| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Section of Oral and Diagnostic Sciences, School of Dental and Oral Surgery, Department of Biomedical Informatics, College of Physicians and Surgeons, Columbia University, 360 West 168th Street, New York, NY 10032; zim{at}columbia.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMethodsConclusionsReferences |
|---|
KEY WORDS: Informatics • education • curriculum
|
Introduction |
|---|
|
TOPAbstractIntroductionMethodsConclusionsReferences |
|---|
The problem of developing a curriculum for biomedical informatics is highly dependent on how we choose to define and practice the field. Dentistry has a traditional struggle to define the field in terms of research, science, and evidence that can be applied to a patient as well as to a population. The dental curriculum is designed to create a medical professional and not a technician. Biomedical informatics is a relatively new field and, as is evident from the confusion in the name, has not differentiated curriculum models for various jobs or roles in the field. Many early leaders in biomedical informatics are physicians, nurses, and dentists, and this tradition of professionalism has influenced the informatics training programs in the United States. As we continue to define bioinformatics competencies in this paper, we will do so in the context of defining a curriculum for a biomedical informatics professional, and not a technician.
What is a profession? And if biomedical informatics is a profession, what are the qualities of professionalism a graduate of a biomedical informatics program should exhibit? Profession and knowledge are closely linked, and our desire to define our field and competencies is tied to the concept of profession and professionalism we have acquired from the medical disciplines. Professions have two qualities: knowledge and public service. Recently, professions have become more closely associated with the acquisition of expert knowledge and less with public good. The control and application of a specialized body of knowledge have come more and more to characterize a profession, as knowledge in all fields has grown and become more complex (Swick, 2000). As the medical profession has discovered, the debate about health care has been dominated not by physicians, individually or collectively, but by business, economic, and political interests. Without a strong sense of the public and social purposes served by professional knowledge, professionals tend to lose their distinctive voice in public debate (Brint, 1994). It is this voice that the profession of biomedical informatics is trying to establish.
The post-doctoral training model and the large number of health care professionals who are the leaders of the field have instilled a tradition of professionalism in biomedical informatics. This professional tradition is rooted in the medical model. Today, many of the biomedical informatics students do not come from the medical profession, and it is a challenge to create this concept of professionalism in all students. At DBMI, we have strong focus on acculturation as well as on competency. The focus of this paper is on core concepts and competencies in biomedical informatics. As we struggle to define the knowledge unique to our field, we must also define our profession.
As biomedical informatics educators and researchers continue to articulate the field, we can look to other fields, such as computer science and medicine, for curriculum models. Insight can be gained from the approaches taken in established academic disciplines. It is traditional to divide the sciences into the basic, such as physics, chemistry, and geology, and the applied, which include the various subfields of engineering and many aspects of medicine and dentistry (The Columbia Encyclopedia, 2002). This definition implies a continuum, ranging from the pure abstract world of mathematics, to empirical studies, to applications that utilize this knowledge to solve practical problems. Biomedical informatics has been positioned at various points along this continuum: as an engineering discipline concerned with developing and evaluating systems (Shahar, 2002); as a modeling discipline concerned with developing formal representation and problem solving methods (Musen, 2002); as a local science that attempts to explain aspects of a domain to design and implement artifacts (Patel and Kaufman, 1998); and as a broad field that ranges from model formulation, to system development and installation, to the study of their effects (Friedman, 1995).
In biomedical informatics, Maojo recommends that we adopt a model, dividing the science continuum into three areas: theory (mathematical constructs), abstraction (empirical validation of models), and design (implementation and assessment of systems) (Maojo et al., 2002). The model serves as an excellent framework to construct curricula for biomedical informatics, and to direct innovations. Derived from this model, we can begin to define the premise of our curriculum—that knowledge and theoretical models are necessary to build testable empirical models, and that understanding and skills of experimentation are necessary for deploying information systems that can be evaluated in a meaningful way (Table 1
) (Johnson, 2003). In building this educational framework, we seek to identify "a set of methods, techniques, and theories that have broad applicability" within the domain of biomedicine (Shortliffe and Johnson, 2002). The resulting set of core informatics competencies is intended to span a wide range of application areas: bioinformatics, imaging informatics, clinical informatics, and public health informatics.
|
|
Methods |
|---|
|
TOPAbstractIntroductionMethodsConclusionsReferences |
|---|
For example, database design is a formal competency and can be applied to storing biological sequences, images of organs, patient encounters, or epidemiologic surveys.
|
The first two domains, formal and empirical, form a foundation and a core set of competencies that are shared horizontally across the four biomedical informatics application domains. It is also useful to further define the competency domains across the ‘hierarchy of meaning’ pyramid (Jordon, 2002). These artifacts fall along a continuum of increasing complexity, in which three divisions can be made: data, knowledge, and systems. (’Systems’ is interpreted as knowledge in the original model.) As seen in Table 2, this provides a useful structure for organizing core competencies, in which mathematical and technical issues are distinguished from cognitive, behavioral, and organizational aspects (Johnson, 2003).
|
Empirical domain
After the student has achieved competency in the formal domain, he/she is prepared to take courses in the second level, the empirical domain. The empirical domain embraces competencies in the general categories cognition, behavioral, and organizational, and the prerequisites are knowledge of statistics and understanding of the scientific method and experimental design, including biomedical research design. In the data level, students learn about (1) data flow diagrams to understand organizational system, (2) data standards development and use, (3) cognitive aspects of vision and understanding, and (4) data security, confidentiality, access, authentication, and authorization. The knowledge level contains competencies that focus on: (1) knowledge acquisition, knowledge representation, and cognitive theories; and (2) decision analysis—Bayesian, belief networks, utilities, influence diagrams, and neural nets. The systems level contains systems lifecycle and systems evaluation.
Many biomedical informatics programs emphasize formal-level courses, and this is evidenced as they ‘borrow’ courses from other programs in the university system, such as computer science. When courses are not created to meet the specific needs of the biomedical informatics student, it leads to courses that do not address the empirical domain concepts described above and seem out of context to students, since the courses are not grounded in an application domain.
Applied domain
The final capstone level is the applied domain. This level is not meant to contain competencies or methods that cut across all four application domains and are common to all students in biomedical informatics. Rather, this level cuts down into the two foundation levels to bring core concepts, methods, and competencies into an application domain that is similar to a medical specialty. The prerequisites for the applied domain come from the biomedical sciences—molecular biology, anatomy, physiology, and public health. Table 3 lists some concepts that students master in their pursuit of competency in one of the four application domains.
|
|
Conclusions |
|---|
|
TOPAbstractIntroductionMethodsConclusionsReferences |
|---|
AMIA identified two types of informatics programs, applied and research, and used the respective workplaces as one means of differentiating between the categories. Informatics researchers are defined as those whose primary responsibility is research and teaching, typically in an academic setting. Applied health informaticians include CIOs, programmers, database administrators, systems analysts, and other professionals whose primary functions include building and implementing systems.
The American Medical Informatics Association Board of Directors charged the AMIA Education Committee to study the issue of core competencies in health informatics. The Education Committee built a Web site to collect input from the members of the Education Special Interest Group. AMIA defined competencies as skills and knowledge that could be considered to be fundamental to informaticians, whether they function in academia, healthcare settings, or industry (AMIA, 2003). Participants in the Web-based survey were asked to emphasize applications, skills, theories, and models when they completed the online forms. Survey participants entered and scored competency statements separately for practicing and academic informaticians. The skill-level qualifiers were: aware, knowledgeable user, designer, and researcher.
The AMIA core competency project will help define the field and further the science of biomedical informatics, but there are other topics to be addressed which are not as tangible as a core set of competencies. The humanistic aspect of the profession is one area that the faculty at Columbia University considers as important as our core competencies. The students in biomedical informatics programs are diverse with respect to their previous educational backgrounds. It is not unusual to have enrolled in a program or research project students who come from computer science degree programs, medical and health professions, cognitive science departments, basic science programs, and other disciplines. One reason for the diversity of the students enrolled in biomedical informatics programs is the lack of biomedical informatics Bachelor’s degree programs. The applicant pool diversity can be seen as a strength of biomedical informatics programs, but this diversity also points to the need for a well-organized acculturation curriculum or program for these students. Collectively, this diverse group of students must develop into a cohesive, yet richly diverse, profession.
The issue of breadth vs. depth in a student’s curriculum is also a challenge for biomedical informatics programs. One principle of the biomedical informatics core competencies is to define a set of knowledge and skills that all students will use, but the diversity of the biomedical informatics specialties makes this goal difficult to achieve. The diversity of knowledge required of bioinformatics, imaging, clinical, and public health informatics specialties is vast. If the core set is too widely defined, then the curriculum may be too crowded or too general. The breadth and depth issue also plays out in differentiating competencies between Master’s and PhD programs.
Biomedical informatics training programs have a strong tradition of post-doctoral training—taking traditionally trained medical professionals, then supplementing the medical education with a biomedical informatics education. This approach was successful when the field was young, but may not be the most appropriate or efficient training model for the future. In this model, biomedical informatics programs rely on students who choose to enter a health profession and, after completing their training, select another career path. I believe that the joint degree program—DDS/PhD or DDS/Master’s—will be more effective, and this is particularly true for dentistry. The development of the novice dentist to reach proficiency in the manual dexterity skills can be severely interrupted during an intensive biomedical informatics training program.
A final issue biomedical informatics may need to address is accreditation of training programs and recertification of professionals. The accreditation process can be a strong driving force in defining a profession, and often is a defining moment in that profession. If biomedical informatics continues to see itself as a profession and not just a science, then accreditation is a natural step in that defining process.
This issue points to the fact that the core competencies should be the set that forms the foundation for all the biomedical informatics competencies. Core competencies of this type will take many years to define if we follow this reductionist approach to defining our field, but I believe that the knowledge and insight we gain during this self-examination process will be worth the effort.
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMethodsConclusionsReferences |
|---|
Brint S (1994). In an age of experts: the changing role of professionals in politics and public life. Princeton, NJ: Princeton University Press.
The Columbia Encyclopedia (2002). 6th ed. New York: Columbia University Press.
Friedman CP (1995). Where’s the science in medical informatics? J Am Med Inf Assoc 2:65–67.
Hasman A, Haux R, Albert A (1996). A systematic view on medical informatics. Comput Methods Prog Biomedicine 51:131–139.
Johnson SB (2003). A framework for the biomedical informatics curriculum. AMIA Fall Symposium 331–335.
Jordon TJ (2002). Medical information: a user’s guide to informatics and decision making. New York: McGraw-Hill.
Maojo V, Martin F, Crespo J, Billhardt H (2002). Theory, abstraction and design in medical informatics. Methods Inf Medicine 41:44–50.
Musen MA (2002). Medical informatics: searching for underlying components. Methods Inf Medicine 41:12–9.
Patel VL, Kaufman DR (1998). Science and practice: a case for medical informatics as a local science of design. J Am Med Inf Assoc 5:489–492.
Shahar Y (2002). Medical informatics: between science and engineering, between academia and industry. Methods Inf Medicine 41:8–11.
Swick HM (2000). Toward a normative definition of medical professionalism. Acad Med 75:612–616.[Medline]
Shortliffe EH, Johnson SB (2002). Medical informatics training and research at Columbia University. IMIA Yearbook of Medical Informatics. Stuttgart, Germany: Schattauer Publishing Company.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| IADR Journals | Advances in Dental Research ® | Journal of Dental Research ® | Critical Reviews (1990-2004) |
