| Sign In to gain access to subscriptions and/or personal tools. |
© 2003 SAGE Publications Dental Informatics: A Work in ProgressPresented at "Dental Informatics & Dental Research: Making the Connection", a conference held in, Bethesda, MD, USA, June 12–13, 2003, sponsored by the University of Pittsburgh Center for Dental Informatics and supported in part by award 1R13DE014611-01 from the National Institute of Dental and Craniofacial Research/National Library of Medicine.
Center for Dental Informatics, University of Pittsburgh, School of Dental Medicine, 3501 Terrace Street, Pittsburgh, PA 15261; titus{at}pitt.edu
Dental informatics is a young scientific discipline that is undergoing continual maturation. Its literature is estimated to consist of approximately 600 papers published between 1975 and 2003, and it is currently growing at a rate of about 50 papers annually. While interest in the discipline is growing, the number of core contributors to dental informatics research remains relatively small. Two major questions for the discipline are: What are the research challenges that dental informatics faces today? and How can the discipline be strengthened and positioned to maximize its success in addressing those challenges? Progress toward research challenges formulated more than ten years ago has been varied. While many new technologies have become available for clinical dental practice, research, and education, many fundamental problems remain to be addressed with informatics research. Recommendations to augment the research capacity in dental informatics include creating a stronger worldwide dental informatics research community, drawing more biomedical informatics researchers to dental research areas, providing career opportunities for dental informatics researchers, addressing grand challenges together as a community, and recruiting subsequent generations of dental informaticians.
Key Words: Dental informatics • dental research/trends • dental practice • dental education • medical information
Dental informatics (Schleyer, 2003a) is a young scientific discipline. It has seen many developments since early computers were first exploited to address problems in dental practice, research, and education. Many signs indicate that dental informatics is gaining strength and coalescing as a discipline; however, dangers and risks exist that can retard or stunt the discipline’s growth and development. The goal of this paper is to stimulate the discussion of two questions: What are the research challenges that dental informatics faces today? and How can the discipline be strengthened and positioned to maximize its success in addressing those challenges? To help answer these questions, this paper is structured into three sections. We first briefly review the growth and size of the dental informatics literature, the individuals who have contributed to it, and the discipline’s achievements. Much of the information in this review is drawn directly from contributions to this conference. Other material is based on our ongoing observation of developments in the discipline during the last 15 years. Second, we discuss a set of research challenges for dental informatics that were postulated more than ten years ago (Lipton, 1992) and assess the progress that has been made toward meeting these challenges. Hindered by the difficulty of quantifying such progress exactly, this assessment is necessarily informal. We then update the list of challenges to reflect contemporary trends, problems, and technologies. Last, we conclude with a set of recommendations intended to help dental informatics grow to its full research potential. In general, this paper is intended to stimulate discussion, not to provide a set of ready-made answers, for the questions posed above.
Dental informatics has developed significantly since the 1960s, when the first uses of informatics approaches to address dental issues were documented (Ledley, 1968). Similar to informatics researchers in medicine (Collen, 1995), researchers in the nascent field of dental informatics were individuals who had been trained in dentistry and other disciplines, such as engineering, or dentists who had partnered with other professionals, such as computer scientists. Interest in dental informatics (Schleyer, 2003a) rose gradually, and the first conferences and workshops on the subject took place (such as the Symposium on Second Generation Clinical Databases and the Electronic Dental Record [1990], the International Conference on Computers in Clinical Dentistry [1991], the International Conference on Computers in Clinical Dentistry [1992], the International Conference on Computers in Clinical Dentistry [1993], and the International Conference on Computer Applications in Dental Education and Practice [1994], and the International Conference on Computer Applications in Dental Education and Practice [1999]). In the early 1990s, some professional organizations—such as the International Medical Informatics Association, the American Medical Informatics Association, and the American Dental Education Association—began to organize interest groups in dental informatics. A key development was the initiation of funding for dental informatics training by the National Institute of Dental and Craniofacial Research (NIDCR) in 1996, which for the first time recognized the need for a formal education of dental informaticians (Schleyer and Spallek, 2001). (The National Library of Medicine [NLM] had been funding similar training in medical informatics since 1972 [Braude, 1991].) Currently, two dental informatics training programs are in operation (Columbia University, 2000; University of Pittsburgh, 2003), and they collectively graduate approximately one or two trainees per year. An examination of the dental informatics literature (Schleyer et al., 2003) presents important clues to the development of the field. In April, 2003, approximately 620 papers in dental informatics had been indexed in MEDLINE since 1975. This number underestimates the true total to some degree, but provides at least some estimate for the size of the relevant literature (Schleyer et al., 2003). More important than the size, however, are some other characteristics of this literature. Compared with other dental specialties, the yearly number of publications in dental informatics is quite small (dental informatics—approximately 50 papers per year; endodontics—approximately 300 papers per year; and oral medicine—approximately 2800 papers per year) (Yang et al., 2001). However, with a 10% annual growth rate for the last ten years, the dental informatics literature seems to be growing more than three times faster than that of the seven dental specialties analyzed by Yang et al.(2001), which grew at a combined rate of 3% annually during the same period. While the metric of publication output is necessarily somewhat imprecise, it nevertheless provides a rudimentary basis for comparing dental informatics with the established dental disciplines. Of approximately 1700 unique authors in dental informatics since 1975, 97.5% have published three or fewer papers. This indicates that only a few individuals have made dental informatics their academic career. Despite these developments in the literature, additional potential for the evolution of the discipline exists. While the early and mid-1990s saw the creation of several dental informatics positions at dental schools, few new positions have become available in the last five years. During the same time, there was no net gain in the number of academic departments exclusively focused on dental informatics. (In the US, between three and five organizational entities devoted to dental informatics exist at dental schools, depending on the definition of "department".) The dental industry is beginning to show interest in informatics expertise, and some larger dental-care settings are looking to applied informaticians to guide and direct the development and implementation of their clinical and administrative information technology (IT) systems. Despite these developments, employment opportunities for recent graduates of dental informatics post-graduate programs have been few and far between. As a result, a growing and well-trained cadre of scientists has little opportunity to contribute to dental practice, research, and education. Nevertheless, significant research challenges in dental informatics exist.
Informatics research: past and present Lipton (1992) published a comprehensive list of research challenges for dental informatics in 1992. Now, more than ten years later, it is useful to review those challenges and assess briefly the progress we have made toward meeting them (see Table  for a summary of Lipton’s recommendations). Without a doubt, we have made progress addressing many of the challenges Lipton posed. For instance, the contributions of dental and craniofacial research to genetic, proteomic, clinical, and epidemiological databases have been substantial. The NIDCR and the Centers for Disease Control and Prevention have established the Dental, Oral and Craniofacial Data Resource Center (NIDCR, 2003a), which collates survey data, data collection instruments, and references, and provides flexible query capabilities. The public availability of high-quality resources such as MEDLINE and MEDLINEplus has made valid and reliable information more accessible to dental clinicians, educators, researchers, patients, and the public. Many investigators have studied how decision-support systems could assist problem-solving in clinical diagnosis and therapy (White, 1996). The Internet has made possible the establishment of large-scale communities where timely and targeted distribution of relevant information is becoming a distinct possibility, if not a reality. Standardization efforts in dental informatics are beginning to show practical results, as evidenced by the recent publication of several ANSI standards by the ADA Standards Committee for Dental Informatics (ADA, 2003a).
While these achievements are significant, major opportunities exist for informatics and IT in dental practice, research, and education. (This assessment is very general and makes no claim to absolute validity. However, it is based on the observation of developments in dental informatics and IT over the last 15 years, and therefore provides a basis for discussing strategies for the future.) Clinical dentistry has seen a slew of informatics and IT innovations, such as computerized charting, digital radiology, the Florida probe, OralCDX (a computer-assisted, brush biopsy test for the detection of oral cancer), computer-based shade-matching, and CEREC (a modular computer-aided design/manufacturing system for creating ceramic restorations). Yet the proportion of general dentists using computers at chairside is still only between about 25% and 30% (data from an ongoing study [2001 Technology Census, 2003]), although approximately 90% of all dentists have computers in their offices (ADA, 2001). Of those who use computers at chairside, few use them to their full potential—for instance, by maintaining completely paperless patient charts or using the computer to analyze clinical patient outcomes. Few commercial offerings of computer-based oral health records have even attempted to implement the vision for electronic patient records articulated by the Institute of Medicine in 1991 (Dick and Steen 1991), and none has succeeded. Aside from a few implementations in closed care-delivery systems, such as the Department of Veterans Affairs (Brown, 1999), there are no systems that demonstrate the free interchange of patient information among a multitude of healthcare providers (dentists, physicians, home healthcare agencies) at multiple locations. Decision support was enthusiastically supported 10–15 years ago (White, 1996), but today, no decision support system is used by a significant number of dental practitioners on a day-to-day basis. Progress in information collection and dissemination has been mixed. The Internet has made information distribution and access technically trivial. While tardy modem connections, copyright issues, and slow-moving content providers are ongoing problems, we can foresee the day when almost all information and computer resources required by biomedical professionals are available electronically, on demand, and independent of location. The more thorny questions to address include when, how, and in what form information should be presented, especially in an information economy where attention is quickly becoming the limited resource (Coiera, 2000). Lipton (1992) suggested DENTLINE, a literature database similar to MEDLINE, focused exclusively on dental issues, as one measure to make information delivery more targeted and relevant for dental professionals. While a system such as DENTLINE has not come into being, it is possible to search only the dental journals in MEDLINE with search interfaces such as PubMed, the Web interface to MEDLINE provided by the NLM. Generally, however, MEDLINE, the Unified Medical Language System, GenBank, and many other major databases continue to serve all biomedical professionals rather than a defined subset. Special information-filtering mechanisms (Bartling, 2003) could achieve the intent of DENTLINE while preserving the advantages of large, homogeneous collections of information in biomedicine. A major portion of Lipton’s proposed goals deal with effective knowledge management, despite the fact that they are not labeled as such. Knowledge management systems maintain data, information, and problem-solving methodologies in an organized, comprehensive, flexible, and accessible fashion. Yet, dentistry has not succeeded in building such systems. We continually create and maintain many separate and independent stores of individual and aggregate data and information. The overwhelming majority of problem-solving methods exist in formats that are inaccessible to computers (such as in research papers, systematic reviews, textbooks, and the heads of scientists, educators, and practitioners). Thus, the absence of effective knowledge management hampers progress in all fields of dentistry and retards the transition of useful innovations to direct patient care. The need to acquire more aggregate data about patients, their diagnoses, their treatment, and the associated outcomes remains as relevant now as it was ten years ago (NIDCR, 2003b). Systematic and large-scale data-gathering on patients remains the exception rather than the rule. While some large and integrated dental care delivery systems are beginning to amass sizable collections of patient data, health services researchers, for the most part, must still make do with the billing databases of insurers as their raw material, inadequate as they may be. The NIDCR recently cleared practice-based research networks (PBRNs) as a new concept for dental research (NIDCR, 2003c), and perhaps this will mark the beginning of a connection between the patient information locked up in dental practices and the researchers who can answer pressing questions using it. However, without some standardization, data from PBRNs will not be as useful as they could be. Standardized vocabularies and information models, a key component in the representation of biomedical data, have shown little progress in dentistry in the last ten years. The ADA’s Current Dental Terminology (CDT) has been updated and augmented regularly, but is limited to treatment procedures. The ADA’s Systematized Nomenclature of Dentistry (SNODENT) project, an effort to create a comprehensive diagnostic vocabulary, has resulted in a terminology that has been neither formally evaluated nor used on a grand scale. Many other areas of dental practice, research, and education would benefit significantly from the availability of standardized vocabularies, and their continued absence is a significant impediment to progress. One can make three key observations from Lipton’s challenges formulated more than a decade ago. First, most of them truly are grand challenges (Sittig et al., 2003), or we would have solved them already. Knowledge management, computer-based oral health records, and dental decision support systems are examples of very difficult and complex research issues for dental informatics. Such challenges can be met only by a large number of scientists exploring many different potential solutions over a long period of time. As the discussion earlier in the paper suggests, this extensive and continual collaboration is what dental informatics currently does not have. Until now, most dental informaticians have been working alone or in very small research groups, and in relative geographic isolation. Qualified researchers in dental informatics are the exception rather than the rule. For most dental schools and research centers, dental informatics has had either low or no priority and thus has received few resources. It is hardly a surprise that we have not been able to address many of these research challenges with more success. A second observation is that while dental informatics is an integral and growing part of the biomedical informatics community, informaticians working in fields other than dentistry will rarely provide ready-made solutions for dental problems, especially in applied informatics (Schleyer, 2003a). For example, much progress has been made toward electronic patient records in medicine, especially in larger care settings, such as academic health centers and hospitals. However, these advances have translated into little or no progress for dental practice. On the other hand, results from theoretical informatics research (Schleyer, 2003a) are often much easier to apply across disciplines, especially when analogous problems are identified. For instance, the image-processing algorithms that make up the core of the OralCDX system (Sciubba, 1999) drew on the rich set of imaging methods that were pioneered in other disciplines. Promising advances from biomedical informatics in general should be applied in dentistry when appropriate; however, when they fail to solve the problem, solutions specific to dentistry should be developed. Last, it is obvious that technology has developed with more speed and in more different directions than could have been foreseen in the early 1990s. These developments make it easier to meet some of the listed research challenges. For instance, before the widespread diffusion of the Internet throughout society, a concept such as the National Health Information Infrastructure (US Department of Health and Human Services, 2003) was truly a dream. Now, with more than 40% of all dental practices being connected to the Internet, that dream has at least a technological platform on which it can be built. However, many issues can be resolved only by the concerted effort of the dental informatics research community, not by relying on the beneficial effects of external trends.
Research challenges: an updated view
Much of the discussion in Lipton’s challenges focused either directly or indirectly on the informatics and information technology environment in dental practice. It is in this environment that large gains can be made with dental informatics innovations. However, it is also one of the most difficult settings in which to implement them. The dental IT industry provides many tools and solutions, but much of the underlying development is not grounded in a holistic view of dental practice, its problems, and its needs (Schleyer, 2003b). Since only about 25–30% of general dentists use computers at chairside, there is a compelling need for research that makes computer technology more useful and usable in the clinical environment. This research would address both hardware and software interfaces, information needs and information presentation, workflow, and device integration. Once the interface between the dentist and the computer system is improved, the door for a whole slew of innovations—such as decision support, real-time information dissemination, and continuing education in the context of practice and collaboration—is opened. Parallel to this research focus on dental care delivery in the operatory, we have to propel the development of computer-based oral health records forward. Such records must be designed for the local, regional, national, and international context. Several companies provide systems for dental practices that capture and manage a rich set of data about patients, and these systems can serve as useful precursors for computer-based oral health records. In all implementations of computer-based oral health records, research must address issues regarding information architecture, forms of data and information representation, security, privacy, controlled vocabularies, interfaces to other systems, and many other considerations. Local, regional, national, and international networks for communicating patient information will bring us closer to the vision of comprehensive and longitudinal electronic patient records (Dick and Steen, 1991). This vision requires resolving a whole host of research questions, including but not limited to authentication and authorization, communication, storage and retrieval, and ownership. Concepts such as practice-based research networks can be built on much of the same infrastructure, even if they have a different purpose and are implemented differently. For instance, data from PBRNs will probably have to be merged into centralized databases, while ad hoc patient record queries could be executed in a distributed manner. Involving patients more in their own care can have a variety of benefits (Agency for Health Care Policy and Research, 1997), such as greater knowledge about treatment options, increased satisfaction with the decision-making process, and improved health outcomes. Some entities, such as the Department of Veterans Affairs, allow their patients to access their health records through the Internet (Veterans Health Administration, 2003). However, in general, the growing area of consumer health informatics has received scant attention in dentistry, and many opportunities remain to be explored formally. Possible research areas include patient access to all or part of their own dental records, improved communication between dentists and patients, health-related educational software, and patient-specific health and wellness information.
This conference was primarily focused on informatics in the context of dental research. The papers in these proceedings provide ample evidence of the necessity for close integration between the two areas. Informatics is a thread that runs through the fabric of all areas of dental research. Documents such as the NIDCR Strategic Plan FY 2003-–2008 (NIDCR, 2003b), the 2003 ADA Research Agenda (ADA, 2003b), and the NIH Roadmap (NIH, 2003) lay down impressive challenges in our quest to improve general and oral health. Dental informatics applications can help researchers meet the goals set forth in these documents. Informatics and IT have already contributed substantially to the progress of dental research, ranging from computer programs that assist researchers in the acquisition, storage, management, and retrieval of research information to computer-based analysis tools that permit many research projects to be performed faster and better than ever before. However, in none of these areas has an endpoint been reached. For example, as the scale and scope of dental research change, more attention must be focused on helping increasingly large, multi-disciplinary and distant research teams collaborate using electronic tools. Computer-based tools for dental research must be developed in accordance with state-of-the-art software engineering principles, so that they can evolve and be re-used by as many researchers as possible. As discussed above, knowledge management and dissemination are crucial if we are to enhance the efficiency and efficacy of dental research. Decision support systems for research can help researchers make optimal decisions in research design and experimentation. Policies and guidelines from funding agencies, such as the NIDCR, can ensure that informatics and IT serve dental research in the best possible manner. In part, this vision also requires a change in our current model of dental research. Top researchers typically maintain, in their heads, a large database of data, information, problem-solving methods, troubleshooting approaches, and creative insights. This knowledge must be made explicit and accessible so that other researchers and software applications can use it. In light of the need to address many research problems using increasingly multidisciplinary approaches, we must ensure that knowledge integration does not become the next bottleneck in the pursuit of research. Researchers, tools, software applications, and databases must be considered part of a system whose performance we strive to optimize. Traditionally, research has benefited significantly from intense competition between and among individuals, research groups, and institutions. However, this may not continue to be the best model. Cooperation, rather than competition, may hold the promise of greater efficiency in addressing important research questions.
The dental education community personifies the connection between research and practice, in addition to its many other roles. Thus, many of the research issues discussed above are brought into sharp focus, and are most immediately felt, at dental schools. Research issues specific to dental education center on teaching and learning, areas that received scant attention in Lipton’s vision (1992). The main research questions in dental education center on content creation and management, learning methodologies, lifelong professional development, and the optimal contribution of computer-based tools to the teaching and learning process. Many innovative tools have emerged from education research, but we are still struggling to determine how those tools can be best used in the pursuit of our educational mission (Rosenberg et al., 2003; Schleyer and Johnson, 2003). Advances in the mainstream of education research, such as adaptive hypermedia (Brusilovsky, 1996), intelligent tutoring systems (Corbett et al., 1997), simulations (Johnson et al., 1997, 1998; Dev et al., 2002), and decision support systems for education (Tsai et al., 1998) need to be applied, evaluated, and validated in biomedical environments.
Several research issues cut across clinical practice, research, and education. Standards in general and controlled vocabularies in particular are a major factor in advancing the state of the art. The development, application, and evaluation of vocabulary standards in dentistry are in their infancy (Barac’h and Schleyer, 2002). This needs to change if we are to capitalize on an improved data communication and data collection infrastructure. Standards should include not only controlled vocabularies, but also communication (Dove, 1996), information models and data representation (ANSI/ADA, 2001), and quality assurance (ANSI/ADA, 2002). Standards should be viewed as means to an end, not goals in themselves. Participation of broad segments of the dental community, including dental researchers, in efforts such as the ANSI-accredited Standards Committee for Dental Informatics (ADA, 2003c) is crucial if we are to bring about standards that serve a useful purpose. Security, privacy, confidentiality, and data access also cross-cut most research issues. In the last few years, new regulations, such as the Health Insurance Portability and Accountability Act (HIPAA) and other Federal regulations, have significantly strengthened the framework in which health information is managed. While many consider compliance with these regulations a necessary evil, the HIPAA actually facilitates the development of systems such as the National Health Information Infrastructure (US Department of Health and Human Services, 2003) because it provides strict guidelines for security, privacy, and confidentiality. However, regulations such as HIPAA do not answer all research questions related to security and confidentiality, and therefore further research on these issues is needed (Masys and Baker, 1997). A last cross-cutting issue is evaluation research. Evaluation of innovations in informatics and IT has often taken a backseat to other, more pressing aspects of research and development (Friedman and Wyatt, 1996). However, without formal evaluation it is difficult to know whether an innovation conveys any benefits, and if so, which ones. As health professionals, we would never accept a drug that did not undergo a clinical trial. Yet we do not have the same scruples regarding the computer programs we use in clinical practice, research, and education. Evaluation in informatics is a complex subject (Friedman and Wyatt, 1996), and continual effort at developing new evaluation approaches is needed (Schleyer and Johnson, 2003).
In light of the magnitude and difficulty of the research challenges discussed above, the resources of the dental informatics community seem inadequate. To help dental informatics "catch fire" and augment its ability to accomplish its research goals, some of the following recommendations may be helpful:
(1) Create a more focused, worldwide community of dental informaticians.
(2) Get more biomedical informaticians interested in dental problems.
(3) Provide career opportunities and career paths for dental informatics researchers.
(4) Address grand challenges together as a community.
(5) Recruit subsequent generations of dental informaticians. These recommendations are far from complete, but some of them may help dental informatics advance in its evolution as a research discipline.
Dental informatics is contributing to progress in dental practice, research, and education every day. The difficult part is to make these contributions explicit and visible. However, doing so will help members of the informatics research community take ownership and pride in their collective interest, and spur them on to greater achievements. At the same time, those who benefit from progress in informatics will appreciate and value the efforts of informaticians more. No one "owns" dental informatics. It is a scientific discipline that lives and thrives through the contributions of many, not the control of just a few. Once that is realized, dental informatics will be empowered to live up to its full potential.
The author thanks Louis Abbey, Luis Actis, Kathryn Atchison, Bill Bartling, Robert Collins, Andriani Daskalaki, Bruce Donoff, Stuart Gansky, Gregg Gilbert, Robert Greenes, Amy Gregg, Colette Hochstein, Lynn Johnson, Ralph Katz, Bill Kotowicz, Robert Lapp, Robert Ledley, Cliff Lieberman, Robert Selwitz, Kishore Shetty, and Greg Zeller for their helpful comments on this paper. The development of this manuscript was supported in part by award 1R13DE014611-01 from the National Institute of Dental and Craniofacial Research/National Library of Medicine.
Publication supported by Software of Excellence (Auckland, NZ)
Advances in Dental Research, Vol. 17, No. 1,
9-15 (2003)
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
Abstract
Introduction