Ten critical rules for applied informatics positions: What every Chief Medical Informatics Officer (CMIO) should know
1. Avoid organizations where top executives are doubtful, ambivalent, or in conflict about value and role of informatics.
2. Avoid responsibility without authority.
3. Avoid 'internal consultant' positions.
4. Beware incompatibilities in problem-solving, thinking, and leadership style. Avoid cultures of mismanagement.
5. Avoid management roles that lack direct control of resources.
6. Reporting should be to senior medical leadership or medical records (HIMS).
7. Report to one, and only one person.
8. Make sure support for the Medical Informatics role extends to the Board.
9. Have a direct, open channel to organization's CEO.
10. Be aware of territorial issues and have the authority to handle them when needed.
For an informaticist contemplating working in applied healthcare as opposed to research settings, especially outside of an academic medical center, it is critical that the position considered be structured appropriately. In positions such as "VP for Medical Informatics", "Director of Medical Informatics", "Medical Director of Information Systems" or similar, factors such as job title, reporting, budget, direct reports and matrixed staff, resources, and other areas will determine the long-range success and satisfaction to both the informaticist and the organization.
Medical informatics leadership positions are of increasing strategic importance to healthcare organizations. That is, highly specialized clinical IT is becoming recognized as critical to quality, error-prevention, cost-effectiveness, and even the long-term survival of healthcare organizations. As such, informatics leadership positions should be structured accordingly, in alignment with the position's strategic importance and value.
In my discussions with those who've held such positions, I've noted that the position structure (and other factors such as compensation relative to other executives) may be a good a barometer of how senior leadership views the importance of medical informaticists. What follows from that observation is the suggestion to avoid organizations where senior leaders view medical informatics as non-strategic or of minor importance, and have set up marginal informatics positions consistent with that out-of-date view.
I will try in this section to outline some guidelines, based on personal experience of myself and others who have held such positions, on critical points that will make or break such positions. As one reader pointed out, no job will satisfy all these rules. On the other hand, being aware of these issues, especially during negotiations for a position, is highly advantageous.
One informaticist told me informally at an AMIA Annual Symposium several years ago that he had been offered a hospital role, but was uncomfortable about it for reasons he was not quite sure of. He then did research, found this Web page, and realized the position violated many of the recommendations on this web site. The hospital refused to negotiate on any of these issues, however. As a result, the person abruptly declined the position and gave the URL of this page as the reason, a rather ironic twist to the usual hiring process. (It is not known if anyone at that hospital actually bothered to follow up by reading this Web page!)
It is my opinion that ignoring or taking for granted the points here can lead to waste of potential, resources, time, and years of labor that could have been spent more productively for more enlightened organizations. Life is too short and medical informatics skills too valuable to take a position where you're underemployed or misemployed, or strongly dislike your job.
Rule 1: Avoid positions in organizations where top executives are doubtful, ambivalent, or in conflict with one another about the value and role of informatics expertise. Seek organizations where there is executive consensus.
Although many at lower levels may not realize it, this can include Board members, who have the ultimate authority and set the ultimate ideology and direction for healthcare organizations, especially outside academia.
An illustration of the importance of executive consensus on informatics is as follows. In a large regional healthcare system, the CEO and the executive VP/COO seemed not at all to agree on the value of, and role for, an Ivy league-trained Medical Informatics professional. The CEO, himself a physician, had hired the informaticist in a previous role as Sr. VP for Medical Affairs. He had done so since the organization's clinical information technology was in disarray due to leadership problems. Expensive business consultants and even an industrial psychologist had been brought in, but were unable to facilitate improvements or change. The CEO seemed to understand the value of informatics expertise in invigorating clinical computing projects that were in difficulty, breaking up old ideas with visionary IT thinking tempered by a clinician's work ethics and insights, and so forth. Unfortunately, the VP/COO, who also oversaw MIS, apparently did not share these views.
This manifested itself after the informaticist wrote an op-ed to a popular healthcare MIS journal about the importance and value of informaticists in hospital IT projects. The informaticist and other clinicians working on clinical computing projects were heartened to see the op-ed actually published in a journal directed towards the healthcare MIS world.
A short time later, the Healthcare Advisory Board, a think-tank that scans the periodic literature for new ideas and topics of relevance to healthcare organizations, reviewed the article and found it interesting. The Advisory Board is a membership organization representing over two thousand member hospitals, health systems, physician practices, insurers, pharmaceutical companies and medical technology firms, and whose publications and white papers on trends in the industry are widely used in setting policy and making purchasing decisions.
The Advisory Board wrote to this organization's VP/COO (who was their listed contact person) about the op-ed on informatics, seeking more information on a topic that they found of interest. Despite good progress at this hospital system on many fronts in which the informaticist had taken a leadership role (e.g., strategy, EPR, GMPI, procedural medicine databases, etc.), and acknowledgment by the organization's medical staff that the 'new thinking' of formal medical informatics had changed the environment very positively, this VP/COO's reaction for whatever reason was to severely downplay to the Advisory Board the value and role of a professional informaticist. The informaticist was represented as a relatively unimportant 'internal consultant' to this VP/COO and to his MIS department, who were the 'real' movers and shakers.
(One reason might have been that this VP/COO and the MIS director reporting to him oversaw some of the failed clinical IT projects that were later made successful by the informaticist. I leave it up to the reader to surmise what might motivate such a reaction.)
To compound the unfortunate marginalization of informatics as a field that occurred here, the letter from the Advisory Board was never shown to the informaticist and was thrown away by the VP/COO. The informaticist found out about the Advisory Board's interest completely by accident several days later. The Advisory Board wasn't directed by the VP/COO to the informaticist author of the article that attracted their attention, which would have been a basic common courtesy. Finally, due to ambivalence about informatics, other senior executives did nothing upon being informed of the breach of etiquette (at best) represented by this revealing incident.
At worst, on the other hand, thousands of Advisory Board-subscribing healthcare organizations were potentially denied hearing about informatics in a positive light due predominantly to the partiality of one executive and the ambivalence of others. Further, due to incidents like this occurring repeatedly, the informaticist resigned from this organization, depriving the clinicians of informatics expertise.
This VP/COO also repeatedly obstructed the informaticist's presentation to an panel of senior industry CIO's who were advising the organization on IT at the board's request. These CIO's had very little knowledge of healthcare. Extremely verbose, complex and intimidating diagrams and charts on healthcare IT from a large, expensive management consultant firm were shown to them by the VP/COO and MIS director. These documents confused the CIO's, by their own acknowledgement. In fact, the informaticist thought the quality of these documents was appalling, which is a serious matter since they were the product of a multimillion-dollar consulting engagement. The informaticist's critiques, however, were glossed over by the organization. That is a story for another time.
A concise, clean, clear presentation assembled by the informaticist on 'healthcare computing for business IT professionals' was suppressed by the VP/COO using a number of subtle and direct tactics, such as "forgetting to put it on the schedule" or allowing other discussions (once it was on schedule) to run overtime, then forgetting to put the presentation on schedule for the following meeting. In effect, the organization was deprived of the CIO advisory panel's full value. Although possible explanations for this behavior range from ineptitude and sophistry to high-level political intrigue and sabotage, God only knows the true reasons for such behavior. This executive often micromanaged, could be ingratiating when he needed to be, but was generally cold, authoritarian and a bully. This combination of characteristics can be very destructive to innovation spearheaded by creative people. Bright people should beware such executives and the "intellectual hospice" atmosphere they sometimes create.
Not surprisingly, other executives did little when informed about this matter, since this was just the "way of business." Even the CEO remarked that this was a good VP/COO, that such views about IT excellence were a form of "extremism" and that "other views had to be taken into account" (i.e., views of those with little or no knowledge, skills or experience in computing or medicine, the chiropractors of clinical computing).
Even under the most favorable of circumstances, this executive did not let up on such territorial tactics. Despite this executive's starving the budget of an EPR project (electronic patient record) for a large primary-care clinic, causing key personnel to resign, the medical team successfully implemented the EPR on-time and under-budget anyway. They did this through advanced informatics thinking, ingenuity, medical will and true collaboration. The reaction of this executive at a meeting with other senior executives about minor fine-tuning and future directions for this project were that "the project was improperly managed in that the style was too collaborative." The delivery of such rhetoric had to be seen to be appreciated for its breathtaking combination of smooth self-confidence and patent absurdity.
To make matters worse, the executive team then gave a key EPR staff member, a Senior Resident who'd done an excellent job writing and programming custom templates for the EPR system, a difficult time on promised payment for his services. They believed such a customization function was trivial and wasteful, and essentially reneged on their agreements with the Resident. When challenged by the informaticist and others that this person's services were essential, the views were met with indifference, if not disdain, for facts and logic. In fact, the executive team clung persistently to a mind-numbing leap of logic: they seemed to believe that just as home computers were "plug and play", so was clinical IT. Their attitudes seemed to reflect a belief that the EPR team and resident were basically deceiving them.
These attitudes and actions inflamed the Resident and the entire EPR team. The services of this resident were lost as a result, a problem in a clinical computing project requiring iterative development and frequent change. This is a concept that seems beyond the cognitive grasp of this type of executive. This is a typical example of the problems caused by progress-inhibiting bureaucrats who seem common in healthcare. (In a sense, current healthcare turmoil is beneficial in that it may cause bureaucrats who cannot handle rapid technologic change, or who are incompetent, to seek jobs elsewhere.)
Unfortunately, healthcare IT is never plug-and-play, and in IT a person is either part of the solution or part of the problem. Such executives are the latter. One thing is certain: patient care ultimately becomes the unfortunate victim of such behaviors and beliefs. The presentation on healthcare IT to the panel of senior industry CIO's was actually never given because the informaticist left the organization due to its culture of mismanagement.
Positions that identify informaticists as leaders, but do not offer real decision-making authority, may be symptomatic of an ideological problem or careless attitudes among an organization's leaders regarding the role of expertise.
Informaticists should aim for positions at the Assistant V.P. level or higher. Director-level (or worse, manager-level) positions often lack real authority and executive presence. Such characteristics are often essential for success in a cross-disciplinary, cross-territorial area such as applied medical informatics. If a position being considered is at the director or manager level, an inquiry as to why it is rated at such a low level may be helpful in understanding the true feelings of the senior executive team about medical informatics leadership.
A related problem in informatics positions is micromanagement. Matrixed leadership works if properly implemented. Micromanagement, where decisions on matters within a professional's areas of responsibility are interfered with by autocratic superiors-in-title, is not matrixed management. This behavior effectively puts the wrong people in the pilot's seat and infuriates competent professionals.
As expressed by a retired president of a unit of Goodyear, Inc., Lee Fiedler, "managers who try to tell employees what and how to do every little thing will end up with only mediocre people, because the talented ones won't submit to control." (Wall St. Journal, 4 Jan 00, p. B1).
During evaluation of a potential position it may be quite valuable to inquire among medical and IT staff if micromanagement by non-technical executives is a common problem. Micromanagement of information technology by non-technical personnel is often a symptom of flawed beliefs and attitudes about the role of knowledge and expertise in technical fields (e.g., clinical computing, and medicine itself).
Micromanagement is more precisely characterized as a severe form of mismanagement. Regarding the health of clinical computing positions, micromanagement is usually symptomatic of a "terminal disease."
Career advancement routes for "Directors of Medical Informatics" are not yet well-defined. A candidate for such a position should consider the issue of career advancement very carefully and raise questions about it before accepting these positions.
If you start as an 'internal consultant', an organization may be strongly motivated to let you remain as an internal consultant. Good external consultants are very expensive. Unfortunately, an 'internal consultant' position does little for a person's career advancement.
Such roles usually have no direct reports (i.e., MIS, performance improvement, or other people reporting to you). It must be remembered that the number of direct reports a person has had is a key factor evaluated for advancement in healthcare management roles. Internal consultant roles are therefore not good long-term prospects for clinical people who have made the sacrifices to become informatics specialists.
In addition, informaticists should think carefully about organizations that believe informatics physicians do not need, or should not have, direct reports. This may be a litmus test of the true beliefs of the organizational leaders about informatics.
In the worst case, 'internal consultant' can become 'glorified errand runner.' Such glass ceilings are best avoided.
An excellent taxonomy of these areas, one of the best and most succinct I've seen, is in the book The Real Team, Richard Marcinko, Pocket Books, Simon and Schuster, 1999. Highly recommended. According to the author, this taxonomy is itself borrowed from a work called Enlightened Leadership by Crug and Oakley.
Orientation: leaders can be problem-oriented, that is, put the focus on what's wrong, look for blame, stifle creativity, keep people stuck in boxes, and other undesirable characteristics. Problem-oriented leaders see events only through the narrow lens of "conventional wisdom" and have a laser-locked focus on what may go wrong. Solution-oriented leaders, on the other hand, focus on what's working, develop openness, energy and enthusiasm, and encourage creativity. Medical informaticists tend towards the latter. It is not difficult to imagine how such a person will feel working in an environment where leaders are of the former persuasion.
Thinking: leaders can be a reactive thinkers, that is, resist change, assume things cannot be done, focus on finding problems to fix, shift blame or responsibility, be poor listeners, do things "right" rather than do the "right things", and micromanage due to feelings of low personal esteem and the need to control. Creative thinkers, on the other hand, take on responsibility with enthusiasm, are "can-do" oriented, can "work smart" (get results through creative solutions), and build on successes and strengths of others without stealing credit. Once again, an impedance mismatch on this axis is to be avoided.
Leadership style: leaders may have a reactive style, needing to have all the answers, telling rather than listening, use bullying and create an environment of fear, suffer analysis paralysis, operate in self-protect mode, and fear loss of control. Creative leaders collaborate rather than control, are less ego-driven to appear 'omniscient', listen to others who have more expertise in various specialties, generate commitment, and know that hiring good people, supporting them, and letting them do their jobs is the best way to get results.
To these three areas, I add a fourth: leadership mental health. Emotionally unstable or disturbed people make poor leaders. One should seek organizations where the senior leaders exhibit maturity and stability.
It should be realized that "organizational culture", a term that came into common use in the early 1980's, often is in reality a euphemism for the tastes, idiosyncrasies, and sometimes psychopathologies of a small cadre of senior executives. It permits unprofessional or uncivil behavior to hide behind a protective lexical facade of "culture." However, an "organizational culture" is not the same as a "societal culture", where factors such as style, taste, mores, and so forth stem from a long historical base of traditions and the contributions of many. Lee Iacocca provides one of the best illustrations I've seen on this issue in his book "Iacocca: An Autobiography" (Bantam Books, 1986), where he describes severe management problems in the American automobile industry in the 1970's.
An excellent illustration of reactive and untrusting leadership in healthcare comes from a personal experience at one hospital. This story doesn't involve medical informatics per se. However, the issues involved are similar, involving "new" technology.
In a hospital with rather anti-academic, anti-intellectual management biases (e.g., the organization actually did not permit the Ph.D. title to be displayed on name badges or business cards), I had a 4th floor office overlooking a traffic circle and a flat, elevated roof with a clear view to a distant horizon. Such a roof is an excellent site for a shortwave radio antenna. I am an extra-class amateur radio enthusiast (the lower classes are novice, technician, general, and advanced) licensed by the Federal Communications Commission after a series of difficult written examinations. I saw the opportunity to provide a service to the community by setting up an amateur radio station at the hospital, as I had seen done by other physician-amateur radio enthusiasts over the years.
Such a station would put out about 50 watts
of power in the frequency range of 3 to 30 MHz, the shortwave bands. (Ordinary
AM radio is at about 1 MHz and FM radio is at about 100 MHz for comparison).
This is less power than it takes to light an ordinary 60-watt light bulb. I
would use the station occasionally on weekends for personal use, such as
talking to amateur radio friends in the
Amateur radio stations have been proven over the years to provide excellent regional, national and international coverage during times of emergency, such as natural disasters, when other services are down. A recent lightning strike, for example, had actually knocked out several emergency communications services and cellular services in this county for several hours.
Having a spare, commercial-quality, state-of-the-art piece of equipment available, I offered to donate the antennas, an unobtrusive horizontal wire or vertical pole, and the electronics to the hospital. Of course, I was concerned about possible interference to other communications (e.g., ambulance) or to medical equipment. I researched the internet, pulling out a dozen stories where amateur radio had been critical to real hospitals after a devastating tornado, hurricane, or other natural disaster, and where such installations had not been causing interference to other hospital functions.
I also knew that a well-known hospital in a nearby city had an extensive ham radio station for many years and operated without interference to other communications or medical equipment. (I had been taken to the roof of that hospital as a guest years ago to see the extensive ham radio antennas). Lastly, I provided information that a very large hospital chain in earthquake-prone California had an extensive ham radio network set up across dozens of its hospitals, and that the director of its ham radio operations, a friend of a former classmate, indicated no problems with interference. My hospital’s own radio telecommunications team did not expect interference, and I agreed to remove the installation after a testing period if any such interference appeared.
The response of this hospital’s leadership, however, was entirely problem-oriented and reactive. They feared interference with existing "microwave radio facilities" and biomedical equipment. Before any such installation could even be attempted, they insisted on calling in a "consultant" to evaluate the proposal, not trusting their own biomedical engineering, telecommunications, or resident extra-class ham radio physician to be capable of rendering such opinions. However, since 'budgets were tight', there was no money to pay for an expensive consultant. So, we're very sorry, permission denied.
Administration was reassured that shortwave emissions were not microwaves. They were told that, in fact, the amount of microwaves emitted from the shortwave transmitter in question was measurable, and on the order of the amount the hospital would receive from a microwave transmitter placed on the moon (using a rough calculation known from earth-moon-earth communications actually sometimes used by hams).
This was not listened to, however. Despite the near-zero cost factor, despite the potential value to the community in an emergency as documented from news releases easily available in the press and on the Internet, despite many other hospitals having such installations without problems, despite the offer to test and remove if any interference occurred, the installation of an amateur radio station was blocked.
In summary, medical informaticists tend to be solution-oriented, very creative people. In many hospitals and healthcare organizations, problem-oriented, reactive leaders seem very common.
In the worst case, the beliefs and styles of such leaders may engender a culture of mismanagement, where poor management is tolerated even when apparent to many people. Often, solution-oriented, creative thinkers are told by such organizations that they "don't fit into the environment." No surprise here. Patient care then suffers as the victim of such organizational cultures.
In medical informatics positions in hospitals, ignoring signs of trouble and incompatibility in these three areas, orientation, thinking, and leadership style, will probably be deleterious to a position's longevity.
"Doctors don't manage projects" is a corollary of "doctors don't do things with computers." Unfortunately, without direct control of resources (such as hiring, firing, and budgets), a person is an 'internal consultant', not a leader, despite any titles or representations to the contrary. This can be referred to, in a term coined by a friend, as a "director of nothing" position. This reduces effectiveness and certainly reduces job satisfaction and career-advancement opportunities.
Physicians and other clinicians, being the primary enablers of healthcare, should report to personnel who understand clinical issues, matrixing with those who facilitate care. Reporting to non-medical personnel should be done only if the chemistry between the two parties is very good, and the senior party is well-rounded and has experience in working as a direct supervisor to medical personnel. Unfortunately, this is not very common.
Multiple reporting is not matrixed management. Matrixed management should involve a single report, plus collaborative relationships with others. Multiple reporting creates multiple points of possible disagreement, discord and failure. It is therefore suboptimal from both a management-engineering and human perspective.
At a Christmas party, an informaticist was introduced to a prominent Board member as a physician and medical computer scientist from an Ivy-league university who was spearheading innovation in use of computers in clinical care. In response, the Board member asked if the informaticist could 'come over his home and fix his PC' so he could log in to his investment broker.
This board member and others, including the Board chairman, also ignored invitations to come to the informaticist's office for dialog or to tour the new clinical computing facilities. This was not exactly an endorsement of beliefs about a serious role for medical informatics. The attitudes were observed by, and filtered down to, the organization's day-to-day executives.
Also make sure the CEO and other senior executives understand what medical informatics is about, and its strategic value. Direct them to this web site, for example.
Medical informaticists may unfortunately be viewed by healthcare executives as "clinical consultants who know a little about computers", as opposed to leaders in clinical computing. Applied informatics positions in hospitals and other healthcare organizations may thus be structured according to a medical-consultant paradigm, leading to a mismatch between informaticist temperament (often oriented towards leadership) and organizational expectations and empowerment. This mismatch may cause considerable informaticist dissatisfaction, poor utilization of expertise, and lost opportunities for the organization.
Senior executive understanding of medical informatics is especially important if territorial issues arise with MIS or with other executives. It is, after all, up to the CEO to redraw territorial boundaries when necessary to resolve such issues. Human nature is such that there are some who are psychologically unwilling or unable to suspend self-interest in the interest of the enterprise. Leaders need to possess, and know how to use, a wide dynamic range of managerial methodologies in such situations, appropriate to the issues in each specific case. Once again, a dogmatic belief by senior executives that "one shoe fits all" (e.g., a politically-correct, consensus-building approach only, or a top-down autocratic one) invites problems. A good senior executive or CEO possesses a wide managerial dynamic range and will stand up for what's right for patient care.
Actually, CEO's best serve by taking an active
role in healthcare IT. An excellent example of such a CEO is Michael Morgan of
the St. Edward Mercy Health Network,
The territoriality and control issues of the mainframe era did not diminish with the appearance of the PC. It is my observation that these issues have instead been amplified and spread over a larger territory by the great increase in the availability of computers. As some of the stories about informatics failures illustrate, these issues can be very contentious.
Once again, without needed authority or support to manage territorial issues, an informaticist is an 'internal consultant' whose job satisfaction and longevity is problematic at best.
The book Territorial Games: Understanding and Ending Turf Wars at Work (A. Simmons, American Management Association/AMACOM, 1998) is a must-read for anyone contemplating work in a cross-disciplinary healthcare field such as Medical Informatics. Its section "Ten Territorial Games", taxonomizing common tactics that often impair or destroy clinical computing initiatives, is material that everyone working in applied medical informatics should study and master.
The ten territoriality dysfunction categories described in the book are: "occupation" (possession is nine-tenths of the law), "information manipulation", "intimidation", "powerful alliances", "invisible wall", "strategic noncompliance", "discreditation", "shunning", "camouflage", and "filibuster". An eleventh, also described in the book and the worst of all, is "sabotage." These categories capture very well the tactics that this author and colleagues have seen impair or destroy healthcare information technology progress.