The Guide to Health Informatics 3rd Edition

If physiology literally means ‘the logic of life’, and pathology is ‘the logic of disease’, then health informatics is the logic of healthcare. It is the study of how clinical knowledge is created, shaped, shared and applied. It is the rational study of the way we think about healthcare, and the way that treatments are defined, selected and evolved. Ultimately, it is the study of how we organise ourselves, both patients and professionals, to create and run healthcare organisations. With such a pivotal role, the study of informatics is as fundamental to the practice of medicine and the delivery of healthcare in this century as anatomy or pathology was in the last.

Health informatics is thus as much about computers as cardiology is about stethoscopes (Coiera 1995). Rather than drugs, X-ray machines or surgical instruments, the tools of informatics are more likely to be clinical guidelines, decision support systems, formal health languages, electronic records, or communication systems like social media. These tools, however, are only a means to an end, which is the delivery of the best possible healthcare.

Although the name ‘health informatics’ only came into use around 1973 (Protti 1995), it is a study that is as old as healthcare itself. It was born the day that a clinician first wrote down some impressions about a patient’s illness, and used these to learn how to treat their next patient. Informatics has grown considerably as a clinical discipline in recent years fuelled, in part no doubt, by the advances in computer technology. What has fundamentally changed now is our ability to describe and manipulate health knowledge at a highly abstract level, and to store vast quantities of raw data. We now also have access to rich communication systems to support the process of healthcare.

We can formally say that health informatics is the study of information and communication processes and systems in healthcare. Health informatics is particularly focused on:

1. Understanding the fundamental nature of these information and communication processes, and describing the principles which shape them,
2. Developing interventions which can improve upon existing information and communication processes,
3. Developing methods and principles which allow such interventions to be designed,
4. Evaluating the impact of these interventions on the way individuals or organizations work, or on the outcome of the work.

Specific subspecialties of health informatics include clinical informatics, which focuses on the use of information in support of patient care and bioinformatics, which focuses on the use of genomic and other biological information.

The rise of health informatics

Perhaps the greatest change in clinical thinking over the last two centuries has been the ascendancy of the scientific method. Since its acceptance, it has become the lens through which we see the world, and governs everything from the way we view disease and battle it.

It is now hard to imagine just how controversial the introduction of theory and experimental method into medicine once was. Then, it was strongly opposed by empiricists, who believed that observation, rather than theoretical conjecture, was the only basis for rational practice.

With this perspective, it is almost uncanny to hear again the old empiricists’ argument that ‘healthcare is an art’, and not a place for unnecessary speculation or formalisation. This time, the empiricists are fighting against those who wish to develop formal theoretical methods to regulate the communal practice of healthcare. Words like quality and safety, clinical audit, clinical guidelines, indicators, outcome measures, healthcare rationing and evidence-based practice now define the new intellectual battleground.

While the advance of science pushes clinical knowledge down to a fine-grained molecular and genetic level, it is events at the other end of the scale are forcing us to change the most. Firstly, the enterprise of healthcare has become so large that it now consumes more national resource than any country is willing to bear. Despite sometimes heroic efforts to control this growth in resource consumption, healthcare budgets continues to expand. There is thus a social and economic imperative to transform healthcare and minimise its drain on social resources.

The structure of clinical practice is also coming under pressure from within. The scientific method, long the backbone of medicine, is now in some ways under threat. The reason for this is not that experimental science is unable to answer our questions about the nature of disease and its treatment. Rather, it is almost too good at its job. As clinical research ploughs ahead in laboratories and clinics across the world, like some great information generating machine, health practitioners are being swamped by its results. So much research is now published each week that it can literally take decades for the results of clinical trials to translate into changes in clinical practice.

So, healthcare workers find themselves practising with ever restricting resources and unable, even if they had the time, to keep abreast of the knowledge of best practice hidden in the literature. As a consequence, the scientific basis of clinical practice trails far behind that of clinical research. Consumers struggle even more, and have to contend with conflicting messages and information they find online, such as in social media.

Two hundred years ago, enlightened physicians understood that empiricism needed to be replaced by a more formal and testable way of characterising disease and its treatment. The tool they used then was the scientific method. Today we are in analogous situation. Now the demand is that we replace the organisational processes and structures that force the arbitrary selection amongst treatments with ones that can be formalised, tested, and applied rationally.

Modern healthcare has also moved away from seeing disease in isolation, to understanding that illness occurs at a complex system level. Infection is not simply the result of the invasion of a pathogenic organism, but the complex interaction of an individual’s immune system, bacterial flora, nutritional status, environmental and genetic endowments. By seeing things at a system level, we come ever closer to understanding what it really means to be diseased, and how that state can be reversed.

We now need to make the same conceptual leap and begin to see the great systems of knowledge that enmesh the delivery of healthcare. These systems produce our knowledge, tools, languages and methods. Thus, a new treatment is never created and tested in intellectual isolation. It gains significance as part of a greater system of knowledge, since it occurs in the context of previous treatments and insights, as well as the context of a society’s resources and needs. Further, our work does not finish when we scientifically prove a treatment works. We must try to disseminate this new knowledge and help others to understand, apply, and adapt it.

These then are the challenges for healthcare. Can we put together rational structures for the way clinical evidence is pooled, communicated and applied to routine care? Can we develop organisational processes and structures that minimise the resources we use, the harms we create, and maximise the benefits delivered? And finally, what tools and methods need to be developed to help achieve these aims in a manner that is practicable, testable and in keeping with the fundamental goal of healthcare – the relief from disease? The role of health informatics is to develop a systems science for healthcare that provides a rational basis to answer these questions, as well as to create the tools to achieve these goals.

The scope of informatics is thus enormous. It finds application in the design of clinical decision support systems for practitioners, consumer decision aids and online health services, in the development of computer tools for research, and in the study of the very essence of healthcare – its corpus of knowledge. Yet the modern discipline of health informatics is still relatively young. Many other groups within healthcare are also addressing the issues raised here and not always in a co-ordinated fashion. Indeed, these groups are not always even aware that their efforts are connected, nor that their concerns are ones of informatics.

The science of what works

I want to let you in on a secret. There are really only three questions that matter in informatics. At the beginning of any new informatics endeavor, you just need to ask:

1 – What is the problem that we are trying to solve?

2 – How will we know when we have succeeded?

3 – Is technology the best solution or are there simpler alternatives?

If you make sure these questions are asked then you will be thought of as a wise indeed. If you know enough to answer them, you might be held up as an informatics guru.

Reading this at the very beginning of your informatics journey, you might be surprised by the triviality of these questions. Re-reading them at the end of your journey through this book, you may now understand why little else matters, and also understand how rare it is for these questions to be asked in the real world – and what the almost inevitable consequences of not asking them are.

With this framing, we need to understand three things about any informatics intervention- its possibility, its practicability, and its desirability. Possibility reflects the science of informatics – what in theory can be achieved? Practicability addresses the potential for successfully engineering a system or introducing a new process – what can actually be done given the constraints of the real world? Desirability looks at the fundamental motivation for using a given process or technology.

These criteria are suggested because we need to evolve a framework to judge the claims made for new technologies and those who seek to profit from them. Just as there is a long-standing, sometimes uneasy, symbiosis between the pharmaceutical industry and medicine, there is a newer and consequently less examined relationship between healthcare and the computing and telecommunication industries. Clinicians should judge the claims of these newcomers in the same cautious way that they examine claims about a new drug and perhaps more so, given that clinicians are far more knowledgeable about pharmacology than they are about informatics and telecommunications.

Overview of the book

The first goal of this book is to present a unifying set of basic informatics principles which influence everything from the delivery of care to an individual patient through to the design of whole healthcare systems. Its next goal is to present the breadth of issues which concern informatics, show how they are related, and to encourage research into understanding the common principles that connect them. The book is organised into a number of parts that revolve around the two distinct but interwoven strands of information and communication systems. While the unique character of each strand is explored individually, there is also an emphasis on understanding the rich way in which they can interact and complement each other.

Part 1 – Basic Concepts in Informatics

This first part of the book offers an intuitive understanding of the basic theoretical concepts needed to understand informatics – the notions of what constitutes a model, what one means by information, and what defines a system. Each concept is used to develop an understanding of the basic nature of information and communication systems. A recurring theme of the book, first articulated here, is the need to understand the limitations imposed upon us whenever we create a model of the world, or use it to design a technology. Understanding these limitations defines the ultimate limits of possibility for informatics, irrespective of whichever technology one may wish to apply in its service.

Part 2 – Informatics Skills

Building upon the concepts in Part One, the second part of the book looks at the practical lessons that can be drawn from informatics to guide everyday clinical activity. Every clinical action, every treatment choice and investigation, is shaped by the available information and how effectively that information is communicated. Five basic clinical informatics skills are explored, each with their own individual chapter:

1. Communicating effectively is based upon understanding cognitive models of information processing, and is constantly challenged by the limits of human attention, and the imperfection of models;
2. Structuring information, with a particular focus on the patient record, is shown to be dependent upon the task at hand, the channel used to communicate the message, and the agent who will receive the message;
3. Questioning others to find information is essential in clinical practice to fill the ever present gaps in every individual’s knowledge;
4. Searching for knowledge describes the broader strategic process of knowing where to ask questions, evaluating answers, and refining questions in the light of previous actions, and occurs in many different settings, from when patient’s are interviewed and examined, through to when treatment options are canvassed;
5. Making Decisions occurs when all the available information needed has been assembled using the other informatics skills, and attempts to come up with the best alternative to solve a problem like selecting a treatment, based both upon the evidence from science, as well as the wishes and needs of individuals.

Part 3 – Information Systems in Healthcare

The chapters in this Part provide the technical core upon which all other parts depend. We introduce clinical information systems and their role in supporting the model, measure and manage cycle. Secondly it is shown that it is not always necessary to completely formalize this cycle, especially when flexibility in decision-making is needed. Consequently, many information processes are left in an unstructured or informal state, and more likely to be supported by communication processes.

The electronic health record is introduced next, and is the first major technical system discussed in the book. The benefits and limitations of existing paper-based systems are compared to their electronic counterparts. Since the electronic patient record feeds so many different clinical systems, later topics including decision support, protocol-based care, population surveillance and clinical audit are all also introduced here.

The next two chapters cover the foundational informatics topics of how to design, evaluate and then implement working technological systems into complex socio-technical organizations. It is often a conundrum that well designed systems do not deliver the benefits expected. The evaluation chapter introduces the concept of the value of information, and uses it to explain the value chain that starts with information creation and extends to ultimate benefit from its use. Understanding where a system is meant to deliver value along this chain becomes a recurrent motif in later chapters This theme is expanded in the Chapter in implementation, which looks at system implementation as one of fitting technologies into complex adaptive organizations. The unexpected outcomes of technology sometimes can only be explained by stepping back and taking such wider system view.

System safety is deeply linked to design and implementation decisions, and the potential downsides of information and communication technologies are explored next, given how closely related the concepts are in these chapters. The final chapter in this section takes another systemic perspective on clinical systems and the value of information, this time coming from economics. Whilst evaluation methods tell us much about the value of information, economics brings its own equally valid insights.

Having completed this Part, one should be able to move on to any of the other sections in any order, as each explores a more specialized topic area.

Part 4 – Guideline and Protocol-based Systems

In this part, the various forms and uses of clinical guidelines, care plans and protocols are introduced. The different roles that computer-based protocol systems can play in clinical practice are outlined in the second chapter. These cover both traditional ‘passive’ support where protocols are kept as a reference, and active systems in which the computer uses the protocol to assist in the delivery of care. For example, protocols incorporated into the electronic record can generate clinical alerts or make treatment recommendations. The growing evidence base for the benefit of such technologies is also summarized, emphasizing that benefits are more likely to be easily demonstrated in process rather than clinical outcome improvements. The third chapter reviews the process of protocol creation, dissemination and application, and explores how informatics can create tools to assist at each of these stages.

Part 5 – Communication Systems

While interpersonal communication skills are fundamental to patient care, the process of communication has, for a long time, not been well supported technologically. Now, with the widespread availability of communication systems supporting mobility, voice mail, electronic mail and social media, new possibilities arise. The chapters in this section introduce the basic types of communication services and explain the different benefits of each.

The next chapter is probably the most technical of the book, covering information and communication networks, and healthcare specific networks such as health information exchanges. It is the place in this text where interoperability standards are covered in detail, as well as topics that relate to how information is accessed across networks, including privacy and consent.

Social media are a different class of communication system and their importance is underlined with a discrete chapter devoted to them. The chapter introduces basic concepts from social networking theory, the social determination of health, and then explores how social media are being harnessed across the spectrum of health care services.

The final chapter in this Part examines clinical communication from the perspective of telemedicine and m-health technologies. The potential of such systems for different areas of healthcare is described, along with the accumulating evidence base for their success, again using the value chain as one way of understanding sometimes unexpected negative results.

Part 6 – Language, Coding and Classification

If the data contained in electronic patient record systems is to be analysed, then it needs to be accessible in some regular way. This is usually thwarted by the variations in health terminology used by different individuals, institutions and nations. To remedy the problem, large dictionaries of standardised clinical terms have been created.

The chapters in this Part introduce the basic ideas of clinical concepts, terms, codes and classifications, and demonstrate their various uses. The inherent advantages and limitations of using different terms and codes are discussed in the second chapter. The last chapter looks at some more advanced issues in coding, describing the theoretical limitations to coding. It introduces natural language processing and text mining methods, and explains how the statistical approach to language management is complementary, and sometimes preferred, to the more formal semantic approaches use din clinical terminologies and ontologies.

Part 7 – Clinical Decision Support and Analytics

Clinical decision support systems (CDSS) are historically one of the most powerful classes of informatics intervention we have at our disposal. These computer programs range from systems that simply present data to aid a human make a decision, some generate prompts or alerts when a clinician’s decision appears problematic, through to systems with the capability of making decisions entirely on their own. In the first chapter, the focus is on the different applications for CDSS, particularly to see where clear successes can be identified. The next chapter takes a more technological focus, and looks at the computational reasoning processes that underpin CDSS. The final Chapter in this part looks at how CDSS knowledge is created, through machine learning, data analytic and computational discovery methods.

Part 8 – Specialized applications for health informatics

The final chapters in book explore some of the specialised ways that decision technologies are applied in clinical practice. They find application in creating intelligent patient monitors, or autonomous therapeutic devices like self-adjusting patient ventilators. Along with communication technologies, CDSS are essential components of public health and biosurveillance systems. In the field of bioinformatics, human genomic and metabolic knowledge is harnessed using computer techniques, and reframes many classes of clinical decision as questions of genetics. When such bioinformatics knowledge is used in clinical practice is often described as personalized or precision medicine, and this topic is covered in its own chapter. The book concludes, not on a minor topic, but on one of the most transformational ones both for informatics as well as health care delivery – the rise of consumer ownership and involvement in the process of care, and the role that informatics has to play in making this necessity a reality.

 

References

Coiera, E. (1995). “Medical informatics.” BMJ: British Medical Journal 310(6991): 1381.

Protti, D. (1995). “The synergism of health/medical informatics revisited.” Methods of information in medicine 34(5): 441-445.

 

 © Enrico Coiera 1997 – 2014