Tuesday, May 12, 2020

Staying Healthy in a Pandemic

I assume that most people in the world, like myself, are assessing what is their personal risk for complications from Covid-19, should they become infected by the virus. I am one of those people I would consider to be on the cusp of risk. Being over age 60 and with borderline hypertension controlled by diet and exercise, I am at the beginning of where the curve starts to ascend for those risk factors. However, I am also very health-conscious, which I like to believe gives me some mitigation against those risks.

Fortunately in this pandemic, I have been able to live quite healthfully. My wife and I have been eating very well, having at least one big salad per day and enjoying her excellent vegetarian cooking. Being confined mostly to home, I have also taken to slightly upping my running routine back to where it was 5-10 years ago, running about five miles every other day. (And of course I keep my social distancing while running, giving a wide berth around anyone else on the road or trail with me.) Between improving my healthy life-style and traveling less (where like most, I tend to eat more), I have actually lost about 5-6 pounds since the start of pandemic.

I fully acknowledge that none of my lifestyle actions will completely mitigate against the risks, were I to become infected. So naturally, I am still hopeful for better treatments and ultimately a vaccine. In the meantime, I hope for the kind of testing, tracing, and quarantining that other countries have implemented much better than my country. And although it seems like a distant dream, I aspire to someday return to my former slightly less healthy lifestyle that allowed me to enjoy so much of the rest of the world. 

Sunday, April 26, 2020

Loss of an Early Informatics Visionary

I am saddened to learn of the passing of Dr. Burton "Bud" Rose due to complications of Covid-19. Dr. Rose was the creator of the well-known UpToDate online medical information system, which is used by clinicians around the world. I played a tiny role in the development of UpToDate by programming its first search capability. In the late 1980s, I was a postdoctoral fellow in medical informatics at Brigham & Women's Hospital, working in the lab of my mentor, Dr. Robert Greenes. Dr. Rose, a kidney specialist, came to Dr. Greenes seeking help to add a searching capability to a collection of "cards" of information about kidney diseases he had collated in an Apple Hypercard Stack. My research had been focusing, then as now, on information retrieval (search) systems. It was relatively straightforward to connect a simple system I had programmed to index and retrieve from the information in the cards. It was a marvel at the time to be able to type in a few words and get medical information, years before the onset of the World Wide Web and Google.

I ultimately finished my fellowship and moved on to Oregon in 1990, and the development of UpToDate was taken over by a fellowship colleague, Dr. Joseph Rush, who stayed on the project for years as it matured into a commercial product that expanded to all of medicine. In 2008, UpToDate was acquired by the large publisher, Kluwer. I had not seen Dr. Rose in many years, but he continued to be a vibrant clinician and educator until his recent retirement.

UpToDate is still widely used and revered in medical settings around the world. I believe its real value is in its content. While its modern search functionality is excellent, what really draws clinicians to it is the quality of its content that can be used to make clinical decisions based on rapid access to high-quality information.

Wednesday, April 22, 2020

Virtual Informatics Course for Medical Students Progresses

My virtual informatics course for medical students is starting its third offering this week, and the uptake has been great. We hope to keep offering the course through the OHSU spring academic quarter, which runs through early June.

The primary impetus for the course is that medical students have been sidelined from clinical experiences due to the need to protect their health as well as conserve personal protective equipment (PPE) for physicians, nurses, and others taking care of patients.

The usual 10-week course has also been organized into a 4-week block format for medical students. Students are required only to complete the weekly multiple-choice assessments and not a term paper or final exam. The course has been offered not only to OHSU medical students but also to any medical student from any US allopathic or osteopathic medical school. External students register for the course through their own institutions, who send us lists of students to enroll in the course.

The medical student course has been offered in weekly waves. The first course started with 17 OHSU medical students. The second two offerings include 62 medical students from 11 different medical schools: Dartmouth College (2), Northwestern University (1), University of Iowa (6), University of North Dakota (2), University of Rochester (15), City University of New York School of Medicine (2), Emory University (12), University of Miami (3), Philadelphia College of Osteopathic Medicine (9), Quinnipac University (1), and Stony Brook University School of Medicine (9). We anticipate additional waves of students from additional medical schools over the next few weeks.

Thursday, April 16, 2020

TREC-COVID: A New Information Retrieval Challenge for Covid-19

Calling all information retrieval (IR) and biomedical informatics researchers interested in IR! My colleagues and I are pleased to announce  a new research challenge related to Covid-19. TREC-COVID aims to develop and evaluate methods to optimize search engines for the current and rapidly expanding number of scientific papers about Covid-19 and related topics. A group of information retrieval (IR) researchers from the Allen Institute for Artificial Intelligence (AI2), the National Institute of Standards and Technology (NIST), the National Library of Medicine (NLM), Oregon Health and Science University (OHSU), and the University of Texas Health Science Center at Houston (UTHealth) have organized the challenge. A press release and official Web site for the project have been posted. Although not official to the project, I am also maintaining a page about the project.

TREC-COVID applies well-known IR evaluation methods from the NIST Text Retrieval Conference (TREC), an annual challenge evaluation that evaluates retrieval methods with data from news sources, Web sites, social media, and biomedical publications. In an IR challenge evaluation, there is typically a collection of documents or other content, a set of topics based on real-world information needs, and relevance assessments to determine which documents are relevant to each topic. Different research teams submit runs of the topics over the collection from their own search systems, from which metrics derived from recall and precision are calculated using the relevance judgments.

The document collection for TREC-COVID comes from AI2, which has created the COVID-19 Open Research Dataset (CORD-19), a free resource of scholarly articles about COVID-19 and other coronaviruses. CORD-19 is updated weekly, although fixed versions will be used for each round of TREC-COVID. It includes not only articles published in journals but also those posted on preprint servers, including bioRxiv, medRxiv, and others.

Because the dataset (along with the world's corpus of scientific literature on Covid-19) is being updated frequently, there will be multiple rounds of the challenge, with later ones focused on identifying newly emerging research. There may also be other IR-related tasks, such as question-answering and fact-checking. The search topics for the first round are based on those submitted to a variety of sources and were developed by myself, Kirk Roberts of UTHealth and Dina Demner-Fushman of NLM. Relevance judgments will be done by people with medical expertise, such as medical students and NLM indexers. I am overseeing the initial relevance judging process, which is being carried out by OHSU medical students who are currently sidelined from clinical activities due to the Covid-19 crisis.

Saturday, April 11, 2020

The Easiest of Times, the Hardest of Times

To paraphrase Charles Dickens,  these Covid-19 times are the easiest and hardest of times.

For me personally, the Covid-19 crisis so far has been relatively easy. Because of this, I have gratitude and also note my fortunes could change at any time. So far, none of my immediate family, friends, or colleagues has become infected or fallen ill. We are comfortably ensconced in our house, have access to just about all of life's essentials, and can enjoy the outdoors, including my own running, with careful physical distancing. Spring is arriving, and the weather over the last few days has been wonderful.

Likewise, my work life has for the most part gone on as usual. I actually have extra work in managing for the future of my department in the emerging financial recession and its impact on my academic medical center. But since my life is already highly virtual, my work has been relatively easy to carry on. For many years, almost all of the teaching I have been doing is already online. My other academic work, especially my informatics research, is also mostly virtual. I have also used this opportunity to offer up additional virtual teaching to other programs within and outside my university, and have become involved in one of many Covid-related informatics research initiatives.

But these times are still extremely hard. It is sad to watch and read the news, and see the statistics. It is difficult to see those succumbing to the virus and that impact on their families and friends, especially since better early management could have prevented some of that. I cannot say enough words of gratitude I have for frontline workers who are the true heroes of this pandemic, from those in healthcare to those in public safety, grocery stores, and other essential businesses. It is also difficult to see the economic impact, especially on those who cannot easily convert to remote work like I can. I also have tremendous worry not only for the recovery of public health and economy as well as inadequacies it has exposed in our healthcare system and the larger social support that society must provide.

It is easy to express my usual optimism being in my situation. While we must honor and protect those who have been impacted by this pandemic, we must also think of how we must also restructure society to insure not only a better approach for crisis times but also when times are good yet not everyone can benefit. 

Thursday, April 2, 2020

A Virtual Course in Biomedical and Health Informatics for Medical Students

Medical students from around the US (and the world) have had their education displaced by the Covid-19 pandemic. In many places, there is either desire to keep them away from risk or to preserve personal protecting equipment (PPE) to physicians, nurses, and others directly involved in patient care. As such, the medical education community has worked to identify virtual educational experiences for medical students.

Our contribution is a virtual course in biomedical and health informatics. Readers familiar with my work will recognize the content of this course as emanating from the introductory course in the Oregon Health & Science University (OHSU) Biomedical Informatics Graduate Program. This course is also used in OHSU's offering as part of the American Medical Informatics Association 10x10 ("ten by ten") program. The syllabus for the course details on how medical schools can enroll their students.

We are implementing the course as a 4-week medical student elective, which is awarded 2 credits at OHSU. The course has about 40 hours of lecture, and we anticipate another 40 hours spent on discussion forums, multiple-choice self-assessments for each unit, and optional readings. The course is graded as pass-fail, and passing requires completion of all of 10 units and their self-assessments over the 4 weeks of the course. Due to high demand, we are limiting enrollment to students in US-based allopathic and osteopathic medical schools.

One new offering of the course will begin each week starting Monday, April 6. We will enroll as many students as we have in a single section, and make all of the content available to them for the duration of the 4 weeks. We will make use of the discussion forums built into our LMS to answer questions they have, and raise a few questions for them to answer. At the end of 4 weeks, the course will end, and those who have completed all of the work will receive a passing grade, which we will report back to the contact from each school.

We are asking each medical school handle student enrollment and credit themselves. In other words, we will make the course available through our learning management system (LMS) at OHSU, but we will ask each school to provide us a list of students to enroll and each will get a login to the course. After the course is done, we will report back to the schools on whether each student completed the course or not. We would like for medical schools that participate to handle giving students credit (probably through some sort of self-study elective).

We also prefer that there be a single point of contact for each school with which we communicate. To capture this information, we have created an online survey that asks for the point of contact (please use a university email address), estimated number of students (initially up to 20 per school - we may be able to accommodate more later), and preferred dates (which we may need to change to balance load). After the survey is completed, someone from our staff will contact the schools to work out the details.

In addition, for those interested in less than a full course on informatics, we have an open Web site that provides some of the materials and is being used by some medical schools.

Monday, March 30, 2020

Keeping Evidence-Based in the Midst of a Pandemic

The Covid-19 pandemic requires urgent scientific knowledge about how to best diagnose, treat, and prevent the spread of the SARS-CoV-2 virus. This is at odds with the deliberate nature of evidence-based medicine (EBM), where it is important to use more deliberate methods to discern the best evidence.

Another challenge is to disseminate the results of research as quickly as possible. The availability of preprint servers and other modern Internet tools allow us to publish first and peer review later. But of course that raises worry that inadvertent error or even deliberate falsehoods might taint the quickly expanding evidence base.

How do we achieve a balance? We have already seen the downside of actions moving ahead of the science. Probably the best example of this is the drug hydroxychloroquine. While this drug may prove of value in preventing and treating SARS-CoV-2, it does have significant adverse effects, especially when taken in doses that exceed the normal therapeutic level. In addition, it is a drug whose availability for other diseases it treats, such as lupus, must be maintained for those patients.

Clearly hydroxychloroquine should be studied, but it should ideally be done in as controlled a way as possible, lest we not cause harm or generate false hope. We may not be able to perform classic double-blind, placebo-controlled randomized controlled trials, but we should still enroll and track patients in highly controlled manners. We cannot forget that this is a disease from which the majority of patients fully recover, so we need to make certain that improvements due to any treatment are not just due to normal recovery from the disease. There must be some sort of control group and a diligent follow-up to insure no missing data in control or experimental groups.

In my view, there are a number of critical questions to answer about SARS-CoV-2:
  • How well do tests diagnose active infection with the disease?
  • How well do tests diagnose serum antibodies indicating immunity?
  • What treatments are available for the disease?
  • Are there any preventive treatments for the disease, from drugs to immunizations?
  • What is the best way to prevent spread in the general population?
  • What is the best way to protect healthcare workers treatment patients with the disease?
All of these can be answered with the usual EBM methods of controlled studies that have served us well. They can also be augmented with large-scale data sources from which we are learning to do better observational studies. We can also carry out systematic reviews, with meta-analysis when appropriate, to collate the results of many studies.

Unfortunately, the deliberate pace of EBM must be balanced with the urgency to develop treatments, vaccinations, and methods to curtail spread of the virus. Likewise, the rapid publication of results on preprint servers and other sources must be followed with peer review and collation into systematic reviews and meta-analyses. Hopefully this will give us the best evidence based for treating and preventing this disease.

Wednesday, March 25, 2020

SARS-CoV-2: How Can I Help?

When the history of the global SARS-CoV-2 pandemic is written, the real heroes will be the frontline workers who cared for those in need and/or kept society functioning. This of course includes healthcare workers but also those working as first-responders and in public safety, or in grocery stores, gas stations, telecommunications infrastructure companies, and other essential businesses. They will certainly come off looking much better than political leaders or even “captains” of industry. I hope that whatever economic recovery plan is implemented that these workers will be appropriately rewarded and that society will have a better appreciation for the essential jobs they do.

It is natural for me to wonder how I can best contribute. As I “retired” from clinical practice some time ago, my skills as a clinician are probably not up to the task. However, there are probably some skills I can contribute, and I will consider those options going forward.

Fortunately there are some non-directly clinical contributions I can make, and these are keeping me busy here and now. One is teaching.  While society is first and foremost dealing with the crisis at hand, we cannot put all education on hold. The situation is particularly challenging for medical students. One might think that the current crisis gives them the opportunity to learn on the front lines. The reality, however, is that there is not enough personal protective equipment (PPE) to protect them. As such, we need to find other ways to maintain their learning trajectory.

A number of medical educators have come up with innovative approaches, and I have thrown my own contribution into the mix. As one who teaches a well-known virtual course that is an introduction to biomedical and health informatics, we are packaging up an offering that we intend to make available as a medical school elective. Because the course is mostly asynchronous, we can scale it up pretty quickly. I don’t just want to throw the materials out there, and still maintain some sort of interaction and connection with learners, but we can offer the course to many students (including those beyond medical students). We plan to launch the first offering to Oregon Health & Science University (OHSU) medical students next week.

I also have an opportunity to advance research related to SARS-CoV-2 in the form of organizing an information retrieval (IR) challenge evaluation. The goal of this retrieval challenge is both to help develop systems capable of identifying relevant information for the current pandemic, but also to scientifically study how retrieval methods can be quickly developed for such situations in the future. The task will follow the "Cranfield" evaluation procedures that are used in the Text Retrieval Conference (TREC) and related challenge evaluations.

This effort is made possibly by work of the Allen Institute for AI and some collaborators who have assembled an open dataset, the COVID-19 Open Research Dataset (CORD-19). This collection of biomedical literature articles currently contains over 40,000 articles and will be updated weekly. Some colleagues and I will be organizing an IR challenge for search engines that find relevant COVID-related articles within this collection. This challenge will provide:
  • A benchmark set of important COVID-related queries (e.g., coronavirus risk factors, COVID-19 ibuprofen)
  • A set of manual judgments for CORD-19 articles on these queries
  • An ongoing leaderboard for comparison of IR systems 
We are even collecting candidate queries in a crowdsourcing manner by asking people to suggest them on Twitter using the hashtag, #COVIDSearch.

The current plan is to run the competition in weekly batches, where that week's snapshot of CORD-19 is used as the corpus and the results of systems participating in that batch are pooled for manual assessment. We will likely use the Kaggle platform to create a “leaderboard” of those whose methods are most effective. The challenge may in the future expand to more detailed tasks such as information-filtering, question-answering, fact-checking, and argument mining.

I make no pretensions that the work I am doing is in any way comparable to front-line healthcare and other essential workers, but I am glad that I can make these contributions that will keep education and research functioning during this tremendous worldwide crisis.

Saturday, March 21, 2020

SARS-CoV-2: The Course Ahead

As the frequency of my postings in this blog has declined in recent years, I have noted several times that the blog started in the frenzied early days of the Health Information Technology for Economic & Clinical Health (HITECH) Act, which was part of the American Recovery and Reinvestment Act (ARRA) and that was instituted in an attempt to blunt the Great Recession of 2008. HITECH was part of ARRA, and of course gave us the big investment that has greatly expanded the adoption of electronic health records (EHRs). A small part of HITECH included investment to build the capacity of the health IT workforce.

Now, of course, we are headed into new economic recessionary times due to SARS-CoV-2, also known as Covid-19 as well as the Novel Coronavirus. Will this be ARRA 2.0?

Before I say anything about reactions to SARS-CoV-2, let me clearly state my sorrow for those most directly affected. Obviously the most sorrow is for those whose lives have been directly impacted by the disease it is causing and also by the disease's impact on their loved ones. There is also sorrow for what is happening to those whose lives are otherwise substantially affected, with threats to their livelihoods or other aspects of their ability to obtain food, shelter, and health care. There is also the impact for those on the front lines, of course in healthcare settings, but also in public safety, grocery stores, and other places of “essential” work. And to a lesser extent the rest of us, obviously minuscule compared to those directly impacted, but with major alterations to our daily lives.

With sorrow does come some opportunity for gratitude. While this is clearly impacting my life, at the end of the day, what I hold most dear - family, friends, and colleagues - are all still there and appreciated for their presence and support. We also owe gratitude for the global Internet, which enables us to keep connected by email, social media, and perhaps most importantly, videoconferencing. A decade ago, the bandwidth and reach of the Internet would not have allowed this level of connection. I am also grateful for my knowledge and experience in online teaching, and how I might put it to work keeping students and faculty connected during these trying times. The latter will likely be a major part of my work effort going forward.

I am certain I will much more to write about in the days ahead. While I had not hoped it would take a crisis to revitalize my blog, it will no doubt do so.

Sunday, February 23, 2020

Adding a New Competency in Clinical Informatics for Medical Education

One of the most widely cited papers I have written in the last decade has been one on competencies in clinical informatics for medical education [1]. For the most part, these 13 competencies have stood the test of time, from knowing how to use the electronic health record and information retrieval systems as well as applying clinical decision support, patient privacy, personal health records, telemedicine, and more. All of these aspects of clinical informatics are essential skills for the 21st century clinicians.

But another area of required competence has come to the fore in recent years, which is the explosion of machine learning and artificial/augmented intelligence in medicine. While the impact of these in real-world clinical practice is still small, the long-term effect is likely to be substantial. Certainly clinicians should be familiar with the myriad of issues related to algorithms and models, including ethical concerns.

In the process of updating our chapter for the forthcoming 2nd edition of the textbook, Health Systems Science [2], my co-author Dr. Jesse Ehrenfeld and I took the opportunity to make this revision to the competencies by adding a 14th one:
14. Apply machine learning applications in clinical care
a. Discuss the applications of artificial/augmented intelligence in clinical settings
b. Describe the limitations and potential biases of data and algorithms

As with the original competencies, we encourage others to improve upon them. But it is also important to add this critical new one to the full set, which are listed below.

References

1. Hersh, WR, Gorman, PN, et al. (2014). Beyond information retrieval and EHR use: competencies in clinical informatics for medical education. Advances in Medical Education and Practice. 5: 205-212. https://www.dovepress.com/beyond-information-retrieval-and-electronic-health-record-use-competen-peer-reviewed-article-AMEP.
2. Skochelak, SE, Hawkins, RE, et al., Eds. (2017). Health Systems Science. New York, NY, Elsevier.

Appendix - Competencies in Clinical Informatics for Medical Education, circa 2020

1.    Find, search, and apply knowledge-based information to patient care and other clinical tasks
a.    Information retrieval/search - choose correct sources for specific task, search using advanced features, apply results
b.    Evaluate information resources (literature, databases, etc.) for their quality, funding sources, biases
c.    Identify tools to assess patient safety (e.g., medication interactions)
d.    Utilize knowledge-based tools to answer clinical questions at the point of care (e.g., text resources, calculators)
e.    Formulate an answerable clinical question
f.    Determine the costs/charges of medications and tests
g.    Identify deviations from normal (labs/x-rays/results) and develop a list of causes of the deviation

2.    Effectively read from, and write to, the electronic health record for patient care and other clinical activities
a.    Graph, display, and trend vital signs and lab values over time
b.    Adopt a uniform method of reviewing a patient record
c.    Create and maintain an accurate problem list
d.    Recognize medical safety issues related to poor chart maintenance
e.    Identify a normal range of results for a specific patient
f.    Access and compare radiographs over time
g.    Identify inaccuracies in the problem list/history/med list/allergies
h.    Create useable notes
i.    Write orders and prescriptions
j.    List common errors with data entry (drop down lists, copy and paste, etc.)

3.    Use and guide implementation of clinical decision support (CDS)
a.    Recognize different types of CDS
b.    Be able to use different types of CDS
c.    Work with clinical and informatics colleagues to guide clinical decision support use in clinical settings

4.    Provide care using population health management approaches
a.    Utilize patient record (data collection and data entry) to assist with disease management
b.    Create reports for populations in different healthcare delivery systems
c.    Use and apply data in accountable care, care coordination, and the primary care medical home settings

5.    Protect patient privacy and security
a.    Use security features of information systems
b.    Adhere to HIPAA privacy and security regulation
c.    Describe and manage ethical issues in privacy and security

6.    Use information technology to improve patient safety
a.    Perform a root-cause analysis to uncover patient safety problems
b.    Familiarity with safety issues
c.    Use resources to solve safety issues

7.    Engage in quality measurement selection and improvement
a.    Recognize the types and limitations of different types of quality measures
b.    Determine the pros and cons of a quality measure, how to measure it, and how to use it to change care

8.    Use health information exchange (HIE) to identify and access patient information across clinical settings
a.    Recognize issues of dispersed patient information across clinical locations
b.    Participate in the use of HIE to improve clinical care

9.    Engage patients to improve their health and care delivery though personal health records and patient portals
a.    Instruct patients in proper use of a personal health record (PHR)
b.    Write an e-message to a patient using a patient portal
c.    Demonstrate appropriate written communication with all members of the healthcare team
d.    Integrate technology into patient education (e.g., decision making tools, diagrams, patient education)
e.    Educate patients to discern quality of online medical resources (Web sites, apps, patient support groups, social media, etc.)
f.    Maintain patient engagement while using an EHR (eye contact, body language, etc.)

10.    Maintain professionalism through use of information technology tools
a.    Describe and manage ethics of media use (cloud storage issues, texting, cell phones, social media professionalism)

11.    Provide clinical care via telemedicine and refer patients as indicated
a.    Be able to function clinically in telemedicine/telehealth environments

12.    Apply personalized/precision medicine
a.    Recognize growing role of genomics and personalized medicine in care
b.    Identify resources enabling access to actionable information related to precision medicine

13.    Participate in practice-based clinical and translational research
a.    Use EHR alerts and other tools to identify patients and populations eligible for participation in clinical trials
b.    Participate in practice-based research to advance medical knowledge

14.    Apply machine learning applications in clinical care
a.    Discuss the applications of artificial/augmented intelligence in clinical settings
b.    Describe the limitations and potential biases of data and algorithms