Guide to Health Informatics

Note xiii Preface xiv Preface to the first edition xvi Acknowledgements xviii Publishers' acknowledgements xx Introduction to health informatics xxi PART 1 BASIC CONCEPTS IN INFORMATICS 1 Models 3(9) 1.1 Models are abstractions of the real world 4(3) 1.2 Models can be used as templates 7(2) 1.3 The way we model the world influences the way we affect the world 9(1) Conclusions 10 (1) Discussion points 11(1) 2 Information 12(10) 2.1 Information is inferred from data and knowledge 12(1) 2.2 Models are built from symbols 13(1) 2.3 Inferences are drawn when data are interpreted according to a model 14(2) 2.4 Assumptions in a model define the limits to knowledge 16(2) 2.5 Computational models permit the automation of data interpretation 18(2) Conclusions 20 Discussion points 20(2) 3 Information systems 22(13) 3.1 A system is a set of interacting components 22(1) 3.2 A system has an internal structure that transforms inputs into outputs for a specific purpose 23(6) 3.3 Information systems contain data and models 29(2) Conclusions 31(1) Discussion points 31(4) Part 2 INFORMATICS SKILLS 4 Communicating 35(9) 4.1 The structure of a message determines how it will be understood 36(1) 4.2 The message that is sent may not be the message that is received 37(4) 4.3 Grice's conversational maxims provide a set of rules for conducting message exchanges 41(1) Conclusions 42(1) Discussion points 42(2) 5 Structuring 44(11) 5.1 Messages are structured to achieve a specific task using available resources to suit the needs of the receiver 44(5) 5.2 The patient record can have many different structures 49(4) Conclusions 53(1) Discussion points 53(2) 6 Questioning 55 (11) 6.1 Clinicians have many gaps and inconsistencies in their clinical knowledge 56(3) 6.2 Well-formed questions seek answers that will have a direct impact on clinical care 59(1) 6.3 Questions to computer knowledge sources are structured according to the rules of logic 60(2) 6.4 Well-formed questions are both accurate and specific 62(3) Conclusions 65 (1) Discussion points 65(1) 7 Searching 7.1 Successful searching for knowledge requires well-structured questions to be asked of well-informed agents 66(1) 7.2 Search strategies are optimized to minimize cost and maximize benefit 67(1) 7.3 The set of all possible options forms a search space 68(1) 7.4 Search strategies are designed to find the answer in the fewest possible steps 69(8) 7.5 The answer is evaluated to see if it is well formed, specific, accurate and reliable 77(2) Conclusions 79(1) Discussion points 79(2) 8 Making decisions 81(20) 8.1 Problem-solving is reasoning from the facts to create alternatives, and then housing one alternative 81(2) 8.2 Hypotheses are generated by making inferences from the given data 83(5) 8.3 Decision trees can be used to determine the most likely outcome when there are several alternatives 88(1) 8.4 Heuristic reasoning guides most clinical decisions but is prone to biases and limited by cognitive resources 89(4) 8.5 An individual's preferences for one outcome over another can be represented mathematically as a utility 93(4) Conclusions s96 Discussion points 97(4) PART 3 INFORMATION SYSTEMS IN HEALTHCARE 9 Information management systems 101(10) 9.1 Information systems are designed to manage activities 101(2) 9.2 There are three distinct information management loops 103(2) 9.3 Formal and informal information systems 105(4) Discussion points 109(2) 10 The electronic medical record 111(13) 10.1 The EMR is not a simple replacement of the paper record 112(1) 10.2 The paper-based medical record 113(4) 10.3 The EMR 117(5) Conclusions 122(1) Discussion points 123(1) 11 Designing and evaluating information systems 124(19) 11.1 Design and evaluation are linked processes 125(3) 11.2 The formative assessment cycle defines clinical needs 128(1) 11.3 Summative evaluations attempt to determine the measurable impact of a system once it is in routine use 129(1) 11.4 Interaction design focuses on the way people interact with technology 130(4) 11.5 Designing for change 134(2) 11.6 Designing the information management cycle 136(3) Discussion Points 139(4) PART 4 PROTOCOL-BASED SYSTEMS 12 Protocols and evidence-based healthcare 143(13) 12.1 Protocols 145(3) 12.2 The structure of protocols 148(2) 12.3 Care pathways 150(1) 12.4 The protocol life cycle 151(1) 12.5 Departures from a protocol help drive protocol refinement 152(1) 12.6 The application of protocols 153(1) Discussion points 154(2) 13 Computer-based protocol systems in healthcare 156(15) 13.1 Passive protocol systems 156(2) 13.2 Active protocol systems 158(6) 13.3 Protocol representations and languages 164(5) Conclusions 169(1) Discussion points 169(2) 14 Disseminating and applying protocols 171(9) 14.1 The uptake of clinical guidelines will remain low as long as the costs perceived by clinicians outweigh the benefits 172(1) 14.2 The clinical impact of a guideline is determined both by its efficacy as well as its adoption rate 173(1) 14.3 Strategies for improving the uptake of evidence into practice may alter either actual or perceived costs and benefits 174(4) 14.4 Socio-technical barriers limit the use of evidence in clinical settings 178(1) Discussion points 178(2) 15 Designing protocols 180(11) 1.1 Protocol construction and maintenance 180(3) 1.2 The design of protocols 183(2) 15.3 Protocol design principles 185(2) Discussion points 187(4) PART 5 LANGUAGE, CODING AND CLASSIFICATION 16 Terms, codes, and classification 191(10) 16.1 Language establishes a common ground 191(1) 16.2 Common terms are needed to permit assessment of clinical activities 192(1) 16.3 Terms, codes, groups and hierachies 193 (3) 16.4 Compositional terminologies create complicated concepts from simple terms 196(1) 16.5 Using coding systems 197(3) Discussion Points 200(1) 17 Healthcare terminologies and classification systems 201(16) 17.1 The International Classification of Diseases 202(3) 17.2 Diagnosis related groups 205(3) 17.3 Read Codes 208(2) 17.4 SNOMED 210(1) 17.5 SNOMED Clinical Terms 210(3) 17.6 The Unified Medical Language System (UMLS) 213(2) 17.7 Comparing coding systems is not easy 215(1) Discussion points 216(1) 18 The trouble with coding 217(14) 18.1 Universal terminological systems are impossible to build 218(4) 18.2 Building and maintaining terminologies is similar to software engineering 222(1) 18.3 Compositional terminologies may be easier to maintain over time despite higher initial building costs 223(3) 18.4 The way forward 226(2) Discussion points 228(3) PART 6 COMMUNICATION SYSTEMS IN HEALTHCARE 19 Communication system basics 231(13) 19.1 The communication space accounts for the bulk of information transactions in healthcare 232(1) 19.2 A communication system includes people, messages, mediating technologies and organizational structures 233(3) 19.3 Shared time or space defines the basic contexts of communication system use 236(3) 19.4 Communication services 239(3) Conclusions 242(1) Discussion points 242(2) 20 Communication technology 244(17) 20.1 Machine communication is governed by a set of layered protocols 244(2) 20.2 Communication channels can be dedicated or shared 246(3) 20.3 Wireline communication systems 249(3) 20.4 Wireless communication systems 252(3) 20.5 HL7 defines standards for the electronic exchange of clinical messages 255(3) 20.6 Computer and communication systems are merging 258(1) Discussion points 259(2) 21 Clinical communication and telemedicine 261(24) 21.1 Telemedicine supports clinical care with communication technologies 261(1) 21.2 The evidence for the effectiveness of telemedicine remains weak 262(5) 21.3 Communication needs in healthcare vary widely 267(1) 21.4 Communication and home healthcare 268(2) 21.5 Communication and primary care 270(3) 21.6 Communication and hospitals 273(3) 21.7 Researching clinical communication 276(5) Discussion points 281(4) Part 7 THE INTERNET 22 The Internet and World Wide Web 285(18) 22.1 The Internet has evolved through four stages 286(1) 22.2 The Internet as a technological phenomenon 287(1) 22.3 The Internet as a social phenomenon 288(1) 22.4 The Internet as a commercial phenomenon 289(1) 22.5 The Internet as an enterprise phenomenon 290(1) 22.6 Communication on the Internet 291(3) 22.7 The World Wide Web 294(5) 22.8 Security on the Internet 299(1) 22.9 Future Web advances 300(2) Discussion points 302(1) 23 Web health services 303(16) 23.1 The Web can support rapid publication and distribution of clinical information resources 304(2) 23.2 The electronic patient record can be built using Web technologies 306(2) 23.3 The dissemination of peer-reviewed scientific knowledge is enhanced through use of the Web 308(1) 23.4 Online systems can support continuing education and decision-making 309(2) 23.5 Patients may access healthcare information on the Web 311(3) 23.6 Notification systems offer a rapid way of communicating with the clinical community 314(1) 23.7 The Internet has given rise to new types of healthcare service 315(2) Discussion points 317(2) 24 Information economics and the Internet 319(12) 24.1 Information has value 320(4) 24.2 Information on the Web is associated with search costs 324(3) Conclusion 327(1) Discussion points 327(4) PART 8 DECISION SUPPORT SYSTEMS 25 Clinical decision support systems 331(14) 25.1 All can support both the creation and the use of clinical knowledge 332(1) 25.2 Reasoning with clinical knowledge 333(4) 25.3 Machine learning systems can create new clinical knowledge 337(1) 25.4 Clinical decision support systems have repeatedly demonstrated their worth when evaluated 338(5) Conclusions 343(1) Discussion Points 343(2) 26 Intelligent systems 345(10) 26.1 Before reasoning about the world, knowledge must be captured and represented 346(2) 26.2 Rule-based expert systems 348(1) 26.3 Belief networks 349(1) 26.4 Neural networks 350(2) 26.5 Model-based systems 352(1) 26.6 The choice of reasoning and representation methods should be based on the needs of the task 352(1) 26.7 Intelligent decision support systems have their limits 353(1) Discussion points 354(1) 27 Intelligent monitoring and control 355(12) 27.1 Automated interpretation and control systems can assist in situations with high cognitive loads or varying expertise 355(2) 27.2 Intelligent systems require access to additional data in the EMR before they can perform many complex functions 357(1) 27.3 There are different levels of signal interpretation, each of which requires increasing amounts of clinical knowledge 358(6) 27.4 Intelligent monitoring systems use a variety of methods for interpretation 364(1) 27.5 Use of intelligent monitors can produce new types of error because of automation bias in the user 364(1) Conclusions 365(1) Discussion points 366(1) 28 Biosurveillance 367(12) 28.1 Event reporting = detection + recognition + communication 368(2) 28.2 Infectious disease surveillance systems play a key role in bioagent detection 370(2) 28.3 Clinical education alone is unlikely to enhance event detection and recognition 372(1) 28.4 Online evidence retrieval and CDSS can help support education and decision-making 373(1) 28.5 The Web will need to be used in combination with other communication technologies to support biosurveillance 374(3) Conclusions 377(1) Discussion points 378(1) 29 Bioinformatics 379(18) 29.1 Genome science is rich in sequence data but poor in functional knowledge 380(3) 29.2 Genome data can allow patient treatments to be highly tailored to the individual 383(2) 29.3 Bioinformatics can answer many questions about the role of genes in human disease, but is limited by our ability to model biological processes 385(1) 29.4 Bioinformatics is made possible by the development of new measurement and analysis technologies 386(9) Conclusions 395(1) Discussion points 395(2) Glossary 397(11) References 408(25) Index 433