The Story of How our Health Informatics Textbook Came into Being

Bob HoytI have been asked many times how and why I became interested in Health Informatics and how that led to the writing and self-publication of our textbook, Health Informatics: Practical Guide. The textbook is now in its 7th edition and has been adopted by a large number of universities for their health informatics courses. More co-authors have come on board, and we are now looking at publishing other textbooks. Thus we thought this would be a good point to tell the story.

My background in health informatics began as a purely clinical experience; many years of practicing general Internal Medicine in private practice and military settings. That changed in November 1999 when one of the first medical apps was released for the Palm Pilot PDA. Epocrates™ is a drug look-up program that could essentially replace the 3500-page Physician Desk Reference (PDR). Multiple other medical apps followed Epocrates, and it became patently obvious that physicians now had a digital library, instead of paper-based "pearls" stuffed into their pockets. Contrast this innovation with my private practice experience in a very rural community for 15 years where specialty care and a medical library were ninety miles away.

Another influence on my future Informatics career was the advent of the web-based PubMed for literature reviews. In private practice, I published eight medical articles, but it required a hundred- and eighty-mile round trip to the library to search the paper-based Index Medicus. Now, thanks to the World Wide Web, I can do a literature search in my lounge chair.

These innovations helped to change my way of thinking about medical education and medical care and my enthusiasm only increased with the rapid improvements in the Internet, computers and mobile devices. I joined the Navy in 1992 and began leveraging technology as much as possible. I became head of the Navy Informatics Action Team and we evaluated health information technologies and developed policies and procedures.

At the tail end of my Navy career, I was fortunate enough to be involved with clinical and informatics research. Personally, I do not recommend trying to learn biostatistics after age 60, which was the case with me. We were able to publish some interesting and unique studies as a result of this experience. At the Robert E Mitchell Center for Prisoner of War studies, we applied artificial neural networks (ANNs) to an extended family history of more than 300 Vietnam era ex-POWs to predict the future of diabetes, hypertension and heart disease in their offspring. The study was one of the few to digitize a family history for analysis and the only one to mine it with ANNs. Another interesting study that comes to mind was based on providing simple pocket accelerometers to the pilots of the Navy Blue Angels and comparing the results with the expensive gyroscopic avionics packages on F-18s. Our results showed that the simple accelerometers (Actigraph™) were just as accurate to measure G-forces as the intrinsic avionics.

In 2004 I began teaching several Health Informatics courses at the University of West Florida, in Pensacola, Florida as part of a certificate program. This was challenging because the subject matter was relatively new and educational resources were not optimal. The most prominent textbook was long, expensive, very dry and outdated. Clearly, students would be bored reading about technology developments in the 1950s and 1960s, particularly when brand new developments were not covered. I had never authored a textbook chapter but nevertheless embarked on writing a Health Informatics textbook in 2007.

As my wife, Ann Yoshihashi MD, a retired Navy Endocrinologist, and I researched the challenge of writing a textbook we were taken back by the process of using a commercial publisher. While the finished product by a commercial publisher would be polished and reviewed it would take a long time to be published and the submitter would have very little control over the book or its profits. My resourceful wife suggested we look at self-publication and specifically Lulu.com. What we found was the means to publish our work rapidly (1-2 weeks), receive a majority of the profits and maintain control over the future of our work. The end product was more like a paperback workbook than a hardbound textbook, but we liked that and were comfortable charging much less for our textbook than what the commercial publishers were charging.

We learned from our experiences and decided to keep the cost of the book down, make it available in as many formats as possible and share the profits with the very people who used the textbook. We have shared profits with the University of West Florida, the American Medical Informatics Association (AMIA) and a multitude of small research grants for students and faculty. The textbook is available as a print book, electronic (PDF, Kindle, iBook) versions, rental version and site license. In addition, individual chapters can be purchased by those educational programs that do not need content from the entire book.

We also learned that a website is pivotal for a "textbook ecosystem." With a website, you can describe your work, explain the options to purchase and offer downloads for instructors. This facilitates communicating with a variety of educators around the world that you would otherwise never meet or interact with. The other application we adopted that was extremely helpful for the ecosystem was a newsletter. On our website, viewers can join a newsletter powered by MailChimp and learn about upcoming new editions and any other informatics issues worth sharing.

Another important lesson we have learned is when you are dealing with a new and evolving subject area, instructors are desperately searching for help. In my experience, many of the instructors I communicate with have little to no training in Informatics but are being asked by the college or university to stand up a Health Informatics course. Those who teach graduate students usually have a Ph.D., but not in Informatics. The end result is that they need instructional material just as much as the textbook. In other words, they struggle to create lesson plans and PowerPoints so we provide those as a free download. This has been of great help to most instructors we know.

The acceptance of the textbook has exceeded our expectations. Over 500 Informatics faculty have requested downloads from our website. This includes unusual locations, such as Iran, Iraq, and Malaysia. Another learning point was that making educational (PDF version of the book, Instructor's Manual and PowerPoints) content readily available was an added plus to promote the book. Our experience with commercial editors has not been very favorable. They often take a long time to respond to requests or the discovery of errors in a textbook. Another advantage of self-publication is the ability to reprint a revision if needed. All that is needed is an update to the PDF and submission to the publisher, followed by a notation of, example version 2 for that edition.

We have kept the format similar to all seven editions. Following the learning objectives, there is an Introduction section where we trace the history and background of the subject. The main subject matter is divided into logical sections, followed by future trends, key points, conclusions, and references. Over time, we have deleted some chapters and consolidated others. For example, we used to have separate chapters on patient safety and quality initiatives, but they are now combined because the current feeling is that high-quality care means better patient safety. Similarly, we combined the chapters on information retrieval and medical knowledge resources. Given the rapidly evolving nature of informatics, this must be done to make the textbook current and readable.

Dr. William HershEvery edition began with an overview of Health Informatics to explain the history and definition of this evolving field. In the overview, we also discuss the private and governmental agencies that are intertwined with Informatics. The first chapter and most chapters also have a resources section where readers can find useful references or links to external information. Also, in most chapters, we have made a point to present the challenges and barriers for a given topic, such as electronic health records. Lastly, we have tried to present a balanced scorecard by presenting adverse unintended consequences the field has seen, such as "alert fatigue" related to clinical decision support.

Because Informatics is an "information science" there are three chapters devoted to health data. We include a chapter on Introduction to Data Science to help explain this new field and how it overlaps significantly with Informatics. Because we have a large international audience we included a chapter on International Health Informatics so readers are aware of the multiple Informatics initiatives taking place in other countries.

In June 2018, the seventh edition was launched. The major change was to add Dr. William Hersh as co-editor and author. Dr. Hersh, based at the Oregon Health & Sciences University, is one of the best known and published Informatics experts in the world. He routinely lectures on informatics topics and publishes on a regular basis. In addition, he is the author of Informatics Professor, a widely read Informatics blog. He is the sole author of three chapters and a co-author of six chapters.

Another change for the seventh edition which was extremely important was that we used a textbook designer for the first time. The only significant downside of self-publishing is the amount of text and image formatting that goes into a textbook before the final submission of the PDF to the publisher. Microsoft Word is not intended for a 500-page textbook, with multiple tables and images. Our new publisher Creative Publishing used Adobe InDesign and the finished product was better formatted and designed and overall it saved us a huge amount of time.

It is our plan to turn over the editorship and authorship completely to Dr. Hersh for the eight edition. In this way, the book lives on and will be in the capable hands of a renowned Informatics expert.

In the past two years, I have focused my attention on three areas. One area is the continued interest in publishing articles dealing with informatics-related topics. Another area of interest is in promoting the use of electronic health records for more than just training. In my opinion, they can be used for a variety of educational purposes, analytics, and research. I have several articles and blogs on this issue and I am the current lead on the educational EHR, known as LibreHealth.

As described in this article, the goal is to have an open source, affordable hands-on EHR for clinical and non-clinical students. The third area of continued interest is in biomedical data science. We now recognize the importance of data mining and analytics in every industry, to include healthcare. What we need is better training and more user-friendly software and tools for clinical and non-clinical students. Traditionally, students needed to either learn computer programming and calculus to be considered a data scientist. This should change in the future with better data science tools and I want to be part of the transition.