Review of Biomedical Informatics

Abstract

When it was first published in 1990, those of us practicing clinical informatics were happy to see Shortliffe's Medical Informatics. However, we were disappointed that it had, from the beginning, a highly academic and theoretical orientation, and did not prove useful to those of us in the practice of pathology informatics. Over the years, subsequent editions were renamed “Biomedical Informatics,” which more accurately reflected its emphasis on numerical and theoretical domains. Informatics consists of a number of related, but quite distinct fields. This book could be best characterized as covering academic and research informatics – the stuff of academic promotions, grants, and international meetings. The practice of clinical or practical informatics is concerned with concrete knowledge of implementation of real systems, change management, prioritizing essential issues such as robust patient identification, the 7-layer Healthcare Information and Management Systems Society (HIMSS) Analytics hierarchy (the absolute foundation of which is laboratory, radiology and pharmacy systems), dealing with recalcitrant physicians, and related topics. Given the large amount of information included, and the costs of printing, the typeface is relatively small (albeit easier to read than the 3rd edition). Fortunately, Springer has also produced an electronic version. We did not have an opportunity to review this latter work. If you are considering purchasing this book, I would recommend the electronic version, because of the ability to enlarge the typeface, as well as search ability. The book provides an extensive discussion of many important theoretical aspects of bioinformatics. It provides reasonable coverage of some aspects of the nonpathology informatics community, as highlighted below. Indeed, if one were examining any of the topic areas noted in the table of contents, it would probably be worthwhile to consult this book, in addition to other more comprehensive references. Topics covered in this book include acquiring and using biomedical data, probabilistic reasoning, cognitive science, computer architectures, software engineering, standards, natural language processing, imaging informatics, ethics, and evaluation of health information resources. Applications mentioned include electronic health records, infrastructure, information management, patient centered care, public health, consumer and personal health, telehealth, patient monitoring, radiology imaging, information retrieval and digital libraries, clinical decision support, healthcare education, bioinformatics, translational bioinformatics, clinical research, health information technology policy, and the future. Many of these are excellent discussions, from an academic perspective (e.g. referencing only papers indexed in the National Library of Medicine's PubMed database, and therefore omitting much practical informatics literature). The first six chapters delve into aspects of fundamental computer science that have been applied to medical applications. Many of us in clinical informatics are unfamiliar with this scientific basis, and I recommend a review of this material. The chapter on standards is particularly noteworthy, as it discusses most of the important terminology and syntax standards. It includes recent significant developments, such as the Fast Healthcare Interoperability Resources version of Health Level 7. It would be helpful to have more extensive coverage on topics such as the Systematized Nomenclature of Medicine [(SNOMED) (developed over 4 decades by pathologists, and now adopted worldwide)] and the [(Logical Observation Identifiers Names and Codes) (now federally mandated for conveying lab test names)]. The chapter does a good job, however, with a large array of important topics. Extracting knowledge from medical natural language will improve the effectiveness of our electronic medical records (EMRs) systems; chapter 8 provides a useful discussion of ongoing progress in this field; we hope to soon see wider application of these tools in commercial clinical systems. The two chapters on imaging informatics are principally focused on radiology imaging, but our colleagues in pathology imaging informatics may be able to apply some of the concepts presented. As we select, build and implement systems, chapter 9 reminds us that these systems apply to people, and must be positioned in an appropriate ethical context. The chapter on electronic health record systems provides a comprehensive overview of many of the topics that should be considered in implementing EMRs systems, and summarizes much knowledge that has been developed in 40 years of developing such systems in academic centers. These topics all apply to the current era, where we are adapting commercially available systems to our specific health system. However, insufficient coverage is provided of the commercial, vendor supported systems that now serve virtually all hospitals and health networks in the United States. Even institutions that developed their own highly successful and forward looking (EMRs, also referred to as Electronic Health Records) (e.g. Brigham and Women's hospital) are now replacing those home grown, one-off implementations with commercially supported systems. In recent years, massive efforts at self-development (e.g. Cedars Sinai; Kaiser Permanente) have failed, and have been superseded by commercial systems. Public health information systems are a large and complex field, summarized in the too brief chapter 16. An example of immunization tracking is provided, but it would also have been useful to include the extensive data flows from hospitals for syndromic surveillance (e.g. chief complaints from patients in emergency departments), and of reportable disease (electronic laboratory reporting from laboratories for communicable disease). One would expect a medical informatics book to reference and cover the extensive practical work of the past 20 years on implementation and tailoring of clinical information systems, embodied in the meetings and listservs of organizations such as the Association of Medical Directors of Information Systems (AMDIS), or the HIMSS. However, such references, authors, and correspondingly such coverage, appear to be largely absent from this book. The otherwise useful article on retrieving information from the medical literature makes the invalid assumption that only journals listed in PubMed are worthwhile consulting. In pathology informatics, in particular, many of our most important reference materials – such as annual tables of the several categories of software vendors serving laboratory medicine and pathology – are found in the monthly journal College of American Pathologists (CAP) today. The references in general shortchange important older contributions (such as Lincoln and Korpman's definition of medical information science - Computers, health care, and medical information science, Science, 1980), and fail to reference pioneering work, instead referring to more recent commentary. The most important work in clinical informatics, SNOMED is listed 3 times-each one incorrectly. The most troubling problem in clinical informatics, that of achieving a robust, reliable identification of the patient, is not treated. Curiously, the single index entry for “biometric identifier” has NO page number next to it. The section on unique health identifiers does not mention the development of the Voluntary Universal Health Identifier, the best possibility so far for a widely useable identifier to partner with biometric authentication. Major domains of clinical informatics, with important sets of challenges, are not mentioned. The discussion omits– pathology and laboratory informatics and pharmacy informatics, Pharmacy and pathology are not in the index, nor is there discussion of pharmacy or laboratory workflow. The two chapters on radiology emphasize image handling, rather than workflow. Sparse coverage is given to the emergence of the subspecialty of clinical informatics, and the mention of this in chapter 19 fails to note that the proposal was first advanced by the American Board of Pathology in 1991, long prior to the American Medical Informatics Association taking up the cause in 2005. Many major players in clinical informatics are not listed in the index. The most important companies providing enterprise wide informatics tools (Cerner, Epic, McKesson, Meditech, Siemens, and others) are not (or barely) mentioned. Problems routinely encountered in the clinical practice of informatics are sometimes given little coverage. For example, the section on E-mail omits discussion of many of the techniques and systems used by health organizations and physicians to ensure secure and robust communication between patients and providers. This book is useful for reference, with respect to several of the topics listed in the table of contents. However, it is not a primary source for practical, clinical informatics. If one is faculty or student in an academic informatics department, or wishing to draw upon this academic corpus to consider applications in clinical informatics practice, then this would be a useful book. Likewise, the conferences presented by (and publications of) the American Medical Informatics Association emphasize research and academic approaches. If, however, your aim is to facilitate the use of information tools in front line practice of clinical informatics – with real systems, real issues, and recalcitrant physician users, then the reader is better advised to seek out the many excellent books published by the HIMSS http://www.himss.org, conferences sponsored by HIMSS, AMDIS, http://www.amdis.org, the Association for Pathology Informatics (http://www.pathologyinformatics.org), the American Society for Clinical Pathology (http://www.ascp.org). The CAP (http://www.cap.org) offers an extensive array of articles on clinical informatics in its monthly journal, CAP today. National meetings of some laboratory and pathology organizations, such as the US and Canadian Association of Pathology, the American Association of Blood Banks, the CAP, the American Pathology Foundation, the American Association for Clinical Chemistry, the American Society of Microbiology, and the Association for Molecular Pathology offer occasional courses introducing their attendees to clinical informatics. One should place a high priority on attending the user conferences pertaining to the electronic health record and laboratory information systems operating in one's organization. These conferences include EPIC, Cerner, Meditech, Soft, Orchard, Sunquest, and many others (as reflected in the vendor and system tables published 6 times/year in CAP today).