H. Dominic Covvey, National Institutes of Health Informatics
Maybe we should just change words so things sound better!
Sittig’s Grand Challenges
Almost exactly 20 years ago, Dean Sittig wrote an important article about the Grand Challenges in medical informatics (1). He cited significant challenges that needed to be addressed for medical informatics to really advance. These were sophisticated challenges, including developing a unified controlled medical vocabulary, automatic coding of records, automated analysis of medical records and comprehensive decision support systems, to name a few. With challenges like these, it is no wonder we still have a lot of work to do.
Only six years ago, Sittig wrote yet another article, this time about the grand challenges in clinical decision support on its own (2). Again, he revealed very important challenges that we have not yet fully addressed. A few of the challenges in this latter article included: improving the user interface, summarizing patient information, creating combined recommendations for patients with comorbidities, and using free text information to drive clinical decision support. But all of these are high-level challenges, and it’s understandable that our progress on them is slow. Are there more basic challenges that block or hinder our progress on these and other advancements?
As a teacher, one learns that it is crucial to teach well, but that even the best teaching is made irrelevant if no one listens, really listens. One ends up repeating the same lessons, giving the same advice but not seeing true progress in one’s students. One little piece of humor states that a professor is someone who talks in someone else’s sleep. Maybe that’s the problem!
One other possibility is that we give problems names that identify them as problems, like ‘poor project management’ or ‘limited competency’. Maybe we need to make them sound more digestible. Maybe we could use what one author called “doublespeak”. This appears to derive from ‘doublethink’, a word that George Orwell created in his book “Nineteen Eighty-Four”. ‘Doublethink’ means that a person accepts two mutually contradictory beliefs as correct – meaning that one can appear to accept distasteful government concepts. ‘Doublespeak’, on the other hand, is the use of deliberately ambiguous speech to make people feel a bit more comfortable about a distasteful matter. Anybody involved in administration has used words like ‘nonperforming assets’ to mean ‘bad loans’, ‘downsizing’ or ‘rightsizing’ to mean ‘firing’, or other mealy-mouthed expressions to avoid the unpleasantries of reality.
We’ll use the word ‘challenge’, instead of the more coarse term ‘problem’, to go along with the doublespeak trend. Let’s also use the word ‘persistent’ here to connote ‘noxious stuff sticking around’. In the realm of warfare, some poison gases, radiological substances and weaponized biological agents are considered ‘persistent’. What this means is that the bad stuff will still be there when a soldier enters an area; it doesn’t dissipate. In World War 1, mustard gas and chlorine gas would settle into shell craters or foxholes, providing a death trap for a combatant. Let’s use ‘persistent’ in this way.
Our Not-So-Grand But Persistent Challenges
What about those basic, but persistent problems – whoops, ‘challenges’ – that all of us face? They don’t meet the high intellectual standards of Dean Sittig’s ‘challenges’, but they are everywhere. The list below indicates what quite a few of us see as being ‘persistent challenges’ that bite us on our butts over and over again. Associated with each item in the list, I’ll suggest a doublespeak term that will serve as perfume or mint mouthwash to cover up its negativity. Maybe, then, we can internalize these, protected by their enteric coating from any digestive reactions they may cause. Maybe that will help minimize the claim that we really don’t have problems.
Not-so-grand Challenge 1:
There seems to be an issue with learning that starts with the idea that everything is simple, that anybody can do it, and that skills, knowledge, experience, attitudes and values can be ignored. Even if not ignored, the need for depth can be avoided. It seems that many practitioners in eHealth feel comfortable with not developing or maintaining their competence. Shooting from the hip seems to be tolerated and having a desire to learn, curiosity and skepticism are of minimal importance. People routinely fail to define what excellence is and to track it. Work is approached without the benefit of the lessons of the past or the use of the methodologies that have been shown to be not only appropriate, but also essential. The challenge here, in plain-speak, is ignorance, but maybe we can make it more palatable to talk about if we call it ‘Knowledge Tolerance’ or ‘Box-Free Thinking’. Whatever we call it, we need to incite the intellectual juices in all of us so that we demand and participate in learning processes and use what we learn.
Not-so-grand Challenge 2:
We still seem to have problems managing projects, particularly, large projects. Sometimes the issue is a kind of megalomania in creating projects that are simply too grand for the time or money or other resources we have. Within projects, it is still rare to see ‘agile’ approaches, such as agile development, agile planning, agile budgeting and even agile thinking. The agility is necessary because of the complexity of what we try to do and the fact that it is, or smacks of being, true innovation, rather than cut and dried construction. Agility means recognizing that projects never truly leave one phase completely while beginning another and that there is a constant feedback to earlier stages. It means that maintenance (especially perfective maintenance) is really just a continued part of development and implementation, as a challenging and potentially unending priority list is addressed. Agile approaches mean that budgeting is forever and that benefits accrue only as the solution starts to approximate what was really needed. I’ll throw into this category also the need for, and often the poverty of realization of, true teamwork, collaboration and meaningful communication. Now this is a mean challenge, so maybe we should be very careful with selecting its term. I’d suggest ‘Crowd-Sourced Management’ or, as an alternative, ‘Management by Subjectives’.
Not-so-grand Challenge 3:
Despite everything that’s been said, many in eHealth still focus on the technology. One would think that the need to move off a technology focus had been beaten to death and that issues like human factors, usability, cultural fit, human impacts, user interfaces and so on had been raised in our consciousness to the point where they were front and center. But we still hear a lot of focus on technology and a minimalist recognition of the impacts of systems on people. Let’s call this ‘Machine Affectionate’. Oh, I also thought of ‘Misanthropic Idolatry’, but that does bear perhaps too much negativity for our sensitive intellects.
Not-so-grand Challenge 4:
Leaders seem to make very little in the way of resources available to their teams so that team members can learn what they need to know and do what they need to do. Investment in staff advancement and learning is often an afterthought, if ‘thunk’ at all. This failure to nurture people sort of freezes them in the state they were when recruited. Perhaps also there is a reticence to give, recognize and share credit and to use the full potential of people, enabling pride. Leaders need to see themselves as mentors, helping, teaching and not just gold-plating their own egos. In this case there are two terms: for the failure to maximize the potential of one’s incumbents, we could call it ‘Free-Range Nurture’ (in other words, staff get whatever they can on their own), while we can call the results of the failure to develop people ‘Non-Accruing Assets’.
Not-so-grand Challenge 5:
Often, problems that are truly complex are wished into simplicity. Real needs are replaced by interpretations that masquerade as requirements, but don’t address the problem’s depth and breadth. To paraphrase Brenda Zimmerman in her book “Edgeware”(3), we need to “put on our complexity glasses” and recognize that many of the things we deal with are truly complex, at the level of raising children versus the level of baking a cake. The failure to wear complexity glasses often leads to the rejection of solutions or to serious adoption problems. Then, of course, we try to force the simple solution on the not-so-simple people and environments and true disasters result. This under-recognition of true complexity can be double-spoken, perhaps, as ‘Conceptual Downsizing’ (meaning making little of a problem) and can be considered the result of ‘Intellectual Attenuation’ (lacking or not using our gray matter).
Not-so-grand Challenge 6:
It is still true that very few projects are subject to formal usability analysis to determine their warts and to fix things before they are imposed on people. Something like the approach of the Centers for Disease Control to approving new therapeutics might make the point even more clear. For instance, there must be an initial trial to make sure the medication is safe, and then further trials to assess if the medication actually does what it’s supposed to. Systems are like therapeutics to fix or at least ameliorate inefficiencies and effectiveness problems in the healthcare system. Despite that realization, usability isn’t usually adequately assessed and adoption is forced all too often. In a way, developers are taking huge risks (mainly with their ‘subjects’, or should we say ‘victims’) and just jumping right in, so maybe the terms ‘Heroic Immersion’ or the ‘Light Brigade Strategy’ would be good ones. Heroic is, after all, a term sometimes associated with surgeons who take great risks with their patients.
Not-so-grand Challenge 7:
We seem, in Canada, to exhibit the NIH or ‘not invented here’ syndrome. Work done in another Province is promptly and continually ignored by other Provinces. Policy varies from Province to Province, disallowing the possibility of a consistent approach by a vendor across the entire country. It is very hard to imagine, unless we are driven strictly by an egotistical need for developing things ourselves, why we do not at least adopt and adapt solutions developed elsewhere. It is as if we differentiate ourselves by our technologies and by how we do things! It must be that there is a perception of lesser intelligence elsewhere, wherever elsewhere is. The cost of this differentiation is high and it prevents or at least makes difficult achieving common, interoperable systems, standards and approaches. Nonetheless, we will need to use a more gentle term. Let’s call this ‘Context-Free Clamor’ (I admit that this one is just a pathetic play on ‘context-free grammar’ in linguistics), meaning that we fail to recognize that we share a context and seem just to want to make a fuss about our local abilities.
Not-so-grand Challenge 8:
Despite our need for information about what works and the effects a system we develop has, it is rare to see proper qualitative or quantitative evaluations. Of course, we claim that whatever we do has worked! Failure is not only not an option, it never occurs! It is sort of like the dearth of scientific publications on negative results. The comeuppance of this is that we cannot learn from each other. Building on what has been done becomes difficult, because we do not know its effects. The result is that decision-making about what to invest in and how much to invest cannot be based on real evidence. What’s interesting, is that the lack of evaluation seems to give the message of success. Even though our evaluations could be of great value to others, our egos dominate and egos never want to look less-than-perfect. It is time for humility, but we exhibit hubris. Time for truth, but we prevaricate by being silent. Despite this, individual projects move forward, ignoring valid evidence of their deficiencies. Perhaps the term that would fit this would be ‘Audacious Advancement’. We move ruthlessly forward without the necessary knowledge!
Not-so-grand Challenge 9:
There is a lot of talk about privacy legislation and regulation. Confidentiality is stated as one of our responsibilities and ethical behavior generally, related to information, is purportedly ingrained in everything that we do. Yes, but we sometimes do not have the systems in place to detect violations of individuals’ privacy, and, worse still, when violations are detected we seldom enforce them with the harsh punishments that are appropriate, such as publication and dismissal. Privacy and confidentiality become mere words, a description of an hypothetical world detached from reality. If we took privacy and confidentiality seriously, security would be far better than it is and violations would be highly publicized and visibly punished. But, that seldom seems to be the case. A softer term for this would be ‘Informational Transparency’. It mixes the positive idea of transparency, for example relative to contract processes, with the word ‘information’, which should not be exposed.
Not-so-grand Challenge 10:
Every one of us is aware of the FUBAR-ity of government processes, especially related to the submission, evaluation and approval of contracts. It seems every Province has suffered to some degree from violations of ethics or procedure or even law when it comes to how contracts are treated. This applies to other areas of government involvement as well, some of which have been addressed in this publication. In my experience with at least one Province, until a contract is provided to a supplier, no review on a broad scale is permitted, and, once the contract has been let, there is no way of changing it. However, despite our knowledge of the inappropriateness of how government proceeds, things stay pretty much the same. It is the exception rather than the rule and many things are overlooked. Let’s call this ‘Exceptional Oversight’, where there are a lot of exceptions to what would be called fair and valid approaches and a lot of things at the elite level of government are overlooked.
Not-so-grand Challenge 11:
Perhaps it is time to address the issue of the use of consultants. Few corners of our health system pay people well enough that a person who can become a private sector consultant would be willing to accept the income restrictions of health system employment. However, companies have no such bounds. Having worked in the consulting field for 15 years, I learned how one gets around limits put on hourly fees and saw how expenses could be expanded to cover effort overruns. I learned that hospitals will pay through the nose for services that essentially duplicate those provided to others. I learned how consultants can provide reports that ensure that they will be hired again. I saw how consultants were used to do jobs too dirty for people in the institutions, the consultants serving as hired guns to eliminate personnel problems. In addition, few consultants really increase the ability of an institution to deal with the same problem in the future, but rather set the stage so that they can be hired again to do something that should’ve been learned and internalized by the client. And of course, institutions pay not only for the consultants’ time but also for the research and development and other costs of the companies that provide services. I thought of several terms that might apply here. They include my favorite ‘Financial Altruism’ (meaning that you graciously give away you money) and runners-up ‘Rational Abnegation’ (meaning you put a stopper in your judgment, but a bit too academic, perhaps), and of course an old favorite, Alan Greenspan’s term, ‘Irrational Exuberance’.
Not-so-grand Challenge 12:
Those steeped in the concepts and tools of computer science recognize that the systems we field today in healthcare are architectural throwbacks to another era. For example, concepts that are now decades-old, including Service-Oriented Architectures, Aspects and Workflow Engines are still extremely rare in products. Furthermore, many products are actually franken-solutions, comprising departmental systems and other components glued together, after acquisition, into what claim to be integrated systems. True integration is still not commonplace and the concept of Active Data Dictionaries extremely rare. The result is that our tools in health care have difficulty supporting change and customization without expensive reprogramming. This is a disease of our technology, and, without being machine affectionate, we need to introduce true adaptability into our systems. The term I would suggest to describe our failure to evolve would be ‘Paleontology-Oriented’, meaning that we have a fascination with what amount to be dinosaurs… a great hobby, but a poor substitute for modern health informatics solutions.
Implications and Need for Action
So what does all this possibly mean to us?
Well, if any of us have knowledge tolerance, we can rest assured that it is correctable and that correcting it will make us better at everything we do. In situations where crowd-sourced management exists, we can educate leadership and insist on a change to more adaptive agile approaches or perhaps move into a leadership position where we can make these things happen. For those who are machine affectionate, the handwriting on the wall is that we should be focused on people and people issues and recognize that the technology is secondary at best. It’s time to put down the Universal Manual of Acronyms and pick up books on psychology and sociology. However, when we do put emphasis on technology, we must avoid being paleontology-oriented, especially when its etiology is knowledge tolerance. Then there are the many immersed in free range nurture environments who are becoming non-accruing assets. They should become militant to get their leaders and organizations to support their development and advancement, recognizing that, unless something is done, they will be like little batteries that will run down and become less useful. In looking at projects we must avoid conceptual downsizing so that the challenges we face simply fit with what we can, or want, to do. Rather, we must address the actual complex challenges we face. Of course, all of us will eventually deal with heroic immersion or the light brigade strategy, where teams throw themselves at users and impose poor solutions – fighting this is crucial if for no other reason than to save the face of eHealth. It is also time that we move past context free clamor and start using what others have done; build on the shoulders of giants or whomever. Crucial to our good name as professionals, audacious advancement should be replaced by systematic and objective evaluation of the work we do and that information provided to others so they can base their own decisions on it. All of us need to move past reading the words and talking the talk and move on to walking the walk related to avoiding and defeating informational transparency. Then, too, it is high time that exceptional oversight is replaced by fair and transparent processes overseen by professionals with actual expertise, standards and criteria. Also, it would make great sense to move away from financial altruism and engage capable individuals in the health system rather than hiring them at great cost from outside. Finally, we must pressure system developers/vendors to move off being paleontology-oriented and to invest in development environments and the building of systems based on modern concepts.
Will these things guarantee that eHealth succeeds? No, but they will make what we do better, more defensible, lower risk and more valuable.
The grand challenges proposed by Dean Sittig absolutely must be addressed, but I think that our ability to address them depends on our addressing these not-so-grand challenges. Corrected, they are the foundations on which everything is based. Without that, there will be no wonder that so many things will be undone and so much scrutiny and doubt will be bestowed on the promise of eHealth!
(Admittedly, some of my doublespeak terms are lame. If you come up with better ones, write them in the comments below!)
Are you interested in discussing these 12 Challenges more directly? Join us for Health Informatics Bootcamp 2: The Challenges in Advancing eHealth. This seven-part online series is an intensive, highly interactive course focusing on the 12 key challenges in advancing eHealth.
1. Sittig, 1994, Grand challenges in Medical Informatics, JAMIA 1994, 1: 412-413.
2. Sittig, et al., Grand challenges in clinical decision support, Journal of Biomedical Informatics 41 (2008) 387–392.
3. Edgeware: Insights from Complexity Science for Health Care Leaders, Brenda Zimmerman, Curt Lindberg, Paul Plsek, (Book), 2008.
Dominic Covvey (FACMI, FHIMSS, FCIPS, SMIEEE, ITCP) is an Adjunct Professor at the University of Waterloo and the University of Ontario Institute of Technology. He is also the President and Director of the National Institutes of Health Informatics. He was the Founding Director of the Waterloo Institute for Health Informatics Research at the University of Waterloo (2003-2010). His research is in the representation and analysis of healthcare workflow, the definition of competencies and curricula in Health Informatics and the design of the Electronic Health Record.