writings seem to imply that informatics is a profession mainly for those with clinical or other healthcare backgrounds. He noted that I point to research and other observations that clearly show than an understanding of the clinical environment, its thinking, and its workflows are essential for career success in this field.

This individual asked, is there a role for non-clinicians in this field? My reply, as always, was a definite YES! Not only has our workforce research and the experience of others shown that there are plenty of opportunities for work for those who do not have clinical backgrounds, we also know that many of the 250+ alumni of our graduate program, a number of whom are non-clinicians, are gainfully employed.

This is not the first time I had been asked this question. In fact, we felt compelled to write about it several years ago in an issue of our department newsletter, noting even then that were plenty of jobs for non-clinicians in a variety of informatics settings.

However, it is clear that those without healthcare backgrounds must understand clinical environments. They need to understand its operations, it workflows, and even its thinking. But that can be learned, and for many jobs it is sufficient to not have formal training in a healthcare profession.

Now it is true that non-clinicians might end up in different jobs and follow different career paths than clinicians. Of course, that is the case even among the different types of clinicians. The best example of that is the position of Chief Medical Information Officer. This position is almost always filled by a physician. However, there are many other informatics jobs in healthcare settings that other physicians, other healthcare professionals (e.g., nurses, pharmacists, lab techs, healthcare administrators, etc.), and non-clinicians fill.

Some readers of this blog have seen my figure that provides an analogy from Bayesian statistics, i.e., what you do in a career after an informatics education is a function of both what you brought into the education and of what knowledge and skills you gained in the education. Ok, so the analogy is not perfect, but I hope it makes the point that informatics is a large and diverse field, and there are roles for people of many backgrounds who are passionate about using information to improve health.

What Level of Training Is Needed for Health IT and Clinical Informatics Jobs?

There will likely be a large number of jobs that result from the funding in the American Recovery and Reinvestment Act (ARRA, aka the stimulus bill) of 2009 providing incentives for electronic health record (EHR) adoption. Dr. Charles Friedman of the Office of the National Coordinator for Health IT (ONC) recently estimated that 60,000 people will be required for these jobs (PHIN 2009 Meeting Town Hall, September 2, 2009). ONC also recently posted on its Web site a set of job roles and competencies for this massive scaling up of EHR use that was developed in a workshop in August.

One question that arises is, who will provide all this education and training? A number of people have advocated that it be carried out by community colleges. A recent article in Healthcare IT News interviewed two people, a health insurance company executive and a president of a community college association, who advocated for community colleges to play that role.

In a rebuttal commentary, however, I replied that I was not so sure. There is no doubt that plenty of jobs in health IT will be for those educated in community colleges, such as the "informatics technicians" noted in a recent CNN posting about "emerging jobs poised for growth." But this is in distinction to the emerging clinical informatics role, which requires a combination of understanding the clinical environment and its workflows, ability to use advanced information analysis (more so than IT or computer science skills), and a myriad of business and soft skills. As the director of an informatics graduate program, I acknowledge my bias, but I advocated in my commentary that these programs, slightly re-orienting and focusing their curricula, may be better suited for training up this workforce. Since the proposed training must necessarily be short-term, I noted in my commentary that we are re-configuring our Graduate Certificate program into a 6-month program when pursued as a full-time student.

One line of evidence supporting my view comes from the Health IT Compensation Survey (Vendome, 2009). This year's survey features a wealth of data that goes way beyond compensation, and provides an interesting synopsis of the job functions and educational backgrounds of a wide variety of people who work in the industry. They segment those they survey into job setting (i.e., hospital, company, etc.), and across every segment, they subdivide people into leadership, clinical, and non-clinical positions.

Those in hospitals make up the largest segment in the survey, so I will focus on them. Among the leaders, 18% have doctoral or professional degrees, 48% have master's degrees, and all but 4% of the rest have bachelor's degrees. They subdivide the clinical and non-clinical professionals into "high authority" and "low authority." The breakdown of degrees within these groups is:

• Clinical/High Authority: 34% have doctoral or professional degrees, 29% have master's degrees, and 30% have bachelor's degrees
• Clinical/Low Authority: 20% have doctoral or professional degrees, 31% have master's degrees, and 35% have bachelor's degrees
• Non-Clinical/High Authority: 1% have doctoral or professional degrees, 36% have master's degrees, and 38% have bachelor's degrees
• Non-Clinical/Low Authority: 1% have doctoral or professional degrees, 24% have master's degrees, and 51% have bachelor's degrees

Clearly a majority of health IT professionals, especially those with clinical roles, have at least a bachelor's degree, and many have more. This is not surprising, as effective training for health IT requires expertise in the clinical environment, IT and information skills, and the ability to work with people in organizations. As I note in my commentary, that is a tall order for any educational program, let alone a two-year associate degree.

I do realize that community colleges play a strong role in rapidly adapting to skills needs in communities, and that many of their students are those who have bachelor's or even graduate degrees and return to attain new skills. And there is no question that some of the jobs in health IT will require the kinds of skills that community colleges already teach, such as those in pure IT. I acknowledge that the person hired to harden a server to prevent its security from being compromised probably does not need courses in change management. But many others who work in health IT do!

The reality is that few community colleges have expertise on their faculty in clinical informatics, which is not the mere addition of computer science, health information management, and health care courses as many seem to think. Informatics is what arises at the unique intersection of those areas, and the expertise for teaching it currently resides mostly in graduate-level informatics programs.